From e6194923237f3952b955c343b65b211f36bce01c Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Thu, 13 Jul 2017 09:13:30 +0200 Subject: esp: Fix memleaks on error paths. We leak the temporary allocated resources in error paths, fix this by freeing them. Fixes: fca11ebde3f ("esp4: Reorganize esp_output") Fixes: 383d0350f2c ("esp6: Reorganize esp_output") Fixes: 3f29770723f ("ipsec: check return value of skb_to_sgvec always") Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 13 ++++++++----- net/ipv6/esp6.c | 9 +++++---- 2 files changed, 13 insertions(+), 9 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 0cbee0a..dbb31a9 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -381,7 +381,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; if (!esp->inplace) { int allocsize; @@ -392,7 +392,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * spin_lock_bh(&x->lock); if (unlikely(!skb_page_frag_refill(allocsize, pfrag, GFP_ATOMIC))) { spin_unlock_bh(&x->lock); - goto error; + goto error_free; } skb_shinfo(skb)->nr_frags = 1; @@ -409,7 +409,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; } if ((x->props.flags & XFRM_STATE_ESN)) @@ -442,8 +442,9 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * if (sg != dsg) esp_ssg_unref(x, tmp); - kfree(tmp); +error_free: + kfree(tmp); error: return err; } @@ -695,8 +696,10 @@ skip_cow: sg_init_table(sg, nfrags); err = skb_to_sgvec(skb, sg, 0, skb->len); - if (unlikely(err < 0)) + if (unlikely(err < 0)) { + kfree(tmp); goto out; + } skb->ip_summed = CHECKSUM_NONE; diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9ed3547..392def1 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -345,7 +345,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; if (!esp->inplace) { int allocsize; @@ -356,7 +356,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info spin_lock_bh(&x->lock); if (unlikely(!skb_page_frag_refill(allocsize, pfrag, GFP_ATOMIC))) { spin_unlock_bh(&x->lock); - goto error; + goto error_free; } skb_shinfo(skb)->nr_frags = 1; @@ -373,7 +373,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); if (unlikely(err < 0)) - goto error; + goto error_free; } if ((x->props.flags & XFRM_STATE_ESN)) @@ -406,8 +406,9 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info if (sg != dsg) esp_ssg_unref(x, tmp); - kfree(tmp); +error_free: + kfree(tmp); error: return err; } -- cgit v1.1 From 2a596fc9d974bb040eda9ab70bf8756fcaaa6afe Mon Sep 17 00:00:00 2001 From: Jonathan Liu Date: Mon, 10 Jul 2017 16:55:04 +1000 Subject: drm/sun4i: Implement drm_driver lastclose to restore fbdev console The drm_driver lastclose callback is called when the last userspace DRM client has closed. Call drm_fbdev_cma_restore_mode to restore the fbdev console otherwise the fbdev console will stop working. Fixes: 9026e0d122ac ("drm: Add Allwinner A10 Display Engine support") Cc: stable@vger.kernel.org Tested-by: Olliver Schinagl Reviewed-by: Chen-Yu Tsai Signed-off-by: Jonathan Liu Signed-off-by: Maxime Ripard --- drivers/gpu/drm/sun4i/sun4i_drv.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/sun4i/sun4i_drv.c b/drivers/gpu/drm/sun4i/sun4i_drv.c index abc7d8f..a45a627 100644 --- a/drivers/gpu/drm/sun4i/sun4i_drv.c +++ b/drivers/gpu/drm/sun4i/sun4i_drv.c @@ -25,12 +25,20 @@ #include "sun4i_framebuffer.h" #include "sun4i_tcon.h" +static void sun4i_drv_lastclose(struct drm_device *dev) +{ + struct sun4i_drv *drv = dev->dev_private; + + drm_fbdev_cma_restore_mode(drv->fbdev); +} + DEFINE_DRM_GEM_CMA_FOPS(sun4i_drv_fops); static struct drm_driver sun4i_drv_driver = { .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME | DRIVER_ATOMIC, /* Generic Operations */ + .lastclose = sun4i_drv_lastclose, .fops = &sun4i_drv_fops, .name = "sun4i-drm", .desc = "Allwinner sun4i Display Engine", -- cgit v1.1 From bc240eec4b074f5dc2753f295e980e66b72c90fb Mon Sep 17 00:00:00 2001 From: Logan Gunthorpe Date: Mon, 26 Jun 2017 13:50:41 -0600 Subject: ntb: use correct mw_count function in ntb_tool and ntb_transport After converting to the new API, both ntb_tool and ntb_transport are using ntb_mw_count to iterate through ntb_peer_get_addr when they should be using ntb_peer_mw_count. This probably isn't an issue with the Intel and AMD drivers but this will matter for any future driver with asymetric memory window counts. Signed-off-by: Logan Gunthorpe Acked-by: Allen Hubbe Signed-off-by: Jon Mason Fixes: 443b9a14ecbe ("NTB: Alter MW API to support multi-ports devices") --- drivers/ntb/ntb_transport.c | 2 +- drivers/ntb/test/ntb_tool.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c index 9a03c58..b29558d 100644 --- a/drivers/ntb/ntb_transport.c +++ b/drivers/ntb/ntb_transport.c @@ -1059,7 +1059,7 @@ static int ntb_transport_probe(struct ntb_client *self, struct ntb_dev *ndev) int node; int rc, i; - mw_count = ntb_mw_count(ndev, PIDX); + mw_count = ntb_peer_mw_count(ndev); if (!ndev->ops->mw_set_trans) { dev_err(&ndev->dev, "Inbound MW based NTB API is required\n"); diff --git a/drivers/ntb/test/ntb_tool.c b/drivers/ntb/test/ntb_tool.c index f002bf4..a69815c 100644 --- a/drivers/ntb/test/ntb_tool.c +++ b/drivers/ntb/test/ntb_tool.c @@ -959,7 +959,7 @@ static int tool_probe(struct ntb_client *self, struct ntb_dev *ntb) tc->ntb = ntb; init_waitqueue_head(&tc->link_wq); - tc->mw_count = min(ntb_mw_count(tc->ntb, PIDX), MAX_MWS); + tc->mw_count = min(ntb_peer_mw_count(tc->ntb), MAX_MWS); for (i = 0; i < tc->mw_count; i++) { rc = tool_init_mw(tc, i); if (rc) -- cgit v1.1 From 3840538ad384fb7891adeeaf36624f870c51fc0e Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 18 Jul 2017 14:56:17 +0200 Subject: netfilter: ipt_CLUSTERIP: fix use-after-free of proc entry When we delete a netns with a CLUSTERIP rule, clusterip_net_exit() is called first, removing /proc/net/ipt_CLUSTERIP. Then clusterip_config_entry_put() is called from clusterip_tg_destroy(), and tries to remove its entry under /proc/net/ipt_CLUSTERIP/. Fix this by checking that the parent directory of the entry to remove hasn't already been deleted. The following triggers a KASAN splat (stealing the reproducer from 202f59afd441, thanks to Jianlin Shi and Xin Long): ip netns add test ip link add veth0_in type veth peer name veth0_out ip link set veth0_in netns test ip netns exec test ip link set lo up ip netns exec test ip link set veth0_in up ip netns exec test iptables -I INPUT -d 1.2.3.4 -i veth0_in -j \ CLUSTERIP --new --clustermac 89:d4:47:eb:9a:fa --total-nodes 3 \ --local-node 1 --hashmode sourceip-sourceport ip netns del test Fixes: ce4ff76c15a8 ("netfilter: ipt_CLUSTERIP: make proc directory per net namespace") Signed-off-by: Sabrina Dubroca Reviewed-by: Xin Long Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/ipt_CLUSTERIP.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 7d72dec..efaa04d 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -117,7 +117,8 @@ clusterip_config_entry_put(struct net *net, struct clusterip_config *c) * functions are also incrementing the refcount on their own, * so it's safe to remove the entry even if it's in use. */ #ifdef CONFIG_PROC_FS - proc_remove(c->pde); + if (cn->procdir) + proc_remove(c->pde); #endif return; } @@ -815,6 +816,7 @@ static void clusterip_net_exit(struct net *net) #ifdef CONFIG_PROC_FS struct clusterip_net *cn = net_generic(net, clusterip_net_id); proc_remove(cn->procdir); + cn->procdir = NULL; #endif nf_unregister_net_hook(net, &cip_arp_ops); } -- cgit v1.1 From f7fb77fc12352d15180dc3c08ffba10573d5167d Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 18 Jul 2017 20:03:05 +0200 Subject: netfilter: nft_compat: check extension hook mask only if set If the x_tables extension comes with no hook mask, skip this validation. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_compat.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index f5a7cb6..b89f4f6 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -305,7 +305,7 @@ static int nft_target_validate(const struct nft_ctx *ctx, const struct nf_hook_ops *ops = &basechain->ops[0]; hook_mask = 1 << ops->hooknum; - if (!(hook_mask & target->hooks)) + if (target->hooks && !(hook_mask & target->hooks)) return -EINVAL; ret = nft_compat_chain_validate_dependency(target->table, @@ -484,7 +484,7 @@ static int nft_match_validate(const struct nft_ctx *ctx, const struct nf_hook_ops *ops = &basechain->ops[0]; hook_mask = 1 << ops->hooknum; - if (!(hook_mask & match->hooks)) + if (match->hooks && !(hook_mask & match->hooks)) return -EINVAL; ret = nft_compat_chain_validate_dependency(match->table, -- cgit v1.1 From c389377c01bf2d6e2a178e86aef8535931bfbd75 Mon Sep 17 00:00:00 2001 From: "Jason J. Herne" Date: Fri, 21 Jul 2017 03:14:36 +0200 Subject: vfio: ccw: fix bad ptr math for TIC cda translation When we are translating channel data addresses from guest to host address space for TIC instructions we are getting incorrect addresses because of a pointer arithmetic error. We currently calculate the offset of the TIC's cda from the start of the channel program chain (ccw->cda - ccw_head). We then add that to the address of the ccw chain in host memory (iter->ch_ccw). The problem is that iter->ch_ccw is a pointer to struct ccw1 so when we increment it we are actually incrementing by the size of struct ccw1 which is 8 bytes. The intent was to increment by n-bytes, not n*8. The fix: cast iter->ch_ccw to char* so it will be incremented by n*1. Reviewed-by: Dong Jia Shi Signed-off-by: Jason J. Herne Signed-off-by: Dong Jia Shi Message-Id: <20170721011436.76112-1-bjsdjshi@linux.vnet.ibm.com> Signed-off-by: Cornelia Huck --- drivers/s390/cio/vfio_ccw_cp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/s390/cio/vfio_ccw_cp.c b/drivers/s390/cio/vfio_ccw_cp.c index ba6ac83..5ccfdc8 100644 --- a/drivers/s390/cio/vfio_ccw_cp.c +++ b/drivers/s390/cio/vfio_ccw_cp.c @@ -481,7 +481,7 @@ static int ccwchain_fetch_tic(struct ccwchain *chain, ccw_tail = ccw_head + (iter->ch_len - 1) * sizeof(struct ccw1); if ((ccw_head <= ccw->cda) && (ccw->cda <= ccw_tail)) { - ccw->cda = (__u32) (addr_t) (iter->ch_ccw + + ccw->cda = (__u32) (addr_t) (((char *)iter->ch_ccw) + (ccw->cda - ccw_head)); return 0; } -- cgit v1.1 From bc9b934b2fbbd51008a1b52c0cd1b457e6440736 Mon Sep 17 00:00:00 2001 From: Alexander Dahl Date: Tue, 25 Jul 2017 14:00:22 +0200 Subject: memory: atmel-ebi: Fix smc timing return value evaluation Setting optional EBI/SMC properties through device tree always fails due to wrong evaluation of the return value of atmel_ebi_xslate_smc_timings(). If you put some of those properties in your dts file, but not 'atmel,smc-tdf-ns' the local variable 'required' in atmel_ebi_xslate_smc_timings() stays on 'false' after the first 'if' block. This leads to setting 'ret' to -EINVAL in the first run of the following 'for' loop which is then the return value of this function. However if you set 'atmel,smc-tdf-ns' in the dts file and everything in atmel_ebi_xslate_smc_timings() works well, it returns the content of 'required' which is 'true' then. So the function atmel_ebi_xslate_smc_timings() always returns non-zero which lets its call in atmel_ebi_xslate_smc_config() always fail and thus returning -EINVAL, so the EBI configuration for this node fails. Judging from the following code evaluating the local 'required' variable in atmel_ebi_xslate_smc_config() and the call of caps->xlate_config in atmel_ebi_dev_setup() it's probably right to only let the call fail if a negative error code is returned. Signed-off-by: Alexander Dahl Acked-by: Boris Brezillon Signed-off-by: Alexandre Belloni --- drivers/memory/atmel-ebi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/memory/atmel-ebi.c b/drivers/memory/atmel-ebi.c index 99e644c..1cf34d2 100644 --- a/drivers/memory/atmel-ebi.c +++ b/drivers/memory/atmel-ebi.c @@ -263,7 +263,7 @@ static int atmel_ebi_xslate_smc_config(struct atmel_ebi_dev *ebid, } ret = atmel_ebi_xslate_smc_timings(ebid, np, &conf->smcconf); - if (ret) + if (ret < 0) return -EINVAL; if ((ret > 0 && !required) || (!ret && required)) { -- cgit v1.1 From 1f6b53901f5bffbbb0464c732e83bb2a3a784ff7 Mon Sep 17 00:00:00 2001 From: Alexander Dahl Date: Tue, 25 Jul 2017 14:00:23 +0200 Subject: memory: atmel-ebi: Allow t_DF timings of zero ns As reported in [1] and in [2] it's not possible to set the device tree property 'atmel,smc-tdf-ns' to zero, although the SoC allows a setting of 0ns for the t_DF time. Allow this setting by doing the same thing as in the atmel nand controller driver by setting ncycles to ATMEL_SMC_MODE_TDF_MIN if zero is set in the dts. [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2017-March/490966.html [2] http://lists.infradead.org/pipermail/linux-arm-kernel/2017-July/520652.html Suggested-by: Boris Brezillon Signed-off-by: Alexander Dahl Acked-by: Boris Brezillon Signed-off-by: Alexandre Belloni --- drivers/memory/atmel-ebi.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/memory/atmel-ebi.c b/drivers/memory/atmel-ebi.c index 1cf34d2..f8a01ae 100644 --- a/drivers/memory/atmel-ebi.c +++ b/drivers/memory/atmel-ebi.c @@ -120,12 +120,14 @@ static int atmel_ebi_xslate_smc_timings(struct atmel_ebi_dev *ebid, if (!ret) { required = true; ncycles = DIV_ROUND_UP(val, clk_period_ns); - if (ncycles > ATMEL_SMC_MODE_TDF_MAX || - ncycles < ATMEL_SMC_MODE_TDF_MIN) { + if (ncycles > ATMEL_SMC_MODE_TDF_MAX) { ret = -EINVAL; goto out; } + if (ncycles < ATMEL_SMC_MODE_TDF_MIN) + ncycles = ATMEL_SMC_MODE_TDF_MIN; + smcconf->mode |= ATMEL_SMC_MODE_TDF(ncycles); } -- cgit v1.1 From 3fb3b3c4b68e33bb4acf42361b6a0db96115de35 Mon Sep 17 00:00:00 2001 From: Alexander Dahl Date: Tue, 25 Jul 2017 14:00:24 +0200 Subject: memory: atmel-ebi: Fix smc cycle xlate converter The converter function for translating ns timings in register values was initialized with a wrong function pointer. This resulted in wrong register values also for the setup and pulse registers when configuring the EBI interface trough dts. Includes a small fix in a comment of the smc driver, which was probably just a copy'n'paste mistake. Signed-off-by: Alexander Dahl Acked-by: Boris Brezillon Acked-by: Lee Jones Signed-off-by: Alexandre Belloni --- drivers/memory/atmel-ebi.c | 2 +- drivers/mfd/atmel-smc.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/memory/atmel-ebi.c b/drivers/memory/atmel-ebi.c index f8a01ae..ebf69ff 100644 --- a/drivers/memory/atmel-ebi.c +++ b/drivers/memory/atmel-ebi.c @@ -72,7 +72,7 @@ struct atmel_smc_timing_xlate { { .name = nm, .converter = atmel_smc_cs_conf_set_pulse, .shift = pos} #define ATMEL_SMC_CYCLE_XLATE(nm, pos) \ - { .name = nm, .converter = atmel_smc_cs_conf_set_setup, .shift = pos} + { .name = nm, .converter = atmel_smc_cs_conf_set_cycle, .shift = pos} static void at91sam9_ebi_get_config(struct atmel_ebi_dev *ebid, struct atmel_ebi_dev_config *conf) diff --git a/drivers/mfd/atmel-smc.c b/drivers/mfd/atmel-smc.c index 954cf0f..20cc0ea 100644 --- a/drivers/mfd/atmel-smc.c +++ b/drivers/mfd/atmel-smc.c @@ -206,7 +206,7 @@ EXPORT_SYMBOL_GPL(atmel_smc_cs_conf_set_pulse); * parameter * * This function encodes the @ncycles value as described in the datasheet - * (section "SMC Pulse Register"), and then stores the result in the + * (section "SMC Cycle Register"), and then stores the result in the * @conf->setup field at @shift position. * * Returns -EINVAL if @shift is invalid, -ERANGE if @ncycles does not fit in -- cgit v1.1 From eb92b4183d93a6f101a6bd3aaae651de404c119a Mon Sep 17 00:00:00 2001 From: Andreas Klinger Date: Tue, 25 Jul 2017 11:11:14 +0200 Subject: iio: bmp280: properly initialize device for humidity reading If the device is not initialized at least once it happens that the humidity reading is skipped, which means the special value 0x8000 is delivered. For omitting this case the oversampling of the humidity must be set before the oversampling of the temperature und pressure is set as written in the datasheet of the BME280. Furthermore proper error detection is added in case a skipped value is read from the device. This is done also for pressure and temperature reading. Especially it don't make sense to compensate this value and treat it as regular value. Signed-off-by: Andreas Klinger Signed-off-by: Jonathan Cameron --- drivers/iio/pressure/bmp280-core.c | 27 ++++++++++++++++++++++++--- drivers/iio/pressure/bmp280.h | 5 +++++ 2 files changed, 29 insertions(+), 3 deletions(-) diff --git a/drivers/iio/pressure/bmp280-core.c b/drivers/iio/pressure/bmp280-core.c index d82b788..0d2ea3e 100644 --- a/drivers/iio/pressure/bmp280-core.c +++ b/drivers/iio/pressure/bmp280-core.c @@ -282,6 +282,11 @@ static int bmp280_read_temp(struct bmp280_data *data, } adc_temp = be32_to_cpu(tmp) >> 12; + if (adc_temp == BMP280_TEMP_SKIPPED) { + /* reading was skipped */ + dev_err(data->dev, "reading temperature skipped\n"); + return -EIO; + } comp_temp = bmp280_compensate_temp(data, adc_temp); /* @@ -317,6 +322,11 @@ static int bmp280_read_press(struct bmp280_data *data, } adc_press = be32_to_cpu(tmp) >> 12; + if (adc_press == BMP280_PRESS_SKIPPED) { + /* reading was skipped */ + dev_err(data->dev, "reading pressure skipped\n"); + return -EIO; + } comp_press = bmp280_compensate_press(data, adc_press); *val = comp_press; @@ -345,6 +355,11 @@ static int bmp280_read_humid(struct bmp280_data *data, int *val, int *val2) } adc_humidity = be16_to_cpu(tmp); + if (adc_humidity == BMP280_HUMIDITY_SKIPPED) { + /* reading was skipped */ + dev_err(data->dev, "reading humidity skipped\n"); + return -EIO; + } comp_humidity = bmp280_compensate_humidity(data, adc_humidity); *val = comp_humidity; @@ -597,14 +612,20 @@ static const struct bmp280_chip_info bmp280_chip_info = { static int bme280_chip_config(struct bmp280_data *data) { - int ret = bmp280_chip_config(data); + int ret; u8 osrs = BMP280_OSRS_HUMIDITIY_X(data->oversampling_humid + 1); + /* + * Oversampling of humidity must be set before oversampling of + * temperature/pressure is set to become effective. + */ + ret = regmap_update_bits(data->regmap, BMP280_REG_CTRL_HUMIDITY, + BMP280_OSRS_HUMIDITY_MASK, osrs); + if (ret < 0) return ret; - return regmap_update_bits(data->regmap, BMP280_REG_CTRL_HUMIDITY, - BMP280_OSRS_HUMIDITY_MASK, osrs); + return bmp280_chip_config(data); } static const struct bmp280_chip_info bme280_chip_info = { diff --git a/drivers/iio/pressure/bmp280.h b/drivers/iio/pressure/bmp280.h index 2c770e1..6134743 100644 --- a/drivers/iio/pressure/bmp280.h +++ b/drivers/iio/pressure/bmp280.h @@ -96,6 +96,11 @@ #define BME280_CHIP_ID 0x60 #define BMP280_SOFT_RESET_VAL 0xB6 +/* BMP280 register skipped special values */ +#define BMP280_TEMP_SKIPPED 0x80000 +#define BMP280_PRESS_SKIPPED 0x80000 +#define BMP280_HUMIDITY_SKIPPED 0x8000 + /* Regmap configurations */ extern const struct regmap_config bmp180_regmap_config; extern const struct regmap_config bmp280_regmap_config; -- cgit v1.1 From 50b39608efb1e60f334f9b59128fd6d970bfd5a6 Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Thu, 27 Jul 2017 18:18:57 +0200 Subject: iio: trigger: stm32-timer: fix quadrature mode get routine Fixes: 4adec7da0536 ("iio: stm32 trigger: Add quadrature encoder device") SMS bitfiled is mode + 1. After reset, upon boot, SMS = 0. When reading from sysfs, stm32_get_quadrature_mode() returns -1 (e.g. -EPERM) which is wrong error code here. So, check SMS bitfiled matches valid encoder mode, or return -EINVAL. Signed-off-by: Fabrice Gasnier Acked-by: Benjamin Gaignard Signed-off-by: Jonathan Cameron --- drivers/iio/trigger/stm32-timer-trigger.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/iio/trigger/stm32-timer-trigger.c b/drivers/iio/trigger/stm32-timer-trigger.c index d22bc56..6aa73d6 100644 --- a/drivers/iio/trigger/stm32-timer-trigger.c +++ b/drivers/iio/trigger/stm32-timer-trigger.c @@ -571,11 +571,14 @@ static int stm32_get_quadrature_mode(struct iio_dev *indio_dev, { struct stm32_timer_trigger *priv = iio_priv(indio_dev); u32 smcr; + int mode; regmap_read(priv->regmap, TIM_SMCR, &smcr); - smcr &= TIM_SMCR_SMS; + mode = (smcr & TIM_SMCR_SMS) - 1; + if ((mode < 0) || (mode > ARRAY_SIZE(stm32_quadrature_modes))) + return -EINVAL; - return smcr - 1; + return mode; } static const struct iio_enum stm32_quadrature_mode_enum = { -- cgit v1.1 From 1987a08cd989fd9e5690e90a04e70046e93315f4 Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Thu, 27 Jul 2017 18:18:58 +0200 Subject: iio: trigger: stm32-timer: fix write_raw return value Fixes: 4adec7da0536 ("iio: stm32 trigger: Add quadrature encoder device") IIO core expects zero as return value for write_raw() callback in case of success. Signed-off-by: Fabrice Gasnier Acked-by: Benjamin Gaignard Signed-off-by: Jonathan Cameron --- drivers/iio/trigger/stm32-timer-trigger.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/iio/trigger/stm32-timer-trigger.c b/drivers/iio/trigger/stm32-timer-trigger.c index 6aa73d6..107918b 100644 --- a/drivers/iio/trigger/stm32-timer-trigger.c +++ b/drivers/iio/trigger/stm32-timer-trigger.c @@ -406,9 +406,8 @@ static int stm32_counter_write_raw(struct iio_dev *indio_dev, switch (mask) { case IIO_CHAN_INFO_RAW: - regmap_write(priv->regmap, TIM_CNT, val); + return regmap_write(priv->regmap, TIM_CNT, val); - return IIO_VAL_INT; case IIO_CHAN_INFO_SCALE: /* fixed scale */ return -EINVAL; -- cgit v1.1 From 06e3fe89988b1c99a3d9953b1d3b1faf3f047017 Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Thu, 27 Jul 2017 18:18:59 +0200 Subject: iio: trigger: stm32-timer: fix get/set down count direction Fixes: 4adec7da0536 ("iio: stm32 trigger: Add quadrature encoder device") This fixes two issues: - stm32_set_count_direction: to set down direction - stm32_get_count_direction: to get down direction IIO core provides/expects value to be an index of iio_enum items array. This needs to be turned by these routines into TIM_CR1_DIR (e.g. BIT(4)) value. Also, report error when attempting to write direction, when in encoder mode: in this case, direction is read only (given by encoder inputs). Signed-off-by: Fabrice Gasnier Acked-by: Benjamin Gaignard Signed-off-by: Jonathan Cameron --- drivers/iio/trigger/stm32-timer-trigger.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/iio/trigger/stm32-timer-trigger.c b/drivers/iio/trigger/stm32-timer-trigger.c index 107918b..d28aa02 100644 --- a/drivers/iio/trigger/stm32-timer-trigger.c +++ b/drivers/iio/trigger/stm32-timer-trigger.c @@ -594,13 +594,20 @@ static const char *const stm32_count_direction_states[] = { static int stm32_set_count_direction(struct iio_dev *indio_dev, const struct iio_chan_spec *chan, - unsigned int mode) + unsigned int dir) { struct stm32_timer_trigger *priv = iio_priv(indio_dev); + u32 val; + int mode; - regmap_update_bits(priv->regmap, TIM_CR1, TIM_CR1_DIR, mode); + /* In encoder mode, direction is RO (given by TI1/TI2 signals) */ + regmap_read(priv->regmap, TIM_SMCR, &val); + mode = (val & TIM_SMCR_SMS) - 1; + if ((mode >= 0) || (mode < ARRAY_SIZE(stm32_quadrature_modes))) + return -EBUSY; - return 0; + return regmap_update_bits(priv->regmap, TIM_CR1, TIM_CR1_DIR, + dir ? TIM_CR1_DIR : 0); } static int stm32_get_count_direction(struct iio_dev *indio_dev, @@ -611,7 +618,7 @@ static int stm32_get_count_direction(struct iio_dev *indio_dev, regmap_read(priv->regmap, TIM_CR1, &cr1); - return (cr1 & TIM_CR1_DIR); + return ((cr1 & TIM_CR1_DIR) ? 1 : 0); } static const struct iio_enum stm32_count_direction_enum = { -- cgit v1.1 From 90938ca432e6b8f6bb1c22a24984738fc3d906ed Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Thu, 27 Jul 2017 18:19:00 +0200 Subject: iio: trigger: stm32-timer: add enable attribute In order to use encoder mode, timers needs to be enabled (e.g. CEN bit) along with peripheral clock. Add IIO_CHAN_INFO_ENABLE attribute to handle this. Also, in triggered mode, CEN bit is set automatically in hardware. Then clock must be enabled before starting triggered mode. Signed-off-by: Fabrice Gasnier Acked-by: Benjamin Gaignard Signed-off-by: Jonathan Cameron --- drivers/iio/trigger/stm32-timer-trigger.c | 55 +++++++++++++++++++++++-------- 1 file changed, 41 insertions(+), 14 deletions(-) diff --git a/drivers/iio/trigger/stm32-timer-trigger.c b/drivers/iio/trigger/stm32-timer-trigger.c index d28aa02..14e6eb0 100644 --- a/drivers/iio/trigger/stm32-timer-trigger.c +++ b/drivers/iio/trigger/stm32-timer-trigger.c @@ -366,34 +366,32 @@ static int stm32_counter_read_raw(struct iio_dev *indio_dev, int *val, int *val2, long mask) { struct stm32_timer_trigger *priv = iio_priv(indio_dev); + u32 dat; switch (mask) { case IIO_CHAN_INFO_RAW: - { - u32 cnt; - - regmap_read(priv->regmap, TIM_CNT, &cnt); - *val = cnt; + regmap_read(priv->regmap, TIM_CNT, &dat); + *val = dat; + return IIO_VAL_INT; + case IIO_CHAN_INFO_ENABLE: + regmap_read(priv->regmap, TIM_CR1, &dat); + *val = (dat & TIM_CR1_CEN) ? 1 : 0; return IIO_VAL_INT; - } - case IIO_CHAN_INFO_SCALE: - { - u32 smcr; - regmap_read(priv->regmap, TIM_SMCR, &smcr); - smcr &= TIM_SMCR_SMS; + case IIO_CHAN_INFO_SCALE: + regmap_read(priv->regmap, TIM_SMCR, &dat); + dat &= TIM_SMCR_SMS; *val = 1; *val2 = 0; /* in quadrature case scale = 0.25 */ - if (smcr == 3) + if (dat == 3) *val2 = 2; return IIO_VAL_FRACTIONAL_LOG2; } - } return -EINVAL; } @@ -403,6 +401,7 @@ static int stm32_counter_write_raw(struct iio_dev *indio_dev, int val, int val2, long mask) { struct stm32_timer_trigger *priv = iio_priv(indio_dev); + u32 dat; switch (mask) { case IIO_CHAN_INFO_RAW: @@ -411,6 +410,22 @@ static int stm32_counter_write_raw(struct iio_dev *indio_dev, case IIO_CHAN_INFO_SCALE: /* fixed scale */ return -EINVAL; + + case IIO_CHAN_INFO_ENABLE: + if (val) { + regmap_read(priv->regmap, TIM_CR1, &dat); + if (!(dat & TIM_CR1_CEN)) + clk_enable(priv->clk); + regmap_update_bits(priv->regmap, TIM_CR1, TIM_CR1_CEN, + TIM_CR1_CEN); + } else { + regmap_read(priv->regmap, TIM_CR1, &dat); + regmap_update_bits(priv->regmap, TIM_CR1, TIM_CR1_CEN, + 0); + if (dat & TIM_CR1_CEN) + clk_disable(priv->clk); + } + return 0; } return -EINVAL; @@ -506,9 +521,19 @@ static int stm32_set_enable_mode(struct iio_dev *indio_dev, { struct stm32_timer_trigger *priv = iio_priv(indio_dev); int sms = stm32_enable_mode2sms(mode); + u32 val; if (sms < 0) return sms; + /* + * Triggered mode sets CEN bit automatically by hardware. So, first + * enable counter clock, so it can use it. Keeps it in sync with CEN. + */ + if (sms == 6) { + regmap_read(priv->regmap, TIM_CR1, &val); + if (!(val & TIM_CR1_CEN)) + clk_enable(priv->clk); + } regmap_update_bits(priv->regmap, TIM_SMCR, TIM_SMCR_SMS, sms); @@ -681,7 +706,9 @@ static const struct iio_chan_spec_ext_info stm32_trigger_count_info[] = { static const struct iio_chan_spec stm32_trigger_channel = { .type = IIO_COUNT, .channel = 0, - .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) | BIT(IIO_CHAN_INFO_SCALE), + .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) | + BIT(IIO_CHAN_INFO_ENABLE) | + BIT(IIO_CHAN_INFO_SCALE), .ext_info = stm32_trigger_count_info, .indexed = 1 }; -- cgit v1.1 From ff3aa88a4d61468baece3fc2bb54e2a3bea6360f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Stefan=20Br=C3=BCns?= Date: Wed, 26 Jul 2017 23:32:06 +0200 Subject: iio: adc: ina219: Avoid underflow for sleeping time MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Proper support for the INA219 lowered the minimum sampling period from 2*140us to 2*84us. Subtracting 200us later leads to an underflow and an almost infinite udelay later. Using a signed int for the sampling period provides sufficient range (at most 2*8640*1024us), but catches the underflow when comparing with buffer_us. Fixes: 18edac2e22f4 ("iio: adc: Fix integration time/averaging for INA219/220") Signed-off-by: Stefan Brüns Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ina2xx-adc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/ina2xx-adc.c b/drivers/iio/adc/ina2xx-adc.c index 232c0b8..c3f8613 100644 --- a/drivers/iio/adc/ina2xx-adc.c +++ b/drivers/iio/adc/ina2xx-adc.c @@ -644,7 +644,7 @@ static int ina2xx_capture_thread(void *data) { struct iio_dev *indio_dev = data; struct ina2xx_chip_info *chip = iio_priv(indio_dev); - unsigned int sampling_us = SAMPLING_PERIOD(chip); + int sampling_us = SAMPLING_PERIOD(chip); int buffer_us; /* -- cgit v1.1 From 50dbe1f4b453b2860ef0e3d48054b9fd24d5ae97 Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Mon, 24 Jul 2017 18:10:38 +0200 Subject: iio: adc: stm32: fix common clock rate ADC clock input is provided to internal prescaler (that decreases its frequency). It's then used as reference clock for conversions. - Fix common clock rate used then by stm32-adc sub-devices. Take common prescaler into account. Currently, rate is used to set "boost" mode. It may unnecessarily be set. This impacts power consumption. - Fix ADC max clock rate on STM32H7 (fADC from datasheet). Currently, prescaler may be set too low. This can result in ADC reference clock used for conversion to exceed max allowed clock frequency. Fixes: 95e339b6e85d ("iio: adc: stm32: add support for STM32H7") Signed-off-by: Fabrice Gasnier Signed-off-by: Jonathan Cameron --- drivers/iio/adc/stm32-adc-core.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/iio/adc/stm32-adc-core.c b/drivers/iio/adc/stm32-adc-core.c index e09233b..6096763 100644 --- a/drivers/iio/adc/stm32-adc-core.c +++ b/drivers/iio/adc/stm32-adc-core.c @@ -64,7 +64,7 @@ #define STM32H7_CKMODE_MASK GENMASK(17, 16) /* STM32 H7 maximum analog clock rate (from datasheet) */ -#define STM32H7_ADC_MAX_CLK_RATE 72000000 +#define STM32H7_ADC_MAX_CLK_RATE 36000000 /** * stm32_adc_common_regs - stm32 common registers, compatible dependent data @@ -148,14 +148,14 @@ static int stm32f4_adc_clk_sel(struct platform_device *pdev, return -EINVAL; } - priv->common.rate = rate; + priv->common.rate = rate / stm32f4_pclk_div[i]; val = readl_relaxed(priv->common.base + STM32F4_ADC_CCR); val &= ~STM32F4_ADC_ADCPRE_MASK; val |= i << STM32F4_ADC_ADCPRE_SHIFT; writel_relaxed(val, priv->common.base + STM32F4_ADC_CCR); dev_dbg(&pdev->dev, "Using analog clock source at %ld kHz\n", - rate / (stm32f4_pclk_div[i] * 1000)); + priv->common.rate / 1000); return 0; } @@ -250,7 +250,7 @@ static int stm32h7_adc_clk_sel(struct platform_device *pdev, out: /* rate used later by each ADC instance to control BOOST mode */ - priv->common.rate = rate; + priv->common.rate = rate / div; /* Set common clock mode and prescaler */ val = readl_relaxed(priv->common.base + STM32H7_ADC_CCR); @@ -260,7 +260,7 @@ out: writel_relaxed(val, priv->common.base + STM32H7_ADC_CCR); dev_dbg(&pdev->dev, "Using %s clock/%d source at %ld kHz\n", - ckmode ? "bus" : "adc", div, rate / (div * 1000)); + ckmode ? "bus" : "adc", div, priv->common.rate / 1000); return 0; } -- cgit v1.1 From 9beceb54fa2c0b47532dd2b07f37e410641cf9b2 Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Thu, 27 Jul 2017 11:22:04 +0900 Subject: netfilter: x_tables: Fix use-after-free in ipt_do_table. If verdict is NF_STOLEN in the SYNPROXY target, the skb is consumed. However, ipt_do_table() always tries to get ip header from the skb. So that, KASAN triggers the use-after-free message. We can reproduce this message using below command. # iptables -I INPUT -p tcp -j SYNPROXY --mss 1460 [ 193.542265] BUG: KASAN: use-after-free in ipt_do_table+0x1405/0x1c10 [ ... ] [ 193.578603] Call Trace: [ 193.581590] [ 193.584107] dump_stack+0x68/0xa0 [ 193.588168] print_address_description+0x78/0x290 [ 193.593828] ? ipt_do_table+0x1405/0x1c10 [ 193.598690] kasan_report+0x230/0x340 [ 193.603194] __asan_report_load2_noabort+0x19/0x20 [ 193.608950] ipt_do_table+0x1405/0x1c10 [ 193.613591] ? rcu_read_lock_held+0xae/0xd0 [ 193.618631] ? ip_route_input_rcu+0x27d7/0x4270 [ 193.624348] ? ipt_do_table+0xb68/0x1c10 [ 193.629124] ? do_add_counters+0x620/0x620 [ 193.634234] ? iptable_filter_net_init+0x60/0x60 [ ... ] After this patch, only when verdict is XT_CONTINUE, ipt_do_table() tries to get ip header. Also arpt_do_table() is modified because it has same bug. Signed-off-by: Taehee Yoo Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/arp_tables.c | 10 +++++----- net/ipv4/netfilter/ip_tables.c | 9 +++++---- 2 files changed, 10 insertions(+), 9 deletions(-) diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 0bc3c3d..9e9d9af 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -268,14 +268,14 @@ unsigned int arpt_do_table(struct sk_buff *skb, acpar.targinfo = t->data; verdict = t->u.kernel.target->target(skb, &acpar); - /* Target might have changed stuff. */ - arp = arp_hdr(skb); - - if (verdict == XT_CONTINUE) + if (verdict == XT_CONTINUE) { + /* Target might have changed stuff. */ + arp = arp_hdr(skb); e = arpt_next_entry(e); - else + } else { /* Verdict */ break; + } } while (!acpar.hotdrop); xt_write_recseq_end(addend); local_bh_enable(); diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 2a55a40..622ed28 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -352,13 +352,14 @@ ipt_do_table(struct sk_buff *skb, acpar.targinfo = t->data; verdict = t->u.kernel.target->target(skb, &acpar); - /* Target might have changed stuff. */ - ip = ip_hdr(skb); - if (verdict == XT_CONTINUE) + if (verdict == XT_CONTINUE) { + /* Target might have changed stuff. */ + ip = ip_hdr(skb); e = ipt_next_entry(e); - else + } else { /* Verdict */ break; + } } while (!acpar.hotdrop); xt_write_recseq_end(addend); -- cgit v1.1 From f3fd2afed8eee91620d05b69ab94c14793c849d7 Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Fri, 28 Jul 2017 15:10:48 -0700 Subject: ntb: transport shouldn't disable link due to bogus values in SPADs It seems that under certain scenarios the SPAD can have bogus values caused by an agent (i.e. BIOS or other software) that is not the kernel driver, and that causes memory window setup failure. This should not cause the link to be disabled because if we do that, the driver will never recover again. We have verified in testing that this issue happens and prevents proper link recovery. Signed-off-by: Dave Jiang Acked-by: Allen Hubbe Signed-off-by: Jon Mason Fixes: 84f766855f61 ("ntb: stop link work when we do not have memory") --- drivers/ntb/ntb_transport.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c index b29558d..f58d8e3 100644 --- a/drivers/ntb/ntb_transport.c +++ b/drivers/ntb/ntb_transport.c @@ -924,10 +924,8 @@ out1: ntb_free_mw(nt, i); /* if there's an actual failure, we should just bail */ - if (rc < 0) { - ntb_link_disable(ndev); + if (rc < 0) return; - } out: if (ntb_link_is_up(ndev, NULL, NULL) == 1) -- cgit v1.1 From 0eb46345364d7318b11068c46e8a68d5dc10f65e Mon Sep 17 00:00:00 2001 From: Logan Gunthorpe Date: Tue, 25 Jul 2017 14:57:42 -0600 Subject: ntb: ntb_test: ensure the link is up before trying to configure the mws After the link tests, there is a race on one side of the test for the link coming up. It's possible, in some cases, for the test script to write to the 'peer_trans' files before the link has come up. To fix this, we simply use the link event file to ensure both sides see the link as up before continuning. Signed-off-by: Logan Gunthorpe Acked-by: Allen Hubbe Signed-off-by: Jon Mason Fixes: a9c59ef77458 ("ntb_test: Add a selftest script for the NTB subsystem") --- tools/testing/selftests/ntb/ntb_test.sh | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/ntb/ntb_test.sh b/tools/testing/selftests/ntb/ntb_test.sh index 1c12b58..5fc7ad3 100755 --- a/tools/testing/selftests/ntb/ntb_test.sh +++ b/tools/testing/selftests/ntb/ntb_test.sh @@ -333,6 +333,10 @@ function ntb_tool_tests() link_test $LOCAL_TOOL $REMOTE_TOOL link_test $REMOTE_TOOL $LOCAL_TOOL + #Ensure the link is up on both sides before continuing + write_file Y $LOCAL_TOOL/link_event + write_file Y $REMOTE_TOOL/link_event + for PEER_TRANS in $(ls $LOCAL_TOOL/peer_trans*); do PT=$(basename $PEER_TRANS) write_file $MW_SIZE $LOCAL_TOOL/$PT -- cgit v1.1 From 3f5a95ad6c6c05e6b00f0c20e30da66c986564d5 Mon Sep 17 00:00:00 2001 From: Koichiro Den Date: Tue, 1 Aug 2017 23:21:46 +0900 Subject: xfrm: fix null pointer dereference on state and tmpl sort Creating sub policy that matches the same outer flow as main policy does leads to a null pointer dereference if the outer mode's family is ipv4. For userspace compatibility, this patch just eliminates the crash i.e., does not introduce any new sorting rule, which would fruitlessly affect all but the aforementioned case. Signed-off-by: Koichiro Den Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 6c0956d..a792eff 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1620,6 +1620,7 @@ int xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n, unsigned short family, struct net *net) { + int i; int err = 0; struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family); if (!afinfo) @@ -1628,6 +1629,9 @@ xfrm_tmpl_sort(struct xfrm_tmpl **dst, struct xfrm_tmpl **src, int n, spin_lock_bh(&net->xfrm.xfrm_state_lock); /*FIXME*/ if (afinfo->tmpl_sort) err = afinfo->tmpl_sort(dst, src, n); + else + for (i = 0; i < n; i++) + dst[i] = src[i]; spin_unlock_bh(&net->xfrm.xfrm_state_lock); rcu_read_unlock(); return err; @@ -1638,6 +1642,7 @@ int xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n, unsigned short family) { + int i; int err = 0; struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family); struct net *net = xs_net(*src); @@ -1648,6 +1653,9 @@ xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n, spin_lock_bh(&net->xfrm.xfrm_state_lock); if (afinfo->state_sort) err = afinfo->state_sort(dst, src, n); + else + for (i = 0; i < n; i++) + dst[i] = src[i]; spin_unlock_bh(&net->xfrm.xfrm_state_lock); rcu_read_unlock(); return err; -- cgit v1.1 From 4bb0f0e73c8c30917d169c4a0f1ac083690c545b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 1 Aug 2017 12:01:52 -0400 Subject: tracing: Call clear_boot_tracer() at lateinit_sync The clear_boot_tracer function is used to reset the default_bootup_tracer string to prevent it from being accessed after boot, as it originally points to init data. But since clear_boot_tracer() is called via the init_lateinit() call, it races with the initcall for registering the hwlat tracer. If someone adds "ftrace=hwlat" to the kernel command line, depending on how the linker sets up the text, the saved command line may be cleared, and the hwlat tracer never is initialized. Simply have the clear_boot_tracer() be called by initcall_lateinit_sync() as that's for tasks to be called after lateinit. Link: https://bugzilla.kernel.org/show_bug.cgi?id=196551 Cc: stable@vger.kernel.org Fixes: e7c15cd8a ("tracing: Added hardware latency tracer") Reported-by: Zamir SUN Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 42b9355..784fb43 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -8407,4 +8407,4 @@ __init static int clear_boot_tracer(void) } fs_initcall(tracer_init_tracefs); -late_initcall(clear_boot_tracer); +late_initcall_sync(clear_boot_tracer); -- cgit v1.1 From 147d88e0b5eb90191bc5c12ca0a3c410b75a13d2 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 1 Aug 2017 14:02:01 +0300 Subject: tracing: Missing error code in tracer_alloc_buffers() If ring_buffer_alloc() or one of the next couple function calls fail then we should return -ENOMEM but the current code returns success. Link: http://lkml.kernel.org/r/20170801110201.ajdkct7vwzixahvx@mwanda Cc: Sebastian Andrzej Siewior Cc: Ingo Molnar Cc: stable@vger.kernel.org Fixes: b32614c03413 ('tracing/rb: Convert to hotplug state machine') Signed-off-by: Dan Carpenter Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 784fb43..d815fc3 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -8293,6 +8293,7 @@ __init static int tracer_alloc_buffers(void) if (ret < 0) goto out_free_cpumask; /* Used for event triggers */ + ret = -ENOMEM; temp_buffer = ring_buffer_alloc(PAGE_SIZE, RB_FL_OVERWRITE); if (!temp_buffer) goto out_rm_hp_state; -- cgit v1.1 From a7e52ad7ed82e21273eccff93d1477a7b313aabb Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Wed, 2 Aug 2017 14:20:54 -0400 Subject: ring-buffer: Have ring_buffer_alloc_read_page() return error on offline CPU Chunyu Hu reported: "per_cpu trace directories and files are created for all possible cpus, but only the cpus which have ever been on-lined have their own per cpu ring buffer (allocated by cpuhp threads). While trace_buffers_open, the open handler for trace file 'trace_pipe_raw' is always trying to access field of ring_buffer_per_cpu, and would panic with the NULL pointer. Align the behavior of trace_pipe_raw with trace_pipe, that returns -NODEV when openning it if that cpu does not have trace ring buffer. Reproduce: cat /sys/kernel/debug/tracing/per_cpu/cpu31/trace_pipe_raw (cpu31 is never on-lined, this is a 16 cores x86_64 box) Tested with: 1) boot with maxcpus=14, read trace_pipe_raw of cpu15. Got -NODEV. 2) oneline cpu15, read trace_pipe_raw of cpu15. Get the raw trace data. Call trace: [ 5760.950995] RIP: 0010:ring_buffer_alloc_read_page+0x32/0xe0 [ 5760.961678] tracing_buffers_read+0x1f6/0x230 [ 5760.962695] __vfs_read+0x37/0x160 [ 5760.963498] ? __vfs_read+0x5/0x160 [ 5760.964339] ? security_file_permission+0x9d/0xc0 [ 5760.965451] ? __vfs_read+0x5/0x160 [ 5760.966280] vfs_read+0x8c/0x130 [ 5760.967070] SyS_read+0x55/0xc0 [ 5760.967779] do_syscall_64+0x67/0x150 [ 5760.968687] entry_SYSCALL64_slow_path+0x25/0x25" This was introduced by the addition of the feature to reuse reader pages instead of re-allocating them. The problem is that the allocation of a reader page (which is per cpu) does not check if the cpu is online and set up for the ring buffer. Link: http://lkml.kernel.org/r/1500880866-1177-1-git-send-email-chuhu@redhat.com Cc: stable@vger.kernel.org Fixes: 73a757e63114 ("ring-buffer: Return reader page back into existing ring buffer") Reported-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ring_buffer.c | 14 +++++++++----- kernel/trace/ring_buffer_benchmark.c | 2 +- kernel/trace/trace.c | 16 +++++++++++----- 3 files changed, 21 insertions(+), 11 deletions(-) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 529cc50..81279c6 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -4386,15 +4386,19 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); * the page that was allocated, with the read page of the buffer. * * Returns: - * The page allocated, or NULL on error. + * The page allocated, or ERR_PTR */ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) { - struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; + struct ring_buffer_per_cpu *cpu_buffer; struct buffer_data_page *bpage = NULL; unsigned long flags; struct page *page; + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return ERR_PTR(-ENODEV); + + cpu_buffer = buffer->buffers[cpu]; local_irq_save(flags); arch_spin_lock(&cpu_buffer->lock); @@ -4412,7 +4416,7 @@ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); if (!page) - return NULL; + return ERR_PTR(-ENOMEM); bpage = page_address(page); @@ -4467,8 +4471,8 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); * * for example: * rpage = ring_buffer_alloc_read_page(buffer, cpu); - * if (!rpage) - * return error; + * if (IS_ERR(rpage)) + * return PTR_ERR(rpage); * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); * if (ret >= 0) * process_page(rpage, ret); diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c index 9fbcaf5..68ee79a 100644 --- a/kernel/trace/ring_buffer_benchmark.c +++ b/kernel/trace/ring_buffer_benchmark.c @@ -113,7 +113,7 @@ static enum event_status read_page(int cpu) int i; bpage = ring_buffer_alloc_read_page(buffer, cpu); - if (!bpage) + if (IS_ERR(bpage)) return EVENT_DROPPED; ret = ring_buffer_read_page(buffer, &bpage, PAGE_SIZE, cpu, 1); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d815fc3..44004d8 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6598,7 +6598,7 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, { struct ftrace_buffer_info *info = filp->private_data; struct trace_iterator *iter = &info->iter; - ssize_t ret; + ssize_t ret = 0; ssize_t size; if (!count) @@ -6612,10 +6612,15 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, if (!info->spare) { info->spare = ring_buffer_alloc_read_page(iter->trace_buffer->buffer, iter->cpu_file); - info->spare_cpu = iter->cpu_file; + if (IS_ERR(info->spare)) { + ret = PTR_ERR(info->spare); + info->spare = NULL; + } else { + info->spare_cpu = iter->cpu_file; + } } if (!info->spare) - return -ENOMEM; + return ret; /* Do we have previous read data to read? */ if (info->read < PAGE_SIZE) @@ -6790,8 +6795,9 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, ref->ref = 1; ref->buffer = iter->trace_buffer->buffer; ref->page = ring_buffer_alloc_read_page(ref->buffer, iter->cpu_file); - if (!ref->page) { - ret = -ENOMEM; + if (IS_ERR(ref->page)) { + ret = PTR_ERR(ref->page); + ref->page = NULL; kfree(ref); break; } -- cgit v1.1 From 7bab09631c2a303f87a7eb7e3d69e888673b9b7e Mon Sep 17 00:00:00 2001 From: Vladis Dronov Date: Wed, 2 Aug 2017 19:50:14 +0200 Subject: xfrm: policy: check policy direction value The 'dir' parameter in xfrm_migrate() is a user-controlled byte which is used as an array index. This can lead to an out-of-bound access, kernel lockup and DoS. Add a check for the 'dir' value. This fixes CVE-2017-11600. References: https://bugzilla.redhat.com/show_bug.cgi?id=1474928 Fixes: 80c9abaabf42 ("[XFRM]: Extension for dynamic update of endpoint address(es)") Cc: # v2.6.21-rc1 Reported-by: "bo Zhang" Signed-off-by: Vladis Dronov Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index ff61d85..6f5a0dad 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -3308,9 +3308,15 @@ int xfrm_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, struct xfrm_state *x_new[XFRM_MAX_DEPTH]; struct xfrm_migrate *mp; + /* Stage 0 - sanity checks */ if ((err = xfrm_migrate_check(m, num_migrate)) < 0) goto out; + if (dir >= XFRM_POLICY_MAX) { + err = -EINVAL; + goto out; + } + /* Stage 1 - find policy */ if ((pol = xfrm_migrate_policy_find(sel, dir, type, net)) == NULL) { err = -ENOENT; -- cgit v1.1 From 4ff0308f06da5016aafb05330ed37809b54f81ae Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Mon, 7 Aug 2017 08:31:07 +0200 Subject: esp: Fix error handling on layer 2 xmit. esp_output_tail() and esp6_output_tail() can return negative and positive error values. We currently treat only negative values as errors, fix this to treat both cases as error. Fixes: fca11ebde3f0 ("esp4: Reorganize esp_output") Fixes: 383d0350f2cc ("esp6: Reorganize esp_output") Signed-off-by: Steffen Klassert --- net/ipv4/esp4_offload.c | 2 +- net/ipv6/esp6_offload.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c index e066601..5011232 100644 --- a/net/ipv4/esp4_offload.c +++ b/net/ipv4/esp4_offload.c @@ -257,7 +257,7 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb, netdev_features_ esp.seqno = cpu_to_be64(xo->seq.low + ((u64)xo->seq.hi << 32)); err = esp_output_tail(x, skb, &esp); - if (err < 0) + if (err) return err; secpath_reset(skb); diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c index f02f131..1cf437f 100644 --- a/net/ipv6/esp6_offload.c +++ b/net/ipv6/esp6_offload.c @@ -286,7 +286,7 @@ static int esp6_xmit(struct xfrm_state *x, struct sk_buff *skb, netdev_features esp.seqno = cpu_to_be64(xo->seq.low + ((u64)xo->seq.hi << 32)); err = esp6_output_tail(x, skb, &esp); - if (err < 0) + if (err) return err; secpath_reset(skb); -- cgit v1.1 From 7f5770678b2d0cc8f3ffbf7eb73410f2acba7925 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 30 Jul 2017 21:10:44 +0300 Subject: dmaengine: tegra210-adma: fix of_irq_get() error check of_irq_get() may return 0 as well as negative error number on failure, while the driver only checks for the negative values. The driver would then call request_irq(0, ...) in tegra_adma_alloc_chan_resources() and never get valid channel interrupt. Check for 'tdc->irq <= 0' instead and return -ENXIO from the driver's probe iff of_irq_get() returned 0. Fixes: f46b195799b5 ("dmaengine: tegra-adma: Add support for Tegra210 ADMA") Signed-off-by: Sergei Shtylyov Acked-by: Thierry Reding Acked-by: Jon Hunter Signed-off-by: Vinod Koul --- drivers/dma/tegra210-adma.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/dma/tegra210-adma.c b/drivers/dma/tegra210-adma.c index b10cbaa..b26256f 100644 --- a/drivers/dma/tegra210-adma.c +++ b/drivers/dma/tegra210-adma.c @@ -717,8 +717,8 @@ static int tegra_adma_probe(struct platform_device *pdev) tdc->chan_addr = tdma->base_addr + ADMA_CH_REG_OFFSET(i); tdc->irq = of_irq_get(pdev->dev.of_node, i); - if (tdc->irq < 0) { - ret = tdc->irq; + if (tdc->irq <= 0) { + ret = tdc->irq ?: -ENXIO; goto irq_dispose; } -- cgit v1.1 From cb87481ee89dbd6609e227afbf64900fb4e5c930 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Wed, 26 Jul 2017 22:46:27 +1000 Subject: kbuild: linker script do not match C names unless LD_DEAD_CODE_DATA_ELIMINATION is configured The .data and .bss sections were modified in the generic linker script to pull in sections named .data., which are generated by gcc with -ffunction-sections and -fdata-sections options. The problem with this pattern is it can also match section names that Linux defines explicitly, e.g., .data.unlikely. This can cause Linux sections to get moved into the wrong place. The way to avoid this is to use ".." separators for explicit section names (the dot character is valid in a section name but not a C identifier). However currently there are sections which don't follow this rule, so for now just disable the wild card by default. Example: http://marc.info/?l=linux-arm-kernel&m=150106824024221&w=2 Cc: # 4.9 Fixes: b67067f1176df ("kbuild: allow archs to select link dead code/data elimination") Signed-off-by: Nicholas Piggin Signed-off-by: Masahiro Yamada --- include/asm-generic/vmlinux.lds.h | 38 ++++++++++++++++++++++++++------------ 1 file changed, 26 insertions(+), 12 deletions(-) diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index da0be9a..9623d78 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -60,6 +60,22 @@ #define ALIGN_FUNCTION() . = ALIGN(8) /* + * LD_DEAD_CODE_DATA_ELIMINATION option enables -fdata-sections, which + * generates .data.identifier sections, which need to be pulled in with + * .data. We don't want to pull in .data..other sections, which Linux + * has defined. Same for text and bss. + */ +#ifdef CONFIG_LD_DEAD_CODE_DATA_ELIMINATION +#define TEXT_MAIN .text .text.[0-9a-zA-Z_]* +#define DATA_MAIN .data .data.[0-9a-zA-Z_]* +#define BSS_MAIN .bss .bss.[0-9a-zA-Z_]* +#else +#define TEXT_MAIN .text +#define DATA_MAIN .data +#define BSS_MAIN .bss +#endif + +/* * Align to a 32 byte boundary equal to the * alignment gcc 4.5 uses for a struct */ @@ -198,12 +214,9 @@ /* * .data section - * LD_DEAD_CODE_DATA_ELIMINATION option enables -fdata-sections generates - * .data.identifier which needs to be pulled in with .data, but don't want to - * pull in .data..stuff which has its own requirements. Same for bss. */ #define DATA_DATA \ - *(.data .data.[0-9a-zA-Z_]*) \ + *(DATA_MAIN) \ *(.ref.data) \ *(.data..shared_aligned) /* percpu related */ \ MEM_KEEP(init.data) \ @@ -434,16 +447,17 @@ VMLINUX_SYMBOL(__security_initcall_end) = .; \ } -/* .text section. Map to function alignment to avoid address changes +/* + * .text section. Map to function alignment to avoid address changes * during second ld run in second ld pass when generating System.map - * LD_DEAD_CODE_DATA_ELIMINATION option enables -ffunction-sections generates - * .text.identifier which needs to be pulled in with .text , but some - * architectures define .text.foo which is not intended to be pulled in here. - * Those enabling LD_DEAD_CODE_DATA_ELIMINATION must ensure they don't have - * conflicting section names, and must pull in .text.[0-9a-zA-Z_]* */ + * + * TEXT_MAIN here will match .text.fixup and .text.unlikely if dead + * code elimination is enabled, so these sections should be converted + * to use ".." first. + */ #define TEXT_TEXT \ ALIGN_FUNCTION(); \ - *(.text.hot .text .text.fixup .text.unlikely) \ + *(.text.hot TEXT_MAIN .text.fixup .text.unlikely) \ *(.ref.text) \ MEM_KEEP(init.text) \ MEM_KEEP(exit.text) \ @@ -613,7 +627,7 @@ BSS_FIRST_SECTIONS \ *(.bss..page_aligned) \ *(.dynbss) \ - *(.bss .bss.[0-9a-zA-Z_]*) \ + *(BSS_MAIN) \ *(COMMON) \ } -- cgit v1.1 From 312a3d0918bb7d65862fbbd3e2f2f4630e4d6f56 Mon Sep 17 00:00:00 2001 From: Cao jin Date: Wed, 2 Aug 2017 10:31:06 +0800 Subject: kbuild: trivial cleanups on the comments This is a bunch of trivial fixes and cleanups. Signed-off-by: Cao jin Signed-off-by: Masahiro Yamada --- Makefile | 10 +++++----- scripts/Kbuild.include | 7 +++---- scripts/Makefile.build | 8 ++++---- scripts/basic/Makefile | 2 +- 4 files changed, 13 insertions(+), 14 deletions(-) diff --git a/Makefile b/Makefile index 6eba23b..473f1aa 100644 --- a/Makefile +++ b/Makefile @@ -442,7 +442,7 @@ export RCS_TAR_IGNORE := --exclude SCCS --exclude BitKeeper --exclude .svn \ # =========================================================================== # Rules shared between *config targets and build targets -# Basic helpers built in scripts/ +# Basic helpers built in scripts/basic/ PHONY += scripts_basic scripts_basic: $(Q)$(MAKE) $(build)=scripts/basic @@ -505,7 +505,7 @@ ifeq ($(KBUILD_EXTMOD),) endif endif endif -# install and module_install need also be processed one by one +# install and modules_install need also be processed one by one ifneq ($(filter install,$(MAKECMDGOALS)),) ifneq ($(filter modules_install,$(MAKECMDGOALS)),) mixed-targets := 1 @@ -964,7 +964,7 @@ export KBUILD_VMLINUX_MAIN := $(core-y) $(libs-y2) $(drivers-y) $(net-y) $(virt- export KBUILD_VMLINUX_LIBS := $(libs-y1) export KBUILD_LDS := arch/$(SRCARCH)/kernel/vmlinux.lds export LDFLAGS_vmlinux -# used by scripts/pacmage/Makefile +# used by scripts/package/Makefile export KBUILD_ALLDIRS := $(sort $(filter-out arch/%,$(vmlinux-alldirs)) arch Documentation include samples scripts tools) vmlinux-deps := $(KBUILD_LDS) $(KBUILD_VMLINUX_INIT) $(KBUILD_VMLINUX_MAIN) $(KBUILD_VMLINUX_LIBS) @@ -992,8 +992,8 @@ include/generated/autoksyms.h: FORCE ARCH_POSTLINK := $(wildcard $(srctree)/arch/$(SRCARCH)/Makefile.postlink) # Final link of vmlinux with optional arch pass after final link - cmd_link-vmlinux = \ - $(CONFIG_SHELL) $< $(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) ; \ +cmd_link-vmlinux = \ + $(CONFIG_SHELL) $< $(LD) $(LDFLAGS) $(LDFLAGS_vmlinux) ; \ $(if $(ARCH_POSTLINK), $(MAKE) -f $(ARCH_POSTLINK) $@, true) vmlinux: scripts/link-vmlinux.sh vmlinux_prereq $(vmlinux-deps) FORCE diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index dd8e2dd..9ffd3dd 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -85,8 +85,8 @@ TMPOUT := $(if $(KBUILD_EXTMOD),$(firstword $(KBUILD_EXTMOD))/) # try-run # Usage: option = $(call try-run, $(CC)...-o "$$TMP",option-ok,otherwise) -# Exit code chooses option. "$$TMP" is can be used as temporary file and -# is automatically cleaned up. +# Exit code chooses option. "$$TMP" serves as a temporary file and is +# automatically cleaned up. try-run = $(shell set -e; \ TMP="$(TMPOUT).$$$$.tmp"; \ TMPO="$(TMPOUT).$$$$.o"; \ @@ -261,7 +261,6 @@ make-cmd = $(call escsq,$(subst \#,\\\#,$(subst $$,$$$$,$(cmd_$(1))))) any-prereq = $(filter-out $(PHONY),$?) $(filter-out $(PHONY) $(wildcard $^),$^) # Execute command if command has changed or prerequisite(s) are updated. -# if_changed = $(if $(strip $(any-prereq) $(arg-check)), \ @set -e; \ $(echo-cmd) $(cmd_$(1)); \ @@ -315,7 +314,7 @@ if_changed_rule = $(if $(strip $(any-prereq) $(arg-check) ), \ $(rule_$(1)), @:) ### -# why - tell why a a target got build +# why - tell why a target got built # enabled by make V=2 # Output (listed in the order they are checked): # (1) - due to target is PHONY diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 4a9a2ce..f6152c7 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -229,8 +229,8 @@ ifeq ("$(origin RECORDMCOUNT_WARN)", "command line") endif # Due to recursion, we must skip empty.o. # The empty.o file is created in the make process in order to determine -# the target endianness and word size. It is made before all other C -# files, including recordmcount. +# the target endianness and word size. It is made before all other C +# files, including recordmcount. sub_cmd_record_mcount = \ if [ $(@) != "scripts/mod/empty.o" ]; then \ $(objtree)/scripts/recordmcount $(RECORDMCOUNT_FLAGS) "$(@)"; \ @@ -245,13 +245,13 @@ sub_cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH "$(LD)" "$(NM)" "$(RM)" "$(MV)" \ "$(if $(part-of-module),1,0)" "$(@)"; recordmcount_source := $(srctree)/scripts/recordmcount.pl -endif +endif # BUILD_C_RECORDMCOUNT cmd_record_mcount = \ if [ "$(findstring $(CC_FLAGS_FTRACE),$(_c_flags))" = \ "$(CC_FLAGS_FTRACE)" ]; then \ $(sub_cmd_record_mcount) \ fi; -endif +endif # CONFIG_FTRACE_MCOUNT_RECORD ifdef CONFIG_STACK_VALIDATION ifneq ($(SKIP_STACK_VALIDATION),1) diff --git a/scripts/basic/Makefile b/scripts/basic/Makefile index ec10d93..0372b33 100644 --- a/scripts/basic/Makefile +++ b/scripts/basic/Makefile @@ -1,5 +1,5 @@ ### -# Makefile.basic lists the most basic programs used during the build process. +# This Makefile lists the most basic programs used during the build process. # The programs listed herein are what are needed to do the basic stuff, # such as fix file dependencies. # This initial step is needed to avoid files to be recompiled -- cgit v1.1 From 4e433fc4d1a93310d31e7671faca8660a2890908 Mon Sep 17 00:00:00 2001 From: Cao jin Date: Tue, 8 Aug 2017 21:20:50 +0800 Subject: fixdep: trivial: typo fix and correction Signed-off-by: Cao jin Signed-off-by: Masahiro Yamada --- scripts/basic/fixdep.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/basic/fixdep.c b/scripts/basic/fixdep.c index fff818b..bbf62cb 100644 --- a/scripts/basic/fixdep.c +++ b/scripts/basic/fixdep.c @@ -25,7 +25,7 @@ * * So we play the same trick that "mkdep" played before. We replace * the dependency on autoconf.h by a dependency on every config - * option which is mentioned in any of the listed prequisites. + * option which is mentioned in any of the listed prerequisites. * * kconfig populates a tree in include/config/ with an empty file * for each config symbol and when the configuration is updated @@ -34,7 +34,7 @@ * the config symbols are rebuilt. * * So if the user changes his CONFIG_HIS_DRIVER option, only the objects - * which depend on "include/linux/config/his/driver.h" will be rebuilt, + * which depend on "include/config/his/driver.h" will be rebuilt, * so most likely only his driver ;-) * * The idea above dates, by the way, back to Michael E Chastain, AFAIK. @@ -75,7 +75,7 @@ * and then basically copies the ..d file to stdout, in the * process filtering out the dependency on autoconf.h and adding * dependencies on include/config/my/option.h for every - * CONFIG_MY_OPTION encountered in any of the prequisites. + * CONFIG_MY_OPTION encountered in any of the prerequisites. * * It will also filter out all the dependencies on *.ver. We need * to make sure that the generated version checksum are globally up -- cgit v1.1 From 0fa375e6bc9023211eead30a6a79963c45a563da Mon Sep 17 00:00:00 2001 From: Jeffy Chen Date: Wed, 9 Aug 2017 18:41:03 +0800 Subject: drm/rockchip: Fix suspend crash when drm is not bound Currently we are allocating drm_device in rockchip_drm_bind, so if the suspend/resume code access it when drm is not bound, we would hit this crash: [ 253.402836] Unable to handle kernel NULL pointer dereference at virtual address 00000028 [ 253.402837] pgd = ffffffc06c9b0000 [ 253.402841] [00000028] *pgd=0000000000000000, *pud=0000000000000000 [ 253.402844] Internal error: Oops: 96000005 [#1] PREEMPT SMP [ 253.402859] Modules linked in: btusb btrtl btbcm btintel bluetooth ath10k_pci ath10k_core ar10k_ath ar10k_mac80211 cfg80211 ip6table_filter asix usbnet mii [ 253.402864] CPU: 4 PID: 1331 Comm: cat Not tainted 4.4.70 #15 [ 253.402865] Hardware name: Google Scarlet (DT) [ 253.402867] task: ffffffc076c0ce00 ti: ffffffc06c2c8000 task.ti: ffffffc06c2c8000 [ 253.402871] PC is at rockchip_drm_sys_suspend+0x20/0x5c Add sanity checks to prevent that. Reported-by: Brian Norris Signed-off-by: Jeffy Chen Signed-off-by: Sean Paul Link: https://patchwork.kernel.org/patch/9890297/ --- drivers/gpu/drm/rockchip/rockchip_drm_drv.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c index c6b1b7f..c16bc0a 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c @@ -275,11 +275,15 @@ static void rockchip_drm_fb_resume(struct drm_device *drm) static int rockchip_drm_sys_suspend(struct device *dev) { struct drm_device *drm = dev_get_drvdata(dev); - struct rockchip_drm_private *priv = drm->dev_private; + struct rockchip_drm_private *priv; + + if (!drm) + return 0; drm_kms_helper_poll_disable(drm); rockchip_drm_fb_suspend(drm); + priv = drm->dev_private; priv->state = drm_atomic_helper_suspend(drm); if (IS_ERR(priv->state)) { rockchip_drm_fb_resume(drm); @@ -293,8 +297,12 @@ static int rockchip_drm_sys_suspend(struct device *dev) static int rockchip_drm_sys_resume(struct device *dev) { struct drm_device *drm = dev_get_drvdata(dev); - struct rockchip_drm_private *priv = drm->dev_private; + struct rockchip_drm_private *priv; + + if (!drm) + return 0; + priv = drm->dev_private; drm_atomic_helper_resume(drm, priv->state); rockchip_drm_fb_resume(drm); drm_kms_helper_poll_enable(drm); -- cgit v1.1 From 2406b296a3a80ba1c78707f205556f2388d474ff Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 21 Jul 2017 22:56:19 +0200 Subject: gpu: ipu-v3: add DRM dependency The new PRE/PRG driver code causes a link failure when IPUv3 is built-in, but DRM is built as a module: drivers/gpu/ipu-v3/ipu-pre.o: In function `ipu_pre_configure': ipu-pre.c:(.text.ipu_pre_configure+0x18): undefined reference to `drm_format_info' drivers/gpu/ipu-v3/ipu-prg.o: In function `ipu_prg_format_supported': ipu-prg.c:(.text.ipu_prg_format_supported+0x8): undefined reference to `drm_format_info' Adding a Kconfig dependency on DRM means we don't run into this problem any more. If DRM is disabled altogether, the IPUv3 driver is built without PRE/PRG support. Fixes: ea9c260514c1 ("gpu: ipu-v3: add driver for Prefetch Resolve Gasket") Link: https://patchwork.kernel.org/patch/9636665/ Signed-off-by: Arnd Bergmann [p.zabel@pengutronix.de: changed the dependency from DRM to DRM || !DRM, since the link failure only happens when DRM=m and IPUV3_CORE=y. Modified the commit message to reflect this.] Signed-off-by: Philipp Zabel --- drivers/gpu/ipu-v3/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/ipu-v3/Kconfig b/drivers/gpu/ipu-v3/Kconfig index 08766c6..87a20b3 100644 --- a/drivers/gpu/ipu-v3/Kconfig +++ b/drivers/gpu/ipu-v3/Kconfig @@ -1,6 +1,7 @@ config IMX_IPUV3_CORE tristate "IPUv3 core support" depends on SOC_IMX5 || SOC_IMX6Q || ARCH_MULTIPLATFORM + depends on DRM || !DRM # if DRM=m, this can't be 'y' select GENERIC_IRQ_CHIP help Choose this if you have a i.MX5/6 system and want to use the Image -- cgit v1.1 From 5be5dd38d4628fdbff7359f235f7cdf0cf9655f1 Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Fri, 5 Aug 2016 11:55:18 +0200 Subject: drm/imx: ipuv3-plane: fix YUV framebuffer scanout on the base plane Historically, only RGB framebuffers could be assigned to the primary plane. This changed with universal plane support. Since no colorspace conversion was set up for the IPUv3 full plane, assigning YUV frame buffers to the primary plane caused incorrect output. Fix this by enabling color space conversion also for the primary plane. Signed-off-by: Philipp Zabel --- drivers/gpu/drm/imx/ipuv3-plane.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c index 6276bb8..d384598 100644 --- a/drivers/gpu/drm/imx/ipuv3-plane.c +++ b/drivers/gpu/drm/imx/ipuv3-plane.c @@ -545,15 +545,13 @@ static void ipu_plane_atomic_update(struct drm_plane *plane, return; } + ics = ipu_drm_fourcc_to_colorspace(fb->format->format); switch (ipu_plane->dp_flow) { case IPU_DP_FLOW_SYNC_BG: - ipu_dp_setup_channel(ipu_plane->dp, - IPUV3_COLORSPACE_RGB, - IPUV3_COLORSPACE_RGB); + ipu_dp_setup_channel(ipu_plane->dp, ics, IPUV3_COLORSPACE_RGB); ipu_dp_set_global_alpha(ipu_plane->dp, true, 0, true); break; case IPU_DP_FLOW_SYNC_FG: - ics = ipu_drm_fourcc_to_colorspace(state->fb->format->format); ipu_dp_setup_channel(ipu_plane->dp, ics, IPUV3_COLORSPACE_UNKNOWN); /* Enable local alpha on partial plane */ -- cgit v1.1 From b80b32b6d5e79798b85cd4644206aaa069059390 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Mon, 14 Aug 2017 08:29:18 -0400 Subject: ext4: fix clang build regression Arnd Bergmann As Stefan pointed out, I misremembered what clang can do specifically, and it turns out that the variable-length array at the end of the structure did not work (a flexible array would have worked here but not solved the problem): fs/ext4/mballoc.c:2303:17: error: fields must have a constant size: 'variable length array in structure' extension will never be supported ext4_grpblk_t counters[blocksize_bits + 2]; This reverts part of my previous patch, using a fixed-size array again, but keeping the check for the array overflow. Fixes: 2df2c3402fc8 ("ext4: fix warning about stack corruption") Reported-by: Stefan Agner Tested-by: Chandan Rajendra Signed-off-by: Arnd Bergmann Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 5a10526..7010856 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2300,7 +2300,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) EXT4_MAX_BLOCK_LOG_SIZE); struct sg { struct ext4_group_info info; - ext4_grpblk_t counters[blocksize_bits + 2]; + ext4_grpblk_t counters[EXT4_MAX_BLOCK_LOG_SIZE + 2]; } sg; group--; @@ -2309,6 +2309,9 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) " 2^0 2^1 2^2 2^3 2^4 2^5 2^6 " " 2^7 2^8 2^9 2^10 2^11 2^12 2^13 ]\n"); + i = (blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) + + sizeof(struct ext4_group_info); + grinfo = ext4_get_group_info(sb, group); /* Load the group info in memory only if not already loaded. */ if (unlikely(EXT4_MB_GRP_NEED_INIT(grinfo))) { @@ -2320,7 +2323,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) buddy_loaded = 1; } - memcpy(&sg, ext4_get_group_info(sb, group), sizeof(sg)); + memcpy(&sg, ext4_get_group_info(sb, group), i); if (buddy_loaded) ext4_mb_unload_buddy(&e4b); -- cgit v1.1 From 32aaf194201e98db4235b7b71ac62a22e2ac355f Mon Sep 17 00:00:00 2001 From: Tahsin Erdogan Date: Mon, 14 Aug 2017 08:30:06 -0400 Subject: ext4: add missing xattr hash update When updating an extended attribute, if the padded value sizes are the same, a shortcut is taken to avoid the bulk of the work. This was fine until the xattr hash update was moved inside ext4_xattr_set_entry(). With that change, the hash update got missed in the shortcut case. Thanks to ZhangYi (yizhang089@gmail.com) for root causing the problem. Fixes: daf8328172df ("ext4: eliminate xattr entry e_hash recalculation for removes") Reported-by: Miklos Szeredi Signed-off-by: Tahsin Erdogan Signed-off-by: Theodore Ts'o --- fs/ext4/xattr.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 82a5af9..3dd9701 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1543,7 +1543,7 @@ static int ext4_xattr_set_entry(struct ext4_xattr_info *i, /* Clear padding bytes. */ memset(val + i->value_len, 0, new_size - i->value_len); } - return 0; + goto update_hash; } /* Compute min_offs and last. */ @@ -1707,6 +1707,7 @@ static int ext4_xattr_set_entry(struct ext4_xattr_info *i, here->e_value_size = cpu_to_le32(i->value_len); } +update_hash: if (i->value) { __le32 hash = 0; @@ -1725,7 +1726,8 @@ static int ext4_xattr_set_entry(struct ext4_xattr_info *i, here->e_name_len, &crc32c_hash, 1); } else if (is_block) { - __le32 *value = s->base + min_offs - new_size; + __le32 *value = s->base + le16_to_cpu( + here->e_value_offs); hash = ext4_xattr_hash_entry(here->e_name, here->e_name_len, value, -- cgit v1.1 From 90b05b0598c698252d5f7dcdce335c5a546e9047 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 2 Aug 2017 13:47:42 +0900 Subject: gpio: reject invalid gpio before getting gpio_desc Check user-given gpio number and reject it before calling gpio_to_desc() because gpio_to_desc() is for kernel driver and it expects given gpio number is valid (means 0 to 511). If given number is invalid, gpio_to_desc() calls WARN() and dump registers and stack for debug. This means user can easily kick WARN() just by writing invalid gpio number (e.g. 512) to /sys/class/gpio/export. Fixes: 0e9a5edf5d01 ("gpio: fix deferred probe detection for legacy API") Signed-off-by: Masami Hiramatsu Reviewed-by: Andy Shevchenko Signed-off-by: Linus Walleij --- drivers/gpio/gpiolib-sysfs.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/gpio/gpiolib-sysfs.c b/drivers/gpio/gpiolib-sysfs.c index 16fe974..fc80add 100644 --- a/drivers/gpio/gpiolib-sysfs.c +++ b/drivers/gpio/gpiolib-sysfs.c @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -432,6 +433,11 @@ static struct attribute *gpiochip_attrs[] = { }; ATTRIBUTE_GROUPS(gpiochip); +static struct gpio_desc *gpio_to_valid_desc(int gpio) +{ + return gpio_is_valid(gpio) ? gpio_to_desc(gpio) : NULL; +} + /* * /sys/class/gpio/export ... write-only * integer N ... number of GPIO to export (full access) @@ -450,7 +456,7 @@ static ssize_t export_store(struct class *class, if (status < 0) goto done; - desc = gpio_to_desc(gpio); + desc = gpio_to_valid_desc(gpio); /* reject invalid GPIOs */ if (!desc) { pr_warn("%s: invalid GPIO %ld\n", __func__, gpio); @@ -493,7 +499,7 @@ static ssize_t unexport_store(struct class *class, if (status < 0) goto done; - desc = gpio_to_desc(gpio); + desc = gpio_to_valid_desc(gpio); /* reject bogus commands (gpio_unexport ignores them) */ if (!desc) { pr_warn("%s: invalid GPIO %ld\n", __func__, gpio); -- cgit v1.1 From 3f13b6a24f192da3096389b82202f16eff2c11ee Mon Sep 17 00:00:00 2001 From: Gregory CLEMENT Date: Wed, 12 Jul 2017 13:22:29 +0200 Subject: gpio: mvebu: Fix cause computation in irq handler When switching to regmap, the way to compute the irq cause was reorganized. However while doing it, a typo was introduced: a 'xor' replaced a 'and'. This lead to wrong behavior in the interrupt handler ans one of the symptom was wrong irq handler called on the Armada 388 GP: "->handle_irq(): c016303c, handle_bad_irq+0x0/0x278 ->irq_data.chip(): c0b0ec0c, 0xc0b0ec0c ->action(): (null) IRQ_NOPROBE set IRQ_NOREQUEST set unexpected IRQ trap at vector 00 irq 0, desc: ee804800, depth: 1, count: 0, unhandled: 0" Fixes: 2233bf7a92e7 ("gpio: mvebu: switch to regmap for register access") Signed-off-by: Gregory CLEMENT Signed-off-by: Linus Walleij --- drivers/gpio/gpio-mvebu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpio/gpio-mvebu.c b/drivers/gpio/gpio-mvebu.c index e338c37..45c65f8 100644 --- a/drivers/gpio/gpio-mvebu.c +++ b/drivers/gpio/gpio-mvebu.c @@ -557,7 +557,7 @@ static void mvebu_gpio_irq_handler(struct irq_desc *desc) edge_cause = mvebu_gpio_read_edge_cause(mvchip); edge_mask = mvebu_gpio_read_edge_mask(mvchip); - cause = (data_in ^ level_mask) | (edge_cause & edge_mask); + cause = (data_in & level_mask) | (edge_cause & edge_mask); for (i = 0; i < mvchip->chip.ngpio; i++) { int irq; -- cgit v1.1 From 491ab4700d1b64f5cf2f9055e01613a923df5fab Mon Sep 17 00:00:00 2001 From: Nikhil Mahale Date: Wed, 9 Aug 2017 09:23:01 +0530 Subject: drm: Fix framebuffer leak Do not leak framebuffer if client provided crtc id found invalid. Signed-off-by: Nikhil Mahale Cc: stable@vger.kernel.org Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/1502250781-5779-1-git-send-email-nmahale@nvidia.com --- drivers/gpu/drm/drm_plane.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c index 5dc8c43..e40c12f 100644 --- a/drivers/gpu/drm/drm_plane.c +++ b/drivers/gpu/drm/drm_plane.c @@ -601,6 +601,7 @@ int drm_mode_setplane(struct drm_device *dev, void *data, crtc = drm_crtc_find(dev, plane_req->crtc_id); if (!crtc) { + drm_framebuffer_put(fb); DRM_DEBUG_KMS("Unknown crtc ID %d\n", plane_req->crtc_id); return -ENOENT; -- cgit v1.1 From 7f5d6dac548b983702dd7aac1d463bd88dff50a8 Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Mon, 14 Aug 2017 12:07:21 +0200 Subject: drm/atomic: Handle -EDEADLK with out-fences correctly complete_crtc_signaling is freeing fence_state, but when retrying num_fences and fence_state are not zero'd. This caused duplicate fd's in the fence_state array, followed by a BUG_ON in fs/file.c because we reallocate freed memory, and installing over an existing fd, or potential other fun. Zero fence_state and num_fences correctly in the retry loop, which allows kms_atomic_transition to pass. Fixes: beaf5af48034 ("drm/fence: add out-fences support") Cc: Gustavo Padovan Cc: Brian Starkey (v10) Cc: Sean Paul Cc: Daniel Vetter Cc: Jani Nikula Cc: David Airlie Signed-off-by: Maarten Lankhorst Cc: # v4.10+ Testcase: kms_atomic_transitions.plane-all-modeset-transition-fencing (with CONFIG_DEBUG_WW_MUTEX_SLOWPATH=y) Link: https://patchwork.freedesktop.org/patch/msgid/20170814100721.13340-1-maarten.lankhorst@linux.intel.com Reviewed-by: Daniel Vetter #intel-gfx on irc --- drivers/gpu/drm/drm_atomic.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index c0f336d..b43939f 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -2167,10 +2167,10 @@ int drm_mode_atomic_ioctl(struct drm_device *dev, struct drm_atomic_state *state; struct drm_modeset_acquire_ctx ctx; struct drm_plane *plane; - struct drm_out_fence_state *fence_state = NULL; + struct drm_out_fence_state *fence_state; unsigned plane_mask; int ret = 0; - unsigned int i, j, num_fences = 0; + unsigned int i, j, num_fences; /* disallow for drivers not supporting atomic: */ if (!drm_core_check_feature(dev, DRIVER_ATOMIC)) @@ -2211,6 +2211,8 @@ retry: plane_mask = 0; copied_objs = 0; copied_props = 0; + fence_state = NULL; + num_fences = 0; for (i = 0; i < arg->count_objs; i++) { uint32_t obj_id, count_props; -- cgit v1.1 From a23318feeff662c8d25d21623daebdd2e55ec221 Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Wed, 9 Aug 2017 15:28:22 +0200 Subject: i2c: designware: Fix system suspend The commit 8503ff166504 ("i2c: designware: Avoid unnecessary resuming during system suspend"), may suggest to the PM core to try out the so called direct_complete path for system sleep. In this path, the PM core treats a runtime suspended device as it's already in a proper low power state for system sleep, which makes it skip calling the system sleep callbacks for the device, except for the ->prepare() and the ->complete() callbacks. However, the PM core may unset the direct_complete flag for a parent device, in case its child device are being system suspended before. In this scenario, the PM core invokes the system sleep callbacks, no matter if the device is runtime suspended or not. Particularly in cases of an existing i2c slave device, the above path is triggered, which breaks the assumption that the i2c device is always runtime resumed whenever the dw_i2c_plat_suspend() is being called. More precisely, dw_i2c_plat_suspend() calls clk_core_disable() and clk_core_unprepare(), for an already disabled/unprepared clock, leading to a splat in the log about clocks calls being wrongly balanced and breaking system sleep. To still allow the direct_complete path in cases when it's possible, but also to keep the fix simple, let's runtime resume the i2c device in the ->suspend() callback, before continuing to put the device into low power state. Note, in cases when the i2c device is attached to the ACPI PM domain, this problem doesn't occur, because ACPI's ->suspend() callback, assigned to acpi_subsys_suspend(), already calls pm_runtime_resume() for the device. It should also be noted that this change does not fix commit 8503ff166504 ("i2c: designware: Avoid unnecessary resuming during system suspend"). Because for the non-ACPI case, the system sleep support was already broken prior that point. Cc: # v4.4+ Signed-off-by: Ulf Hansson Acked-by: Rafael J. Wysocki Tested-by: John Stultz Tested-by: Jarkko Nikula Acked-by: Jarkko Nikula Reviewed-by: Mika Westerberg Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-designware-platdrv.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c index 143a8fd..441afc7 100644 --- a/drivers/i2c/busses/i2c-designware-platdrv.c +++ b/drivers/i2c/busses/i2c-designware-platdrv.c @@ -430,7 +430,7 @@ static void dw_i2c_plat_complete(struct device *dev) #endif #ifdef CONFIG_PM -static int dw_i2c_plat_suspend(struct device *dev) +static int dw_i2c_plat_runtime_suspend(struct device *dev) { struct platform_device *pdev = to_platform_device(dev); struct dw_i2c_dev *i_dev = platform_get_drvdata(pdev); @@ -452,11 +452,21 @@ static int dw_i2c_plat_resume(struct device *dev) return 0; } +#ifdef CONFIG_PM_SLEEP +static int dw_i2c_plat_suspend(struct device *dev) +{ + pm_runtime_resume(dev); + return dw_i2c_plat_runtime_suspend(dev); +} +#endif + static const struct dev_pm_ops dw_i2c_dev_pm_ops = { .prepare = dw_i2c_plat_prepare, .complete = dw_i2c_plat_complete, SET_SYSTEM_SLEEP_PM_OPS(dw_i2c_plat_suspend, dw_i2c_plat_resume) - SET_RUNTIME_PM_OPS(dw_i2c_plat_suspend, dw_i2c_plat_resume, NULL) + SET_RUNTIME_PM_OPS(dw_i2c_plat_runtime_suspend, + dw_i2c_plat_resume, + NULL) }; #define DW_I2C_DEV_PMOPS (&dw_i2c_dev_pm_ops) -- cgit v1.1 From 984277a041d5ee4a65aaadf0307d67a7c401e11c Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Fri, 11 Aug 2017 14:44:55 +0300 Subject: i2c: designware: Fix oops from i2c_dw_irq_handler_slave When i2c-designware is initialized in slave mode the i2c-designware-slave.c: i2c_dw_irq_handler_slave() can hit a NULL pointer dereference when I2C slave backend is not registered but code is accessing the struct dw_i2c_dev.slave without testing is it NULL. We might get spurious interrupts from other devices or from IRQ core during unloading the driver when CONFIG_DEBUG_SHIRQ is set. Existing check for enable and IRQ status is not enough since device can be power gated and those bits may read 1. Fix this by handling the interrupt only when also struct dw_i2c_dev.slave is set. Signed-off-by: Jarkko Nikula Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-designware-slave.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-designware-slave.c b/drivers/i2c/busses/i2c-designware-slave.c index 0548c7e..4b62a38 100644 --- a/drivers/i2c/busses/i2c-designware-slave.c +++ b/drivers/i2c/busses/i2c-designware-slave.c @@ -272,7 +272,7 @@ static int i2c_dw_irq_handler_slave(struct dw_i2c_dev *dev) slave_activity = ((dw_readl(dev, DW_IC_STATUS) & DW_IC_STATUS_SLAVE_ACTIVITY) >> 6); - if (!enabled || !(raw_stat & ~DW_IC_INTR_ACTIVITY)) + if (!enabled || !(raw_stat & ~DW_IC_INTR_ACTIVITY) || !dev->slave) return 0; dev_dbg(dev->dev, -- cgit v1.1 From 4e2d93de070ceaca5097f7ee5c311731b83208a0 Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Wed, 9 Aug 2017 15:24:44 +0300 Subject: i2c: designware: Fix standard mode speed when configuring the slave mode Code sets bit DW_IC_CON_SPEED_FAST (0x4) always when configuring the slave mode. This results incorrect register value DW_IC_CON_SPEED_HIGH (0x6) when OR'ed together with DW_IC_CON_SPEED_STD (0x2). Remove this and let the code set the speed mode bits according to clock frequency or default to fast mode. Signed-off-by: Jarkko Nikula Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-designware-platdrv.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c index 441afc7..57248bc 100644 --- a/drivers/i2c/busses/i2c-designware-platdrv.c +++ b/drivers/i2c/busses/i2c-designware-platdrv.c @@ -198,8 +198,7 @@ static void i2c_dw_configure_slave(struct dw_i2c_dev *dev) dev->functionality = I2C_FUNC_SLAVE | DW_IC_DEFAULT_FUNCTIONALITY; dev->slave_cfg = DW_IC_CON_RX_FIFO_FULL_HLD_CTRL | - DW_IC_CON_RESTART_EN | DW_IC_CON_STOP_DET_IFADDRESSED | - DW_IC_CON_SPEED_FAST; + DW_IC_CON_RESTART_EN | DW_IC_CON_STOP_DET_IFADDRESSED; dev->mode = DW_IC_SLAVE; -- cgit v1.1 From f4b17a14faeec4160f97ad75ea7534f571f12404 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Wed, 9 Aug 2017 11:21:28 +0200 Subject: i2c: core: Make comment about I2C table requirement to reflect the code I2C drivers were required to have an I2C device ID table even if were for devices that would only be registered using a specific firmware interface (e.g: OF or ACPI). But commit da10c06a044b ("i2c: Make I2C ID tables non-mandatory for DT'ed devices") changed the I2C core to relax the requirement and allow drivers to avoid defining this table. Unfortunately it only took into account drivers for OF-only devices and forgot about ACPI-only ones, and this was fixed by commit c64ffff7a9d1 ("i2c: core: Allow empty id_table in ACPI case as well"). But the latter didn't update the original comment, so it doesn't reflect what the code does now. Signed-off-by: Javier Martinez Canillas Signed-off-by: Wolfram Sang --- drivers/i2c/i2c-core-base.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/i2c/i2c-core-base.c b/drivers/i2c/i2c-core-base.c index 12822a4..56e4658 100644 --- a/drivers/i2c/i2c-core-base.c +++ b/drivers/i2c/i2c-core-base.c @@ -353,8 +353,8 @@ static int i2c_device_probe(struct device *dev) } /* - * An I2C ID table is not mandatory, if and only if, a suitable Device - * Tree match table entry is supplied for the probing device. + * An I2C ID table is not mandatory, if and only if, a suitable OF + * or ACPI ID table is supplied for the probing device. */ if (!driver->id_table && !i2c_acpi_match_device(dev->driver->acpi_match_table, client) && -- cgit v1.1 From 42543aeb48e3701b49e0a83654248afc38feb88f Mon Sep 17 00:00:00 2001 From: Anton Vasilyev Date: Thu, 10 Aug 2017 18:15:45 +0300 Subject: i2c: simtec: use release_mem_region instead of release_resource Use api pair of request_mem_region and release_mem_region instead of release_resource. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Anton Vasilyev Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-simtec.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/i2c/busses/i2c-simtec.c b/drivers/i2c/busses/i2c-simtec.c index b4685bb9..adca51a 100644 --- a/drivers/i2c/busses/i2c-simtec.c +++ b/drivers/i2c/busses/i2c-simtec.c @@ -127,8 +127,7 @@ static int simtec_i2c_probe(struct platform_device *dev) iounmap(pd->reg); err_res: - release_resource(pd->ioarea); - kfree(pd->ioarea); + release_mem_region(pd->ioarea->start, size); err: kfree(pd); @@ -142,8 +141,7 @@ static int simtec_i2c_remove(struct platform_device *dev) i2c_del_adapter(&pd->adap); iounmap(pd->reg); - release_resource(pd->ioarea); - kfree(pd->ioarea); + release_mem_region(pd->ioarea->start, resource_size(pd->ioarea)); kfree(pd); return 0; -- cgit v1.1 From f1c0b7e448b9e66dd9a7343bab58a3d3a477e104 Mon Sep 17 00:00:00 2001 From: Brendan Higgins Date: Fri, 28 Jul 2017 18:00:12 -0700 Subject: i2c: aspeed: fixed potential null pointer dereference Before I skipped null checks when the master is in the STOP state; this fixes that. Signed-off-by: Brendan Higgins Acked-by: Joel Stanley Signed-off-by: Wolfram Sang Fixes: f327c686d3ba ("i2c: aspeed: added driver for Aspeed I2C") --- drivers/i2c/busses/i2c-aspeed.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/i2c/busses/i2c-aspeed.c b/drivers/i2c/busses/i2c-aspeed.c index f193483..6fdf923 100644 --- a/drivers/i2c/busses/i2c-aspeed.c +++ b/drivers/i2c/busses/i2c-aspeed.c @@ -410,10 +410,11 @@ static bool aspeed_i2c_master_irq(struct aspeed_i2c_bus *bus) } /* We are in an invalid state; reset bus to a known state. */ - if (!bus->msgs && bus->master_state != ASPEED_I2C_MASTER_STOP) { + if (!bus->msgs) { dev_err(bus->dev, "bus in unknown state"); bus->cmd_err = -EIO; - aspeed_i2c_do_stop(bus); + if (bus->master_state != ASPEED_I2C_MASTER_STOP) + aspeed_i2c_do_stop(bus); goto out_no_complete; } msg = &bus->msgs[bus->msgs_index]; -- cgit v1.1 From a0ffc51e20e90e0c1c2491de2b4b03f48b6caaba Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 15 Aug 2017 11:57:06 +0200 Subject: drm/atomic: If the atomic check fails, return its value first The last part of drm_atomic_check_only is testing whether we need to fail with -EINVAL when modeset is not allowed, but forgets to return the value when atomic_check() fails first. This results in -EDEADLK being replaced by -EINVAL, and the sanity check in drm_modeset_drop_locks kicks in: [ 308.531734] ------------[ cut here ]------------ [ 308.531791] WARNING: CPU: 0 PID: 1886 at drivers/gpu/drm/drm_modeset_lock.c:217 drm_modeset_drop_locks+0x33/0xc0 [drm] [ 308.531828] Modules linked in: [ 308.532050] CPU: 0 PID: 1886 Comm: kms_atomic Tainted: G U W 4.13.0-rc5-patser+ #5225 [ 308.532082] Hardware name: NUC5i7RYB, BIOS RYBDWi35.86A.0246.2015.0309.1355 03/09/2015 [ 308.532124] task: ffff8800cd9dae00 task.stack: ffff8800ca3b8000 [ 308.532168] RIP: 0010:drm_modeset_drop_locks+0x33/0xc0 [drm] [ 308.532189] RSP: 0018:ffff8800ca3bf980 EFLAGS: 00010282 [ 308.532211] RAX: dffffc0000000000 RBX: ffff8800ca3bfaf8 RCX: 0000000013a171e6 [ 308.532235] RDX: 1ffff10019477f69 RSI: ffffffffa8ba4fa0 RDI: ffff8800ca3bfb48 [ 308.532258] RBP: ffff8800ca3bf998 R08: 0000000000000000 R09: 0000000000000003 [ 308.532281] R10: 0000000079dbe066 R11: 00000000f760b34b R12: 0000000000000001 [ 308.532304] R13: dffffc0000000000 R14: 00000000ffffffea R15: ffff880096889680 [ 308.532328] FS: 00007ff00959cec0(0000) GS:ffff8800d4e00000(0000) knlGS:0000000000000000 [ 308.532359] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 308.532380] CR2: 0000000000000008 CR3: 00000000ca2e3000 CR4: 00000000003406f0 [ 308.532402] Call Trace: [ 308.532440] drm_mode_atomic_ioctl+0x19fa/0x1c00 [drm] [ 308.532488] ? drm_atomic_set_property+0x1220/0x1220 [drm] [ 308.532565] ? avc_has_extended_perms+0xc39/0xff0 [ 308.532593] ? lock_downgrade+0x610/0x610 [ 308.532640] ? drm_atomic_set_property+0x1220/0x1220 [drm] [ 308.532680] drm_ioctl_kernel+0x154/0x1a0 [drm] [ 308.532755] drm_ioctl+0x624/0x8f0 [drm] [ 308.532858] ? drm_atomic_set_property+0x1220/0x1220 [drm] [ 308.532976] ? drm_getunique+0x210/0x210 [drm] [ 308.533061] do_vfs_ioctl+0xd92/0xe40 [ 308.533121] ? ioctl_preallocate+0x1b0/0x1b0 [ 308.533160] ? selinux_capable+0x20/0x20 [ 308.533191] ? do_fcntl+0x1b1/0xbf0 [ 308.533219] ? kasan_slab_free+0xa2/0xb0 [ 308.533249] ? f_getown+0x4b/0xa0 [ 308.533278] ? putname+0xcf/0xe0 [ 308.533309] ? security_file_ioctl+0x57/0x90 [ 308.533342] SyS_ioctl+0x4e/0x80 [ 308.533374] entry_SYSCALL_64_fastpath+0x18/0xad [ 308.533405] RIP: 0033:0x7ff00779e4d7 [ 308.533431] RSP: 002b:00007fff66a043d8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 308.533481] RAX: ffffffffffffffda RBX: 000000e7c7ca5910 RCX: 00007ff00779e4d7 [ 308.533560] RDX: 00007fff66a04430 RSI: 00000000c03864bc RDI: 0000000000000003 [ 308.533608] RBP: 00007ff007a5fb00 R08: 000000e7c7ca4620 R09: 000000e7c7ca5e60 [ 308.533647] R10: 0000000000000001 R11: 0000000000000246 R12: 0000000000000070 [ 308.533685] R13: 0000000000000000 R14: 0000000000000000 R15: 000000e7c7ca5930 [ 308.533770] Code: ff df 55 48 89 e5 41 55 41 54 53 48 89 fb 48 83 c7 50 48 89 fa 48 c1 ea 03 80 3c 02 00 74 05 e8 94 d4 16 e7 48 83 7b 50 00 74 02 <0f> ff 4c 8d 6b 58 48 b8 00 00 00 00 00 fc ff df 4c 89 ea 48 c1 [ 308.534086] ---[ end trace 77f11e53b1df44ad ]--- Solve this by adding the missing return. This is also a bugfix because we could end up rejecting updates with -EINVAL because of a early -EDEADLK, while if atomic_check ran to completion it might have downgraded the modeset to a fastset. Signed-off-by: Maarten Lankhorst Testcase: kms_atomic Link: https://patchwork.freedesktop.org/patch/msgid/20170815095706.23624-1-maarten.lankhorst@linux.intel.com Fixes: d34f20d6e2f2 ("drm: Atomic modeset ioctl") Cc: # v4.0+ Reviewed-by: Daniel Vetter --- drivers/gpu/drm/drm_atomic.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index b43939f..aed25c4 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -1655,6 +1655,9 @@ int drm_atomic_check_only(struct drm_atomic_state *state) if (config->funcs->atomic_check) ret = config->funcs->atomic_check(state->dev, state); + if (ret) + return ret; + if (!state->allow_modeset) { for_each_new_crtc_in_state(state, crtc, crtc_state, i) { if (drm_atomic_crtc_needs_modeset(crtc_state)) { @@ -1665,7 +1668,7 @@ int drm_atomic_check_only(struct drm_atomic_state *state) } } - return ret; + return 0; } EXPORT_SYMBOL(drm_atomic_check_only); -- cgit v1.1 From 2926a2aa5c14fb2add75e6584845b1c03022235f Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 14 Aug 2017 17:19:26 +0200 Subject: iommu: Fix wrong freeing of iommu_device->dev The struct iommu_device has a 'struct device' embedded into it, not as a pointer, but the whole struct. In the conversion of the iommu drivers to use struct iommu_device it was forgotten that the relase function for that struct device simply calls kfree() on the pointer. This frees memory that was never allocated and causes memory corruption. To fix this issue, use a pointer to struct device instead of embedding the whole struct. This needs some updates in the iommu sysfs code as well as the Intel VT-d and AMD IOMMU driver. Reported-by: Sebastian Ott Fixes: 39ab9555c241 ('iommu: Add sysfs bindings for struct iommu_device') Cc: stable@vger.kernel.org # >= v4.11 Signed-off-by: Joerg Roedel --- drivers/iommu/amd_iommu_types.h | 4 +++- drivers/iommu/intel-iommu.c | 4 +++- drivers/iommu/iommu-sysfs.c | 32 ++++++++++++++++++++------------ include/linux/iommu.h | 12 +++++++++++- 4 files changed, 37 insertions(+), 15 deletions(-) diff --git a/drivers/iommu/amd_iommu_types.h b/drivers/iommu/amd_iommu_types.h index 294a409..d6b873b 100644 --- a/drivers/iommu/amd_iommu_types.h +++ b/drivers/iommu/amd_iommu_types.h @@ -574,7 +574,9 @@ struct amd_iommu { static inline struct amd_iommu *dev_to_amd_iommu(struct device *dev) { - return container_of(dev, struct amd_iommu, iommu.dev); + struct iommu_device *iommu = dev_to_iommu_device(dev); + + return container_of(iommu, struct amd_iommu, iommu); } #define ACPIHID_UID_LEN 256 diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 687f18f..3e8636f 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -4736,7 +4736,9 @@ static void intel_disable_iommus(void) static inline struct intel_iommu *dev_to_intel_iommu(struct device *dev) { - return container_of(dev, struct intel_iommu, iommu.dev); + struct iommu_device *iommu_dev = dev_to_iommu_device(dev); + + return container_of(iommu_dev, struct intel_iommu, iommu); } static ssize_t intel_iommu_show_version(struct device *dev, diff --git a/drivers/iommu/iommu-sysfs.c b/drivers/iommu/iommu-sysfs.c index c58351e..36d1a7c 100644 --- a/drivers/iommu/iommu-sysfs.c +++ b/drivers/iommu/iommu-sysfs.c @@ -62,32 +62,40 @@ int iommu_device_sysfs_add(struct iommu_device *iommu, va_list vargs; int ret; - device_initialize(&iommu->dev); + iommu->dev = kzalloc(sizeof(*iommu->dev), GFP_KERNEL); + if (!iommu->dev) + return -ENOMEM; - iommu->dev.class = &iommu_class; - iommu->dev.parent = parent; - iommu->dev.groups = groups; + device_initialize(iommu->dev); + + iommu->dev->class = &iommu_class; + iommu->dev->parent = parent; + iommu->dev->groups = groups; va_start(vargs, fmt); - ret = kobject_set_name_vargs(&iommu->dev.kobj, fmt, vargs); + ret = kobject_set_name_vargs(&iommu->dev->kobj, fmt, vargs); va_end(vargs); if (ret) goto error; - ret = device_add(&iommu->dev); + ret = device_add(iommu->dev); if (ret) goto error; + dev_set_drvdata(iommu->dev, iommu); + return 0; error: - put_device(&iommu->dev); + put_device(iommu->dev); return ret; } void iommu_device_sysfs_remove(struct iommu_device *iommu) { - device_unregister(&iommu->dev); + dev_set_drvdata(iommu->dev, NULL); + device_unregister(iommu->dev); + iommu->dev = NULL; } /* * IOMMU drivers can indicate a device is managed by a given IOMMU using @@ -102,14 +110,14 @@ int iommu_device_link(struct iommu_device *iommu, struct device *link) if (!iommu || IS_ERR(iommu)) return -ENODEV; - ret = sysfs_add_link_to_group(&iommu->dev.kobj, "devices", + ret = sysfs_add_link_to_group(&iommu->dev->kobj, "devices", &link->kobj, dev_name(link)); if (ret) return ret; - ret = sysfs_create_link_nowarn(&link->kobj, &iommu->dev.kobj, "iommu"); + ret = sysfs_create_link_nowarn(&link->kobj, &iommu->dev->kobj, "iommu"); if (ret) - sysfs_remove_link_from_group(&iommu->dev.kobj, "devices", + sysfs_remove_link_from_group(&iommu->dev->kobj, "devices", dev_name(link)); return ret; @@ -121,5 +129,5 @@ void iommu_device_unlink(struct iommu_device *iommu, struct device *link) return; sysfs_remove_link(&link->kobj, "iommu"); - sysfs_remove_link_from_group(&iommu->dev.kobj, "devices", dev_name(link)); + sysfs_remove_link_from_group(&iommu->dev->kobj, "devices", dev_name(link)); } diff --git a/include/linux/iommu.h b/include/linux/iommu.h index 2cb54ad..176f756 100644 --- a/include/linux/iommu.h +++ b/include/linux/iommu.h @@ -240,7 +240,7 @@ struct iommu_device { struct list_head list; const struct iommu_ops *ops; struct fwnode_handle *fwnode; - struct device dev; + struct device *dev; }; int iommu_device_register(struct iommu_device *iommu); @@ -265,6 +265,11 @@ static inline void iommu_device_set_fwnode(struct iommu_device *iommu, iommu->fwnode = fwnode; } +static inline struct iommu_device *dev_to_iommu_device(struct device *dev) +{ + return (struct iommu_device *)dev_get_drvdata(dev); +} + #define IOMMU_GROUP_NOTIFY_ADD_DEVICE 1 /* Device added */ #define IOMMU_GROUP_NOTIFY_DEL_DEVICE 2 /* Pre Device removed */ #define IOMMU_GROUP_NOTIFY_BIND_DRIVER 3 /* Pre Driver bind */ @@ -589,6 +594,11 @@ static inline void iommu_device_set_fwnode(struct iommu_device *iommu, { } +static inline struct iommu_device *dev_to_iommu_device(struct device *dev) +{ + return NULL; +} + static inline void iommu_device_unregister(struct iommu_device *iommu) { } -- cgit v1.1 From c88c094985ad38c6f71ab5b0d3d68757db14c8b2 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Wed, 16 Aug 2017 07:31:03 +0200 Subject: sata: ahci-da850: Fix some error handling paths in 'ahci_da850_probe()' 'rc' is known to be 0 at this point. If 'platform_get_resource()' or 'devm_ioremap()' fail, return -ENOMEM instead of 0 which means success. tj: Changed error code from -ENOMEM to -ENODEV for get_resource failure as suggested by Sergei. Signed-off-by: Christophe JAILLET Signed-off-by: Tejun Heo Cc: Sergei Shtylyov --- drivers/ata/ahci_da850.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/ata/ahci_da850.c b/drivers/ata/ahci_da850.c index 1a50cd3..9b34dff 100644 --- a/drivers/ata/ahci_da850.c +++ b/drivers/ata/ahci_da850.c @@ -216,12 +216,16 @@ static int ahci_da850_probe(struct platform_device *pdev) return rc; res = platform_get_resource(pdev, IORESOURCE_MEM, 1); - if (!res) + if (!res) { + rc = -ENODEV; goto disable_resources; + } pwrdn_reg = devm_ioremap(dev, res->start, resource_size(res)); - if (!pwrdn_reg) + if (!pwrdn_reg) { + rc = -ENOMEM; goto disable_resources; + } da850_sata_init(dev, pwrdn_reg, hpriv->mmio, mpy); -- cgit v1.1 From 211b7aac5443d347ff7b6dabf54702b40228cfaf Mon Sep 17 00:00:00 2001 From: "Balasubramaniam, Hari Chand" Date: Tue, 15 Aug 2017 10:05:46 +0800 Subject: drm/i915: Initialize 'data' in intel_dsi_dcs_backlight.c variable 'data' may be used uninitialized in this function. thus, 'function dcs_get_backlight' will return unwanted value/fail. Thus, adding NULL initialized to 'data' variable will solve the return failure happening. v2: Change commit message to reflect upstream with proper message Fixes: 90198355b83c ("drm/i915/dsi: Add DCS control for Panel PWM") Cc: Jani Nikula Cc: Daniel Vetter Cc: Yetunde Adebisi Cc: Deepak M Cc: Jani Nikula Signed-off-by: Balasubramaniam, Hari Chand Reviewed-by: Jani Nikula Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/1502762746-191826-1-git-send-email-hari.chand.balasubramaniam@intel.com (cherry picked from commit d59814a5b4852442e1d03c569a4542f8b08356a7) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_dsi_dcs_backlight.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_dsi_dcs_backlight.c b/drivers/gpu/drm/i915/intel_dsi_dcs_backlight.c index 6e09ceb..150a156 100644 --- a/drivers/gpu/drm/i915/intel_dsi_dcs_backlight.c +++ b/drivers/gpu/drm/i915/intel_dsi_dcs_backlight.c @@ -46,7 +46,7 @@ static u32 dcs_get_backlight(struct intel_connector *connector) struct intel_encoder *encoder = connector->encoder; struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); struct mipi_dsi_device *dsi_device; - u8 data; + u8 data = 0; enum port port; /* FIXME: Need to take care of 16 bit brightness level */ -- cgit v1.1 From 7c648bde211baeda7a029bd6be4957e8be48d8c9 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Fri, 11 Aug 2017 14:39:07 +0300 Subject: drm/i915/vbt: ignore extraneous child devices for a port MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ever since we've parsed VBT child devices, starting from 6acab15a7b0d ("drm/i915: use the HDMI DDI buffer translations from VBT"), we've ignored the child device information if more than one child device references the same port. The rationale for this seems lost in time. Since commit 311a20949f04 ("drm/i915: don't init DP or HDMI when not supported by DDI port") we started using this information more to skip HDMI/DP init if the port wasn't there per VBT child devices. However, at the same time it added port defaults without further explanation. Thus, if the child device info was skipped due to multiple child devices referencing the same port, the device info would be retrieved from the somewhat arbitrary defaults. Finally, when commit bb1d132935c2 ("drm/i915/vbt: split out defaults that are set when there is no VBT") stopped initializing the defaults whenever VBT is present, thus trusting the VBT more, we stopped initializing ports which were referenced by more than one child device. Apparently at least Asus UX305UA, UX305U, and UX306U laptops have VBT child device blocks which cause this behaviour. Arguably they were shipped with a broken VBT. Relax the rules for multiple references to the same port, and use the first child device info to reference a port. Retain the logic to debug log about this, though. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101745 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=196233 Fixes: bb1d132935c2 ("drm/i915/vbt: split out defaults that are set when there is no VBT") Tested-by: Oliver Weißbarth Reported-by: Oliver Weißbarth Reported-by: Didier G Reported-by: Giles Anderson Cc: Manasi Navare Cc: Ville Syrjälä Cc: Paulo Zanoni Cc: # v4.12+ Reviewed-by: Ville Syrjälä Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20170811113907.6716-1-jani.nikula@intel.com Signed-off-by: Jani Nikula (cherry picked from commit b5273d72750555a673040070bfb23c454a7cd3ef) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_bios.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index 639d45c..7ea7fd1 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -1120,8 +1120,8 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, bool is_dvi, is_hdmi, is_dp, is_edp, is_crt; uint8_t aux_channel, ddc_pin; /* Each DDI port can have more than one value on the "DVO Port" field, - * so look for all the possible values for each port and abort if more - * than one is found. */ + * so look for all the possible values for each port. + */ int dvo_ports[][3] = { {DVO_PORT_HDMIA, DVO_PORT_DPA, -1}, {DVO_PORT_HDMIB, DVO_PORT_DPB, -1}, @@ -1130,7 +1130,10 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, {DVO_PORT_CRT, DVO_PORT_HDMIE, DVO_PORT_DPE}, }; - /* Find the child device to use, abort if more than one found. */ + /* + * Find the first child device to reference the port, report if more + * than one found. + */ for (i = 0; i < dev_priv->vbt.child_dev_num; i++) { it = dev_priv->vbt.child_dev + i; @@ -1140,11 +1143,11 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, if (it->common.dvo_port == dvo_ports[port][j]) { if (child) { - DRM_DEBUG_KMS("More than one child device for port %c in VBT.\n", + DRM_DEBUG_KMS("More than one child device for port %c in VBT, using the first.\n", port_name(port)); - return; + } else { + child = it; } - child = it; } } } -- cgit v1.1 From 31c1a7b8f966470ce136710a95afcf5822fecef8 Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Tue, 15 Aug 2017 20:04:03 -0700 Subject: drm/i915/cnl: Fix LSPCON support. When LSPCON support was extended to CNL one part was missed on lspcon_init. So, instead of adding check per platform on lspcon_init let's use HAS_LSPCON that is already there for that purpose. Fixes: ff15947e0f02 ("drm/i915/cnl: LSPCON support is gen9+") Cc: Shashank Sharma Cc: Paulo Zanoni Signed-off-by: Rodrigo Vivi Reviewed-by: Jani Nikula Reviewed-by: Shashank Sharma Link: https://patchwork.freedesktop.org/patch/msgid/20170816030403.11368-1-rodrigo.vivi@intel.com (cherry picked from commit acf58d4e965d40fc014252292b0911b4c9fe6697) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_lspcon.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c index 5abef482..beb9baa 100644 --- a/drivers/gpu/drm/i915/intel_lspcon.c +++ b/drivers/gpu/drm/i915/intel_lspcon.c @@ -210,8 +210,8 @@ bool lspcon_init(struct intel_digital_port *intel_dig_port) struct drm_device *dev = intel_dig_port->base.base.dev; struct drm_i915_private *dev_priv = to_i915(dev); - if (!IS_GEN9(dev_priv)) { - DRM_ERROR("LSPCON is supported on GEN9 only\n"); + if (!HAS_LSPCON(dev_priv)) { + DRM_ERROR("LSPCON is not supported on this platform\n"); return false; } -- cgit v1.1 From 47f078339be902e97d0ad828ca1d614a5a95334b Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 16 Aug 2017 18:21:40 +0100 Subject: Revert "staging: fsl-mc: be consistent when checking strcmp() return" The previous fix removed the equal to zero comparisons by the strcmps and now the function always returns true. Revert this change to restore the original correctly functioning code. Detected by CoverityScan, CID#1452267 ("Constant expression result") This reverts commit b93ad9a067e1515af42da7d56bc61f1a25075f94. Fixes: b93ad9a067e1 ("staging: fsl-mc: be consistent when checking strcmp() return") Signed-off-by: Colin Ian King Signed-off-by: Greg Kroah-Hartman --- drivers/staging/fsl-mc/bus/fsl-mc-allocator.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/staging/fsl-mc/bus/fsl-mc-allocator.c b/drivers/staging/fsl-mc/bus/fsl-mc-allocator.c index b37a6f4..8ea3920 100644 --- a/drivers/staging/fsl-mc/bus/fsl-mc-allocator.c +++ b/drivers/staging/fsl-mc/bus/fsl-mc-allocator.c @@ -16,9 +16,9 @@ static bool __must_check fsl_mc_is_allocatable(const char *obj_type) { - return strcmp(obj_type, "dpbp") || - strcmp(obj_type, "dpmcp") || - strcmp(obj_type, "dpcon"); + return strcmp(obj_type, "dpbp") == 0 || + strcmp(obj_type, "dpmcp") == 0 || + strcmp(obj_type, "dpcon") == 0; } /** -- cgit v1.1 From 61f0c3c7a0fd37f252b0f23c237973ddf4c21ea7 Mon Sep 17 00:00:00 2001 From: weiping zhang Date: Tue, 8 Aug 2017 01:26:57 +0800 Subject: scsi: megaraid_sas: fix error handle in megasas_probe_one megasas_mgmt_info.max_index has increased by 1 before megasas_io_attach, if megasas_io_attach return error, then goto fail_io_attach, megasas_mgmt_info.instance has a wrong index here. So first reduce max_index and then set that instance to NULL. Signed-off-by: weiping zhang Acked-by: Sumit Saxena Signed-off-by: Martin K. Petersen --- drivers/scsi/megaraid/megaraid_sas_base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c index 316c3df..71c4746 100644 --- a/drivers/scsi/megaraid/megaraid_sas_base.c +++ b/drivers/scsi/megaraid/megaraid_sas_base.c @@ -6228,8 +6228,8 @@ static int megasas_probe_one(struct pci_dev *pdev, fail_start_aen: fail_io_attach: megasas_mgmt_info.count--; - megasas_mgmt_info.instance[megasas_mgmt_info.max_index] = NULL; megasas_mgmt_info.max_index--; + megasas_mgmt_info.instance[megasas_mgmt_info.max_index] = NULL; instance->instancet->disable_intr(instance); megasas_destroy_irqs(instance); -- cgit v1.1 From 82f0fd06d4a84ff32cc2cf9503764219316ec728 Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Mon, 31 Jul 2017 22:02:55 +0530 Subject: scsi: csiostor: fail probe if fw does not support FCoE Fail probe if FCoE capability is not enabled in the firmware. Signed-off-by: Varun Prakash Signed-off-by: Martin K. Petersen --- drivers/scsi/csiostor/csio_hw.c | 4 +++- drivers/scsi/csiostor/csio_init.c | 12 ++++++++---- 2 files changed, 11 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/csiostor/csio_hw.c b/drivers/scsi/csiostor/csio_hw.c index 2029ad2..5be0086 100644 --- a/drivers/scsi/csiostor/csio_hw.c +++ b/drivers/scsi/csiostor/csio_hw.c @@ -3845,8 +3845,10 @@ csio_hw_start(struct csio_hw *hw) if (csio_is_hw_ready(hw)) return 0; - else + else if (csio_match_state(hw, csio_hws_uninit)) return -EINVAL; + else + return -ENODEV; } int diff --git a/drivers/scsi/csiostor/csio_init.c b/drivers/scsi/csiostor/csio_init.c index ea0c310..dcd0741 100644 --- a/drivers/scsi/csiostor/csio_init.c +++ b/drivers/scsi/csiostor/csio_init.c @@ -969,10 +969,14 @@ static int csio_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) pci_set_drvdata(pdev, hw); - if (csio_hw_start(hw) != 0) { - dev_err(&pdev->dev, - "Failed to start FW, continuing in debug mode.\n"); - return 0; + rv = csio_hw_start(hw); + if (rv) { + if (rv == -EINVAL) { + dev_err(&pdev->dev, + "Failed to start FW, continuing in debug mode.\n"); + return 0; + } + goto err_lnode_exit; } sprintf(hw->fwrev_str, "%u.%u.%u.%u\n", -- cgit v1.1 From c80267324938a5517fd31fa4bbd2d63c564401f9 Mon Sep 17 00:00:00 2001 From: Raghava Aditya Renukunta Date: Fri, 4 Aug 2017 03:51:41 -0700 Subject: scsi: aacraid: Fix out of bounds in aac_get_name_resp We terminate the aac_get_name_resp on a byte that is outside the bounds of the structure. Extend the return response by one byte to remove the out of bounds reference. Fixes: b836439faf04 ("aacraid: 4KB sector support") Reported-by: Dan Carpenter Signed-off-by: David Carroll Signed-off-by: Raghava Aditya Renukunta Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/aacraid/aachba.c | 9 +++++++-- drivers/scsi/aacraid/aacraid.h | 2 +- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/aacraid/aachba.c b/drivers/scsi/aacraid/aachba.c index 4591113..a1a2c71 100644 --- a/drivers/scsi/aacraid/aachba.c +++ b/drivers/scsi/aacraid/aachba.c @@ -549,7 +549,9 @@ static void get_container_name_callback(void *context, struct fib * fibptr) if ((le32_to_cpu(get_name_reply->status) == CT_OK) && (get_name_reply->data[0] != '\0')) { char *sp = get_name_reply->data; - sp[sizeof(((struct aac_get_name_resp *)NULL)->data)] = '\0'; + int data_size = FIELD_SIZEOF(struct aac_get_name_resp, data); + + sp[data_size - 1] = '\0'; while (*sp == ' ') ++sp; if (*sp) { @@ -579,12 +581,15 @@ static void get_container_name_callback(void *context, struct fib * fibptr) static int aac_get_container_name(struct scsi_cmnd * scsicmd) { int status; + int data_size; struct aac_get_name *dinfo; struct fib * cmd_fibcontext; struct aac_dev * dev; dev = (struct aac_dev *)scsicmd->device->host->hostdata; + data_size = FIELD_SIZEOF(struct aac_get_name_resp, data); + cmd_fibcontext = aac_fib_alloc_tag(dev, scsicmd); aac_fib_init(cmd_fibcontext); @@ -593,7 +598,7 @@ static int aac_get_container_name(struct scsi_cmnd * scsicmd) dinfo->command = cpu_to_le32(VM_ContainerConfig); dinfo->type = cpu_to_le32(CT_READ_NAME); dinfo->cid = cpu_to_le32(scmd_id(scsicmd)); - dinfo->count = cpu_to_le32(sizeof(((struct aac_get_name_resp *)NULL)->data)); + dinfo->count = cpu_to_le32(data_size - 1); status = aac_fib_send(ContainerCommand, cmd_fibcontext, diff --git a/drivers/scsi/aacraid/aacraid.h b/drivers/scsi/aacraid/aacraid.h index d31a9bc..ee2667e 100644 --- a/drivers/scsi/aacraid/aacraid.h +++ b/drivers/scsi/aacraid/aacraid.h @@ -2274,7 +2274,7 @@ struct aac_get_name_resp { __le32 parm3; __le32 parm4; __le32 parm5; - u8 data[16]; + u8 data[17]; }; #define CT_CID_TO_32BITS_UID 165 -- cgit v1.1 From 70e42fd02c46e2aa9ab07b766d418637e3a51de7 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Wed, 9 Aug 2017 12:00:22 +0900 Subject: scsi: sd_zbc: Write unlock zone from sd_uninit_cmnd() Releasing a zone write lock only when the write commnand that acquired the lock completes can cause deadlocks due to potential command reordering if the lock owning request is requeued and not executed. This problem exists only with the scsi-mq path as, unlike the legacy path, requests are moved out of the dispatch queue before being prepared and so before locking a zone for a write command. Since sd_uninit_cmnd() is now always called when a request is requeued, call sd_zbc_write_unlock_zone() from that function for write requests that acquired a zone lock instead of from sd_done(). Acquisition of a zone lock by a write command is indicated using the new command flag SCMD_ZONE_WRITE_LOCK. Signed-off-by: Damien Le Moal Reviewed-by: Christoph Hellwig Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/sd.c | 3 +++ drivers/scsi/sd_zbc.c | 9 +++++---- include/scsi/scsi_cmnd.h | 1 + 3 files changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index bea36ad..e2647f2 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -1277,6 +1277,9 @@ static void sd_uninit_command(struct scsi_cmnd *SCpnt) { struct request *rq = SCpnt->request; + if (SCpnt->flags & SCMD_ZONE_WRITE_LOCK) + sd_zbc_write_unlock_zone(SCpnt); + if (rq->rq_flags & RQF_SPECIAL_PAYLOAD) __free_page(rq->special_vec.bv_page); diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c index 96855df..8aa5477 100644 --- a/drivers/scsi/sd_zbc.c +++ b/drivers/scsi/sd_zbc.c @@ -294,6 +294,9 @@ int sd_zbc_write_lock_zone(struct scsi_cmnd *cmd) test_and_set_bit(zno, sdkp->zones_wlock)) return BLKPREP_DEFER; + WARN_ON_ONCE(cmd->flags & SCMD_ZONE_WRITE_LOCK); + cmd->flags |= SCMD_ZONE_WRITE_LOCK; + return BLKPREP_OK; } @@ -302,9 +305,10 @@ void sd_zbc_write_unlock_zone(struct scsi_cmnd *cmd) struct request *rq = cmd->request; struct scsi_disk *sdkp = scsi_disk(rq->rq_disk); - if (sdkp->zones_wlock) { + if (sdkp->zones_wlock && cmd->flags & SCMD_ZONE_WRITE_LOCK) { unsigned int zno = sd_zbc_zone_no(sdkp, blk_rq_pos(rq)); WARN_ON_ONCE(!test_bit(zno, sdkp->zones_wlock)); + cmd->flags &= ~SCMD_ZONE_WRITE_LOCK; clear_bit_unlock(zno, sdkp->zones_wlock); smp_mb__after_atomic(); } @@ -335,9 +339,6 @@ void sd_zbc_complete(struct scsi_cmnd *cmd, case REQ_OP_WRITE_ZEROES: case REQ_OP_WRITE_SAME: - /* Unlock the zone */ - sd_zbc_write_unlock_zone(cmd); - if (result && sshdr->sense_key == ILLEGAL_REQUEST && sshdr->asc == 0x21) diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h index a1266d3..6af198d 100644 --- a/include/scsi/scsi_cmnd.h +++ b/include/scsi/scsi_cmnd.h @@ -57,6 +57,7 @@ struct scsi_pointer { /* for scmd->flags */ #define SCMD_TAGGED (1 << 0) #define SCMD_UNCHECKED_ISA_DMA (1 << 1) +#define SCMD_ZONE_WRITE_LOCK (1 << 2) struct scsi_cmnd { struct scsi_request req; -- cgit v1.1 From cbe7dfa26eee4819db7b5846181d56fd0cece0ea Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Sun, 13 Aug 2017 19:44:22 +0200 Subject: Revert "scsi: default to scsi-mq" Defaulting to scsi-mq in 4.13-rc has shown various regressions on setups that we didn't previously consider. Fixes for them are in progress, but too invasive to make it in this cycle. So for now revert the commit that defaults to blk-mq for SCSI. For 4.14 we'll plan to try again with these fixes. This reverts commit 5c279bd9e40624f4ab6e688671026d6005b066fa. Signed-off-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/Kconfig | 11 +++++++++++ drivers/scsi/scsi.c | 4 ++++ 2 files changed, 15 insertions(+) diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index f4538d7..d145e0d 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -47,6 +47,17 @@ config SCSI_NETLINK default n depends on NET +config SCSI_MQ_DEFAULT + bool "SCSI: use blk-mq I/O path by default" + depends on SCSI + ---help--- + This option enables the new blk-mq based I/O path for SCSI + devices by default. With the option the scsi_mod.use_blk_mq + module/boot option defaults to Y, without it to N, but it can + still be overridden either way. + + If unsure say N. + config SCSI_PROC_FS bool "legacy /proc/scsi/ support" depends on SCSI && PROC_FS diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 3d38c6d..1bf274e 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -800,7 +800,11 @@ MODULE_LICENSE("GPL"); module_param(scsi_logging_level, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(scsi_logging_level, "a bit mask of logging levels"); +#ifdef CONFIG_SCSI_MQ_DEFAULT bool scsi_use_blk_mq = true; +#else +bool scsi_use_blk_mq = false; +#endif module_param_named(use_blk_mq, scsi_use_blk_mq, bool, S_IWUSR | S_IRUGO); static int __init init_scsi(void) -- cgit v1.1 From 71eb2ac58aa52f7948deda4d9a75754b6898e5c9 Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Wed, 16 Aug 2017 19:40:32 +0530 Subject: scsi: cxgb4i: call neigh_event_send() to update MAC address If nud_state is not valid then call neigh_event_send() to update MAC address. Signed-off-by: Varun Prakash Signed-off-by: Martin K. Petersen --- drivers/scsi/cxgbi/cxgb4i/cxgb4i.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index a69a9ac..1d02cf9 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -1635,6 +1635,9 @@ static int init_act_open(struct cxgbi_sock *csk) goto rel_resource; } + if (!(n->nud_state & NUD_VALID)) + neigh_event_send(n, NULL); + csk->atid = cxgb4_alloc_atid(lldi->tids, csk); if (csk->atid < 0) { pr_err("%s, NO atid available.\n", ndev->name); -- cgit v1.1 From 733f6563979d96dec180c350abb8ac67cc0367ba Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Tue, 15 Aug 2017 17:34:44 +0300 Subject: i2c: designware: Remove needless pm_runtime_put_noidle() call I guess pm_runtime_put_noidle() call in i2c_dw_probe_slave() was copied by accident from similar master mode adapter registration code. It is unbalanced due missing pm_runtime_get_noresume() but harmless since it doesn't decrease dev->power.usage_count below zero. In theory we can hit similar needless runtime suspend/resume cycle during slave mode adapter registration that was happening when registering the master mode adapter. See commit cd998ded5c12 ("i2c: designware: Prevent runtime suspend during adapter registration"). However, since we are slave, we can consider it as a wrong configuration if we have other slaves attached under this adapter and can omit the pm_runtime_get_noresume()/pm_runtime_put_noidle() calls for simplicity. Signed-off-by: Jarkko Nikula Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-designware-slave.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-designware-slave.c b/drivers/i2c/busses/i2c-designware-slave.c index 4b62a38..25fa339 100644 --- a/drivers/i2c/busses/i2c-designware-slave.c +++ b/drivers/i2c/busses/i2c-designware-slave.c @@ -382,7 +382,6 @@ int i2c_dw_probe_slave(struct dw_i2c_dev *dev) ret = i2c_add_numbered_adapter(adap); if (ret) dev_err(dev->dev, "failure adding adapter: %d\n", ret); - pm_runtime_put_noidle(dev->dev); return ret; } -- cgit v1.1 From 2a86cdd2e7d3c75580f41f89f9b9211e225573cc Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Tue, 15 Aug 2017 17:34:45 +0300 Subject: i2c: designware: Fix runtime PM for I2C slave mode I2C slave controller must be powered and active all the time when I2C slave backend is registered in order to let master address and communicate with us. Now if the controller is runtime PM capable it will be suspended after probe and cannot ever respond to the master or generate interrupts. Fix this by resuming the controller when I2C slave backend is registered and let it suspend after unregistering. Signed-off-by: Jarkko Nikula Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-designware-slave.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/i2c/busses/i2c-designware-slave.c b/drivers/i2c/busses/i2c-designware-slave.c index 25fa339..78d8fb7 100644 --- a/drivers/i2c/busses/i2c-designware-slave.c +++ b/drivers/i2c/busses/i2c-designware-slave.c @@ -177,6 +177,8 @@ static int i2c_dw_reg_slave(struct i2c_client *slave) return -EBUSY; if (slave->flags & I2C_CLIENT_TEN) return -EAFNOSUPPORT; + pm_runtime_get_sync(dev->dev); + /* * Set slave address in the IC_SAR register, * the address to which the DW_apb_i2c responds. @@ -205,6 +207,7 @@ static int i2c_dw_unreg_slave(struct i2c_client *slave) dev->disable_int(dev); dev->disable(dev); dev->slave = NULL; + pm_runtime_put(dev->dev); return 0; } -- cgit v1.1 From 98529b9272e06a7767034fb8a32e43cdecda240a Mon Sep 17 00:00:00 2001 From: Lv Zheng Date: Wed, 16 Aug 2017 15:29:49 +0800 Subject: ACPI: EC: Fix regression related to wrong ECDT initialization order Commit 2a5708409e4e (ACPI / EC: Fix a gap that ECDT EC cannot handle EC events) introduced acpi_ec_ecdt_start(), but that function is invoked before acpi_ec_query_init(), which is too early. This causes the kernel to crash if an EC event occurs after boot, when ec_query_wq is not valid: BUG: unable to handle kernel NULL pointer dereference at 0000000000000102 ... Workqueue: events acpi_ec_event_handler task: ffff9f539790dac0 task.stack: ffffb437c0e10000 RIP: 0010:__queue_work+0x32/0x430 Normally, the DSDT EC should always be valid, so acpi_ec_ecdt_start() is actually a no-op in the majority of cases. However, commit c712bb58d827 (ACPI / EC: Add support to skip boot stage DSDT probe) caused the probing of the DSDT EC as the "boot EC" to be skipped when the ECDT EC is valid and uncovered the bug. Fix this issue by invoking acpi_ec_ecdt_start() after acpi_ec_query_init() in acpi_ec_init(). Link: https://jira01.devtools.intel.com/browse/LCK-4348 Fixes: 2a5708409e4e (ACPI / EC: Fix a gap that ECDT EC cannot handle EC events) Fixes: c712bb58d827 (ACPI / EC: Add support to skip boot stage DSDT probe) Reported-by: Wang Wendy Tested-by: Feng Chenzhou Signed-off-by: Lv Zheng [ rjw: Changelog ] Signed-off-by: Rafael J. Wysocki --- drivers/acpi/ec.c | 17 +++++++---------- drivers/acpi/internal.h | 1 - drivers/acpi/scan.c | 1 - 3 files changed, 7 insertions(+), 12 deletions(-) diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 62068a5..ae3d6d1 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -1741,7 +1741,7 @@ error: * functioning ECDT EC first in order to handle the events. * https://bugzilla.kernel.org/show_bug.cgi?id=115021 */ -int __init acpi_ec_ecdt_start(void) +static int __init acpi_ec_ecdt_start(void) { acpi_handle handle; @@ -2003,20 +2003,17 @@ static inline void acpi_ec_query_exit(void) int __init acpi_ec_init(void) { int result; + int ecdt_fail, dsdt_fail; /* register workqueue for _Qxx evaluations */ result = acpi_ec_query_init(); if (result) - goto err_exit; - /* Now register the driver for the EC */ - result = acpi_bus_register_driver(&acpi_ec_driver); - if (result) - goto err_exit; + return result; -err_exit: - if (result) - acpi_ec_query_exit(); - return result; + /* Drivers must be started after acpi_ec_query_init() */ + ecdt_fail = acpi_ec_ecdt_start(); + dsdt_fail = acpi_bus_register_driver(&acpi_ec_driver); + return ecdt_fail && dsdt_fail ? -ENODEV : 0; } /* EC driver currently not unloadable */ diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h index 58dd7ab..3f5af4d 100644 --- a/drivers/acpi/internal.h +++ b/drivers/acpi/internal.h @@ -185,7 +185,6 @@ typedef int (*acpi_ec_query_func) (void *data); int acpi_ec_init(void); int acpi_ec_ecdt_probe(void); int acpi_ec_dsdt_probe(void); -int acpi_ec_ecdt_start(void); void acpi_ec_block_transactions(void); void acpi_ec_unblock_transactions(void); int acpi_ec_add_query_handler(struct acpi_ec *ec, u8 query_bit, diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 3389729..70fd550 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -2084,7 +2084,6 @@ int __init acpi_scan_init(void) acpi_gpe_apply_masked_gpes(); acpi_update_all_gpes(); - acpi_ec_ecdt_start(); acpi_scan_initialized = true; -- cgit v1.1 From 1a92a80ad386a1a6e3b36d576d52a1a456394b70 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 24 Jul 2017 14:28:00 +1000 Subject: powerpc/mm: Ensure cpumask update is ordered There is no guarantee that the various isync's involved with the context switch will order the update of the CPU mask with the first TLB entry for the new context being loaded by the HW. Be safe here and add a memory barrier to order any subsequent load/store which may bring entries into the TLB. The corresponding barrier on the other side already exists as pte updates use pte_xchg() which uses __cmpxchg_u64 which has a sync after the atomic operation. Cc: stable@vger.kernel.org Signed-off-by: Benjamin Herrenschmidt Reviewed-by: Nicholas Piggin [mpe: Add comments in the code] Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/mmu_context.h | 18 ++++++++++++++++++ arch/powerpc/include/asm/pgtable-be-types.h | 1 + arch/powerpc/include/asm/pgtable-types.h | 1 + 3 files changed, 20 insertions(+) diff --git a/arch/powerpc/include/asm/mmu_context.h b/arch/powerpc/include/asm/mmu_context.h index 0c76675..35bec1c 100644 --- a/arch/powerpc/include/asm/mmu_context.h +++ b/arch/powerpc/include/asm/mmu_context.h @@ -90,6 +90,24 @@ static inline void switch_mm_irqs_off(struct mm_struct *prev, /* Mark this context has been used on the new CPU */ if (!cpumask_test_cpu(smp_processor_id(), mm_cpumask(next))) { cpumask_set_cpu(smp_processor_id(), mm_cpumask(next)); + + /* + * This full barrier orders the store to the cpumask above vs + * a subsequent operation which allows this CPU to begin loading + * translations for next. + * + * When using the radix MMU that operation is the load of the + * MMU context id, which is then moved to SPRN_PID. + * + * For the hash MMU it is either the first load from slb_cache + * in switch_slb(), and/or the store of paca->mm_ctx_id in + * copy_mm_to_paca(). + * + * On the read side the barrier is in pte_xchg(), which orders + * the store to the PTE vs the load of mm_cpumask. + */ + smp_mb(); + new_on_cpu = true; } diff --git a/arch/powerpc/include/asm/pgtable-be-types.h b/arch/powerpc/include/asm/pgtable-be-types.h index 9c0f5db..67e7e3d9 100644 --- a/arch/powerpc/include/asm/pgtable-be-types.h +++ b/arch/powerpc/include/asm/pgtable-be-types.h @@ -87,6 +87,7 @@ static inline bool pte_xchg(pte_t *ptep, pte_t old, pte_t new) unsigned long *p = (unsigned long *)ptep; __be64 prev; + /* See comment in switch_mm_irqs_off() */ prev = (__force __be64)__cmpxchg_u64(p, (__force unsigned long)pte_raw(old), (__force unsigned long)pte_raw(new)); diff --git a/arch/powerpc/include/asm/pgtable-types.h b/arch/powerpc/include/asm/pgtable-types.h index 8bd3b13..369a164 100644 --- a/arch/powerpc/include/asm/pgtable-types.h +++ b/arch/powerpc/include/asm/pgtable-types.h @@ -62,6 +62,7 @@ static inline bool pte_xchg(pte_t *ptep, pte_t old, pte_t new) { unsigned long *p = (unsigned long *)ptep; + /* See comment in switch_mm_irqs_off() */ return pte_val(old) == __cmpxchg_u64(p, pte_val(old), pte_val(new)); } #endif -- cgit v1.1 From f299aec6ebd747298e35934cff7709c6b119ca52 Mon Sep 17 00:00:00 2001 From: Charles Milette Date: Fri, 18 Aug 2017 16:30:34 -0400 Subject: staging: rtl8188eu: add RNX-N150NUB support Add support for USB Device Rosewill RNX-N150NUB. VendorID: 0x0bda, ProductID: 0xffef Signed-off-by: Charles Milette Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/staging/rtl8188eu/os_dep/usb_intf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/staging/rtl8188eu/os_dep/usb_intf.c b/drivers/staging/rtl8188eu/os_dep/usb_intf.c index d283341..56cd4e5 100644 --- a/drivers/staging/rtl8188eu/os_dep/usb_intf.c +++ b/drivers/staging/rtl8188eu/os_dep/usb_intf.c @@ -45,6 +45,7 @@ static struct usb_device_id rtw_usb_id_tbl[] = { {USB_DEVICE(0x2001, 0x3311)}, /* DLink GO-USB-N150 REV B1 */ {USB_DEVICE(0x2357, 0x010c)}, /* TP-Link TL-WN722N v2 */ {USB_DEVICE(0x0df6, 0x0076)}, /* Sitecom N150 v2 */ + {USB_DEVICE(USB_VENDER_ID_REALTEK, 0xffef)}, /* Rosewill RNX-N150NUB */ {} /* Terminating entry */ }; -- cgit v1.1 From 1d2226e45040ed4aee95b633cbd64702bf7fc2a1 Mon Sep 17 00:00:00 2001 From: KT Liao Date: Fri, 18 Aug 2017 16:58:15 -0700 Subject: Input: elan_i2c - add ELAN0602 ACPI ID to support Lenovo Yoga310 Add ELAN0602 to the list of known ACPI IDs to enable support for ELAN touchpads found in Lenovo Yoga310. Signed-off-by: KT Liao Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index 714cf7f..cfbc8ba 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -1247,6 +1247,7 @@ static const struct acpi_device_id elan_acpi_id[] = { { "ELAN0000", 0 }, { "ELAN0100", 0 }, { "ELAN0600", 0 }, + { "ELAN0602", 0 }, { "ELAN0605", 0 }, { "ELAN0608", 0 }, { "ELAN0605", 0 }, -- cgit v1.1 From ec667683c532c93fb41e100e5d61a518971060e2 Mon Sep 17 00:00:00 2001 From: Aaron Ma Date: Fri, 18 Aug 2017 12:17:21 -0700 Subject: Input: trackpoint - add new trackpoint firmware ID Synaptics add new TP firmware ID: 0x2 and 0x3, for now both lower 2 bits are indicated as TP. Change the constant to bitwise values. This makes trackpoint to be recognized on Lenovo Carbon X1 Gen5 instead of it being identified as "PS/2 Generic Mouse". Signed-off-by: Aaron Ma Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/trackpoint.c | 3 ++- drivers/input/mouse/trackpoint.h | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/input/mouse/trackpoint.c b/drivers/input/mouse/trackpoint.c index 20b5b21..0871010 100644 --- a/drivers/input/mouse/trackpoint.c +++ b/drivers/input/mouse/trackpoint.c @@ -265,7 +265,8 @@ static int trackpoint_start_protocol(struct psmouse *psmouse, unsigned char *fir if (ps2_command(&psmouse->ps2dev, param, MAKE_PS2_CMD(0, 2, TP_READ_ID))) return -1; - if (param[0] != TP_MAGIC_IDENT) + /* add new TP ID. */ + if (!(param[0] & TP_MAGIC_IDENT)) return -1; if (firmware_id) diff --git a/drivers/input/mouse/trackpoint.h b/drivers/input/mouse/trackpoint.h index 5617ed3..8805575 100644 --- a/drivers/input/mouse/trackpoint.h +++ b/drivers/input/mouse/trackpoint.h @@ -21,8 +21,9 @@ #define TP_COMMAND 0xE2 /* Commands start with this */ #define TP_READ_ID 0xE1 /* Sent for device identification */ -#define TP_MAGIC_IDENT 0x01 /* Sent after a TP_READ_ID followed */ +#define TP_MAGIC_IDENT 0x03 /* Sent after a TP_READ_ID followed */ /* by the firmware ID */ + /* Firmware ID includes 0x1, 0x2, 0x3 */ /* -- cgit v1.1 From 0c264af7be2013266c5b4c644f3f366399ee490a Mon Sep 17 00:00:00 2001 From: Takashi Sakamoto Date: Sun, 20 Aug 2017 15:54:26 +0900 Subject: ALSA: firewire: fix NULL pointer dereference when releasing uninitialized data of iso-resource When calling 'iso_resource_free()' for uninitialized data, this function causes NULL pointer dereference due to its 'unit' member. This occurs when unplugging audio and music units on IEEE 1394 bus at failure of card registration. This commit fixes the bug. The bug exists since kernel v4.5. Fixes: 324540c4e05c ('ALSA: fireface: postpone sound card registration') at v4.12 Fixes: 8865a31e0fd8 ('ALSA: firewire-motu: postpone sound card registration') at v4.12 Fixes: b610386c8afb ('ALSA: firewire-tascam: deleyed registration of sound card') at v4.7 Fixes: 86c8dd7f4da3 ('ALSA: firewire-digi00x: delayed registration of sound card') at v4.7 Fixes: 6c29230e2a5f ('ALSA: oxfw: delayed registration of sound card') at v4.7 Fixes: 7d3c1d5901aa ('ALSA: fireworks: delayed registration of sound card') at v4.7 Fixes: 04a2c73c97eb ('ALSA: bebob: delayed registration of sound card') at v4.7 Fixes: b59fb1900b4f ('ALSA: dice: postpone card registration') at v4.5 Cc: # v4.5+ Signed-off-by: Takashi Sakamoto Signed-off-by: Takashi Iwai --- sound/firewire/iso-resources.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/sound/firewire/iso-resources.c b/sound/firewire/iso-resources.c index f0e4d50..066b5df 100644 --- a/sound/firewire/iso-resources.c +++ b/sound/firewire/iso-resources.c @@ -210,9 +210,14 @@ EXPORT_SYMBOL(fw_iso_resources_update); */ void fw_iso_resources_free(struct fw_iso_resources *r) { - struct fw_card *card = fw_parent_device(r->unit)->card; + struct fw_card *card; int bandwidth, channel; + /* Not initialized. */ + if (r->unit == NULL) + return; + card = fw_parent_device(r->unit)->card; + mutex_lock(&r->mutex); if (r->allocated) { -- cgit v1.1 From dbd7396b4f24e0c3284fcc05f5def24f52c09884 Mon Sep 17 00:00:00 2001 From: Takashi Sakamoto Date: Sun, 20 Aug 2017 15:55:02 +0900 Subject: ALSA: firewire-motu: destroy stream data surely at failure of card initialization When failing sound card registration after initializing stream data, this module leaves allocated data in stream data. This commit fixes the bug. Fixes: 9b2bb4f2f4a2 ('ALSA: firewire-motu: add stream management functionality') Cc: # v4.12+ Signed-off-by: Takashi Sakamoto Signed-off-by: Takashi Iwai --- sound/firewire/motu/motu.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/firewire/motu/motu.c b/sound/firewire/motu/motu.c index bf779cf..59a2704 100644 --- a/sound/firewire/motu/motu.c +++ b/sound/firewire/motu/motu.c @@ -128,6 +128,7 @@ static void do_registration(struct work_struct *work) return; error: snd_motu_transaction_unregister(motu); + snd_motu_stream_destroy_duplex(motu); snd_card_free(motu->card); dev_info(&motu->unit->device, "Sound card registration failed: %d\n", err); -- cgit v1.1 From f00fd7ae4f409abb7b2e5d099248832548199f0c Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Sun, 30 Jul 2017 16:14:42 -0600 Subject: PATCH] iio: Fix some documentation warnings The kerneldoc description for the trig_readonly field of struct iio_dev lacked a colon, leading to this doc build warning: ./include/linux/iio/iio.h:603: warning: No description found for parameter 'trig_readonly' A similar issue for iio_trigger_set_immutable() in trigger.h yielded: ./include/linux/iio/trigger.h:151: warning: No description found for parameter 'indio_dev' ./include/linux/iio/trigger.h:151: warning: No description found for parameter 'trig' Fix the formatting and silence the warnings. Signed-off-by: Jonathan Corbet Signed-off-by: Jonathan Cameron --- include/linux/iio/iio.h | 2 +- include/linux/iio/trigger.h | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/linux/iio/iio.h b/include/linux/iio/iio.h index d68bec2..c380daa 100644 --- a/include/linux/iio/iio.h +++ b/include/linux/iio/iio.h @@ -535,7 +535,7 @@ struct iio_buffer_setup_ops { * @scan_timestamp: [INTERN] set if any buffers have requested timestamp * @scan_index_timestamp:[INTERN] cache of the index to the timestamp * @trig: [INTERN] current device trigger (buffer modes) - * @trig_readonly [INTERN] mark the current trigger immutable + * @trig_readonly: [INTERN] mark the current trigger immutable * @pollfunc: [DRIVER] function run on trigger being received * @pollfunc_event: [DRIVER] function run on events trigger being received * @channels: [DRIVER] channel specification structure table diff --git a/include/linux/iio/trigger.h b/include/linux/iio/trigger.h index ea08302..7142d8d 100644 --- a/include/linux/iio/trigger.h +++ b/include/linux/iio/trigger.h @@ -144,8 +144,8 @@ void devm_iio_trigger_unregister(struct device *dev, /** * iio_trigger_set_immutable() - set an immutable trigger on destination * - * @indio_dev - IIO device structure containing the device - * @trig - trigger to assign to device + * @indio_dev: IIO device structure containing the device + * @trig: trigger to assign to device * **/ int iio_trigger_set_immutable(struct iio_dev *indio_dev, struct iio_trigger *trig); -- cgit v1.1 From fdd0d32eb95f135041236a6885d9006315aa9a1d Mon Sep 17 00:00:00 2001 From: Dragos Bogdan Date: Fri, 4 Aug 2017 01:37:27 +0300 Subject: iio: imu: adis16480: Fix acceleration scale factor for adis16480 According to the datasheet, the range of the acceleration is [-10 g, + 10 g], so the scale factor should be 10 instead of 5. Signed-off-by: Dragos Bogdan Acked-by: Lars-Peter Clausen Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/imu/adis16480.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/adis16480.c b/drivers/iio/imu/adis16480.c index 8cf84d3..1289842 100644 --- a/drivers/iio/imu/adis16480.c +++ b/drivers/iio/imu/adis16480.c @@ -696,7 +696,7 @@ static const struct adis16480_chip_info adis16480_chip_info[] = { .gyro_max_val = IIO_RAD_TO_DEGREE(22500), .gyro_max_scale = 450, .accel_max_val = IIO_M_S_2_TO_G(12500), - .accel_max_scale = 5, + .accel_max_scale = 10, }, [ADIS16485] = { .channels = adis16485_channels, -- cgit v1.1 From a359bb2a55f384bb93349ddf9d30b20b37e02e8a Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Thu, 3 Aug 2017 11:22:17 +0200 Subject: iio: trigger: stm32-timer: fix get trigger mode Fix reading trigger mode, when other bit-fields are set. SMCR register value must be masked to read SMS (slave mode selection) only. Fixes: 9eba381 ("iio: make stm32 trigger driver use INDIO_HARDWARE_TRIGGERED mode") Signed-off-by: Fabrice Gasnier Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/trigger/stm32-timer-trigger.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/trigger/stm32-timer-trigger.c b/drivers/iio/trigger/stm32-timer-trigger.c index 14e6eb0..25ad6ab 100644 --- a/drivers/iio/trigger/stm32-timer-trigger.c +++ b/drivers/iio/trigger/stm32-timer-trigger.c @@ -485,7 +485,7 @@ static int stm32_get_trigger_mode(struct iio_dev *indio_dev, regmap_read(priv->regmap, TIM_SMCR, &smcr); - return smcr == TIM_SMCR_SMS ? 0 : -EINVAL; + return (smcr & TIM_SMCR_SMS) == TIM_SMCR_SMS ? 0 : -EINVAL; } static const struct iio_enum stm32_trigger_mode_enum = { -- cgit v1.1 From f1664eaacec31035450132c46ed2915fd2b2049a Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Sat, 12 Aug 2017 09:09:21 -0700 Subject: iio: hid-sensor-trigger: Fix the race with user space powering up sensors It has been reported for a while that with iio-sensor-proxy service the rotation only works after one suspend/resume cycle. This required a wait in the systemd unit file to avoid race. I found a Yoga 900 where I could reproduce this. The problem scenerio is: - During sensor driver init, enable run time PM and also set a auto-suspend for 3 seconds. This result in one runtime resume. But there is a check to avoid a powerup in this sequence, but rpm is active - User space iio-sensor-proxy tries to power up the sensor. Since rpm is active it will simply return. But sensors were not actually powered up in the prior sequence, so actaully the sensors will not work - After 3 seconds the auto suspend kicks If we add a wait in systemd service file to fire iio-sensor-proxy after 3 seconds, then now everything will work as the runtime resume will actually powerup the sensor as this is a user request. To avoid this: - Remove the check to match user requested state, this will cause a brief powerup, but if the iio-sensor-proxy starts immediately it will still work as the sensors are ON. - Also move the autosuspend delay to place when user requested turn off of sensors, like after user finished raw read or buffer disable Signed-off-by: Srinivas Pandruvada Tested-by: Bastien Nocera Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/common/hid-sensors/hid-sensor-trigger.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c index 16ade0a..0e4b379 100644 --- a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c +++ b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c @@ -111,8 +111,6 @@ static int _hid_sensor_power_state(struct hid_sensor_common *st, bool state) s32 poll_value = 0; if (state) { - if (!atomic_read(&st->user_requested_state)) - return 0; if (sensor_hub_device_open(st->hsdev)) return -EIO; @@ -161,6 +159,9 @@ static int _hid_sensor_power_state(struct hid_sensor_common *st, bool state) &report_val); } + pr_debug("HID_SENSOR %s set power_state %d report_state %d\n", + st->pdev->name, state_val, report_val); + sensor_hub_get_feature(st->hsdev, st->power_state.report_id, st->power_state.index, sizeof(state_val), &state_val); @@ -182,6 +183,7 @@ int hid_sensor_power_state(struct hid_sensor_common *st, bool state) ret = pm_runtime_get_sync(&st->pdev->dev); else { pm_runtime_mark_last_busy(&st->pdev->dev); + pm_runtime_use_autosuspend(&st->pdev->dev); ret = pm_runtime_put_autosuspend(&st->pdev->dev); } if (ret < 0) { @@ -285,8 +287,6 @@ int hid_sensor_setup_trigger(struct iio_dev *indio_dev, const char *name, /* Default to 3 seconds, but can be changed from sysfs */ pm_runtime_set_autosuspend_delay(&attrb->pdev->dev, 3000); - pm_runtime_use_autosuspend(&attrb->pdev->dev); - return ret; error_unreg_trigger: iio_trigger_unregister(trig); -- cgit v1.1 From 541ee9b24fca587f510fe1bc58508d5cf40707af Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Wed, 16 Aug 2017 19:02:50 +0200 Subject: iio: magnetometer: st_magn: fix status register address for LSM303AGR Fixes: 97865fe41322 (iio: st_sensors: verify interrupt event to status) Signed-off-by: Lorenzo Bianconi Reviewed-by: Linus Walleij Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/magnetometer/st_magn_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/magnetometer/st_magn_core.c b/drivers/iio/magnetometer/st_magn_core.c index 8e1b086..c11f0da 100644 --- a/drivers/iio/magnetometer/st_magn_core.c +++ b/drivers/iio/magnetometer/st_magn_core.c @@ -358,7 +358,7 @@ static const struct st_sensor_settings st_magn_sensors_settings[] = { .mask_int1 = 0x01, .addr_ihl = 0x63, .mask_ihl = 0x04, - .addr_stat_drdy = ST_SENSORS_DEFAULT_STAT_ADDR, + .addr_stat_drdy = 0x67, }, .multi_read_bit = false, .bootime = 2, -- cgit v1.1 From 8b35a5f87a73842601cd376e0f5b9b25831390f4 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Wed, 16 Aug 2017 19:02:51 +0200 Subject: iio: magnetometer: st_magn: remove ihl property for LSM303AGR Remove IRQ active low support for LSM303AGR since the sensor does not support that capability for data-ready line Fixes: a9fd053b56c6 (iio: st_sensors: support active-low interrupts) Signed-off-by: Lorenzo Bianconi Reviewed-by: Linus Walleij Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/magnetometer/st_magn_core.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/iio/magnetometer/st_magn_core.c b/drivers/iio/magnetometer/st_magn_core.c index c11f0da..c385636 100644 --- a/drivers/iio/magnetometer/st_magn_core.c +++ b/drivers/iio/magnetometer/st_magn_core.c @@ -356,8 +356,6 @@ static const struct st_sensor_settings st_magn_sensors_settings[] = { .drdy_irq = { .addr = 0x62, .mask_int1 = 0x01, - .addr_ihl = 0x63, - .mask_ihl = 0x04, .addr_stat_drdy = 0x67, }, .multi_read_bit = false, -- cgit v1.1 From d912366a59c5384df436fd007667d6e574128b44 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Sun, 20 Aug 2017 09:29:03 -0700 Subject: Input: soc_button_array - silence -ENOENT error on Dell XPS13 9365 The Dell XPS13 9365 has an INT33D2 ACPI node with no GPIOs, causing the following error in dmesg: [ 7.172275] soc_button_array: probe of INT33D2:00 failed with error -2 This commit silences this, by returning -ENODEV when there are no GPIOs. BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=196679 Signed-off-by: Hans de Goede Signed-off-by: Dmitry Torokhov --- drivers/input/misc/soc_button_array.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/misc/soc_button_array.c b/drivers/input/misc/soc_button_array.c index f600f3a..23520df 100644 --- a/drivers/input/misc/soc_button_array.c +++ b/drivers/input/misc/soc_button_array.c @@ -331,7 +331,7 @@ static int soc_button_probe(struct platform_device *pdev) error = gpiod_count(dev, NULL); if (error < 0) { dev_dbg(dev, "no GPIO attached, ignoring...\n"); - return error; + return -ENODEV; } priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); -- cgit v1.1 From 8c97023cf0518f172b8cb7a9fffc28b89401abbf Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 26 Jul 2017 15:36:23 +0200 Subject: Kbuild: use -fshort-wchar globally Commit 971a69db7dc0 ("Xen: don't warn about 2-byte wchar_t in efi") added the --no-wchar-size-warning to the Makefile to avoid this harmless warning: arm-linux-gnueabi-ld: warning: drivers/xen/efi.o uses 2-byte wchar_t yet the output is to use 4-byte wchar_t; use of wchar_t values across objects may fail Changing kbuild to use thin archives instead of recursive linking unfortunately brings the same warning back during the final link. The kernel does not use wchar_t string literals at this point, and xen does not use wchar_t at all (only efi_char16_t), so the flag has no effect, but as pointed out by Jan Beulich, adding a wchar_t string literal would be bad here. Since wchar_t is always defined as u16, independent of the toolchain default, always passing -fshort-wchar is correct and lets us remove the Xen specific hack along with fixing the warning. Link: https://patchwork.kernel.org/patch/9275217/ Fixes: 971a69db7dc0 ("Xen: don't warn about 2-byte wchar_t in efi") Signed-off-by: Arnd Bergmann Acked-by: David Vrabel Signed-off-by: Masahiro Yamada --- Makefile | 2 +- drivers/xen/Makefile | 3 --- 2 files changed, 1 insertion(+), 4 deletions(-) diff --git a/Makefile b/Makefile index 473f1aa..4dac110 100644 --- a/Makefile +++ b/Makefile @@ -396,7 +396,7 @@ LINUXINCLUDE := \ KBUILD_CPPFLAGS := -D__KERNEL__ KBUILD_CFLAGS := -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \ - -fno-strict-aliasing -fno-common \ + -fno-strict-aliasing -fno-common -fshort-wchar \ -Werror-implicit-function-declaration \ -Wno-format-security \ -std=gnu89 $(call cc-option,-fno-PIE) diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile index 8feab810..7f188b8 100644 --- a/drivers/xen/Makefile +++ b/drivers/xen/Makefile @@ -7,9 +7,6 @@ obj-y += xenbus/ nostackp := $(call cc-option, -fno-stack-protector) CFLAGS_features.o := $(nostackp) -CFLAGS_efi.o += -fshort-wchar -LDFLAGS += $(call ld-option, --no-wchar-size-warning) - dom0-$(CONFIG_ARM64) += arm-device.o dom0-$(CONFIG_PCI) += pci.o dom0-$(CONFIG_USB_SUPPORT) += dbgp.o -- cgit v1.1 From 801d2e9f1cfb24c15ec821288b6bc2e3107f6fcd Mon Sep 17 00:00:00 2001 From: Shuah Khan Date: Fri, 18 Aug 2017 15:54:41 -0600 Subject: Makefile: add kselftest-clean to PHONY target list kselftest-clean isn't in the PHONY target list. Add it. Signed-off-by: Shuah Khan Signed-off-by: Masahiro Yamada --- Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/Makefile b/Makefile index 4dac110..c913d80 100644 --- a/Makefile +++ b/Makefile @@ -1184,6 +1184,7 @@ PHONY += kselftest kselftest: $(Q)$(MAKE) -C tools/testing/selftests run_tests +PHONY += kselftest-clean kselftest-clean: $(Q)$(MAKE) -C tools/testing/selftests clean -- cgit v1.1 From 2bfbe7881ee03d12bbedb05f804d139c88df30a0 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sat, 19 Aug 2017 22:30:02 +0100 Subject: kbuild: Do not use hyphen in exported variable name This definition in Makefile.dtbinst: export dtbinst-root ?= $(obj) should define and export dtbinst-root when handling the root dts directory, and do nothing in the subdirectories. However some shells, including dash, will not pass through environment variables whose name includes a hyphen. Usually GNU make does not use a shell to recurse, but if e.g. $(srctree) contains '~' it will use a shell here. Rename the variable to dtbinst_root. References: https://bugs.debian.org/833561 Fixes: 323a028d39cdi ("dts, kbuild: Implement support for dtb vendor subdirs") Signed-off-by: Ben Hutchings Signed-off-by: Masahiro Yamada --- scripts/Makefile.dtbinst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/Makefile.dtbinst b/scripts/Makefile.dtbinst index 34614a4..993fb85 100644 --- a/scripts/Makefile.dtbinst +++ b/scripts/Makefile.dtbinst @@ -14,7 +14,7 @@ src := $(obj) PHONY := __dtbs_install __dtbs_install: -export dtbinst-root ?= $(obj) +export dtbinst_root ?= $(obj) include include/config/auto.conf include scripts/Kbuild.include @@ -27,7 +27,7 @@ dtbinst-dirs := $(dts-dirs) quiet_cmd_dtb_install = INSTALL $< cmd_dtb_install = mkdir -p $(2); cp $< $(2) -install-dir = $(patsubst $(dtbinst-root)%,$(INSTALL_DTBS_PATH)%,$(obj)) +install-dir = $(patsubst $(dtbinst_root)%,$(INSTALL_DTBS_PATH)%,$(obj)) $(dtbinst-files): %.dtb: $(obj)/%.dtb $(call cmd,dtb_install,$(install-dir)) -- cgit v1.1 From 4459398b6d9e3055ced5de9820364b3bdd79ac25 Mon Sep 17 00:00:00 2001 From: Keerthy Date: Mon, 31 Jul 2017 10:01:36 +0530 Subject: soc: ti: knav: Add a NULL pointer check for kdev in knav_pool_create knav_pool_create is an exported function. In the event of a call before knav_queue_probe, we encounter a NULL pointer dereference in the following line. Hence return -EPROBE_DEFER to the caller till the kdev pointer is non-NULL. Signed-off-by: Keerthy Acked-by: Santosh Shilimkar Signed-off-by: Arnd Bergmann --- drivers/soc/ti/knav_qmss_queue.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/soc/ti/knav_qmss_queue.c b/drivers/soc/ti/knav_qmss_queue.c index 279e7c5..39225de 100644 --- a/drivers/soc/ti/knav_qmss_queue.c +++ b/drivers/soc/ti/knav_qmss_queue.c @@ -745,6 +745,9 @@ void *knav_pool_create(const char *name, bool slot_found; int ret; + if (!kdev) + return ERR_PTR(-EPROBE_DEFER); + if (!kdev->dev) return ERR_PTR(-ENODEV); -- cgit v1.1 From a93c11527528c951b8d8db638162128a09e09ec2 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 17 Aug 2017 13:55:41 +0300 Subject: drm/i915/bxt: use NULL for GPIO connection ID The commit 213e08ad60ba ("drm/i915/bxt: add bxt dsi gpio element support") enables GPIO support for Broxton based platforms. While using that API we might get into troubles in the future, because we can't rely on label name in the driver since vendor firmware might provide any GPIO pin there, e.g. "reset", and even mark it in _DSD (in which case the request will fail). To avoid inconsistency and potential issues we have two options: a) generate GPIO ACPI mapping table and supply it via acpi_dev_add_driver_gpios(), or b) just pass NULL as connection ID. The b) approach is much simpler and would work since the driver relies on GPIO indices only. Moreover, the _CRS fallback mechanism, when requesting GPIO, has been made stricter, and supplying non-NULL connection ID when neither _DSD, nor GPIO ACPI mapping is present, is making request fail. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101921 Fixes: f10e4bf6632b ("gpio: acpi: Even more tighten up ACPI GPIO lookups") Cc: Mika Kahola Cc: Jani Nikula Tested-by: Mika Kahola Signed-off-by: Andy Shevchenko Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20170817105541.63914-1-andriy.shevchenko@linux.intel.com (cherry picked from commit cd55a1fbd21a820b7dd85a208b3170aa0b06adfa) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_dsi_vbt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_dsi_vbt.c b/drivers/gpu/drm/i915/intel_dsi_vbt.c index 7158c7c..91c07b0 100644 --- a/drivers/gpu/drm/i915/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/intel_dsi_vbt.c @@ -306,7 +306,7 @@ static void bxt_exec_gpio(struct drm_i915_private *dev_priv, if (!gpio_desc) { gpio_desc = devm_gpiod_get_index(dev_priv->drm.dev, - "panel", gpio_index, + NULL, gpio_index, value ? GPIOD_OUT_LOW : GPIOD_OUT_HIGH); -- cgit v1.1 From d41a3c2be178783c85e05025265ab58fbb4d4ce1 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Mon, 7 Aug 2017 13:19:19 +0100 Subject: drm/i915: Clear lost context-switch interrupts across reset During a global reset, we disable the irq. As we disable the irq, the hardware may be raising a GT interrupt that we then ignore, leaving it pending in the GTIIR. After the reset, we then re-enable the irq, triggering the pending interrupt. However, that interrupt was for the stale state from before the reset, and the contents of the CSB buffer are now invalid. v2: Add a comment to make it clear that the double clear is purely my paranoia. Reported-by: "Dong, Chuanxiao" Fixes: 821ed7df6e2a ("drm/i915: Update reset path to fix incomplete requests") Signed-off-by: Chris Wilson Cc: "Dong, Chuanxiao" Cc: Tvrtko Ursulin Cc: Michal Winiarski Cc: Michel Thierry Link: https://patchwork.freedesktop.org/patch/msgid/20170807121919.30165-1-chris@chris-wilson.co.uk Link: https://patchwork.freedesktop.org/patch/msgid/20170818090509.5363-1-chris@chris-wilson.co.uk Reviewed-by: Michel Thierry (cherry picked from commit 64f09f00caf0a7cb40a8c0b85789bacba0f51d9e) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_lrc.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 7404cf2..2afa4da 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -1221,6 +1221,14 @@ static int intel_init_workaround_bb(struct intel_engine_cs *engine) return ret; } +static u8 gtiir[] = { + [RCS] = 0, + [BCS] = 0, + [VCS] = 1, + [VCS2] = 1, + [VECS] = 3, +}; + static int gen8_init_common_ring(struct intel_engine_cs *engine) { struct drm_i915_private *dev_priv = engine->i915; @@ -1245,9 +1253,22 @@ static int gen8_init_common_ring(struct intel_engine_cs *engine) DRM_DEBUG_DRIVER("Execlists enabled for %s\n", engine->name); - /* After a GPU reset, we may have requests to replay */ + GEM_BUG_ON(engine->id >= ARRAY_SIZE(gtiir)); + + /* + * Clear any pending interrupt state. + * + * We do it twice out of paranoia that some of the IIR are double + * buffered, and if we only reset it once there may still be + * an interrupt pending. + */ + I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]), + GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift); + I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]), + GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift); clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); + /* After a GPU reset, we may have requests to replay */ submit = false; for (n = 0; n < ARRAY_SIZE(engine->execlist_port); n++) { if (!port_isset(&port[n])) -- cgit v1.1 From d83c2dbaa90a9bd6346e234d9802080a9c7b2fea Mon Sep 17 00:00:00 2001 From: Shawn Lin Date: Fri, 18 Aug 2017 09:16:08 +0800 Subject: mmc: block: prevent propagating R1_OUT_OF_RANGE for open-ending mode We to some extent should tolerate R1_OUT_OF_RANGE for open-ending mode as it is expected behaviour and most of the backup partition tables should be located near some of the last blocks which will always make open-ending read exceed the capacity of cards. Fixes: 9820a5b11101 ("mmc: core: for data errors, take response of stop cmd into account") Fixes: a04e6bae9e6f ("mmc: core: check also R1 response for stop commands") Signed-off-by: Shawn Lin Reviewed-by: Wolfram Sang Tested-by: Shawn Guo Signed-off-by: Ulf Hansson --- drivers/mmc/core/block.c | 49 ++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 43 insertions(+), 6 deletions(-) diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c index f1bbfd3..80d1ec6 100644 --- a/drivers/mmc/core/block.c +++ b/drivers/mmc/core/block.c @@ -1371,12 +1371,46 @@ static inline void mmc_apply_rel_rw(struct mmc_blk_request *brq, R1_CC_ERROR | /* Card controller error */ \ R1_ERROR) /* General/unknown error */ -static bool mmc_blk_has_cmd_err(struct mmc_command *cmd) +static void mmc_blk_eval_resp_error(struct mmc_blk_request *brq) { - if (!cmd->error && cmd->resp[0] & CMD_ERRORS) - cmd->error = -EIO; + u32 val; - return cmd->error; + /* + * Per the SD specification(physical layer version 4.10)[1], + * section 4.3.3, it explicitly states that "When the last + * block of user area is read using CMD18, the host should + * ignore OUT_OF_RANGE error that may occur even the sequence + * is correct". And JESD84-B51 for eMMC also has a similar + * statement on section 6.8.3. + * + * Multiple block read/write could be done by either predefined + * method, namely CMD23, or open-ending mode. For open-ending mode, + * we should ignore the OUT_OF_RANGE error as it's normal behaviour. + * + * However the spec[1] doesn't tell us whether we should also + * ignore that for predefined method. But per the spec[1], section + * 4.15 Set Block Count Command, it says"If illegal block count + * is set, out of range error will be indicated during read/write + * operation (For example, data transfer is stopped at user area + * boundary)." In another word, we could expect a out of range error + * in the response for the following CMD18/25. And if argument of + * CMD23 + the argument of CMD18/25 exceed the max number of blocks, + * we could also expect to get a -ETIMEDOUT or any error number from + * the host drivers due to missing data response(for write)/data(for + * read), as the cards will stop the data transfer by itself per the + * spec. So we only need to check R1_OUT_OF_RANGE for open-ending mode. + */ + + if (!brq->stop.error) { + bool oor_with_open_end; + /* If there is no error yet, check R1 response */ + + val = brq->stop.resp[0] & CMD_ERRORS; + oor_with_open_end = val & R1_OUT_OF_RANGE && !brq->mrq.sbc; + + if (val && !oor_with_open_end) + brq->stop.error = -EIO; + } } static enum mmc_blk_status mmc_blk_err_check(struct mmc_card *card, @@ -1400,8 +1434,11 @@ static enum mmc_blk_status mmc_blk_err_check(struct mmc_card *card, * stop.error indicates a problem with the stop command. Data * may have been transferred, or may still be transferring. */ - if (brq->sbc.error || brq->cmd.error || mmc_blk_has_cmd_err(&brq->stop) || - brq->data.error) { + + mmc_blk_eval_resp_error(brq); + + if (brq->sbc.error || brq->cmd.error || + brq->stop.error || brq->data.error) { switch (mmc_blk_cmd_recovery(card, req, brq, &ecc_err, &gen_err)) { case ERR_RETRY: return MMC_BLK_RETRY; -- cgit v1.1 From 4a4eefcd0e49f9f339933324c1bde431186a0a7d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 3 Aug 2017 13:05:11 +0200 Subject: KVM: s390: sthyi: fix sthyi inline assembly The sthyi inline assembly misses register r3 within the clobber list. The sthyi instruction will always write a return code to register "R2+1", which in this case would be r3. Due to that we may have register corruption and see host crashes or data corruption depending on how gcc decided to allocate and use registers during compile time. Fixes: 95ca2cb57985 ("KVM: s390: Add sthyi emulation") Cc: # 4.8+ Reviewed-by: Janosch Frank Signed-off-by: Heiko Carstens Reviewed-by: David Hildenbrand Reviewed-by: Cornelia Huck Signed-off-by: Christian Borntraeger --- arch/s390/kvm/sthyi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/s390/kvm/sthyi.c b/arch/s390/kvm/sthyi.c index 926b524..2773a2f 100644 --- a/arch/s390/kvm/sthyi.c +++ b/arch/s390/kvm/sthyi.c @@ -394,7 +394,7 @@ static int sthyi(u64 vaddr) "srl %[cc],28\n" : [cc] "=d" (cc) : [code] "d" (code), [addr] "a" (addr) - : "memory", "cc"); + : "3", "memory", "cc"); return cc; } -- cgit v1.1 From 857b8de96795646c5891cf44ae6fb19b9ff74bf9 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 3 Aug 2017 14:27:30 +0200 Subject: KVM: s390: sthyi: fix specification exception detection sthyi should only generate a specification exception if the function code is zero and the response buffer is not on a 4k boundary. The current code would also test for unknown function codes if the response buffer, that is currently only defined for function code 0, is not on a 4k boundary and incorrectly inject a specification exception instead of returning with condition code 3 and return code 4 (unsupported function code). Fix this by moving the boundary check. Fixes: 95ca2cb57985 ("KVM: s390: Add sthyi emulation") Cc: # 4.8+ Reviewed-by: Janosch Frank Signed-off-by: Heiko Carstens Reviewed-by: David Hildenbrand Reviewed-by: Cornelia Huck Signed-off-by: Christian Borntraeger --- arch/s390/kvm/sthyi.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/s390/kvm/sthyi.c b/arch/s390/kvm/sthyi.c index 2773a2f..a2e5c24 100644 --- a/arch/s390/kvm/sthyi.c +++ b/arch/s390/kvm/sthyi.c @@ -425,7 +425,7 @@ int handle_sthyi(struct kvm_vcpu *vcpu) VCPU_EVENT(vcpu, 3, "STHYI: fc: %llu addr: 0x%016llx", code, addr); trace_kvm_s390_handle_sthyi(vcpu, code, addr); - if (reg1 == reg2 || reg1 & 1 || reg2 & 1 || addr & ~PAGE_MASK) + if (reg1 == reg2 || reg1 & 1 || reg2 & 1) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); if (code & 0xffff) { @@ -433,6 +433,9 @@ int handle_sthyi(struct kvm_vcpu *vcpu) goto out; } + if (addr & ~PAGE_MASK) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + /* * If the page has not yet been faulted in, we want to do that * now and not after all the expensive calculations. -- cgit v1.1 From deecd4d71b12626db48544faa66bb897e2cafd07 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Thu, 27 Jul 2017 15:56:55 -0500 Subject: objtool: Fix '-mtune=atom' decoding support in objtool 2.0 With '-mtune=atom', which is enabled with CONFIG_MATOM=y, GCC uses some unusual instructions for setting up the stack. Instead of: mov %rsp, %rbp it does: lea (%rsp), %rbp And instead of: add imm, %rsp it does: lea disp(%rsp), %rsp Add support for these instructions to the objtool decoder. Reported-by: Arnd Bergmann Signed-off-by: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0") Link: http://lkml.kernel.org/r/4ea1db896e821226efe1f8e09f270771bde47e65.1501188854.git.jpoimboe@redhat.com [ This is a cherry-picked version of upcoming commit 5b8de48e82ba. ] Signed-off-by: Ingo Molnar --- tools/objtool/arch/x86/decode.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-) diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c index a36c2eb..4559a21a 100644 --- a/tools/objtool/arch/x86/decode.c +++ b/tools/objtool/arch/x86/decode.c @@ -271,7 +271,7 @@ int arch_decode_instruction(struct elf *elf, struct section *sec, case 0x8d: if (rex == 0x48 && modrm == 0x65) { - /* lea -disp(%rbp), %rsp */ + /* lea disp(%rbp), %rsp */ *type = INSN_STACK; op->src.type = OP_SRC_ADD; op->src.reg = CFI_BP; @@ -281,6 +281,30 @@ int arch_decode_instruction(struct elf *elf, struct section *sec, break; } + if (rex == 0x48 && (modrm == 0xa4 || modrm == 0x64) && + sib == 0x24) { + + /* lea disp(%rsp), %rsp */ + *type = INSN_STACK; + op->src.type = OP_SRC_ADD; + op->src.reg = CFI_SP; + op->src.offset = insn.displacement.value; + op->dest.type = OP_DEST_REG; + op->dest.reg = CFI_SP; + break; + } + + if (rex == 0x48 && modrm == 0x2c && sib == 0x24) { + + /* lea (%rsp), %rbp */ + *type = INSN_STACK; + op->src.type = OP_SRC_REG; + op->src.reg = CFI_SP; + op->dest.type = OP_DEST_REG; + op->dest.reg = CFI_BP; + break; + } + if (rex == 0x4c && modrm == 0x54 && sib == 0x24 && insn.displacement.value == 8) { -- cgit v1.1 From dea3eb8b452e36cf2dd572b0a797915ccf452ae6 Mon Sep 17 00:00:00 2001 From: Stephan Mueller Date: Thu, 10 Aug 2017 08:06:18 +0200 Subject: lib/mpi: kunmap after finishing accessing buffer Using sg_miter_start and sg_miter_next, the buffer of an SG is kmap'ed to *buff. The current code calls sg_miter_stop (and thus kunmap) on the SG entry before the last access of *buff. The patch moves the sg_miter_stop call after the last access to *buff to ensure that the memory pointed to by *buff is still mapped. Fixes: 4816c9406430 ("lib/mpi: Fix SG miter leak") Cc: Signed-off-by: Stephan Mueller Signed-off-by: Herbert Xu --- lib/mpi/mpicoder.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/mpi/mpicoder.c b/lib/mpi/mpicoder.c index 5a0f75a..eead4b3 100644 --- a/lib/mpi/mpicoder.c +++ b/lib/mpi/mpicoder.c @@ -364,11 +364,11 @@ MPI mpi_read_raw_from_sgl(struct scatterlist *sgl, unsigned int nbytes) } miter.consumed = lzeros; - sg_miter_stop(&miter); nbytes -= lzeros; nbits = nbytes * 8; if (nbits > MAX_EXTERN_MPI_BITS) { + sg_miter_stop(&miter); pr_info("MPI: mpi too large (%u bits)\n", nbits); return NULL; } @@ -376,6 +376,8 @@ MPI mpi_read_raw_from_sgl(struct scatterlist *sgl, unsigned int nbytes) if (nbytes > 0) nbits -= count_leading_zeros(*buff) - (BITS_PER_LONG - 8); + sg_miter_stop(&miter); + nlimbs = DIV_ROUND_UP(nbytes, BYTES_PER_MPI_LIMB); val = mpi_alloc(nlimbs); if (!val) -- cgit v1.1 From 4de437265eaac18f880d827f8e105b10de9b87a3 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Mon, 14 Aug 2017 14:28:14 +0100 Subject: crypto: chacha20 - fix handling of chunked input Commit 9ae433bc79f9 ("crypto: chacha20 - convert generic and x86 versions to skcipher") ported the existing chacha20 code to use the new skcipher API, and introduced a bug along the way. Unfortunately, the tcrypt tests did not catch the error, and it was only found recently by Tobias. Stefan kindly diagnosed the error, and proposed a fix which is similar to the one below, with the exception that 'walk.stride' is used rather than the hardcoded block size. This does not actually matter in this case, but it's a better example of how to use the skcipher walk API. Fixes: 9ae433bc79f9 ("crypto: chacha20 - convert generic and x86 ...") Cc: # v4.11+ Cc: Steffen Klassert Reported-by: Tobias Brunner Signed-off-by: Ard Biesheuvel Signed-off-by: Herbert Xu --- crypto/chacha20_generic.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/crypto/chacha20_generic.c b/crypto/chacha20_generic.c index 8b3c04d..4a45fa4 100644 --- a/crypto/chacha20_generic.c +++ b/crypto/chacha20_generic.c @@ -91,9 +91,14 @@ int crypto_chacha20_crypt(struct skcipher_request *req) crypto_chacha20_init(state, ctx, walk.iv); while (walk.nbytes > 0) { + unsigned int nbytes = walk.nbytes; + + if (nbytes < walk.total) + nbytes = round_down(nbytes, walk.stride); + chacha20_docrypt(state, walk.dst.virt.addr, walk.src.virt.addr, - walk.nbytes); - err = skcipher_walk_done(&walk, 0); + nbytes); + err = skcipher_walk_done(&walk, walk.nbytes - nbytes); } return err; -- cgit v1.1 From 549f64153c354e69fc19534f7d7e867de1992f95 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Mon, 14 Aug 2017 14:28:15 +0100 Subject: crypto: testmgr - add chunked test cases for chacha20 We failed to catch a bug in the chacha20 code after porting it to the skcipher API. We would have caught it if any chunked tests had been defined, so define some now so we will catch future regressions. Signed-off-by: Ard Biesheuvel Signed-off-by: Herbert Xu --- crypto/testmgr.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/crypto/testmgr.h b/crypto/testmgr.h index 6ceb0e2..d54971d 100644 --- a/crypto/testmgr.h +++ b/crypto/testmgr.h @@ -32675,6 +32675,10 @@ static const struct cipher_testvec chacha20_enc_tv_template[] = { "\x5b\x86\x2f\x37\x30\xe3\x7c\xfd" "\xc4\xfd\x80\x6c\x22\xf2\x21", .rlen = 375, + .also_non_np = 1, + .np = 3, + .tap = { 375 - 20, 4, 16 }, + }, { /* RFC7539 A.2. Test Vector #3 */ .key = "\x1c\x92\x40\xa5\xeb\x55\xd3\x8a" "\xf3\x33\x88\x86\x04\xf6\xb5\xf0" @@ -33049,6 +33053,9 @@ static const struct cipher_testvec chacha20_enc_tv_template[] = { "\xa1\xed\xad\xd5\x76\xfa\x24\x8f" "\x98", .rlen = 1281, + .also_non_np = 1, + .np = 3, + .tap = { 1200, 1, 80 }, }, }; -- cgit v1.1 From 445a582738de6802669aeed9c33ca406c23c3b1f Mon Sep 17 00:00:00 2001 From: Stephan Mueller Date: Wed, 16 Aug 2017 11:56:24 +0200 Subject: crypto: algif_skcipher - only call put_page on referenced and used pages For asynchronous operation, SGs are allocated without a page mapped to them or with a page that is not used (ref-counted). If the SGL is freed, the code must only call put_page for an SG if there was a page assigned and ref-counted in the first place. This fixes a kernel crash when using io_submit with more than one iocb using the sendmsg and sendpage (vmsplice/splice) interface. Cc: Signed-off-by: Stephan Mueller Signed-off-by: Herbert Xu --- crypto/algif_skcipher.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c index 43839b0..903605d 100644 --- a/crypto/algif_skcipher.c +++ b/crypto/algif_skcipher.c @@ -87,8 +87,13 @@ static void skcipher_free_async_sgls(struct skcipher_async_req *sreq) } sgl = sreq->tsg; n = sg_nents(sgl); - for_each_sg(sgl, sg, n, i) - put_page(sg_page(sg)); + for_each_sg(sgl, sg, n, i) { + struct page *page = sg_page(sg); + + /* some SGs may not have a page mapped */ + if (page && page_ref_count(page)) + put_page(page); + } kfree(sreq->tsg); } -- cgit v1.1 From 0f0fc5c090055983163ed8e7b3a73980e128048c Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Tue, 15 Aug 2017 09:30:05 +0100 Subject: Revert "mfd: da9061: Fix to remove BBAT_CONT register from chip model" This patch was applied to the MFD twice, causing unwanted behavour. This reverts commit b77eb79acca3203883e8d8dbc7f2b842def1bff8. Fixes: b77eb79acca3 ("mfd: da9061: Fix to remove BBAT_CONT register from chip model") Reported-by: Steve Twiss Reviewed-by: Steve Twiss Signed-off-by: Lee Jones --- drivers/mfd/da9062-core.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/mfd/da9062-core.c b/drivers/mfd/da9062-core.c index fbe0f24..fe18115 100644 --- a/drivers/mfd/da9062-core.c +++ b/drivers/mfd/da9062-core.c @@ -645,6 +645,9 @@ static const struct regmap_range da9062_aa_readable_ranges[] = { .range_min = DA9062AA_VLDO1_B, .range_max = DA9062AA_VLDO4_B, }, { + .range_min = DA9062AA_BBAT_CONT, + .range_max = DA9062AA_BBAT_CONT, + }, { .range_min = DA9062AA_INTERFACE, .range_max = DA9062AA_CONFIG_E, }, { @@ -721,6 +724,9 @@ static const struct regmap_range da9062_aa_writeable_ranges[] = { .range_min = DA9062AA_VLDO1_B, .range_max = DA9062AA_VLDO4_B, }, { + .range_min = DA9062AA_BBAT_CONT, + .range_max = DA9062AA_BBAT_CONT, + }, { .range_min = DA9062AA_GP_ID_0, .range_max = DA9062AA_GP_ID_19, }, -- cgit v1.1 From 07b3b5e9ed807a0d2077319b8e43a42e941db818 Mon Sep 17 00:00:00 2001 From: Joakim Tjernlund Date: Tue, 22 Aug 2017 08:33:53 +0200 Subject: ALSA: usb-audio: Add delay quirk for H650e/Jabra 550a USB headsets These headsets reports a lot of: cannot set freq 44100 to ep 0x81 and need a small delay between sample rate settings, just like Zoom R16/24. Add both headsets to the Zoom R16/24 quirk for a 1 ms delay between control msgs. Signed-off-by: Joakim Tjernlund Cc: Signed-off-by: Takashi Iwai --- sound/usb/quirks.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 6a03f96..5d2a632 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1309,10 +1309,13 @@ void snd_usb_ctl_msg_quirk(struct usb_device *dev, unsigned int pipe, && (requesttype & USB_TYPE_MASK) == USB_TYPE_CLASS) mdelay(20); - /* Zoom R16/24 needs a tiny delay here, otherwise requests like - * get/set frequency return as failed despite actually succeeding. + /* Zoom R16/24, Logitech H650e, Jabra 550a needs a tiny delay here, + * otherwise requests like get/set frequency return as failed despite + * actually succeeding. */ - if (chip->usb_id == USB_ID(0x1686, 0x00dd) && + if ((chip->usb_id == USB_ID(0x1686, 0x00dd) || + chip->usb_id == USB_ID(0x046d, 0x0a46) || + chip->usb_id == USB_ID(0x0b0e, 0x0349)) && (requesttype & USB_TYPE_MASK) == USB_TYPE_CLASS) mdelay(1); } -- cgit v1.1 From 804e73adf5cf4b3aea3b6ce058f4dc0191143821 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 22 Aug 2017 07:44:52 +0200 Subject: ASoC: rt5670: Fix GPIO headset detection regression RT5670 codec driver and its machine driver for Intel CHT assume the implicit GPIO mapping on the index 0 while BIOS on most devices don't provide it. The recent commit f10e4bf6632b ("gpio: acpi: Even more tighten up ACPI GPIO lookups") restricts such cases and it resulted in a regression where the headset jack setup fails like: rt5670 i2c-10EC5672:00: ASoC: Cannot get gpio at index 0: -2 rt5670 i2c-10EC5672:00: Adding jack GPIO failed For fixing this, we need to provide the GPIO mapping explicitly in the machine driver. Also this patch corrects the string to be passed to gpiolib to match with the pre-given mapping, too. Fixes: f10e4bf6632b ("gpio: acpi: Even more tighten up ACPI GPIO lookups") Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=115531 Reported-by: Pierre-Louis Bossart Tested-by: Pierre-Louis Bossart Signed-off-by: Takashi Iwai Reviewed-by: Andy Shevchenko Signed-off-by: Mark Brown --- sound/soc/codecs/rt5670.c | 2 +- sound/soc/intel/boards/cht_bsw_rt5672.c | 10 ++++++++++ 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/sound/soc/codecs/rt5670.c b/sound/soc/codecs/rt5670.c index 0ec7985..054b613 100644 --- a/sound/soc/codecs/rt5670.c +++ b/sound/soc/codecs/rt5670.c @@ -567,7 +567,7 @@ int rt5670_set_jack_detect(struct snd_soc_codec *codec, rt5670->jack = jack; rt5670->hp_gpio.gpiod_dev = codec->dev; - rt5670->hp_gpio.name = "headphone detect"; + rt5670->hp_gpio.name = "headset"; rt5670->hp_gpio.report = SND_JACK_HEADSET | SND_JACK_BTN_0 | SND_JACK_BTN_1 | SND_JACK_BTN_2; rt5670->hp_gpio.debounce_time = 150; diff --git a/sound/soc/intel/boards/cht_bsw_rt5672.c b/sound/soc/intel/boards/cht_bsw_rt5672.c index bc2a52d..f597d55 100644 --- a/sound/soc/intel/boards/cht_bsw_rt5672.c +++ b/sound/soc/intel/boards/cht_bsw_rt5672.c @@ -184,6 +184,13 @@ static int cht_aif1_hw_params(struct snd_pcm_substream *substream, return 0; } +static const struct acpi_gpio_params headset_gpios = { 0, 0, false }; + +static const struct acpi_gpio_mapping cht_rt5672_gpios[] = { + { "headset-gpios", &headset_gpios, 1 }, + {}, +}; + static int cht_codec_init(struct snd_soc_pcm_runtime *runtime) { int ret; @@ -191,6 +198,9 @@ static int cht_codec_init(struct snd_soc_pcm_runtime *runtime) struct snd_soc_codec *codec = codec_dai->codec; struct cht_mc_private *ctx = snd_soc_card_get_drvdata(runtime->card); + if (devm_acpi_dev_add_driver_gpios(codec->dev, cht_rt5672_gpios)) + dev_warn(runtime->dev, "Unable to add GPIO mapping table\n"); + /* TDM 4 slots 24 bit, set Rx & Tx bitmask to 4 active slots */ ret = snd_soc_dai_set_tdm_slot(codec_dai, 0xF, 0xF, 4, 24); if (ret < 0) { -- cgit v1.1 From fe4600a548f2763dec91b3b27a1245c370ceee2a Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sat, 19 Aug 2017 13:05:58 +0100 Subject: drm: Release driver tracking before making the object available again MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is the same bug as we fixed in commit f6cd7daecff5 ("drm: Release driver references to handle before making it available again"), but now the exposure is via the PRIME lookup tables. If we remove the object/handle from the PRIME lut, then a new request for the same object/fd will generate a new handle, thus for a short window that object is known to userspace by two different handles. Fix this by releasing the driver tracking before PRIME. Fixes: 0ff926c7d4f0 ("drm/prime: add exported buffers to current fprivs imported buffer list (v2)") Signed-off-by: Chris Wilson Cc: David Airlie Cc: Daniel Vetter Cc: Rob Clark Cc: Ville Syrjälä Cc: Thierry Reding Cc: stable@vger.kernel.org Reviewed-by: Daniel Vetter Signed-off-by: Joonas Lahtinen Link: https://patchwork.freedesktop.org/patch/msgid/20170819120558.6465-1-chris@chris-wilson.co.uk --- drivers/gpu/drm/drm_gem.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 8dc1106..cdaac37 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -255,13 +255,13 @@ drm_gem_object_release_handle(int id, void *ptr, void *data) struct drm_gem_object *obj = ptr; struct drm_device *dev = obj->dev; + if (dev->driver->gem_close_object) + dev->driver->gem_close_object(obj, file_priv); + if (drm_core_check_feature(dev, DRIVER_PRIME)) drm_gem_remove_prime_handles(obj, file_priv); drm_vma_node_revoke(&obj->vma_node, file_priv); - if (dev->driver->gem_close_object) - dev->driver->gem_close_object(obj, file_priv); - drm_gem_object_handle_put_unlocked(obj); return 0; -- cgit v1.1 From 88c54cdf61f508ebcf8da2d819f5dfc03e954d1d Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 22 Aug 2017 08:15:13 +0200 Subject: ALSA: core: Fix unexpected error at replacing user TLV When user tries to replace the user-defined control TLV, the kernel checks the change of its content via memcmp(). The problem is that the kernel passes the return value from memcmp() as is. memcmp() gives a non-zero negative value depending on the comparison result, and this shall be recognized as an error code. The patch covers that corner-case, return 1 properly for the changed TLV. Fixes: 8aa9b586e420 ("[ALSA] Control API - more robust TLV implementation") Cc: Signed-off-by: Takashi Iwai --- sound/core/control.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/core/control.c b/sound/core/control.c index 3c6be14..4525e12 100644 --- a/sound/core/control.c +++ b/sound/core/control.c @@ -1137,7 +1137,7 @@ static int snd_ctl_elem_user_tlv(struct snd_kcontrol *kcontrol, mutex_lock(&ue->card->user_ctl_lock); change = ue->tlv_data_size != size; if (!change) - change = memcmp(ue->tlv_data, new_data, size); + change = memcmp(ue->tlv_data, new_data, size) != 0; kfree(ue->tlv_data); ue->tlv_data = new_data; ue->tlv_data_size = size; -- cgit v1.1 From 096622104e14d8a1db4860bd557717067a0515d2 Mon Sep 17 00:00:00 2001 From: Dave Martin Date: Fri, 18 Aug 2017 16:57:01 +0100 Subject: arm64: fpsimd: Prevent registers leaking across exec There are some tricky dependencies between the different stages of flushing the FPSIMD register state during exec, and these can race with context switch in ways that can cause the old task's regs to leak across. In particular, a context switch during the memset() can cause some of the task's old FPSIMD registers to reappear. Disabling preemption for this small window would be no big deal for performance: preemption is already disabled for similar scenarios like updating the FPSIMD registers in sigreturn. So, instead of rearranging things in ways that might swap existing subtle bugs for new ones, this patch just disables preemption around the FPSIMD state flushing so that races of this type can't occur here. This brings fpsimd_flush_thread() into line with other code paths. Cc: stable@vger.kernel.org Fixes: 674c242c9323 ("arm64: flush FP/SIMD state correctly after execve()") Reviewed-by: Ard Biesheuvel Signed-off-by: Dave Martin Signed-off-by: Will Deacon --- arch/arm64/kernel/fpsimd.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c index 06da8ea..c7b4995 100644 --- a/arch/arm64/kernel/fpsimd.c +++ b/arch/arm64/kernel/fpsimd.c @@ -161,9 +161,11 @@ void fpsimd_flush_thread(void) { if (!system_supports_fpsimd()) return; + preempt_disable(); memset(¤t->thread.fpsimd_state, 0, sizeof(struct fpsimd_state)); fpsimd_flush_task_state(current); set_thread_flag(TIF_FOREIGN_FPSTATE); + preempt_enable(); } /* -- cgit v1.1 From 289d07a2dc6c6b6f3e4b8a62669320d99dbe6c3d Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Tue, 11 Jul 2017 15:19:22 +0100 Subject: arm64: mm: abort uaccess retries upon fatal signal When there's a fatal signal pending, arm64's do_page_fault() implementation returns 0. The intent is that we'll return to the faulting userspace instruction, delivering the signal on the way. However, if we take a fatal signal during fixing up a uaccess, this results in a return to the faulting kernel instruction, which will be instantly retried, resulting in the same fault being taken forever. As the task never reaches userspace, the signal is not delivered, and the task is left unkillable. While the task is stuck in this state, it can inhibit the forward progress of the system. To avoid this, we must ensure that when a fatal signal is pending, we apply any necessary fixup for a faulting kernel instruction. Thus we will return to an error path, and it is up to that code to make forward progress towards delivering the fatal signal. Cc: Catalin Marinas Cc: Laura Abbott Cc: stable@vger.kernel.org Reviewed-by: Steve Capper Tested-by: Steve Capper Reviewed-by: James Morse Tested-by: James Morse Signed-off-by: Mark Rutland Signed-off-by: Will Deacon --- arch/arm64/mm/fault.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index 2509e4f..1f22a41 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -435,8 +435,11 @@ retry: * the mmap_sem because it would already be released * in __lock_page_or_retry in mm/filemap.c. */ - if (fatal_signal_pending(current)) + if (fatal_signal_pending(current)) { + if (!user_mode(regs)) + goto no_context; return 0; + } /* * Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk of -- cgit v1.1 From 4a23e56ad6549d0b8c0fac6d9eb752884379c391 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 18 Aug 2017 18:42:30 +0100 Subject: arm64: kaslr: ignore modulo offset when validating virtual displacement In the KASLR setup routine, we ensure that the early virtual mapping of the kernel image does not cover more than a single table entry at the level above the swapper block level, so that the assembler routines involved in setting up this mapping can remain simple. In this calculation we add the proposed KASLR offset to the values of the _text and _end markers, and reject it if they would end up falling in different swapper table sized windows. However, when taking the addresses of _text and _end, the modulo offset (the physical displacement modulo 2 MB) is already accounted for, and so adding it again results in incorrect results. So disregard the modulo offset from the calculation. Fixes: 08cdac619c81 ("arm64: relocatable: deal with physically misaligned ...") Reviewed-by: Catalin Marinas Tested-by: Catalin Marinas Signed-off-by: Ard Biesheuvel Signed-off-by: Will Deacon --- arch/arm64/kernel/head.S | 1 - arch/arm64/kernel/kaslr.c | 12 +++++++++--- 2 files changed, 9 insertions(+), 4 deletions(-) diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S index 973df7d..adb0910 100644 --- a/arch/arm64/kernel/head.S +++ b/arch/arm64/kernel/head.S @@ -354,7 +354,6 @@ __primary_switched: tst x23, ~(MIN_KIMG_ALIGN - 1) // already running randomized? b.ne 0f mov x0, x21 // pass FDT address in x0 - mov x1, x23 // pass modulo offset in x1 bl kaslr_early_init // parse FDT for KASLR options cbz x0, 0f // KASLR disabled? just proceed orr x23, x23, x0 // record KASLR offset diff --git a/arch/arm64/kernel/kaslr.c b/arch/arm64/kernel/kaslr.c index a9710ef..1d95c20 100644 --- a/arch/arm64/kernel/kaslr.c +++ b/arch/arm64/kernel/kaslr.c @@ -75,7 +75,7 @@ extern void *__init __fixmap_remap_fdt(phys_addr_t dt_phys, int *size, * containing function pointers) to be reinitialized, and zero-initialized * .bss variables will be reset to 0. */ -u64 __init kaslr_early_init(u64 dt_phys, u64 modulo_offset) +u64 __init kaslr_early_init(u64 dt_phys) { void *fdt; u64 seed, offset, mask, module_range; @@ -133,9 +133,15 @@ u64 __init kaslr_early_init(u64 dt_phys, u64 modulo_offset) * boundary (for 4KB/16KB/64KB granule kernels, respectively). If this * happens, increase the KASLR offset by the size of the kernel image * rounded up by SWAPPER_BLOCK_SIZE. + * + * NOTE: The references to _text and _end below will already take the + * modulo offset (the physical displacement modulo 2 MB) into + * account, given that the physical placement is controlled by + * the loader, and will not change as a result of the virtual + * mapping we choose. */ - if ((((u64)_text + offset + modulo_offset) >> SWAPPER_TABLE_SHIFT) != - (((u64)_end + offset + modulo_offset) >> SWAPPER_TABLE_SHIFT)) { + if ((((u64)_text + offset) >> SWAPPER_TABLE_SHIFT) != + (((u64)_end + offset) >> SWAPPER_TABLE_SHIFT)) { u64 kimg_sz = _end - _text; offset = (offset + round_up(kimg_sz, SWAPPER_BLOCK_SIZE)) & mask; -- cgit v1.1 From a067d94d37ed590fd17684d18c3edf52110d305a Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Tue, 22 Aug 2017 15:39:00 +0100 Subject: arm64: kaslr: Adjust the offset to avoid Image across alignment boundary With 16KB pages and a kernel Image larger than 16MB, the current kaslr_early_init() logic for avoiding mappings across swapper table boundaries fails since increasing the offset by kimg_sz just moves the problem to the next boundary. This patch rounds the offset down to (1 << SWAPPER_TABLE_SHIFT) if the Image crosses a PMD_SIZE boundary. Fixes: afd0e5a87670 ("arm64: kaslr: Fix up the kernel image alignment") Cc: Ard Biesheuvel Cc: Mark Rutland Cc: Will Deacon Cc: Neeraj Upadhyay Signed-off-by: Catalin Marinas Signed-off-by: Will Deacon --- arch/arm64/kernel/kaslr.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/arch/arm64/kernel/kaslr.c b/arch/arm64/kernel/kaslr.c index 1d95c20..47080c4 100644 --- a/arch/arm64/kernel/kaslr.c +++ b/arch/arm64/kernel/kaslr.c @@ -131,8 +131,7 @@ u64 __init kaslr_early_init(u64 dt_phys) /* * The kernel Image should not extend across a 1GB/32MB/512MB alignment * boundary (for 4KB/16KB/64KB granule kernels, respectively). If this - * happens, increase the KASLR offset by the size of the kernel image - * rounded up by SWAPPER_BLOCK_SIZE. + * happens, round down the KASLR offset by (1 << SWAPPER_TABLE_SHIFT). * * NOTE: The references to _text and _end below will already take the * modulo offset (the physical displacement modulo 2 MB) into @@ -141,11 +140,8 @@ u64 __init kaslr_early_init(u64 dt_phys) * mapping we choose. */ if ((((u64)_text + offset) >> SWAPPER_TABLE_SHIFT) != - (((u64)_end + offset) >> SWAPPER_TABLE_SHIFT)) { - u64 kimg_sz = _end - _text; - offset = (offset + round_up(kimg_sz, SWAPPER_BLOCK_SIZE)) - & mask; - } + (((u64)_end + offset) >> SWAPPER_TABLE_SHIFT)) + offset = round_down(offset, 1 << SWAPPER_TABLE_SHIFT); if (IS_ENABLED(CONFIG_KASAN)) /* -- cgit v1.1 From 3de33e1ba0506723ab25734e098cf280ecc34756 Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Fri, 18 Aug 2017 14:40:53 +0200 Subject: ipv6: accept 64k - 1 packet length in ip6_find_1stfragopt() A packet length of exactly IPV6_MAXPLEN is allowed, we should refuse parsing options only if the size is 64KiB or more. While at it, remove one extra variable and one assignment which were also introduced by the commit that introduced the size check. Checking the sum 'offset + len' and only later adding 'len' to 'offset' doesn't provide any advantage over directly summing to 'offset' and checking it. Fixes: 6399f1fae4ec ("ipv6: avoid overflow of offset in ip6_find_1stfragopt") Signed-off-by: Stefano Brivio Signed-off-by: David S. Miller --- net/ipv6/output_core.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index abb2c30..a338bbc 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -86,7 +86,6 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) while (offset <= packet_len) { struct ipv6_opt_hdr *exthdr; - unsigned int len; switch (**nexthdr) { @@ -112,10 +111,9 @@ int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) exthdr = (struct ipv6_opt_hdr *)(skb_network_header(skb) + offset); - len = ipv6_optlen(exthdr); - if (len + offset >= IPV6_MAXPLEN) + offset += ipv6_optlen(exthdr); + if (offset > IPV6_MAXPLEN) return -EINVAL; - offset += len; *nexthdr = &exthdr->nexthdr; } -- cgit v1.1 From 65159c051c45f269cf40a14f9404248f2d524920 Mon Sep 17 00:00:00 2001 From: Bharat Potnuri Date: Tue, 1 Aug 2017 10:58:35 +0530 Subject: RDMA/uverbs: Initialize cq_context appropriately Initializing cq_context with ev_queue in create_cq(), leads to NULL pointer dereference in ib_uverbs_comp_handler(), if application doesnot use completion channel. This patch fixes the cq_context initialization. Fixes: 1e7710f3f65 ("IB/core: Change completion channel to use the reworked") Cc: stable@vger.kernel.org # 4.12 Signed-off-by: Potnuri Bharat Teja Reviewed-by: Matan Barak Signed-off-by: Doug Ledford (cherry picked from commit 699a2d5b1b880b4e4e1c7d55fa25659322cf5b51) --- drivers/infiniband/core/uverbs_cmd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index c551d2b..55822ae 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1015,7 +1015,7 @@ static struct ib_ucq_object *create_cq(struct ib_uverbs_file *file, cq->uobject = &obj->uobject; cq->comp_handler = ib_uverbs_comp_handler; cq->event_handler = ib_uverbs_cq_event_handler; - cq->cq_context = &ev_file->ev_queue; + cq->cq_context = ev_file ? &ev_file->ev_queue : NULL; atomic_set(&cq->usecnt, 0); obj->uobject.object = cq; -- cgit v1.1 From c5cff8561d2d0006e972bd114afd51f082fee77c Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Mon, 21 Aug 2017 09:47:10 -0700 Subject: ipv6: add rcu grace period before freeing fib6_node We currently keep rt->rt6i_node pointing to the fib6_node for the route. And some functions make use of this pointer to dereference the fib6_node from rt structure, e.g. rt6_check(). However, as there is neither refcount nor rcu taken when dereferencing rt->rt6i_node, it could potentially cause crashes as rt->rt6i_node could be set to NULL by other CPUs when doing a route deletion. This patch introduces an rcu grace period before freeing fib6_node and makes sure the functions that dereference it takes rcu_read_lock(). Note: there is no "Fixes" tag because this bug was there in a very early stage. Signed-off-by: Wei Wang Acked-by: Eric Dumazet Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 30 +++++++++++++++++++++++++++++- net/ipv6/ip6_fib.c | 20 ++++++++++++++++---- net/ipv6/route.c | 14 +++++++++++--- 3 files changed, 56 insertions(+), 8 deletions(-) diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 1a88008..e9c59db 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -70,6 +70,7 @@ struct fib6_node { __u16 fn_flags; int fn_sernum; struct rt6_info *rr_ptr; + struct rcu_head rcu; }; #ifndef CONFIG_IPV6_SUBTREES @@ -167,13 +168,40 @@ static inline void rt6_update_expires(struct rt6_info *rt0, int timeout) rt0->rt6i_flags |= RTF_EXPIRES; } +/* Function to safely get fn->sernum for passed in rt + * and store result in passed in cookie. + * Return true if we can get cookie safely + * Return false if not + */ +static inline bool rt6_get_cookie_safe(const struct rt6_info *rt, + u32 *cookie) +{ + struct fib6_node *fn; + bool status = false; + + rcu_read_lock(); + fn = rcu_dereference(rt->rt6i_node); + + if (fn) { + *cookie = fn->fn_sernum; + status = true; + } + + rcu_read_unlock(); + return status; +} + static inline u32 rt6_get_cookie(const struct rt6_info *rt) { + u32 cookie = 0; + if (rt->rt6i_flags & RTF_PCPU || (unlikely(!list_empty(&rt->rt6i_uncached)) && rt->dst.from)) rt = (struct rt6_info *)(rt->dst.from); - return rt->rt6i_node ? rt->rt6i_node->fn_sernum : 0; + rt6_get_cookie_safe(rt, &cookie); + + return cookie; } static inline void ip6_rt_put(struct rt6_info *rt) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 5cc0ea0..a5ebf86 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -148,11 +148,23 @@ static struct fib6_node *node_alloc(void) return fn; } -static void node_free(struct fib6_node *fn) +static void node_free_immediate(struct fib6_node *fn) +{ + kmem_cache_free(fib6_node_kmem, fn); +} + +static void node_free_rcu(struct rcu_head *head) { + struct fib6_node *fn = container_of(head, struct fib6_node, rcu); + kmem_cache_free(fib6_node_kmem, fn); } +static void node_free(struct fib6_node *fn) +{ + call_rcu(&fn->rcu, node_free_rcu); +} + static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) { int cpu; @@ -601,9 +613,9 @@ insert_above: if (!in || !ln) { if (in) - node_free(in); + node_free_immediate(in); if (ln) - node_free(ln); + node_free_immediate(ln); return ERR_PTR(-ENOMEM); } @@ -1038,7 +1050,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, root, and then (in failure) stale node in main tree. */ - node_free(sfn); + node_free_immediate(sfn); err = PTR_ERR(sn); goto failure; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 94d6a13..a9d3564 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1289,7 +1289,9 @@ static void rt6_dst_from_metrics_check(struct rt6_info *rt) static struct dst_entry *rt6_check(struct rt6_info *rt, u32 cookie) { - if (!rt->rt6i_node || (rt->rt6i_node->fn_sernum != cookie)) + u32 rt_cookie; + + if (!rt6_get_cookie_safe(rt, &rt_cookie) || rt_cookie != cookie) return NULL; if (rt6_check_expired(rt)) @@ -1357,8 +1359,14 @@ static void ip6_link_failure(struct sk_buff *skb) if (rt->rt6i_flags & RTF_CACHE) { if (dst_hold_safe(&rt->dst)) ip6_del_rt(rt); - } else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT)) { - rt->rt6i_node->fn_sernum = -1; + } else { + struct fib6_node *fn; + + rcu_read_lock(); + fn = rcu_dereference(rt->rt6i_node); + if (fn && (rt->rt6i_flags & RTF_DEFAULT)) + fn->fn_sernum = -1; + rcu_read_unlock(); } } } -- cgit v1.1 From 414e7d76af6d3ec2dd4e9079927dbe0e2e4ca914 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 21 Aug 2017 12:59:10 -0700 Subject: net/hsr: Check skb_put_padto() return value skb_put_padto() will free the sk_buff passed as reference in case of errors, but we still need to check its return value and decide what to do. Detected by CoverityScan, CID#1416688 ("CHECKED_RETURN") Fixes: ee1c27977284 ("net/hsr: Added support for HSR v1") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- net/hsr/hsr_device.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index 4e7bdb2..172d830 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -314,7 +314,8 @@ static void send_hsr_supervision_frame(struct hsr_port *master, hsr_sp = skb_put(skb, sizeof(struct hsr_sup_payload)); ether_addr_copy(hsr_sp->MacAddressA, master->dev->dev_addr); - skb_put_padto(skb, ETH_ZLEN + HSR_HLEN); + if (skb_put_padto(skb, ETH_ZLEN + HSR_HLEN)) + return; hsr_forward_skb(skb, master); return; -- cgit v1.1 From 9b40eebcd339c47921ff8b04c77af7c762b74216 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 22 Aug 2017 22:44:13 +0200 Subject: ACPICA: Fix acpi_evaluate_object_typed() Commit 2d2a954375a0 (ACPICA: Update two error messages to emit control method name) causes acpi_evaluate_object_typed() to fail if its pathname argument is NULL, but some callers of that function in the kernel, particularly acpi_nondev_subnode_data_ok(), pass NULL as pathname to it and expect it to work. For this reason, make acpi_evaluate_object_typed() check if its pathname argument is NULL and fall back to using the pathname of its handle argument if that is the case. Reported-by: Sakari Ailus Tested-by: Yang, Hyungwoo Fixes: 2d2a954375a0 (ACPICA: Update two error messages to emit control method name) Signed-off-by: Rafael J. Wysocki --- drivers/acpi/acpica/nsxfeval.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/acpi/acpica/nsxfeval.c b/drivers/acpi/acpica/nsxfeval.c index 538c616..783f4c8 100644 --- a/drivers/acpi/acpica/nsxfeval.c +++ b/drivers/acpi/acpica/nsxfeval.c @@ -100,9 +100,13 @@ acpi_evaluate_object_typed(acpi_handle handle, free_buffer_on_error = TRUE; } - status = acpi_get_handle(handle, pathname, &target_handle); - if (ACPI_FAILURE(status)) { - return_ACPI_STATUS(status); + if (pathname) { + status = acpi_get_handle(handle, pathname, &target_handle); + if (ACPI_FAILURE(status)) { + return_ACPI_STATUS(status); + } + } else { + target_handle = handle; } full_pathname = acpi_ns_get_external_pathname(target_handle); -- cgit v1.1 From b5212f57da145e53df790a7e211d94daac768bf8 Mon Sep 17 00:00:00 2001 From: Sakari Ailus Date: Tue, 22 Aug 2017 23:39:58 +0300 Subject: ACPI: device property: Fix node lookup in acpi_graph_get_child_prop_value() acpi_graph_get_child_prop_value() is intended to find a child node with a certain property value pair. The check if (!fwnode_property_read_u32(fwnode, prop_name, &nr)) continue; is faulty: fwnode_property_read_u32() returns zero on success, not on failure, leading to comparing values only if the searched property was not found. Moreover, the check is made against the parent device node instead of the child one as it should be. Fixes: 79389a83bc38 (ACPI / property: Add support for remote endpoints) Reported-by: Hyungwoo Yang Signed-off-by: Sakari Ailus Cc: 4.12+ # 4.12+ [ rjw: Changelog ] Signed-off-by: Rafael J. Wysocki --- drivers/acpi/property.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/property.c b/drivers/acpi/property.c index 917c789..476a52c 100644 --- a/drivers/acpi/property.c +++ b/drivers/acpi/property.c @@ -1047,7 +1047,7 @@ static struct fwnode_handle *acpi_graph_get_child_prop_value( fwnode_for_each_child_node(fwnode, child) { u32 nr; - if (!fwnode_property_read_u32(fwnode, prop_name, &nr)) + if (fwnode_property_read_u32(child, prop_name, &nr)) continue; if (val == nr) -- cgit v1.1 From 5160a153a00ef24dc3a47c215e21fbc197f60887 Mon Sep 17 00:00:00 2001 From: Andrew Jeffery Date: Tue, 22 Aug 2017 16:06:22 +0930 Subject: net: ftgmac100: Fix oops in probe on failure to find associated PHY netif_napi_del() should be paired with netif_napi_add(), however no such call takes place in ftgmac100_probe(). This triggers a NULL pointer dereference if e.g. no PHY is found by the MDIO probe: [ 2.770000] libphy: Fixed MDIO Bus: probed [ 2.770000] ftgmac100 1e660000.ethernet: Generated random MAC address 66:58:c0:5a:50:b8 [ 2.790000] libphy: ftgmac100_mdio: probed [ 2.790000] ftgmac100 1e660000.ethernet (unnamed net_device) (uninitialized): eth%d: no PHY found [ 2.790000] ftgmac100 1e660000.ethernet: MII Probe failed! [ 2.810000] Unable to handle kernel NULL pointer dereference at virtual address 00000004 [ 2.810000] pgd = 80004000 [ 2.810000] [00000004] *pgd=00000000 [ 2.810000] Internal error: Oops: 805 [#1] ARM [ 2.810000] CPU: 0 PID: 1 Comm: swapper Not tainted 4.10.17-1a4df30c39cf5ee0e3d2528c409787ccbb4a672a #1 [ 2.810000] Hardware name: ASpeed SoC [ 2.810000] task: 9e421b60 task.stack: 9e4a0000 [ 2.810000] PC is at netif_napi_del+0x74/0xa4 [ 2.810000] LR is at ftgmac100_probe+0x290/0x674 [ 2.810000] pc : [<80331004>] lr : [<80292b30>] psr: 60000013 [ 2.810000] sp : 9e4a1d70 ip : 9e4a1d88 fp : 9e4a1d84 [ 2.810000] r10: 9e565000 r9 : ffffffed r8 : 00000007 [ 2.810000] r7 : 9e565480 r6 : 9ec072c0 r5 : 00000000 r4 : 9e5654d8 [ 2.810000] r3 : 9e565530 r2 : 00000000 r1 : 00000000 r0 : 9e5654d8 [ 2.810000] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 2.810000] Control: 00c5387d Table: 80004008 DAC: 00000055 [ 2.810000] Process swapper (pid: 1, stack limit = 0x9e4a0188) [ 2.810000] Stack: (0x9e4a1d70 to 0x9e4a2000) [ 2.810000] 1d60: 9e565000 9e549e10 9e4a1dcc 9e4a1d88 [ 2.810000] 1d80: 80292b30 80330f9c ffffffff 9e4a1d98 80146058 9ec072c0 00009e10 00000000 [ 2.810000] 1da0: 9e549e18 9e549e10 ffffffed 805f81f4 fffffdfb 00000000 00000000 00000000 [ 2.810000] 1dc0: 9e4a1dec 9e4a1dd0 80243df8 802928ac 9e549e10 8062cbd8 8062cbe0 805f81f4 [ 2.810000] 1de0: 9e4a1e24 9e4a1df0 80242178 80243da4 803001d0 802ffa60 9e4a1e24 9e549e10 [ 2.810000] 1e00: 9e549e44 805f81f4 00000000 00000000 805b8840 8058a6b0 9e4a1e44 9e4a1e28 [ 2.810000] 1e20: 80242434 80241f04 00000000 805f81f4 80242344 00000000 9e4a1e6c 9e4a1e48 [ 2.810000] 1e40: 80240148 80242350 9e425bac 9e4fdc90 9e790e94 805f81f4 9e790e60 805f5640 [ 2.810000] 1e60: 9e4a1e7c 9e4a1e70 802425dc 802400d8 9e4a1ea4 9e4a1e80 80240ba8 802425c0 [ 2.810000] 1e80: 8050b6ac 9e4a1e90 805f81f4 ffffe000 805b8838 80616720 9e4a1ebc 9e4a1ea8 [ 2.810000] 1ea0: 80243068 80240a68 805ab24c ffffe000 9e4a1ecc 9e4a1ec0 80244a38 80242fec [ 2.810000] 1ec0: 9e4a1edc 9e4a1ed0 805ab264 80244a04 9e4a1f4c 9e4a1ee0 8058ae70 805ab258 [ 2.810000] 1ee0: 80032c68 801e3fd8 8052f800 8041af2c 9e4a1f4c 9e4a1f00 80032f90 8058a6bc [ 2.810000] 1f00: 9e4a1f2c 9e4a1f10 00000006 00000006 00000000 8052f220 805112f0 00000000 [ 2.810000] 1f20: 9e4a1f4c 00000006 80616720 805cf400 80616720 805b8838 80616720 00000057 [ 2.810000] 1f40: 9e4a1f94 9e4a1f50 8058b040 8058add0 00000006 00000006 00000000 8058a6b0 [ 2.810000] 1f60: 3940bf3d 00000007 f115c2e8 00000000 803fd158 00000000 00000000 00000000 [ 2.810000] 1f80: 00000000 00000000 9e4a1fac 9e4a1f98 803fd170 8058af38 00000000 803fd158 [ 2.810000] 1fa0: 00000000 9e4a1fb0 8000a5e8 803fd164 00000000 00000000 00000000 00000000 [ 2.810000] 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 [ 2.810000] 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000 d11dcae8 af8ddec5 [ 2.810000] [<80331004>] (netif_napi_del) from [<80292b30>] (ftgmac100_probe+0x290/0x674) [ 2.810000] [<80292b30>] (ftgmac100_probe) from [<80243df8>] (platform_drv_probe+0x60/0xc0) [ 2.810000] [<80243df8>] (platform_drv_probe) from [<80242178>] (driver_probe_device+0x280/0x44c) [ 2.810000] [<80242178>] (driver_probe_device) from [<80242434>] (__driver_attach+0xf0/0x104) [ 2.810000] [<80242434>] (__driver_attach) from [<80240148>] (bus_for_each_dev+0x7c/0xb0) [ 2.810000] [<80240148>] (bus_for_each_dev) from [<802425dc>] (driver_attach+0x28/0x30) [ 2.810000] [<802425dc>] (driver_attach) from [<80240ba8>] (bus_add_driver+0x14c/0x268) [ 2.810000] [<80240ba8>] (bus_add_driver) from [<80243068>] (driver_register+0x88/0x104) [ 2.810000] [<80243068>] (driver_register) from [<80244a38>] (__platform_driver_register+0x40/0x54) [ 2.810000] [<80244a38>] (__platform_driver_register) from [<805ab264>] (ftgmac100_driver_init+0x18/0x20) [ 2.810000] [<805ab264>] (ftgmac100_driver_init) from [<8058ae70>] (do_one_initcall+0xac/0x168) [ 2.810000] [<8058ae70>] (do_one_initcall) from [<8058b040>] (kernel_init_freeable+0x114/0x1cc) [ 2.810000] [<8058b040>] (kernel_init_freeable) from [<803fd170>] (kernel_init+0x18/0x104) [ 2.810000] [<803fd170>] (kernel_init) from [<8000a5e8>] (ret_from_fork+0x14/0x2c) [ 2.810000] Code: e594205c e5941058 e2843058 e3a05000 (e5812004) [ 3.210000] ---[ end trace f32811052fd3860c ]--- Signed-off-by: Andrew Jeffery Acked-by: Benjamin Herrenschmidt Signed-off-by: David S. Miller --- drivers/net/ethernet/faraday/ftgmac100.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 34dae51..59da7ac 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -1863,7 +1863,6 @@ err_setup_mdio: err_ioremap: release_resource(priv->res); err_req_mem: - netif_napi_del(&priv->napi); free_netdev(netdev); err_alloc_etherdev: return err; -- cgit v1.1 From 4eb6a3bdb4629b610a39dd222a0170c72ef1c690 Mon Sep 17 00:00:00 2001 From: Nogah Frankel Date: Tue, 22 Aug 2017 10:28:11 +0200 Subject: mlxsw: spectrum_switchdev: Fix mrouter flag update Update the value of the mrouter flag in struct mlxsw_sp_bridge_port when it is being changed. Fixes: c57529e1d5d8 ("mlxsw: spectrum: Replace vPorts with Port-VLAN") Signed-off-by: Nogah Frankel Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 5eb1606..d39ffbf 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -705,6 +705,7 @@ static int mlxsw_sp_port_attr_mc_router_set(struct mlxsw_sp_port *mlxsw_sp_port, bool is_port_mc_router) { struct mlxsw_sp_bridge_port *bridge_port; + int err; if (switchdev_trans_ph_prepare(trans)) return 0; @@ -715,11 +716,17 @@ static int mlxsw_sp_port_attr_mc_router_set(struct mlxsw_sp_port *mlxsw_sp_port, return 0; if (!bridge_port->bridge_device->multicast_enabled) - return 0; + goto out; - return mlxsw_sp_bridge_port_flood_table_set(mlxsw_sp_port, bridge_port, - MLXSW_SP_FLOOD_TYPE_MC, - is_port_mc_router); + err = mlxsw_sp_bridge_port_flood_table_set(mlxsw_sp_port, bridge_port, + MLXSW_SP_FLOOD_TYPE_MC, + is_port_mc_router); + if (err) + return err; + +out: + bridge_port->mrouter = is_port_mc_router; + return 0; } static int mlxsw_sp_port_mc_disabled_set(struct mlxsw_sp_port *mlxsw_sp_port, -- cgit v1.1 From 458be024ef5f72f1697d5d55289c4ed45f8ee910 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 22 Aug 2017 12:28:40 +0200 Subject: tipc: remove subscription references only for pending timers In commit, 139bb36f754a ("tipc: advance the time of deleting subscription from subscriber->subscrp_list"), we delete the subscription from the subscribers list and from nametable unconditionally. This leads to the following bug if the timer running tipc_subscrp_timeout() in another CPU accesses the subscription list after the subscription delete request. [39.570] general protection fault: 0000 [#1] SMP :: [39.574] task: ffffffff81c10540 task.stack: ffffffff81c00000 [39.575] RIP: 0010:tipc_subscrp_timeout+0x32/0x80 [tipc] [39.576] RSP: 0018:ffff88003ba03e90 EFLAGS: 00010282 [39.576] RAX: dead000000000200 RBX: ffff88003f0f3600 RCX: 0000000000000101 [39.577] RDX: dead000000000100 RSI: 0000000000000201 RDI: ffff88003f0d7948 [39.578] RBP: ffff88003ba03ea0 R08: 0000000000000001 R09: ffff88003ba03ef8 [39.579] R10: 000000000000014f R11: 0000000000000000 R12: ffff88003f0d7948 [39.580] R13: ffff88003f0f3618 R14: ffffffffa006c250 R15: ffff88003f0f3600 [39.581] FS: 0000000000000000(0000) GS:ffff88003ba00000(0000) knlGS:0000000000000000 [39.582] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [39.583] CR2: 00007f831c6e0714 CR3: 000000003d3b0000 CR4: 00000000000006f0 [39.584] Call Trace: [39.584] [39.585] call_timer_fn+0x3d/0x180 [39.585] ? tipc_subscrb_rcv_cb+0x260/0x260 [tipc] [39.586] run_timer_softirq+0x168/0x1f0 [39.586] ? sched_clock_cpu+0x16/0xc0 [39.587] __do_softirq+0x9b/0x2de [39.587] irq_exit+0x60/0x70 [39.588] smp_apic_timer_interrupt+0x3d/0x50 [39.588] apic_timer_interrupt+0x86/0x90 [39.589] RIP: 0010:default_idle+0x20/0xf0 [39.589] RSP: 0018:ffffffff81c03e58 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff10 [39.590] RAX: 0000000000000000 RBX: ffffffff81c10540 RCX: 0000000000000000 [39.591] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 [39.592] RBP: ffffffff81c03e68 R08: 0000000000000000 R09: 0000000000000000 [39.593] R10: ffffc90001cbbe00 R11: 0000000000000000 R12: 0000000000000000 [39.594] R13: ffffffff81c10540 R14: 0000000000000000 R15: 0000000000000000 [39.595] :: [39.603] RIP: tipc_subscrp_timeout+0x32/0x80 [tipc] RSP: ffff88003ba03e90 [39.604] ---[ end trace 79ce94b7216cb459 ]--- Fixes: 139bb36f754a ("tipc: advance the time of deleting subscription from subscriber->subscrp_list") Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/subscr.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index 0bf91cd..f2c81f4 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -52,7 +52,6 @@ struct tipc_subscriber { struct list_head subscrp_list; }; -static void tipc_subscrp_delete(struct tipc_subscription *sub); static void tipc_subscrb_put(struct tipc_subscriber *subscriber); /** @@ -197,15 +196,19 @@ static void tipc_subscrb_subscrp_delete(struct tipc_subscriber *subscriber, { struct list_head *subscription_list = &subscriber->subscrp_list; struct tipc_subscription *sub, *temp; + u32 timeout; spin_lock_bh(&subscriber->lock); list_for_each_entry_safe(sub, temp, subscription_list, subscrp_list) { if (s && memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) continue; - tipc_nametbl_unsubscribe(sub); - list_del(&sub->subscrp_list); - tipc_subscrp_delete(sub); + timeout = htohl(sub->evt.s.timeout, sub->swap); + if (timeout == TIPC_WAIT_FOREVER || del_timer(&sub->timer)) { + tipc_nametbl_unsubscribe(sub); + list_del(&sub->subscrp_list); + tipc_subscrp_put(sub); + } if (s) break; @@ -236,14 +239,6 @@ static void tipc_subscrb_delete(struct tipc_subscriber *subscriber) tipc_subscrb_put(subscriber); } -static void tipc_subscrp_delete(struct tipc_subscription *sub) -{ - u32 timeout = htohl(sub->evt.s.timeout, sub->swap); - - if (timeout == TIPC_WAIT_FOREVER || del_timer(&sub->timer)) - tipc_subscrp_put(sub); -} - static void tipc_subscrp_cancel(struct tipc_subscr *s, struct tipc_subscriber *subscriber) { -- cgit v1.1 From fd849b7c41f0fabfe783d0691a63c5518e8ebc99 Mon Sep 17 00:00:00 2001 From: Ying Xue Date: Tue, 22 Aug 2017 12:28:41 +0200 Subject: tipc: fix a race condition of releasing subscriber object No matter whether a request is inserted into workqueue as a work item to cancel a subscription or to delete a subscription's subscriber asynchronously, the work items may be executed in different workers. As a result, it doesn't mean that one request which is raised prior to another request is definitely handled before the latter. By contrast, if the latter request is executed before the former request, below error may happen: [ 656.183644] BUG: spinlock bad magic on CPU#0, kworker/u8:0/12117 [ 656.184487] general protection fault: 0000 [#1] SMP [ 656.185160] Modules linked in: tipc ip6_udp_tunnel udp_tunnel 9pnet_virtio 9p 9pnet virtio_net virtio_pci virtio_ring virtio [last unloaded: ip6_udp_tunnel] [ 656.187003] CPU: 0 PID: 12117 Comm: kworker/u8:0 Not tainted 4.11.0-rc7+ #6 [ 656.187920] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 [ 656.188690] Workqueue: tipc_rcv tipc_recv_work [tipc] [ 656.189371] task: ffff88003f5cec40 task.stack: ffffc90004448000 [ 656.190157] RIP: 0010:spin_bug+0xdd/0xf0 [ 656.190678] RSP: 0018:ffffc9000444bcb8 EFLAGS: 00010202 [ 656.191375] RAX: 0000000000000034 RBX: ffff88003f8d1388 RCX: 0000000000000000 [ 656.192321] RDX: ffff88003ba13708 RSI: ffff88003ba0cd08 RDI: ffff88003ba0cd08 [ 656.193265] RBP: ffffc9000444bcd0 R08: 0000000000000030 R09: 000000006b6b6b6b [ 656.194208] R10: ffff8800bde3e000 R11: 00000000000001b4 R12: 6b6b6b6b6b6b6b6b [ 656.195157] R13: ffffffff81a3ca64 R14: ffff88003f8d1388 R15: ffff88003f8d13a0 [ 656.196101] FS: 0000000000000000(0000) GS:ffff88003ba00000(0000) knlGS:0000000000000000 [ 656.197172] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 656.197935] CR2: 00007f0b3d2e6000 CR3: 000000003ef9e000 CR4: 00000000000006f0 [ 656.198873] Call Trace: [ 656.199210] do_raw_spin_lock+0x66/0xa0 [ 656.199735] _raw_spin_lock_bh+0x19/0x20 [ 656.200258] tipc_subscrb_subscrp_delete+0x28/0xf0 [tipc] [ 656.200990] tipc_subscrb_rcv_cb+0x45/0x260 [tipc] [ 656.201632] tipc_receive_from_sock+0xaf/0x100 [tipc] [ 656.202299] tipc_recv_work+0x2b/0x60 [tipc] [ 656.202872] process_one_work+0x157/0x420 [ 656.203404] worker_thread+0x69/0x4c0 [ 656.203898] kthread+0x138/0x170 [ 656.204328] ? process_one_work+0x420/0x420 [ 656.204889] ? kthread_create_on_node+0x40/0x40 [ 656.205527] ret_from_fork+0x29/0x40 [ 656.206012] Code: 48 8b 0c 25 00 c5 00 00 48 c7 c7 f0 24 a3 81 48 81 c1 f0 05 00 00 65 8b 15 61 ef f5 7e e8 9a 4c 09 00 4d 85 e4 44 8b 4b 08 74 92 <45> 8b 84 24 40 04 00 00 49 8d 8c 24 f0 05 00 00 eb 8d 90 0f 1f [ 656.208504] RIP: spin_bug+0xdd/0xf0 RSP: ffffc9000444bcb8 [ 656.209798] ---[ end trace e2a800e6eb0770be ]--- In above scenario, the request of deleting subscriber was performed earlier than the request of canceling a subscription although the latter was issued before the former, which means tipc_subscrb_delete() was called before tipc_subscrp_cancel(). As a result, when tipc_subscrb_subscrp_delete() called by tipc_subscrp_cancel() was executed to cancel a subscription, the subscription's subscriber refcnt had been decreased to 1. After tipc_subscrp_delete() where the subscriber was freed because its refcnt was decremented to zero, but the subscriber's lock had to be released, as a consequence, panic happened. By contrast, if we increase subscriber's refcnt before tipc_subscrb_subscrp_delete() is called in tipc_subscrp_cancel(), the panic issue can be avoided. Fixes: d094c4d5f5c7 ("tipc: add subscription refcount to avoid invalid delete") Reported-by: Parthasarathy Bhuvaragan Signed-off-by: Ying Xue Signed-off-by: David S. Miller --- net/tipc/subscr.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index f2c81f4..be3d9e3 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -242,7 +242,9 @@ static void tipc_subscrb_delete(struct tipc_subscriber *subscriber) static void tipc_subscrp_cancel(struct tipc_subscr *s, struct tipc_subscriber *subscriber) { + tipc_subscrb_get(subscriber); tipc_subscrb_subscrp_delete(subscriber, s); + tipc_subscrb_put(subscriber); } static struct tipc_subscription *tipc_subscrp_create(struct net *net, -- cgit v1.1 From 78362998f58c7c271e2719dcd0aaced435c801f9 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 22 Aug 2017 15:36:08 +0200 Subject: macsec: add genl family module alias This helps tools such as wpa_supplicant can start even if the macsec module isn't loaded yet. Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: Sabrina Dubroca Signed-off-by: David S. Miller --- drivers/net/macsec.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 5e1ab11..98e4dea 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3521,6 +3521,7 @@ module_init(macsec_init); module_exit(macsec_exit); MODULE_ALIAS_RTNL_LINK("macsec"); +MODULE_ALIAS_GENL_FAMILY("macsec"); MODULE_DESCRIPTION("MACsec IEEE 802.1AE"); MODULE_LICENSE("GPL v2"); -- cgit v1.1 From fd6055a806edc4019be1b9fb7d25262599bca5b1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 22 Aug 2017 09:39:28 -0700 Subject: udp: on peeking bad csum, drop packets even if not at head When peeking, if a bad csum is discovered, the skb is unlinked from the queue with __sk_queue_drop_skb and the peek operation restarted. __sk_queue_drop_skb only drops packets that match the queue head. This fails if the skb was found after the head, using SO_PEEK_OFF socket option. This causes an infinite loop. We MUST drop this problematic skb, and we can simply check if skb was already removed by another thread, by looking at skb->next : This pointer is set to NULL by the __skb_unlink() operation, that might have happened only under the spinlock protection. Many thanks to syzkaller team (and particularly Dmitry Vyukov who provided us nice C reproducers exhibiting the lockup) and Willem de Bruijn who provided first version for this patch and a test program. Fixes: 627d2d6b5500 ("udp: enable MSG_PEEK at non-zero offset") Signed-off-by: Eric Dumazet Reported-by: Dmitry Vyukov Cc: Willem de Bruijn Acked-by: Paolo Abeni Acked-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/core/datagram.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/datagram.c b/net/core/datagram.c index a21ca8d..8c2f448 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -362,7 +362,7 @@ int __sk_queue_drop_skb(struct sock *sk, struct sk_buff_head *sk_queue, if (flags & MSG_PEEK) { err = -ENOENT; spin_lock_bh(&sk_queue->lock); - if (skb == skb_peek(sk_queue)) { + if (skb->next) { __skb_unlink(skb, sk_queue); refcount_dec(&skb->users); if (destructor) -- cgit v1.1 From 744a4cf63e528c29840f45811d6fb93fd129b87d Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 22 Aug 2017 22:46:49 +0200 Subject: net: sched: fix use after free when tcf_chain_destroy is called multiple times The goto_chain termination action takes a reference of a chain. In that case, there is an issue when block_put is called tcf_chain_destroy directly. The follo-up call of tcf_chain_put by goto_chain action free works with memory that is already freed. This was caught by kasan: [ 220.337908] BUG: KASAN: use-after-free in tcf_chain_put+0x1b/0x50 [ 220.344103] Read of size 4 at addr ffff88036d1f2cec by task systemd-journal/261 [ 220.353047] CPU: 0 PID: 261 Comm: systemd-journal Not tainted 4.13.0-rc5jiri+ #54 [ 220.360661] Hardware name: Mellanox Technologies Ltd. Mellanox switch/Mellanox x86 mezzanine board, BIOS 4.6.5 08/02/2016 [ 220.371784] Call Trace: [ 220.374290] [ 220.376355] dump_stack+0xd5/0x150 [ 220.391485] print_address_description+0x86/0x410 [ 220.396308] kasan_report+0x181/0x4c0 [ 220.415211] tcf_chain_put+0x1b/0x50 [ 220.418949] free_tcf+0x95/0xc0 So allow tcf_chain_destroy to be called multiple times, free only in case the reference count drops to 0. Fixes: 5bc1701881e3 ("net: sched: introduce multichain support for filters") Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- net/sched/cls_api.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 9fd44c2..45cd34e 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -215,9 +215,17 @@ static void tcf_chain_flush(struct tcf_chain *chain) static void tcf_chain_destroy(struct tcf_chain *chain) { - list_del(&chain->list); + /* May be already removed from the list by the previous call. */ + if (!list_empty(&chain->list)) + list_del_init(&chain->list); + tcf_chain_flush(chain); - kfree(chain); + + /* There might still be a reference held when we got here from + * tcf_block_put. Wait for the user to drop reference before free. + */ + if (!chain->refcnt) + kfree(chain); } struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, -- cgit v1.1 From 30d65e8f96ad01d9f998039e9af9ce5545e5a4ee Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 22 Aug 2017 22:46:50 +0200 Subject: net: sched: don't do tcf_chain_flush from tcf_chain_destroy tcf_chain_flush needs to be called with RTNL. However, on free_tcf-> tcf_action_goto_chain_fini-> tcf_chain_put-> tcf_chain_destroy-> tcf_chain_flush callpath, it is called without RTNL. This issue was notified by following warning: [ 155.599052] WARNING: suspicious RCU usage [ 155.603165] 4.13.0-rc5jiri+ #54 Not tainted [ 155.607456] ----------------------------- [ 155.611561] net/sched/cls_api.c:195 suspicious rcu_dereference_protected() usage! Since on this callpath, the chain is guaranteed to be already empty by check in tcf_chain_put, move the tcf_chain_flush call out and call it only where it is needed - into tcf_block_put. Fixes: db50514f9a9c ("net: sched: add termination action to allow goto chain") Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- net/sched/cls_api.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 45cd34e..6c5ea84 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -219,8 +219,6 @@ static void tcf_chain_destroy(struct tcf_chain *chain) if (!list_empty(&chain->list)) list_del_init(&chain->list); - tcf_chain_flush(chain); - /* There might still be a reference held when we got here from * tcf_block_put. Wait for the user to drop reference before free. */ @@ -296,8 +294,10 @@ void tcf_block_put(struct tcf_block *block) if (!block) return; - list_for_each_entry_safe(chain, tmp, &block->chain_list, list) + list_for_each_entry_safe(chain, tmp, &block->chain_list, list) { + tcf_chain_flush(chain); tcf_chain_destroy(chain); + } kfree(block); } EXPORT_SYMBOL(tcf_block_put); -- cgit v1.1 From fcd03e362b1cd17de487953aac34f2d4574895cf Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 22 Aug 2017 14:26:47 -0700 Subject: net: phy: Deal with unbound PHY driver in phy_attached_print() Priit reported that stmmac was crashing with the trace below. This is because phy_attached_print() is called too early right after the PHY device has been found, but before it has a driver attached, since that is only done in phy_probe() which occurs later. Fix this by dealing with a possibly NULL phydev->drv point since that can happen here, but could also happen if we voluntarily did an unbind of the PHY device with the PHY driver. sun7i-dwmac 1c50000.ethernet: PTP uses main clock sun7i-dwmac 1c50000.ethernet: no reset control found sun7i-dwmac 1c50000.ethernet: no regulator found sun7i-dwmac 1c50000.ethernet: Ring mode enabled sun7i-dwmac 1c50000.ethernet: DMA HW capability register supported sun7i-dwmac 1c50000.ethernet: Normal descriptors libphy: stmmac: probed Unable to handle kernel NULL pointer dereference at virtual address 00000048 pgd = c0004000 [00000048] *pgd=00000000 Internal error: Oops: 5 [#1] SMP ARM Modules linked in: CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.13.0-rc6-00318-g0065bd7fa384 #1 Hardware name: Allwinner sun7i (A20) Family task: ee868000 task.stack: ee85c000 PC is at phy_attached_print+0x1c/0x8c LR is at stmmac_mdio_register+0x12c/0x200 pc : [] lr : [] psr: 60000013 sp : ee85ddc8 ip : 00000000 fp : c07dfb5c r10: ee981210 r9 : 00000001 r8 : eea73000 r7 : eeaa6dd0 r6 : eeb49800 r5 : 00000000 r4 : 00000000 r3 : 00000000 r2 : 00000000 r1 : 00000000 r0 : eeb49800 Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: 4000406a DAC: 00000051 Process swapper/0 (pid: 1, stack limit = 0xee85c210) Stack: (0xee85ddc8 to 0xee85e000) ddc0: 00000000 00000002 eeb49400 eea72000 00000000 eeb49400 dde0: c045e6b4 00000000 ffffffff eeab0810 00000000 c08051f8 ee9292c0 c016d480 de00: eea725c0 eea73000 eea72000 00000001 eea726c0 c0457d0c 00000040 00000020 de20: 00000000 c045b850 00000001 00000000 ee981200 eeab0810 eeaa6ed0 ee981210 de40: 00000000 c094a4a0 00000000 c0465180 eeaa7550 f08d0000 c9ffb90c 00000032 de60: fffffffa 00000032 ee981210 ffffffed c0a46620 fffffdfb c0a46620 c03f7be8 de80: ee981210 c0a9a388 00000000 00000000 c0a46620 c03f63e0 ee981210 c0a46620 dea0: ee981244 00000000 00000007 000000c6 c094a4a0 c03f6534 00000000 c0a46620 dec0: c03f6490 c03f49ec ee828a58 ee9217b4 c0a46620 eeaa4b00 c0a43230 c03f59fc dee0: c08051f8 c094a49c c0a46620 c0a46620 00000000 c091c668 c093783c c03f6dfc df00: ffffe000 00000000 c091c668 c010177c eefe0938 eefe0935 c085e200 000000c6 df20: 00000005 c0136bc8 60000013 c080b3a4 00000006 00000006 c07ce7b4 00000000 df40: c07d7ddc c07cef28 eefe0938 eefe093e c0a0b2f0 c0a641c0 c0a641c0 c0a641c0 df60: c0937834 00000007 000000c6 c094a4a0 00000000 c0900d88 00000006 00000006 df80: 00000000 c09005a8 00000000 c060ecf4 00000000 00000000 00000000 00000000 dfa0: 00000000 c060ecfc 00000000 c0107738 00000000 00000000 00000000 00000000 dfc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 dfe0: 00000000 00000000 00000000 00000000 00000013 00000000 ffdeffff ffffffff [] (phy_attached_print) from [] (stmmac_mdio_register+0x12c/0x200) [] (stmmac_mdio_register) from [] (stmmac_dvr_probe+0x850/0x96c) [] (stmmac_dvr_probe) from [] (sun7i_gmac_probe+0x120/0x180) [] (sun7i_gmac_probe) from [] (platform_drv_probe+0x50/0xac) [] (platform_drv_probe) from [] (driver_probe_device+0x234/0x2e4) [] (driver_probe_device) from [] (__driver_attach+0xa4/0xa8) [] (__driver_attach) from [] (bus_for_each_dev+0x4c/0x9c) [] (bus_for_each_dev) from [] (bus_add_driver+0x190/0x214) [] (bus_add_driver) from [] (driver_register+0x78/0xf4) [] (driver_register) from [] (do_one_initcall+0x44/0x168) [] (do_one_initcall) from [] (kernel_init_freeable+0x144/0x1d0) [] (kernel_init_freeable) from [] (kernel_init+0x8/0x110) [] (kernel_init) from [] (ret_from_fork+0x14/0x3c) Code: e59021c8 e59d401c e590302c e3540000 (e5922048) ---[ end trace 39ae87c7923562d0 ]--- Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b Tested-By: Priit Laes Fixes: fbca164776e4 ("net: stmmac: Use the right logging function in stmmac_mdio_register") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 1790f7f..2f742ae 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -864,15 +864,17 @@ EXPORT_SYMBOL(phy_attached_info); #define ATTACHED_FMT "attached PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)" void phy_attached_print(struct phy_device *phydev, const char *fmt, ...) { + const char *drv_name = phydev->drv ? phydev->drv->name : "unbound"; + if (!fmt) { dev_info(&phydev->mdio.dev, ATTACHED_FMT "\n", - phydev->drv->name, phydev_name(phydev), + drv_name, phydev_name(phydev), phydev->irq); } else { va_list ap; dev_info(&phydev->mdio.dev, ATTACHED_FMT, - phydev->drv->name, phydev_name(phydev), + drv_name, phydev_name(phydev), phydev->irq); va_start(ap, fmt); -- cgit v1.1 From 33ba43ed0afc13a29b1314e3e45a9938d310ba13 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 23 Aug 2017 00:06:09 +0200 Subject: bpf: fix map value attribute for hash of maps Currently, iproute2's BPF ELF loader works fine with array of maps when retrieving the fd from a pinned node and doing a selfcheck against the provided map attributes from the object file, but we fail to do the same for hash of maps and thus refuse to get the map from pinned node. Reason is that when allocating hash of maps, fd_htab_map_alloc() will set the value size to sizeof(void *), and any user space map creation requests are forced to set 4 bytes as value size. Thus, selfcheck will complain about exposed 8 bytes on 64 bit archs vs. 4 bytes from object file as value size. Contract is that fdinfo or BPF_MAP_GET_FD_BY_ID returns the value size used to create the map. Fix it by handling it the same way as we do for array of maps, which means that we leave value size at 4 bytes and in the allocation phase round up value size to 8 bytes. alloc_htab_elem() needs an adjustment in order to copy rounded up 8 bytes due to bpf_fd_htab_map_update_elem() calling into htab_map_update_elem() with the pointer of the map pointer as value. Unlike array of maps where we just xchg(), we're using the generic htab_map_update_elem() callback also used from helper calls, which published the key/value already on return, so we need to ensure to memcpy() the right size. Fixes: bcc6b1b7ebf8 ("bpf: Add hash of maps support") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- kernel/bpf/hashtab.c | 30 +++++++++++++++++------------- 1 file changed, 17 insertions(+), 13 deletions(-) diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index 4fb4631..d11c818 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -652,12 +652,27 @@ static void pcpu_copy_value(struct bpf_htab *htab, void __percpu *pptr, } } +static bool fd_htab_map_needs_adjust(const struct bpf_htab *htab) +{ + return htab->map.map_type == BPF_MAP_TYPE_HASH_OF_MAPS && + BITS_PER_LONG == 64; +} + +static u32 htab_size_value(const struct bpf_htab *htab, bool percpu) +{ + u32 size = htab->map.value_size; + + if (percpu || fd_htab_map_needs_adjust(htab)) + size = round_up(size, 8); + return size; +} + static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, void *value, u32 key_size, u32 hash, bool percpu, bool onallcpus, struct htab_elem *old_elem) { - u32 size = htab->map.value_size; + u32 size = htab_size_value(htab, percpu); bool prealloc = htab_is_prealloc(htab); struct htab_elem *l_new, **pl_new; void __percpu *pptr; @@ -696,9 +711,6 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, memcpy(l_new->key, key, key_size); if (percpu) { - /* round up value_size to 8 bytes */ - size = round_up(size, 8); - if (prealloc) { pptr = htab_elem_get_ptr(l_new, key_size); } else { @@ -1209,17 +1221,9 @@ const struct bpf_map_ops htab_lru_percpu_map_ops = { static struct bpf_map *fd_htab_map_alloc(union bpf_attr *attr) { - struct bpf_map *map; - if (attr->value_size != sizeof(u32)) return ERR_PTR(-EINVAL); - - /* pointer is stored internally */ - attr->value_size = sizeof(void *); - map = htab_map_alloc(attr); - attr->value_size = sizeof(u32); - - return map; + return htab_map_alloc(attr); } static void fd_htab_map_free(struct bpf_map *map) -- cgit v1.1 From a1a50c8e4c241a505b7270e1a3c6e50d94e794b1 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 22 Aug 2017 15:24:47 -0700 Subject: fsl/man: Inherit parent device and of_node Junote Cai reported that he was not able to get a DSA setup involving the Freescale DPAA/FMAN driver to work and narrowed it down to of_find_net_device_by_node(). This function requires the network device's device reference to be correctly set which is the case here, though we have lost any device_node association there. The problem is that dpaa_eth_add_device() allocates a "dpaa-ethernet" platform device, and later on dpaa_eth_probe() is called but SET_NETDEV_DEV() won't be propagating &pdev->dev.of_node properly. Fix this by inherenting both the parent device and the of_node when dpaa_eth_add_device() creates the platform device. Fixes: 3933961682a3 ("fsl/fman: Add FMan MAC driver") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fman/mac.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/freescale/fman/mac.c b/drivers/net/ethernet/freescale/fman/mac.c index 6e67d22..1c7da16 100644 --- a/drivers/net/ethernet/freescale/fman/mac.c +++ b/drivers/net/ethernet/freescale/fman/mac.c @@ -623,6 +623,8 @@ static struct platform_device *dpaa_eth_add_device(int fman_id, goto no_mem; } + pdev->dev.of_node = node; + pdev->dev.parent = priv->dev; set_dma_ops(&pdev->dev, get_dma_ops(priv->dev)); ret = platform_device_add_data(pdev, &data, sizeof(data)); -- cgit v1.1 From b2a6d1b999a4c13e5997bb864694e77172d45250 Mon Sep 17 00:00:00 2001 From: Martijn Coenen Date: Fri, 28 Jul 2017 13:56:08 +0200 Subject: ANDROID: binder: fix proc->tsk check. Commit c4ea41ba195d ("binder: use group leader instead of open thread")' was incomplete and didn't update a check in binder_mmap(), causing all mmap() calls into the binder driver to fail. Signed-off-by: Martijn Coenen Tested-by: John Stultz Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/android/binder.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/android/binder.c b/drivers/android/binder.c index f7665c3..831cdd7 100644 --- a/drivers/android/binder.c +++ b/drivers/android/binder.c @@ -3362,7 +3362,7 @@ static int binder_mmap(struct file *filp, struct vm_area_struct *vma) const char *failure_string; struct binder_buffer *buffer; - if (proc->tsk != current) + if (proc->tsk != current->group_leader) return -EINVAL; if ((vma->vm_end - vma->vm_start) > SZ_4M) -- cgit v1.1 From 723cd772fde2344a9810eeaf5106787d535ec4a4 Mon Sep 17 00:00:00 2001 From: Brian King Date: Fri, 18 Aug 2017 16:17:32 -0500 Subject: scsi: ipr: Set no_report_opcodes for RAID arrays Since ipr RAID arrays do not support the MAINTENANCE_IN / MI_REPORT_SUPPORTED_OPERATION_CODES, set no_report_opcodes to prevent it from being sent. Signed-off-by: Brian King Signed-off-by: Martin K. Petersen --- drivers/scsi/ipr.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index da5bdbd..f838bd7 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -4945,6 +4945,7 @@ static int ipr_slave_configure(struct scsi_device *sdev) } if (ipr_is_vset_device(res)) { sdev->scsi_level = SCSI_SPC_3; + sdev->no_report_opcodes = 1; blk_queue_rq_timeout(sdev->request_queue, IPR_VSET_RW_TIMEOUT); blk_queue_max_hw_sectors(sdev->request_queue, IPR_VSET_MAX_SECTORS); -- cgit v1.1 From bd46fc406b30d1db1aff8dabaff8d18bb423fdcf Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 17 Aug 2017 10:09:54 +0300 Subject: scsi: sg: off by one in sg_ioctl() If "val" is SG_MAX_QUEUE then we are one element beyond the end of the "rinfo" array so the > should be >=. Fixes: 109bade9c625 ("scsi: sg: use standard lists for sg_requests") Signed-off-by: Dan Carpenter Acked-by: Douglas Gilbert Signed-off-by: Martin K. Petersen --- drivers/scsi/sg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index d7ff71e..84e782d 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1021,7 +1021,7 @@ sg_ioctl(struct file *filp, unsigned int cmd_in, unsigned long arg) read_lock_irqsave(&sfp->rq_list_lock, iflags); val = 0; list_for_each_entry(srp, &sfp->rq_list, entry) { - if (val > SG_MAX_QUEUE) + if (val >= SG_MAX_QUEUE) break; memset(&rinfo[val], 0, SZ_SG_REQ_INFO); rinfo[val].req_state = srp->done + 1; -- cgit v1.1 From ffeaf9aaf97b4bdaf114d6df52f800d71918768c Mon Sep 17 00:00:00 2001 From: fred gao Date: Wed, 16 Aug 2017 15:48:03 +0800 Subject: drm/i915/gvt: Fix the kernel null pointer error once error happens in shadow_indirect_ctx function, the variable wa_ctx->indirect_ctx.obj is not initialized but accessed, so the kernel null point panic occurs. Fixes: 894cf7d15634 ("drm/i915/gvt: i915_gem_object_create() returns an error pointer") Cc: stable@vger.kernel.org # v4.8+ Signed-off-by: fred gao Signed-off-by: Zhenyu Wang --- drivers/gpu/drm/i915/gvt/cmd_parser.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 713848c..e556a46 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -2714,7 +2714,7 @@ static int shadow_indirect_ctx(struct intel_shadow_wa_ctx *wa_ctx) unmap_src: i915_gem_object_unpin_map(obj); put_obj: - i915_gem_object_put(wa_ctx->indirect_ctx.obj); + i915_gem_object_put(obj); return ret; } -- cgit v1.1 From bbba6f9d3da357bbabc6fda81e99ff5584500e76 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 23 Aug 2017 09:30:17 +0200 Subject: ALSA: hda - Add stereo mic quirk for Lenovo G50-70 (17aa:3978) Lenovo G50-70 (17aa:3978) with Conexant codec chip requires the similar workaround for the inverted stereo dmic like other Lenovo models. Bugzilla: https://bugzilla.suse.com/show_bug.cgi?id=1020657 Cc: Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_conexant.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index 8c12899..a81aacf 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -947,6 +947,7 @@ static const struct snd_pci_quirk cxt5066_fixups[] = { SND_PCI_QUIRK(0x17aa, 0x390b, "Lenovo G50-80", CXT_FIXUP_STEREO_DMIC), SND_PCI_QUIRK(0x17aa, 0x3975, "Lenovo U300s", CXT_FIXUP_STEREO_DMIC), SND_PCI_QUIRK(0x17aa, 0x3977, "Lenovo IdeaPad U310", CXT_FIXUP_STEREO_DMIC), + SND_PCI_QUIRK(0x17aa, 0x3978, "Lenovo G50-70", CXT_FIXUP_STEREO_DMIC), SND_PCI_QUIRK(0x17aa, 0x397b, "Lenovo S205", CXT_FIXUP_STEREO_DMIC), SND_PCI_QUIRK_VENDOR(0x17aa, "Thinkpad", CXT_FIXUP_THINKPAD_ACPI), SND_PCI_QUIRK(0x1c06, 0x2011, "Lemote A1004", CXT_PINCFG_LEMOTE_A1004), -- cgit v1.1 From dbeb0c8e84805b78f3979b53a4a12751403e4520 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 23 Aug 2017 16:46:15 +0200 Subject: ARM: at91: don't select CONFIG_ARM_CPU_SUSPEND for old platforms My previous patch fixed a link error for all at91 platforms when CONFIG_ARM_CPU_SUSPEND was not set, however this caused another problem on a configuration that enabled CONFIG_ARCH_AT91 but none of the individual SoCs, and that also enabled CPU_ARM720 as the only CPU: warning: (ARCH_AT91 && SOC_IMX23 && SOC_IMX28 && ARCH_PXA && MACH_MVEBU_V7 && SOC_IMX6 && ARCH_OMAP3 && ARCH_OMAP4 && SOC_OMAP5 && SOC_AM33XX && SOC_DRA7XX && ARCH_EXYNOS3 && ARCH_EXYNOS4 && EXYNOS5420_MCPM && EXYNOS_CPU_SUSPEND && ARCH_VEXPRESS_TC2_PM && ARM_BIG_LITTLE_CPUIDLE && ARM_HIGHBANK_CPUIDLE && QCOM_PM) selects ARM_CPU_SUSPEND which has unmet direct dependencies (ARCH_SUSPEND_POSSIBLE) arch/arm/kernel/sleep.o: In function `cpu_resume': (.text+0xf0): undefined reference to `cpu_arm720_suspend_size' arch/arm/kernel/suspend.o: In function `__cpu_suspend_save': suspend.c:(.text+0x134): undefined reference to `cpu_arm720_do_suspend' This improves the hack some more by only selecting ARM_CPU_SUSPEND for the part that requires it, and changing pm.c to drop the contents of unused init functions so we no longer refer to cpu_resume on at91 platforms that don't need it. Fixes: cc7a938f5f30 ("ARM: at91: select CONFIG_ARM_CPU_SUSPEND") Acked-by: Alexandre Belloni Signed-off-by: Arnd Bergmann --- arch/arm/mach-at91/Kconfig | 2 +- arch/arm/mach-at91/pm.c | 12 ++++++++++++ 2 files changed, 13 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-at91/Kconfig b/arch/arm/mach-at91/Kconfig index d735e5f..195da38 100644 --- a/arch/arm/mach-at91/Kconfig +++ b/arch/arm/mach-at91/Kconfig @@ -1,7 +1,7 @@ menuconfig ARCH_AT91 bool "Atmel SoCs" depends on ARCH_MULTI_V4T || ARCH_MULTI_V5 || ARCH_MULTI_V7 || ARM_SINGLE_ARMV7M - select ARM_CPU_SUSPEND if PM + select ARM_CPU_SUSPEND if PM && ARCH_MULTI_V7 select COMMON_CLK_AT91 select GPIOLIB select PINCTRL diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c index 667fdda..5036f99 100644 --- a/arch/arm/mach-at91/pm.c +++ b/arch/arm/mach-at91/pm.c @@ -608,6 +608,9 @@ static void __init at91_pm_init(void (*pm_idle)(void)) void __init at91rm9200_pm_init(void) { + if (!IS_ENABLED(CONFIG_SOC_AT91RM9200)) + return; + at91_dt_ramc(); /* @@ -620,18 +623,27 @@ void __init at91rm9200_pm_init(void) void __init at91sam9_pm_init(void) { + if (!IS_ENABLED(CONFIG_SOC_AT91SAM9)) + return; + at91_dt_ramc(); at91_pm_init(at91sam9_idle); } void __init sama5_pm_init(void) { + if (!IS_ENABLED(CONFIG_SOC_SAMA5)) + return; + at91_dt_ramc(); at91_pm_init(NULL); } void __init sama5d2_pm_init(void) { + if (!IS_ENABLED(CONFIG_SOC_SAMA5D2)) + return; + at91_pm_backup_init(); sama5_pm_init(); } -- cgit v1.1 From 42bec214d8bd432be6d32a1acb0a9079ecd4d142 Mon Sep 17 00:00:00 2001 From: Sachin Prabhu Date: Thu, 3 Aug 2017 13:09:03 +0530 Subject: cifs: Fix df output for users with quota limits The df for a SMB2 share triggers a GetInfo call for FS_FULL_SIZE_INFORMATION. The values returned are used to populate struct statfs. The problem is that none of the information returned by the call contains the total blocks available on the filesystem. Instead we use the blocks available to the user ie. quota limitation when filling out statfs.f_blocks. The information returned does contain Actual free units on the filesystem and is used to populate statfs.f_bfree. For users with quota enabled, it can lead to situations where the total free space reported is more than the total blocks on the system ending up with df reports like the following # df -h /mnt/a Filesystem Size Used Avail Use% Mounted on //192.168.22.10/a 2.5G -2.3G 2.5G - /mnt/a To fix this problem, we instead populate both statfs.f_bfree with the same value as statfs.f_bavail ie. CallerAvailableAllocationUnits. This is similar to what is done already in the code for cifs and df now reports the quota information for the user used to mount the share. # df --si /mnt/a Filesystem Size Used Avail Use% Mounted on //192.168.22.10/a 2.7G 101M 2.6G 4% /mnt/a Signed-off-by: Sachin Prabhu Signed-off-by: Pierguido Lambri Signed-off-by: Steve French Cc: --- fs/cifs/smb2pdu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 5fb2fc2..97edb4d3 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -3219,8 +3219,8 @@ copy_fs_info_to_kstatfs(struct smb2_fs_full_size_info *pfs_inf, kst->f_bsize = le32_to_cpu(pfs_inf->BytesPerSector) * le32_to_cpu(pfs_inf->SectorsPerAllocationUnit); kst->f_blocks = le64_to_cpu(pfs_inf->TotalAllocationUnits); - kst->f_bfree = le64_to_cpu(pfs_inf->ActualAvailableAllocationUnits); - kst->f_bavail = le64_to_cpu(pfs_inf->CallerAvailableAllocationUnits); + kst->f_bfree = kst->f_bavail = + le64_to_cpu(pfs_inf->CallerAvailableAllocationUnits); return; } -- cgit v1.1 From d3edede29f74d335f81d95a4588f5f136a9f7dcf Mon Sep 17 00:00:00 2001 From: Ronnie Sahlberg Date: Wed, 23 Aug 2017 14:48:14 +1000 Subject: cifs: return ENAMETOOLONG for overlong names in cifs_open()/cifs_lookup() Add checking for the path component length and verify it is <= the maximum that the server advertizes via FileFsAttributeInformation. With this patch cifs.ko will now return ENAMETOOLONG instead of ENOENT when users to access an overlong path. To test this, try to cd into a (non-existing) directory on a CIFS share that has a too long name: cd /mnt/aaaaaaaaaaaaaaa... and it now should show a good error message from the shell: bash: cd: /mnt/aaaaaaaaaaaaaaaa...aaaaaa: File name too long rh bz 1153996 Signed-off-by: Ronnie Sahlberg Signed-off-by: Steve French Cc: --- fs/cifs/dir.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 56366e9..569d3fb 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -194,15 +194,20 @@ cifs_bp_rename_retry: } /* + * Don't allow path components longer than the server max. * Don't allow the separator character in a path component. * The VFS will not allow "/", but "\" is allowed by posix. */ static int -check_name(struct dentry *direntry) +check_name(struct dentry *direntry, struct cifs_tcon *tcon) { struct cifs_sb_info *cifs_sb = CIFS_SB(direntry->d_sb); int i; + if (unlikely(direntry->d_name.len > + tcon->fsAttrInfo.MaxPathNameComponentLength)) + return -ENAMETOOLONG; + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS)) { for (i = 0; i < direntry->d_name.len; i++) { if (direntry->d_name.name[i] == '\\') { @@ -500,10 +505,6 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, return finish_no_open(file, res); } - rc = check_name(direntry); - if (rc) - return rc; - xid = get_xid(); cifs_dbg(FYI, "parent inode = 0x%p name is: %pd and dentry = 0x%p\n", @@ -516,6 +517,11 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, } tcon = tlink_tcon(tlink); + + rc = check_name(direntry, tcon); + if (rc) + goto out_free_xid; + server = tcon->ses->server; if (server->ops->new_lease_key) @@ -776,7 +782,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, } pTcon = tlink_tcon(tlink); - rc = check_name(direntry); + rc = check_name(direntry, pTcon); if (rc) goto lookup_out; -- cgit v1.1 From 93a4c8355e0e448d83f31801b4c72f66e4360975 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Wed, 23 Aug 2017 18:24:50 +0200 Subject: ARM: dts: exynos: add needs-hpd for Odroid-XU3/4 CEC support was added for Exynos5 in 4.13, but for the Odroids we need to set 'needs-hpd' as well since CEC is disabled when there is no HDMI hotplug signal, just as for the exynos4 Odroid-U3. This is due to the level-shifter that is disabled when there is no HPD, thus blocking the CEC signal as well. Same close-but-no-cigar board design as the Odroid-U3. Tested with my Odroid XU4. Signed-off-by: Hans Verkuil Signed-off-by: Krzysztof Kozlowski Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi index f92f957..a183b56 100644 --- a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi +++ b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi @@ -266,6 +266,7 @@ &hdmicec { status = "okay"; + needs-hpd; }; &hsi2c_4 { -- cgit v1.1 From ea0ea2bc6dd8923d86a0fa98743dbeed98645486 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Fri, 18 Aug 2017 16:08:13 -0700 Subject: blk-throttle: cap discard request size discard request usually is very big and easily use all bandwidth budget of a cgroup. discard request size doesn't really mean the size of data written, so it doesn't make sense to account it into bandwidth budget. Jens pointed out treating the size 0 doesn't make sense too, because discard request does have cost. But it's not easy to find the actual cost. This patch simply makes the size one sector. Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- block/blk-throttle.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c index a7285bf..80f5481 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -382,6 +382,14 @@ static unsigned int tg_iops_limit(struct throtl_grp *tg, int rw) } \ } while (0) +static inline unsigned int throtl_bio_data_size(struct bio *bio) +{ + /* assume it's one sector */ + if (unlikely(bio_op(bio) == REQ_OP_DISCARD)) + return 512; + return bio->bi_iter.bi_size; +} + static void throtl_qnode_init(struct throtl_qnode *qn, struct throtl_grp *tg) { INIT_LIST_HEAD(&qn->node); @@ -934,6 +942,7 @@ static bool tg_with_in_bps_limit(struct throtl_grp *tg, struct bio *bio, bool rw = bio_data_dir(bio); u64 bytes_allowed, extra_bytes, tmp; unsigned long jiffy_elapsed, jiffy_wait, jiffy_elapsed_rnd; + unsigned int bio_size = throtl_bio_data_size(bio); jiffy_elapsed = jiffy_elapsed_rnd = jiffies - tg->slice_start[rw]; @@ -947,14 +956,14 @@ static bool tg_with_in_bps_limit(struct throtl_grp *tg, struct bio *bio, do_div(tmp, HZ); bytes_allowed = tmp; - if (tg->bytes_disp[rw] + bio->bi_iter.bi_size <= bytes_allowed) { + if (tg->bytes_disp[rw] + bio_size <= bytes_allowed) { if (wait) *wait = 0; return true; } /* Calc approx time to dispatch */ - extra_bytes = tg->bytes_disp[rw] + bio->bi_iter.bi_size - bytes_allowed; + extra_bytes = tg->bytes_disp[rw] + bio_size - bytes_allowed; jiffy_wait = div64_u64(extra_bytes * HZ, tg_bps_limit(tg, rw)); if (!jiffy_wait) @@ -1034,11 +1043,12 @@ static bool tg_may_dispatch(struct throtl_grp *tg, struct bio *bio, static void throtl_charge_bio(struct throtl_grp *tg, struct bio *bio) { bool rw = bio_data_dir(bio); + unsigned int bio_size = throtl_bio_data_size(bio); /* Charge the bio to the group */ - tg->bytes_disp[rw] += bio->bi_iter.bi_size; + tg->bytes_disp[rw] += bio_size; tg->io_disp[rw]++; - tg->last_bytes_disp[rw] += bio->bi_iter.bi_size; + tg->last_bytes_disp[rw] += bio_size; tg->last_io_disp[rw]++; /* -- cgit v1.1 From 1e6ec9ea89d30739b9447c1860fcb07fc29f3aef Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Wed, 23 Aug 2017 14:54:59 -0700 Subject: Revert "loop: support 4k physical blocksize" There's some stuff still up in the air, let's not get stuck with a subpar ABI. I'll follow up with something better for 4.14. Signed-off-by: Omar Sandoval Signed-off-by: Jens Axboe --- drivers/block/loop.c | 42 ++++++------------------------------------ drivers/block/loop.h | 1 - include/uapi/linux/loop.h | 3 --- 3 files changed, 6 insertions(+), 40 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index ef83349..f321b96 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -221,8 +221,7 @@ static void __loop_update_dio(struct loop_device *lo, bool dio) } static int -figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit, - loff_t logical_blocksize) +figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit) { loff_t size = get_size(offset, sizelimit, lo->lo_backing_file); sector_t x = (sector_t)size; @@ -234,12 +233,6 @@ figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit, lo->lo_offset = offset; if (lo->lo_sizelimit != sizelimit) lo->lo_sizelimit = sizelimit; - if (lo->lo_flags & LO_FLAGS_BLOCKSIZE) { - lo->lo_logical_blocksize = logical_blocksize; - blk_queue_physical_block_size(lo->lo_queue, lo->lo_blocksize); - blk_queue_logical_block_size(lo->lo_queue, - lo->lo_logical_blocksize); - } set_capacity(lo->lo_disk, x); bd_set_size(bdev, (loff_t)get_capacity(bdev->bd_disk) << 9); /* let user-space know about the new size */ @@ -820,7 +813,6 @@ static void loop_config_discard(struct loop_device *lo) struct file *file = lo->lo_backing_file; struct inode *inode = file->f_mapping->host; struct request_queue *q = lo->lo_queue; - int lo_bits = 9; /* * We use punch hole to reclaim the free space used by the @@ -840,11 +832,9 @@ static void loop_config_discard(struct loop_device *lo) q->limits.discard_granularity = inode->i_sb->s_blocksize; q->limits.discard_alignment = 0; - if (lo->lo_flags & LO_FLAGS_BLOCKSIZE) - lo_bits = blksize_bits(lo->lo_logical_blocksize); - blk_queue_max_discard_sectors(q, UINT_MAX >> lo_bits); - blk_queue_max_write_zeroes_sectors(q, UINT_MAX >> lo_bits); + blk_queue_max_discard_sectors(q, UINT_MAX >> 9); + blk_queue_max_write_zeroes_sectors(q, UINT_MAX >> 9); queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, q); } @@ -938,7 +928,6 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, lo->use_dio = false; lo->lo_blocksize = lo_blocksize; - lo->lo_logical_blocksize = 512; lo->lo_device = bdev; lo->lo_flags = lo_flags; lo->lo_backing_file = file; @@ -1104,7 +1093,6 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) int err; struct loop_func_table *xfer; kuid_t uid = current_uid(); - int lo_flags = lo->lo_flags; if (lo->lo_encrypt_key_size && !uid_eq(lo->lo_key_owner, uid) && @@ -1137,26 +1125,9 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) if (err) goto exit; - if (info->lo_flags & LO_FLAGS_BLOCKSIZE) { - if (!(lo->lo_flags & LO_FLAGS_BLOCKSIZE)) - lo->lo_logical_blocksize = 512; - lo->lo_flags |= LO_FLAGS_BLOCKSIZE; - if (LO_INFO_BLOCKSIZE(info) != 512 && - LO_INFO_BLOCKSIZE(info) != 1024 && - LO_INFO_BLOCKSIZE(info) != 2048 && - LO_INFO_BLOCKSIZE(info) != 4096) - return -EINVAL; - if (LO_INFO_BLOCKSIZE(info) > lo->lo_blocksize) - return -EINVAL; - } - if (lo->lo_offset != info->lo_offset || - lo->lo_sizelimit != info->lo_sizelimit || - lo->lo_flags != lo_flags || - ((lo->lo_flags & LO_FLAGS_BLOCKSIZE) && - lo->lo_logical_blocksize != LO_INFO_BLOCKSIZE(info))) { - if (figure_loop_size(lo, info->lo_offset, info->lo_sizelimit, - LO_INFO_BLOCKSIZE(info))) { + lo->lo_sizelimit != info->lo_sizelimit) { + if (figure_loop_size(lo, info->lo_offset, info->lo_sizelimit)) { err = -EFBIG; goto exit; } @@ -1348,8 +1319,7 @@ static int loop_set_capacity(struct loop_device *lo) if (unlikely(lo->lo_state != Lo_bound)) return -ENXIO; - return figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit, - lo->lo_logical_blocksize); + return figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit); } static int loop_set_dio(struct loop_device *lo, unsigned long arg) diff --git a/drivers/block/loop.h b/drivers/block/loop.h index 2c096b9..fecd3f9 100644 --- a/drivers/block/loop.h +++ b/drivers/block/loop.h @@ -49,7 +49,6 @@ struct loop_device { struct file * lo_backing_file; struct block_device *lo_device; unsigned lo_blocksize; - unsigned lo_logical_blocksize; void *key_data; gfp_t old_gfp_mask; diff --git a/include/uapi/linux/loop.h b/include/uapi/linux/loop.h index a3960f9..c8125ec 100644 --- a/include/uapi/linux/loop.h +++ b/include/uapi/linux/loop.h @@ -22,7 +22,6 @@ enum { LO_FLAGS_AUTOCLEAR = 4, LO_FLAGS_PARTSCAN = 8, LO_FLAGS_DIRECT_IO = 16, - LO_FLAGS_BLOCKSIZE = 32, }; #include /* for __kernel_old_dev_t */ @@ -60,8 +59,6 @@ struct loop_info64 { __u64 lo_init[2]; }; -#define LO_INFO_BLOCKSIZE(l) (l)->lo_init[0] - /* * Loop filter types */ -- cgit v1.1 From 143c97cc652949893c8056c679012f0aeccb80e5 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 23 Aug 2017 18:16:11 -0700 Subject: Revert "pty: fix the cached path of the pty slave file descriptor in the master" This reverts commit c8c03f1858331e85d397bacccd34ef409aae993c. It turns out that while fixing the ptmx file descriptor to have the correct 'struct path' to the associated slave pty is a really good thing, it breaks some user space tools for a very annoying reason. The problem is that /dev/ptmx and its associated slave pty (/dev/pts/X) are on different mounts. That was what caused us to have the wrong path in the first place (we would mix up the vfsmount of the 'ptmx' node, with the dentry of the pty slave node), but it also means that now while we use the right vfsmount, having the pty master open also keeps the pts mount busy. And it turn sout that that makes 'pbuilder' very unhappy, as noted by Stefan Lippers-Hollmann: "This patch introduces a regression for me when using pbuilder 0.228.7[2] (a helper to build Debian packages in a chroot and to create and update its chroots) when trying to umount /dev/ptmx (inside the chroot) on Debian/ unstable (full log and pbuilder configuration file[3] attached). [...] Setting up build-essential (12.3) ... Processing triggers for libc-bin (2.24-15) ... I: unmounting dev/ptmx filesystem W: Could not unmount dev/ptmx: umount: /var/cache/pbuilder/build/1340/dev/ptmx: target is busy (In some cases useful info about processes that use the device is found by lsof(8) or fuser(1).)" apparently pbuilder tries to unmount the /dev/pts filesystem while still holding at least one master node open, which is arguably not very nice, but we don't break user space even when fixing other bugs. So this commit has to be reverted. I'll try to figure out a way to avoid caching the path to the slave pty in the master pty. The only thing that actually wants that slave pty path is the "TIOCGPTPEER" ioctl, and I think we could just recreate the path at that time. Reported-by: Stefan Lippers-Hollmann Cc: Eric W Biederman Cc: Christian Brauner Cc: Al Viro Signed-off-by: Linus Torvalds --- drivers/tty/pty.c | 7 ++----- fs/devpts/inode.c | 4 +--- include/linux/devpts_fs.h | 2 +- 3 files changed, 4 insertions(+), 9 deletions(-) diff --git a/drivers/tty/pty.c b/drivers/tty/pty.c index 1fc80ea..284749f 100644 --- a/drivers/tty/pty.c +++ b/drivers/tty/pty.c @@ -793,7 +793,6 @@ static int ptmx_open(struct inode *inode, struct file *filp) struct tty_struct *tty; struct path *pts_path; struct dentry *dentry; - struct vfsmount *mnt; int retval; int index; @@ -806,7 +805,7 @@ static int ptmx_open(struct inode *inode, struct file *filp) if (retval) return retval; - fsi = devpts_acquire(filp, &mnt); + fsi = devpts_acquire(filp); if (IS_ERR(fsi)) { retval = PTR_ERR(fsi); goto out_free_file; @@ -850,7 +849,7 @@ static int ptmx_open(struct inode *inode, struct file *filp) pts_path = kmalloc(sizeof(struct path), GFP_KERNEL); if (!pts_path) goto err_release; - pts_path->mnt = mnt; + pts_path->mnt = filp->f_path.mnt; pts_path->dentry = dentry; path_get(pts_path); tty->link->driver_data = pts_path; @@ -867,7 +866,6 @@ err_path_put: path_put(pts_path); kfree(pts_path); err_release: - mntput(mnt); tty_unlock(tty); // This will also put-ref the fsi tty_release(inode, filp); @@ -876,7 +874,6 @@ out: devpts_kill_index(fsi, index); out_put_fsi: devpts_release(fsi); - mntput(mnt); out_free_file: tty_free_file(filp); return retval; diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 44dfbca..108df2e 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -133,7 +133,7 @@ static inline struct pts_fs_info *DEVPTS_SB(struct super_block *sb) return sb->s_fs_info; } -struct pts_fs_info *devpts_acquire(struct file *filp, struct vfsmount **ptsmnt) +struct pts_fs_info *devpts_acquire(struct file *filp) { struct pts_fs_info *result; struct path path; @@ -142,7 +142,6 @@ struct pts_fs_info *devpts_acquire(struct file *filp, struct vfsmount **ptsmnt) path = filp->f_path; path_get(&path); - *ptsmnt = NULL; /* Has the devpts filesystem already been found? */ sb = path.mnt->mnt_sb; @@ -166,7 +165,6 @@ struct pts_fs_info *devpts_acquire(struct file *filp, struct vfsmount **ptsmnt) * pty code needs to hold extra references in case of last /dev/tty close */ atomic_inc(&sb->s_active); - *ptsmnt = mntget(path.mnt); result = DEVPTS_SB(sb); out: diff --git a/include/linux/devpts_fs.h b/include/linux/devpts_fs.h index 7883e90..277ab9a 100644 --- a/include/linux/devpts_fs.h +++ b/include/linux/devpts_fs.h @@ -19,7 +19,7 @@ struct pts_fs_info; -struct pts_fs_info *devpts_acquire(struct file *, struct vfsmount **ptsmnt); +struct pts_fs_info *devpts_acquire(struct file *); void devpts_release(struct pts_fs_info *); int devpts_new_index(struct pts_fs_info *); -- cgit v1.1 From 47c4ccd3086139d2085e0f1e59c0f26e8ce4ff46 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sun, 20 Aug 2017 19:09:35 +0200 Subject: scsi: qedf: Fix a potential NULL pointer dereference At the beginning of 'qedf_srr_compl()' and of 'qedf_rec_compl()', we check if 'orig_io_req' is NULL. If this happens, a NULL pointer dereference will occur in the error handling path. Fix it by adding an additional label in the error handling path in order to avoid this NULL pointer dereference. [mkp: typo] Fixes: 61d8658b4a43 ("scsi: qedf: Add QLogic FastLinQ offload FCoE driver framework.") Signed-off-by: Christophe JAILLET Acked-by: Chad Dupuis Signed-off-by: Martin K. Petersen --- drivers/scsi/qedf/qedf_els.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/scsi/qedf/qedf_els.c b/drivers/scsi/qedf/qedf_els.c index eb07f1d..59c18ca 100644 --- a/drivers/scsi/qedf/qedf_els.c +++ b/drivers/scsi/qedf/qedf_els.c @@ -489,7 +489,7 @@ static void qedf_srr_compl(struct qedf_els_cb_arg *cb_arg) /* If a SRR times out, simply free resources */ if (srr_req->event == QEDF_IOREQ_EV_ELS_TMO) - goto out_free; + goto out_put; /* Normalize response data into struct fc_frame */ mp_req = &(srr_req->mp_req); @@ -501,7 +501,7 @@ static void qedf_srr_compl(struct qedf_els_cb_arg *cb_arg) if (!fp) { QEDF_ERR(&(qedf->dbg_ctx), "fc_frame_alloc failure.\n"); - goto out_free; + goto out_put; } /* Copy frame header from firmware into fp */ @@ -526,9 +526,10 @@ static void qedf_srr_compl(struct qedf_els_cb_arg *cb_arg) } fc_frame_free(fp); -out_free: +out_put: /* Put reference for original command since SRR completed */ kref_put(&orig_io_req->refcount, qedf_release_cmd); +out_free: kfree(cb_arg); } @@ -780,7 +781,7 @@ static void qedf_rec_compl(struct qedf_els_cb_arg *cb_arg) /* If a REC times out, free resources */ if (rec_req->event == QEDF_IOREQ_EV_ELS_TMO) - goto out_free; + goto out_put; /* Normalize response data into struct fc_frame */ mp_req = &(rec_req->mp_req); @@ -792,7 +793,7 @@ static void qedf_rec_compl(struct qedf_els_cb_arg *cb_arg) if (!fp) { QEDF_ERR(&(qedf->dbg_ctx), "fc_frame_alloc failure.\n"); - goto out_free; + goto out_put; } /* Copy frame header from firmware into fp */ @@ -884,9 +885,10 @@ static void qedf_rec_compl(struct qedf_els_cb_arg *cb_arg) out_free_frame: fc_frame_free(fp); -out_free: +out_put: /* Put reference for original command since REC completed */ kref_put(&orig_io_req->refcount, qedf_release_cmd); +out_free: kfree(cb_arg); } -- cgit v1.1 From 013dae5dbc07aa521a38f1ca2d32123ec674bd5d Mon Sep 17 00:00:00 2001 From: Stephan Gatzka Date: Tue, 22 Aug 2017 14:25:07 +0200 Subject: net: stmmac: socfgpa: Ensure emac bit set in sys manager for MII/GMII/SGMII. When using MII/GMII/SGMII in the Altera SoC, the phy needs to be wired through the FPGA. To ensure correct behavior, the appropriate bit in the System Manager FPGA Interface Group register needs to be set. Signed-off-by: Stephan Gatzka Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 17d4bba..6e35957 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -269,7 +269,10 @@ static int socfpga_dwmac_set_phy_mode(struct socfpga_dwmac *dwmac) ctrl &= ~(SYSMGR_EMACGRP_CTRL_PHYSEL_MASK << reg_shift); ctrl |= val << reg_shift; - if (dwmac->f2h_ptp_ref_clk) { + if (dwmac->f2h_ptp_ref_clk || + phymode == PHY_INTERFACE_MODE_MII || + phymode == PHY_INTERFACE_MODE_GMII || + phymode == PHY_INTERFACE_MODE_SGMII) { ctrl |= SYSMGR_EMACGRP_CTRL_PTP_REF_CLK_MASK << (reg_shift / 2); regmap_read(sys_mgr_base_addr, SYSMGR_FPGAGRP_MODULE_REG, &module); -- cgit v1.1 From cd0a137acbb66208368353723f5f1480995cf1c4 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 22 Aug 2017 15:12:14 -0700 Subject: net: core: Specify skb_pad()/skb_put_padto() SKB freeing Rename skb_pad() into __skb_pad() and make it take a third argument: free_on_error which controls whether kfree_skb() should be called or not, skb_pad() directly makes use of it and passes true to preserve its existing behavior. Do exactly the same thing with __skb_put_padto() and skb_put_padto(). Suggested-by: David Miller Signed-off-by: Florian Fainelli Reviewed-by: Woojung Huh Signed-off-by: David S. Miller --- include/linux/skbuff.h | 41 +++++++++++++++++++++++++++++++++++++---- net/core/skbuff.c | 13 ++++++++----- 2 files changed, 45 insertions(+), 9 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index dbe29b6..d67a818 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -973,7 +973,23 @@ int __must_check skb_to_sgvec_nomark(struct sk_buff *skb, struct scatterlist *sg int __must_check skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset, int len); int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer); -int skb_pad(struct sk_buff *skb, int pad); +int __skb_pad(struct sk_buff *skb, int pad, bool free_on_error); + +/** + * skb_pad - zero pad the tail of an skb + * @skb: buffer to pad + * @pad: space to pad + * + * Ensure that a buffer is followed by a padding area that is zero + * filled. Used by network drivers which may DMA or transfer data + * beyond the buffer end onto the wire. + * + * May return error in out of memory cases. The skb is freed on error. + */ +static inline int skb_pad(struct sk_buff *skb, int pad) +{ + return __skb_pad(skb, pad, true); +} #define dev_kfree_skb(a) consume_skb(a) int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb, @@ -2825,25 +2841,42 @@ static inline int skb_padto(struct sk_buff *skb, unsigned int len) * skb_put_padto - increase size and pad an skbuff up to a minimal size * @skb: buffer to pad * @len: minimal length + * @free_on_error: free buffer on error * * Pads up a buffer to ensure the trailing bytes exist and are * blanked. If the buffer already contains sufficient data it * is untouched. Otherwise it is extended. Returns zero on - * success. The skb is freed on error. + * success. The skb is freed on error if @free_on_error is true. */ -static inline int skb_put_padto(struct sk_buff *skb, unsigned int len) +static inline int __skb_put_padto(struct sk_buff *skb, unsigned int len, + bool free_on_error) { unsigned int size = skb->len; if (unlikely(size < len)) { len -= size; - if (skb_pad(skb, len)) + if (__skb_pad(skb, len, free_on_error)) return -ENOMEM; __skb_put(skb, len); } return 0; } +/** + * skb_put_padto - increase size and pad an skbuff up to a minimal size + * @skb: buffer to pad + * @len: minimal length + * + * Pads up a buffer to ensure the trailing bytes exist and are + * blanked. If the buffer already contains sufficient data it + * is untouched. Otherwise it is extended. Returns zero on + * success. The skb is freed on error. + */ +static inline int skb_put_padto(struct sk_buff *skb, unsigned int len) +{ + return __skb_put_padto(skb, len, true); +} + static inline int skb_add_data(struct sk_buff *skb, struct iov_iter *from, int copy) { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f990eb8..e075566 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1363,18 +1363,20 @@ struct sk_buff *skb_copy_expand(const struct sk_buff *skb, EXPORT_SYMBOL(skb_copy_expand); /** - * skb_pad - zero pad the tail of an skb + * __skb_pad - zero pad the tail of an skb * @skb: buffer to pad * @pad: space to pad + * @free_on_error: free buffer on error * * Ensure that a buffer is followed by a padding area that is zero * filled. Used by network drivers which may DMA or transfer data * beyond the buffer end onto the wire. * - * May return error in out of memory cases. The skb is freed on error. + * May return error in out of memory cases. The skb is freed on error + * if @free_on_error is true. */ -int skb_pad(struct sk_buff *skb, int pad) +int __skb_pad(struct sk_buff *skb, int pad, bool free_on_error) { int err; int ntail; @@ -1403,10 +1405,11 @@ int skb_pad(struct sk_buff *skb, int pad) return 0; free_skb: - kfree_skb(skb); + if (free_on_error) + kfree_skb(skb); return err; } -EXPORT_SYMBOL(skb_pad); +EXPORT_SYMBOL(__skb_pad); /** * pskb_put - add data to the tail of a potentially fragmented buffer -- cgit v1.1 From 49716679248a0bf64d8914711b2d1ea48751853e Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 22 Aug 2017 15:12:15 -0700 Subject: net: dsa: skb_put_padto() already frees nskb The first call of skb_put_padto() will free up the SKB on error, but we return NULL which tells dsa_slave_xmit() that the original SKB should be freed so this would lead to a double free here. The second skb_put_padto() already frees the passed sk_buff reference upon error, so calling kfree_skb() on it again is not necessary. Detected by CoverityScan, CID#1416687 ("USE_AFTER_FREE") Fixes: e71cb9e00922 ("net: dsa: ksz: fix skb freeing") Signed-off-by: Florian Fainelli Reviewed-by: Woojung Huh Signed-off-by: David S. Miller --- net/dsa/tag_ksz.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c index de66ca8..3bd6e2a 100644 --- a/net/dsa/tag_ksz.c +++ b/net/dsa/tag_ksz.c @@ -42,7 +42,8 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev) padlen = (skb->len >= ETH_ZLEN) ? 0 : ETH_ZLEN - skb->len; if (skb_tailroom(skb) >= padlen + KSZ_INGRESS_TAG_LEN) { - if (skb_put_padto(skb, skb->len + padlen)) + /* Let dsa_slave_xmit() free skb */ + if (__skb_put_padto(skb, skb->len + padlen, false)) return NULL; nskb = skb; @@ -60,10 +61,11 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev) skb_transport_header(skb) - skb->head); skb_copy_and_csum_dev(skb, skb_put(nskb, skb->len)); - if (skb_put_padto(nskb, nskb->len + padlen)) { - kfree_skb(nskb); + /* Let skb_put_padto() free nskb, and let dsa_slave_xmit() free + * skb + */ + if (skb_put_padto(nskb, nskb->len + padlen)) return NULL; - } kfree_skb(skb); } -- cgit v1.1 From d6e1ab9ea3514840e4f32957c457b094646c2e9d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 22 Aug 2017 23:22:42 -0700 Subject: nfp: don't hold PF lock while enabling SR-IOV Enabling SR-IOV VFs will cause the PCI subsystem to schedule a work and flush its workqueue. Since the nfp driver schedules its own work we can't enable VFs while holding driver load. Commit 6d48ceb27af1 ("nfp: allocate a private workqueue for driver work") tried to avoid this deadlock by creating a separate workqueue. Unfortunately, due to the architecture of workqueue subsystem this does not guarantee a separate thread of execution. Luckily we can simply take pci_enable_sriov() from under the driver lock. Take pci_disable_sriov() from under the lock too for symmetry. Fixes: 6d48ceb27af1 ("nfp: allocate a private workqueue for driver work") Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_main.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_main.c b/drivers/net/ethernet/netronome/nfp/nfp_main.c index d67969d..3f199db 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_main.c @@ -98,21 +98,20 @@ static int nfp_pcie_sriov_enable(struct pci_dev *pdev, int num_vfs) struct nfp_pf *pf = pci_get_drvdata(pdev); int err; - mutex_lock(&pf->lock); - if (num_vfs > pf->limit_vfs) { nfp_info(pf->cpp, "Firmware limits number of VFs to %u\n", pf->limit_vfs); - err = -EINVAL; - goto err_unlock; + return -EINVAL; } err = pci_enable_sriov(pdev, num_vfs); if (err) { dev_warn(&pdev->dev, "Failed to enable PCI SR-IOV: %d\n", err); - goto err_unlock; + return err; } + mutex_lock(&pf->lock); + err = nfp_app_sriov_enable(pf->app, num_vfs); if (err) { dev_warn(&pdev->dev, @@ -129,9 +128,8 @@ static int nfp_pcie_sriov_enable(struct pci_dev *pdev, int num_vfs) return num_vfs; err_sriov_disable: - pci_disable_sriov(pdev); -err_unlock: mutex_unlock(&pf->lock); + pci_disable_sriov(pdev); return err; #endif return 0; @@ -158,10 +156,10 @@ static int nfp_pcie_sriov_disable(struct pci_dev *pdev) pf->num_vfs = 0; + mutex_unlock(&pf->lock); + pci_disable_sriov(pdev); dev_dbg(&pdev->dev, "Removed VFs.\n"); - - mutex_unlock(&pf->lock); #endif return 0; } -- cgit v1.1 From 326ce603015eefaa86fc6e490f43638e1010a838 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 22 Aug 2017 23:22:43 -0700 Subject: nfp: make sure representors are destroyed before their lower netdev App start/stop callbacks can perform application initialization. Unfortunately, flower app started using them for creating and destroying representors. This can lead to a situation where lower vNIC netdev is destroyed while representors still try to pass traffic. This will most likely lead to a NULL-dereference on the lower netdev TX path. Move the start/stop callbacks, so that representors are created/ destroyed when vNICs are fully initialized. Fixes: 5de73ee46704 ("nfp: general representor implementation") Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_main.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c index 5797dbf..1aca4e5 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c @@ -456,10 +456,6 @@ static int nfp_net_pf_app_start(struct nfp_pf *pf) { int err; - err = nfp_net_pf_app_start_ctrl(pf); - if (err) - return err; - err = nfp_app_start(pf->app, pf->ctrl_vnic); if (err) goto err_ctrl_stop; @@ -484,7 +480,6 @@ static void nfp_net_pf_app_stop(struct nfp_pf *pf) if (pf->num_vfs) nfp_app_sriov_disable(pf->app); nfp_app_stop(pf->app); - nfp_net_pf_app_stop_ctrl(pf); } static void nfp_net_pci_unmap_mem(struct nfp_pf *pf) @@ -559,7 +554,7 @@ err_unmap_ctrl: static void nfp_net_pci_remove_finish(struct nfp_pf *pf) { - nfp_net_pf_app_stop(pf); + nfp_net_pf_app_stop_ctrl(pf); /* stop app first, to avoid double free of ctrl vNIC's ddir */ nfp_net_debugfs_dir_clean(&pf->ddir); @@ -690,6 +685,7 @@ int nfp_net_pci_probe(struct nfp_pf *pf) { struct nfp_net_fw_version fw_ver; u8 __iomem *ctrl_bar, *qc_bar; + struct nfp_net *nn; int stride; int err; @@ -766,7 +762,7 @@ int nfp_net_pci_probe(struct nfp_pf *pf) if (err) goto err_free_vnics; - err = nfp_net_pf_app_start(pf); + err = nfp_net_pf_app_start_ctrl(pf); if (err) goto err_free_irqs; @@ -774,12 +770,20 @@ int nfp_net_pci_probe(struct nfp_pf *pf) if (err) goto err_stop_app; + err = nfp_net_pf_app_start(pf); + if (err) + goto err_clean_vnics; + mutex_unlock(&pf->lock); return 0; +err_clean_vnics: + list_for_each_entry(nn, &pf->vnics, vnic_list) + if (nfp_net_is_data_vnic(nn)) + nfp_net_pf_clean_vnic(pf, nn); err_stop_app: - nfp_net_pf_app_stop(pf); + nfp_net_pf_app_stop_ctrl(pf); err_free_irqs: nfp_net_pf_free_irqs(pf); err_free_vnics: @@ -803,6 +807,8 @@ void nfp_net_pci_remove(struct nfp_pf *pf) if (list_empty(&pf->vnics)) goto out; + nfp_net_pf_app_stop(pf); + list_for_each_entry(nn, &pf->vnics, vnic_list) if (nfp_net_is_data_vnic(nn)) nfp_net_pf_clean_vnic(pf, nn); -- cgit v1.1 From 1691a4c0f4634d50ffeb74373fdeec63495c911e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 22 Aug 2017 23:22:44 -0700 Subject: nfp: avoid buffer leak when representor is missing When driver receives a muxed frame, but it can't find the representor netdev it is destined to it will try to "drop" that frame, i.e. reuse the buffer. The issue is that the replacement buffer has already been allocated at this point, and reusing the buffer from received frame will leak it. Change the code to put the new buffer on the ring earlier and not reuse the old buffer (make the buffer parameter to nfp_net_rx_drop() a NULL). Fixes: 91bf82ca9eed ("nfp: add support for tx/rx with metadata portid") Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 9f77ce0..1ff0c57 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1751,6 +1751,10 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget) continue; } + nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr); + + nfp_net_rx_give_one(dp, rx_ring, new_frag, new_dma_addr); + if (likely(!meta.portid)) { netdev = dp->netdev; } else { @@ -1759,16 +1763,12 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget) nn = netdev_priv(dp->netdev); netdev = nfp_app_repr_get(nn->app, meta.portid); if (unlikely(!netdev)) { - nfp_net_rx_drop(dp, r_vec, rx_ring, rxbuf, skb); + nfp_net_rx_drop(dp, r_vec, rx_ring, NULL, skb); continue; } nfp_repr_inc_rx_stats(netdev, pkt_len); } - nfp_net_dma_unmap_rx(dp, rxbuf->dma_addr); - - nfp_net_rx_give_one(dp, rx_ring, new_frag, new_dma_addr); - skb_reserve(skb, pkt_off); skb_put(skb, pkt_len); -- cgit v1.1 From 2b33bc8aa236b75d6e86a8a79126fd9739e4a5bd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 23 Aug 2017 21:40:32 -0700 Subject: net: dsa: use consume_skb() Two kfree_skb() should be consume_skb(), to be friend with drop monitor (perf record ... -e skb:kfree_skb) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/dsa/tag_ksz.c | 2 +- net/dsa/tag_trailer.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c index 3bd6e2a..fcd90f7 100644 --- a/net/dsa/tag_ksz.c +++ b/net/dsa/tag_ksz.c @@ -67,7 +67,7 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev) if (skb_put_padto(nskb, nskb->len + padlen)) return NULL; - kfree_skb(skb); + consume_skb(skb); } tag = skb_put(nskb, KSZ_INGRESS_TAG_LEN); diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c index b09e562..9c7b1d7 100644 --- a/net/dsa/tag_trailer.c +++ b/net/dsa/tag_trailer.c @@ -40,7 +40,7 @@ static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev) skb_set_network_header(nskb, skb_network_header(skb) - skb->head); skb_set_transport_header(nskb, skb_transport_header(skb) - skb->head); skb_copy_and_csum_dev(skb, skb_put(nskb, skb->len)); - kfree_skb(skb); + consume_skb(skb); if (padlen) { skb_put_zero(nskb, padlen); -- cgit v1.1 From ee6c88bb754e3d363e568da78086adfedb692447 Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Wed, 23 Aug 2017 13:27:13 +0200 Subject: sctp: Avoid out-of-bounds reads from address storage inet_diag_msg_sctp{,l}addr_fill() and sctp_get_sctp_info() copy sizeof(sockaddr_storage) bytes to fill in sockaddr structs used to export diagnostic information to userspace. However, the memory allocated to store sockaddr information is smaller than that and depends on the address family, so we leak up to 100 uninitialized bytes to userspace. Just use the size of the source structs instead, in all the three cases this is what userspace expects. Zero out the remaining memory. Unused bytes (i.e. when IPv4 addresses are used) in source structs sctp_sockaddr_entry and sctp_transport are already cleared by sctp_add_bind_addr() and sctp_transport_new(), respectively. Noticed while testing KASAN-enabled kernel with 'ss': [ 2326.885243] BUG: KASAN: slab-out-of-bounds in inet_sctp_diag_fill+0x42c/0x6c0 [sctp_diag] at addr ffff881be8779800 [ 2326.896800] Read of size 128 by task ss/9527 [ 2326.901564] CPU: 0 PID: 9527 Comm: ss Not tainted 4.11.0-22.el7a.x86_64 #1 [ 2326.909236] Hardware name: Dell Inc. PowerEdge R730/072T6D, BIOS 2.4.3 01/17/2017 [ 2326.917585] Call Trace: [ 2326.920312] dump_stack+0x63/0x8d [ 2326.924014] kasan_object_err+0x21/0x70 [ 2326.928295] kasan_report+0x288/0x540 [ 2326.932380] ? inet_sctp_diag_fill+0x42c/0x6c0 [sctp_diag] [ 2326.938500] ? skb_put+0x8b/0xd0 [ 2326.942098] ? memset+0x31/0x40 [ 2326.945599] check_memory_region+0x13c/0x1a0 [ 2326.950362] memcpy+0x23/0x50 [ 2326.953669] inet_sctp_diag_fill+0x42c/0x6c0 [sctp_diag] [ 2326.959596] ? inet_diag_msg_sctpasoc_fill+0x460/0x460 [sctp_diag] [ 2326.966495] ? __lock_sock+0x102/0x150 [ 2326.970671] ? sock_def_wakeup+0x60/0x60 [ 2326.975048] ? remove_wait_queue+0xc0/0xc0 [ 2326.979619] sctp_diag_dump+0x44a/0x760 [sctp_diag] [ 2326.985063] ? sctp_ep_dump+0x280/0x280 [sctp_diag] [ 2326.990504] ? memset+0x31/0x40 [ 2326.994007] ? mutex_lock+0x12/0x40 [ 2326.997900] __inet_diag_dump+0x57/0xb0 [inet_diag] [ 2327.003340] ? __sys_sendmsg+0x150/0x150 [ 2327.007715] inet_diag_dump+0x4d/0x80 [inet_diag] [ 2327.012979] netlink_dump+0x1e6/0x490 [ 2327.017064] __netlink_dump_start+0x28e/0x2c0 [ 2327.021924] inet_diag_handler_cmd+0x189/0x1a0 [inet_diag] [ 2327.028045] ? inet_diag_rcv_msg_compat+0x1b0/0x1b0 [inet_diag] [ 2327.034651] ? inet_diag_dump_compat+0x190/0x190 [inet_diag] [ 2327.040965] ? __netlink_lookup+0x1b9/0x260 [ 2327.045631] sock_diag_rcv_msg+0x18b/0x1e0 [ 2327.050199] netlink_rcv_skb+0x14b/0x180 [ 2327.054574] ? sock_diag_bind+0x60/0x60 [ 2327.058850] sock_diag_rcv+0x28/0x40 [ 2327.062837] netlink_unicast+0x2e7/0x3b0 [ 2327.067212] ? netlink_attachskb+0x330/0x330 [ 2327.071975] ? kasan_check_write+0x14/0x20 [ 2327.076544] netlink_sendmsg+0x5be/0x730 [ 2327.080918] ? netlink_unicast+0x3b0/0x3b0 [ 2327.085486] ? kasan_check_write+0x14/0x20 [ 2327.090057] ? selinux_socket_sendmsg+0x24/0x30 [ 2327.095109] ? netlink_unicast+0x3b0/0x3b0 [ 2327.099678] sock_sendmsg+0x74/0x80 [ 2327.103567] ___sys_sendmsg+0x520/0x530 [ 2327.107844] ? __get_locked_pte+0x178/0x200 [ 2327.112510] ? copy_msghdr_from_user+0x270/0x270 [ 2327.117660] ? vm_insert_page+0x360/0x360 [ 2327.122133] ? vm_insert_pfn_prot+0xb4/0x150 [ 2327.126895] ? vm_insert_pfn+0x32/0x40 [ 2327.131077] ? vvar_fault+0x71/0xd0 [ 2327.134968] ? special_mapping_fault+0x69/0x110 [ 2327.140022] ? __do_fault+0x42/0x120 [ 2327.144008] ? __handle_mm_fault+0x1062/0x17a0 [ 2327.148965] ? __fget_light+0xa7/0xc0 [ 2327.153049] __sys_sendmsg+0xcb/0x150 [ 2327.157133] ? __sys_sendmsg+0xcb/0x150 [ 2327.161409] ? SyS_shutdown+0x140/0x140 [ 2327.165688] ? exit_to_usermode_loop+0xd0/0xd0 [ 2327.170646] ? __do_page_fault+0x55d/0x620 [ 2327.175216] ? __sys_sendmsg+0x150/0x150 [ 2327.179591] SyS_sendmsg+0x12/0x20 [ 2327.183384] do_syscall_64+0xe3/0x230 [ 2327.187471] entry_SYSCALL64_slow_path+0x25/0x25 [ 2327.192622] RIP: 0033:0x7f41d18fa3b0 [ 2327.196608] RSP: 002b:00007ffc3b731218 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 2327.205055] RAX: ffffffffffffffda RBX: 00007ffc3b731380 RCX: 00007f41d18fa3b0 [ 2327.213017] RDX: 0000000000000000 RSI: 00007ffc3b731340 RDI: 0000000000000003 [ 2327.220978] RBP: 0000000000000002 R08: 0000000000000004 R09: 0000000000000040 [ 2327.228939] R10: 00007ffc3b730f30 R11: 0000000000000246 R12: 0000000000000003 [ 2327.236901] R13: 00007ffc3b731340 R14: 00007ffc3b7313d0 R15: 0000000000000084 [ 2327.244865] Object at ffff881be87797e0, in cache kmalloc-64 size: 64 [ 2327.251953] Allocated: [ 2327.254581] PID = 9484 [ 2327.257215] save_stack_trace+0x1b/0x20 [ 2327.261485] save_stack+0x46/0xd0 [ 2327.265179] kasan_kmalloc+0xad/0xe0 [ 2327.269165] kmem_cache_alloc_trace+0xe6/0x1d0 [ 2327.274138] sctp_add_bind_addr+0x58/0x180 [sctp] [ 2327.279400] sctp_do_bind+0x208/0x310 [sctp] [ 2327.284176] sctp_bind+0x61/0xa0 [sctp] [ 2327.288455] inet_bind+0x5f/0x3a0 [ 2327.292151] SYSC_bind+0x1a4/0x1e0 [ 2327.295944] SyS_bind+0xe/0x10 [ 2327.299349] do_syscall_64+0xe3/0x230 [ 2327.303433] return_from_SYSCALL_64+0x0/0x6a [ 2327.308194] Freed: [ 2327.310434] PID = 4131 [ 2327.313065] save_stack_trace+0x1b/0x20 [ 2327.317344] save_stack+0x46/0xd0 [ 2327.321040] kasan_slab_free+0x73/0xc0 [ 2327.325220] kfree+0x96/0x1a0 [ 2327.328530] dynamic_kobj_release+0x15/0x40 [ 2327.333195] kobject_release+0x99/0x1e0 [ 2327.337472] kobject_put+0x38/0x70 [ 2327.341266] free_notes_attrs+0x66/0x80 [ 2327.345545] mod_sysfs_teardown+0x1a5/0x270 [ 2327.350211] free_module+0x20/0x2a0 [ 2327.354099] SyS_delete_module+0x2cb/0x2f0 [ 2327.358667] do_syscall_64+0xe3/0x230 [ 2327.362750] return_from_SYSCALL_64+0x0/0x6a [ 2327.367510] Memory state around the buggy address: [ 2327.372855] ffff881be8779700: fc fc fc fc 00 00 00 00 00 00 00 00 fc fc fc fc [ 2327.380914] ffff881be8779780: fb fb fb fb fb fb fb fb fc fc fc fc 00 00 00 00 [ 2327.388972] >ffff881be8779800: 00 00 00 00 fc fc fc fc fb fb fb fb fb fb fb fb [ 2327.397031] ^ [ 2327.401792] ffff881be8779880: fc fc fc fc fb fb fb fb fb fb fb fb fc fc fc fc [ 2327.409850] ffff881be8779900: 00 00 00 00 00 04 fc fc fc fc fc fc 00 00 00 00 [ 2327.417907] ================================================================== This fixes CVE-2017-7558. References: https://bugzilla.redhat.com/show_bug.cgi?id=1480266 Fixes: 8f840e47f190 ("sctp: add the sctp_diag.c file") Cc: Xin Long Cc: Vlad Yasevich Cc: Neil Horman Signed-off-by: Stefano Brivio Acked-by: Marcelo Ricardo Leitner Reviewed-by: Xin Long Signed-off-by: David S. Miller --- net/sctp/sctp_diag.c | 7 +++++-- net/sctp/socket.c | 3 +-- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/net/sctp/sctp_diag.c b/net/sctp/sctp_diag.c index 9a64721..e99518e 100644 --- a/net/sctp/sctp_diag.c +++ b/net/sctp/sctp_diag.c @@ -70,7 +70,8 @@ static int inet_diag_msg_sctpladdrs_fill(struct sk_buff *skb, info = nla_data(attr); list_for_each_entry_rcu(laddr, address_list, list) { - memcpy(info, &laddr->a, addrlen); + memcpy(info, &laddr->a, sizeof(laddr->a)); + memset(info + sizeof(laddr->a), 0, addrlen - sizeof(laddr->a)); info += addrlen; } @@ -93,7 +94,9 @@ static int inet_diag_msg_sctpaddrs_fill(struct sk_buff *skb, info = nla_data(attr); list_for_each_entry(from, &asoc->peer.transport_addr_list, transports) { - memcpy(info, &from->ipaddr, addrlen); + memcpy(info, &from->ipaddr, sizeof(from->ipaddr)); + memset(info + sizeof(from->ipaddr), 0, + addrlen - sizeof(from->ipaddr)); info += addrlen; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 1db478e..8d76086 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4538,8 +4538,7 @@ int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc, info->sctpi_ictrlchunks = asoc->stats.ictrlchunks; prim = asoc->peer.primary_path; - memcpy(&info->sctpi_p_address, &prim->ipaddr, - sizeof(struct sockaddr_storage)); + memcpy(&info->sctpi_p_address, &prim->ipaddr, sizeof(prim->ipaddr)); info->sctpi_p_state = prim->state; info->sctpi_p_cwnd = prim->cwnd; info->sctpi_p_srtt = prim->srtt; -- cgit v1.1 From 46f1c52e66dbc0d7a99f7c2a3c9debb497fe7b7b Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 23 Aug 2017 14:41:50 -0700 Subject: nfp: TX time stamp packets before HW doorbell is rung TX completion may happen any time after HW queue was kicked. We can't access the skb afterwards. Move the time stamping before ringing the doorbell. Fixes: 4c3523623dc0 ("net: add driver for Netronome NFP4000/NFP6000 NIC VFs") Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 1ff0c57..66a09e4 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -895,6 +895,8 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev) netdev_tx_sent_queue(nd_q, txbuf->real_len); + skb_tx_timestamp(skb); + tx_ring->wr_p += nr_frags + 1; if (nfp_net_tx_ring_should_stop(tx_ring)) nfp_net_tx_ring_stop(nd_q, tx_ring); @@ -903,8 +905,6 @@ static int nfp_net_tx(struct sk_buff *skb, struct net_device *netdev) if (!skb->xmit_more || netif_xmit_stopped(nd_q)) nfp_net_tx_xmit_more_flush(tx_ring); - skb_tx_timestamp(skb); - return NETDEV_TX_OK; err_unmap: -- cgit v1.1 From 87e9b3778c94694c9e098c91a0cc05725f0e017f Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 23 Aug 2017 19:34:03 -0400 Subject: bnxt_en: Fix .ndo_setup_tc() to include XDP rings. When the number of TX rings is changed in bnxt_setup_tc(), we need to include the XDP rings in the total TX ring count. Fixes: 38413406277f ("bnxt_en: Add support for XDP_TX action.") Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index e7c8539..f2f2bbf 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -7152,6 +7152,7 @@ int bnxt_setup_mq_tc(struct net_device *dev, u8 tc) bp->tx_nr_rings = bp->tx_nr_rings_per_tc; netdev_reset_tc(dev); } + bp->tx_nr_rings += bp->tx_nr_rings_xdp; bp->cp_nr_rings = sh ? max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) : bp->tx_nr_rings + bp->rx_nr_rings; bp->num_stat_ctxs = bp->cp_nr_rings; -- cgit v1.1 From 146ed3c5b87d8c65ec31bc56df26f027fe624b8f Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 23 Aug 2017 19:34:04 -0400 Subject: bnxt_en: Free MSIX vectors when unregistering the device from bnxt_re. Take back ownership of the MSIX vectors when unregistering the device from bnxt_re. Fixes: a588e4580a7e ("bnxt_en: Add interface to support RDMA driver.") Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c index 77da75a..997e10e 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c @@ -84,6 +84,8 @@ static int bnxt_unregister_dev(struct bnxt_en_dev *edev, int ulp_id) max_stat_ctxs = bnxt_get_max_func_stat_ctxs(bp); bnxt_set_max_func_stat_ctxs(bp, max_stat_ctxs + 1); + if (ulp->msix_requested) + edev->en_ops->bnxt_free_msix(edev, ulp_id); } if (ulp->max_async_event_id) bnxt_hwrm_func_rgtr_async_events(bp, NULL, 0); -- cgit v1.1 From a22a6ac2ff8080c87e446e20592725c064229c71 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 23 Aug 2017 19:34:05 -0400 Subject: bnxt_en: Do not setup MAC address in bnxt_hwrm_func_qcaps(). bnxt_hwrm_func_qcaps() is called during probe to get all device resources and it also sets up the factory MAC address. The same function is called when SRIOV is disabled to reclaim all resources. If the MAC address has been overridden by a user administered MAC address, calling this function will overwrite it. Separate the logic that sets up the default MAC address into a new function bnxt_init_mac_addr() that is only called during probe time. Fixes: 4a21b49b34c0 ("bnxt_en: Improve VF resource accounting.") Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 40 +++++++++++++++++++++---------- 1 file changed, 28 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index f2f2bbf..f20b3d2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4647,7 +4647,6 @@ static int bnxt_hwrm_func_qcaps(struct bnxt *bp) pf->port_id = le16_to_cpu(resp->port_id); bp->dev->dev_port = pf->port_id; memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN); - memcpy(bp->dev->dev_addr, pf->mac_addr, ETH_ALEN); pf->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx); pf->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings); pf->max_tx_rings = le16_to_cpu(resp->max_tx_rings); @@ -4687,16 +4686,6 @@ static int bnxt_hwrm_func_qcaps(struct bnxt *bp) vf->max_stat_ctxs = le16_to_cpu(resp->max_stat_ctx); memcpy(vf->mac_addr, resp->mac_address, ETH_ALEN); - mutex_unlock(&bp->hwrm_cmd_lock); - - if (is_valid_ether_addr(vf->mac_addr)) { - /* overwrite netdev dev_adr with admin VF MAC */ - memcpy(bp->dev->dev_addr, vf->mac_addr, ETH_ALEN); - } else { - eth_hw_addr_random(bp->dev); - rc = bnxt_approve_mac(bp, bp->dev->dev_addr); - } - return rc; #endif } @@ -7662,6 +7651,28 @@ void bnxt_restore_pf_fw_resources(struct bnxt *bp) bnxt_subtract_ulp_resources(bp, BNXT_ROCE_ULP); } +static int bnxt_init_mac_addr(struct bnxt *bp) +{ + int rc = 0; + + if (BNXT_PF(bp)) { + memcpy(bp->dev->dev_addr, bp->pf.mac_addr, ETH_ALEN); + } else { +#ifdef CONFIG_BNXT_SRIOV + struct bnxt_vf_info *vf = &bp->vf; + + if (is_valid_ether_addr(vf->mac_addr)) { + /* overwrite netdev dev_adr with admin VF MAC */ + memcpy(bp->dev->dev_addr, vf->mac_addr, ETH_ALEN); + } else { + eth_hw_addr_random(bp->dev); + rc = bnxt_approve_mac(bp, bp->dev->dev_addr); + } +#endif + } + return rc; +} + static void bnxt_parse_log_pcie_link(struct bnxt *bp) { enum pcie_link_width width = PCIE_LNK_WIDTH_UNKNOWN; @@ -7790,7 +7801,12 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) rc = -1; goto init_err_pci_clean; } - + rc = bnxt_init_mac_addr(bp); + if (rc) { + dev_err(&pdev->dev, "Unable to initialize mac address.\n"); + rc = -EADDRNOTAVAIL; + goto init_err_pci_clean; + } rc = bnxt_hwrm_queue_qportcfg(bp); if (rc) { netdev_err(bp->dev, "hwrm query qportcfg failure rc: %x\n", -- cgit v1.1 From 2fe59f507a65dbd734b990a11ebc7488f6f87a24 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Tue, 22 Aug 2017 18:43:48 +1000 Subject: timers: Fix excessive granularity of new timers after a nohz idle When a timer base is idle, it is forwarded when a new timer is added to ensure that granularity does not become excessive. When not idle, the timer tick is expected to increment the base. However there are several problems: - If an existing timer is modified, the base is forwarded only after the index is calculated. - The base is not forwarded by add_timer_on. - There is a window after a timer is restarted from a nohz idle, after it is marked not-idle and before the timer tick on this CPU, where a timer may be added but the ancient base does not get forwarded. These result in excessive granularity (a 1 jiffy timeout can blow out to 100s of jiffies), which cause the rcu lockup detector to trigger, among other things. Fix this by keeping track of whether the timer base has been idle since it was last run or forwarded, and if so then forward it before adding a new timer. There is still a case where mod_timer optimises the case of a pending timer mod with the same expiry time, where the timer can see excessive granularity relative to the new, shorter interval. A comment is added, but it's not changed because it is an important fastpath for networking. This has been tested and found to fix the RCU softlockup messages. Testing was also done with tracing to measure requested versus achieved wakeup latencies for all non-deferrable timers in an idle system (with no lockup watchdogs running). Wakeup latency relative to absolute latency is calculated (note this suffers from round-up skew at low absolute times) and analysed: max avg std upstream 506.0 1.20 4.68 patched 2.0 1.08 0.15 The bug was noticed due to the lockup detector Kconfig changes dropping it out of people's .configs and resulting in larger base clk skew When the lockup detectors are enabled, no CPU can go idle for longer than 4 seconds, which limits the granularity errors. Sub-optimal timer behaviour is observable on a smaller scale in that case: max avg std upstream 9.0 1.05 0.19 patched 2.0 1.04 0.11 Fixes: Fixes: a683f390b93f ("timers: Forward the wheel clock whenever possible") Signed-off-by: Nicholas Piggin Signed-off-by: Thomas Gleixner Tested-by: Jonathan Cameron Tested-by: David Miller Cc: dzickus@redhat.com Cc: sfr@canb.auug.org.au Cc: mpe@ellerman.id.au Cc: Stephen Boyd Cc: linuxarm@huawei.com Cc: abdhalee@linux.vnet.ibm.com Cc: John Stultz Cc: akpm@linux-foundation.org Cc: paulmck@linux.vnet.ibm.com Cc: torvalds@linux-foundation.org Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20170822084348.21436-1-npiggin@gmail.com --- kernel/time/timer.c | 50 +++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 41 insertions(+), 9 deletions(-) diff --git a/kernel/time/timer.c b/kernel/time/timer.c index 8f5d1bf..f2674a0 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -203,6 +203,7 @@ struct timer_base { bool migration_enabled; bool nohz_active; bool is_idle; + bool must_forward_clk; DECLARE_BITMAP(pending_map, WHEEL_SIZE); struct hlist_head vectors[WHEEL_SIZE]; } ____cacheline_aligned; @@ -856,13 +857,19 @@ get_target_base(struct timer_base *base, unsigned tflags) static inline void forward_timer_base(struct timer_base *base) { - unsigned long jnow = READ_ONCE(jiffies); + unsigned long jnow; /* - * We only forward the base when it's idle and we have a delta between - * base clock and jiffies. + * We only forward the base when we are idle or have just come out of + * idle (must_forward_clk logic), and have a delta between base clock + * and jiffies. In the common case, run_timers will take care of it. */ - if (!base->is_idle || (long) (jnow - base->clk) < 2) + if (likely(!base->must_forward_clk)) + return; + + jnow = READ_ONCE(jiffies); + base->must_forward_clk = base->is_idle; + if ((long)(jnow - base->clk) < 2) return; /* @@ -938,6 +945,11 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) * same array bucket then just return: */ if (timer_pending(timer)) { + /* + * The downside of this optimization is that it can result in + * larger granularity than you would get from adding a new + * timer with this expiry. + */ if (timer->expires == expires) return 1; @@ -948,6 +960,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) * dequeue/enqueue dance. */ base = lock_timer_base(timer, &flags); + forward_timer_base(base); clk = base->clk; idx = calc_wheel_index(expires, clk); @@ -964,6 +977,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) } } else { base = lock_timer_base(timer, &flags); + forward_timer_base(base); } ret = detach_if_pending(timer, base, false); @@ -991,12 +1005,10 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) raw_spin_lock(&base->lock); WRITE_ONCE(timer->flags, (timer->flags & ~TIMER_BASEMASK) | base->cpu); + forward_timer_base(base); } } - /* Try to forward a stale timer base clock */ - forward_timer_base(base); - timer->expires = expires; /* * If 'idx' was calculated above and the base time did not advance @@ -1112,6 +1124,7 @@ void add_timer_on(struct timer_list *timer, int cpu) WRITE_ONCE(timer->flags, (timer->flags & ~TIMER_BASEMASK) | cpu); } + forward_timer_base(base); debug_activate(timer, timer->expires); internal_add_timer(base, timer); @@ -1497,10 +1510,16 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) if (!is_max_delta) expires = basem + (u64)(nextevt - basej) * TICK_NSEC; /* - * If we expect to sleep more than a tick, mark the base idle: + * If we expect to sleep more than a tick, mark the base idle. + * Also the tick is stopped so any added timer must forward + * the base clk itself to keep granularity small. This idle + * logic is only maintained for the BASE_STD base, deferrable + * timers may still see large granularity skew (by design). */ - if ((expires - basem) > TICK_NSEC) + if ((expires - basem) > TICK_NSEC) { + base->must_forward_clk = true; base->is_idle = true; + } } raw_spin_unlock(&base->lock); @@ -1611,6 +1630,19 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h) { struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]); + /* + * must_forward_clk must be cleared before running timers so that any + * timer functions that call mod_timer will not try to forward the + * base. idle trcking / clock forwarding logic is only used with + * BASE_STD timers. + * + * The deferrable base does not do idle tracking at all, so we do + * not forward it. This can result in very large variations in + * granularity for deferrable timers, but they can be deferred for + * long periods due to idle. + */ + base->must_forward_clk = false; + __run_timers(base); if (IS_ENABLED(CONFIG_NO_HZ_COMMON) && base->nohz_active) __run_timers(this_cpu_ptr(&timer_bases[BASE_DEF])); -- cgit v1.1 From bd0fdb191c8523a9126bb14ac1b22cb47698ebf5 Mon Sep 17 00:00:00 2001 From: Nicholas Piggin Date: Mon, 13 Mar 2017 03:03:49 +1000 Subject: KVM: PPC: Book3S HV: Use msgsync with hypervisor doorbells on POWER9 When msgsnd is used for IPIs to other cores, msgsync must be executed by the target to order stores performed on the source before its msgsnd (provided the source executes the appropriate sync). Fixes: 1704a81ccebc ("KVM: PPC: Book3S HV: Use msgsnd for IPIs to other cores on POWER9") Signed-off-by: Nicholas Piggin Signed-off-by: Paul Mackerras --- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index c52184a..9c9c983 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -1291,6 +1291,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR) /* Hypervisor doorbell - exit only if host IPI flag set */ cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL bne 3f +BEGIN_FTR_SECTION + PPC_MSGSYNC +END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300) lbz r0, HSTATE_HOST_IPI(r13) cmpwi r0, 0 beq 4f -- cgit v1.1 From 2c4fb78f78b6e420604ee1b05bdfb5c1d637869f Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 18 Aug 2017 12:10:52 +1000 Subject: KVM: PPC: Book3S HV: Workaround POWER9 DD1.0 bug causing IPB bit loss This adds a workaround for a bug in POWER9 DD1 chips where changing the CPPR (Current Processor Priority Register) can cause bits in the IPB (Interrupt Pending Buffer) to get lost. Thankfully it only happens when manually manipulating CPPR which is quite rare. When it does happen it can cause interrupts to be delayed or lost. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kvm/book3s_xive_template.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/kvm/book3s_xive_template.c b/arch/powerpc/kvm/book3s_xive_template.c index 4636ca6..150be86 100644 --- a/arch/powerpc/kvm/book3s_xive_template.c +++ b/arch/powerpc/kvm/book3s_xive_template.c @@ -16,7 +16,16 @@ static void GLUE(X_PFX,ack_pending)(struct kvmppc_xive_vcpu *xc) u8 cppr; u16 ack; - /* XXX DD1 bug workaround: Check PIPR vs. CPPR first ! */ + /* + * DD1 bug workaround: If PIPR is less favored than CPPR + * ignore the interrupt or we might incorrectly lose an IPB + * bit. + */ + if (cpu_has_feature(CPU_FTR_POWER9_DD1)) { + u8 pipr = __x_readb(__x_tima + TM_QW1_OS + TM_PIPR); + if (pipr >= xc->hw_cppr) + return; + } /* Perform the acknowledge OS to register cycle. */ ack = be16_to_cpu(__x_readw(__x_tima + TM_SPC_ACK_OS_REG)); -- cgit v1.1 From bb9b52bd51dcb17b965a30167d0812902c1b9927 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 18 Aug 2017 12:10:58 +1000 Subject: KVM: PPC: Book3S HV: Add missing barriers to XIVE code and document them This adds missing memory barriers to order updates/tests of the virtual CPPR and MFRR, thus fixing a lost IPI problem. While at it also document all barriers in this file. This fixes a bug causing guest IPIs to occasionally get lost. The symptom then is hangs or stalls in the guest. Signed-off-by: Benjamin Herrenschmidt Tested-by: Guilherme G. Piccoli Signed-off-by: Paul Mackerras --- arch/powerpc/kvm/book3s_xive_template.c | 57 +++++++++++++++++++++++++++++++-- 1 file changed, 55 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kvm/book3s_xive_template.c b/arch/powerpc/kvm/book3s_xive_template.c index 150be86..d1ed2c4 100644 --- a/arch/powerpc/kvm/book3s_xive_template.c +++ b/arch/powerpc/kvm/book3s_xive_template.c @@ -17,6 +17,12 @@ static void GLUE(X_PFX,ack_pending)(struct kvmppc_xive_vcpu *xc) u16 ack; /* + * Ensure any previous store to CPPR is ordered vs. + * the subsequent loads from PIPR or ACK. + */ + eieio(); + + /* * DD1 bug workaround: If PIPR is less favored than CPPR * ignore the interrupt or we might incorrectly lose an IPB * bit. @@ -244,6 +250,11 @@ skip_ipi: /* * If we found an interrupt, adjust what the guest CPPR should * be as if we had just fetched that interrupt from HW. + * + * Note: This can only make xc->cppr smaller as the previous + * loop will only exit with hirq != 0 if prio is lower than + * the current xc->cppr. Thus we don't need to re-check xc->mfrr + * for pending IPIs. */ if (hirq) xc->cppr = prio; @@ -390,6 +401,12 @@ X_STATIC int GLUE(X_PFX,h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr) xc->cppr = cppr; /* + * Order the above update of xc->cppr with the subsequent + * read of xc->mfrr inside push_pending_to_hw() + */ + smp_mb(); + + /* * We are masking less, we need to look for pending things * to deliver and set VP pending bits accordingly to trigger * a new interrupt otherwise we might miss MFRR changes for @@ -429,21 +446,37 @@ X_STATIC int GLUE(X_PFX,h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr) * used to signal MFRR changes is EOId when fetched from * the queue. */ - if (irq == XICS_IPI || irq == 0) + if (irq == XICS_IPI || irq == 0) { + /* + * This barrier orders the setting of xc->cppr vs. + * subsquent test of xc->mfrr done inside + * scan_interrupts and push_pending_to_hw + */ + smp_mb(); goto bail; + } /* Find interrupt source */ sb = kvmppc_xive_find_source(xive, irq, &src); if (!sb) { pr_devel(" source not found !\n"); rc = H_PARAMETER; + /* Same as above */ + smp_mb(); goto bail; } state = &sb->irq_state[src]; kvmppc_xive_select_irq(state, &hw_num, &xd); state->in_eoi = true; - mb(); + + /* + * This barrier orders both setting of in_eoi above vs, + * subsequent test of guest_priority, and the setting + * of xc->cppr vs. subsquent test of xc->mfrr done inside + * scan_interrupts and push_pending_to_hw + */ + smp_mb(); again: if (state->guest_priority == MASKED) { @@ -470,6 +503,14 @@ again: } + /* + * This barrier orders the above guest_priority check + * and spin_lock/unlock with clearing in_eoi below. + * + * It also has to be a full mb() as it must ensure + * the MMIOs done in source_eoi() are completed before + * state->in_eoi is visible. + */ mb(); state->in_eoi = false; bail: @@ -504,6 +545,18 @@ X_STATIC int GLUE(X_PFX,h_ipi)(struct kvm_vcpu *vcpu, unsigned long server, /* Locklessly write over MFRR */ xc->mfrr = mfrr; + /* + * The load of xc->cppr below and the subsequent MMIO store + * to the IPI must happen after the above mfrr update is + * globally visible so that: + * + * - Synchronize with another CPU doing an H_EOI or a H_CPPR + * updating xc->cppr then reading xc->mfrr. + * + * - The target of the IPI sees the xc->mfrr update + */ + mb(); + /* Shoot the IPI if most favored than target cppr */ if (mfrr < xc->cppr) __x_writeq(0, __x_trig_page(&xc->vp_ipi_data)); -- cgit v1.1 From 8a4b5784fac2af93ad6a8c08fb56d021e0c0416b Mon Sep 17 00:00:00 2001 From: Lorenzo Colitti Date: Wed, 23 Aug 2017 17:14:39 +0900 Subject: net: xfrm: don't double-hold dst when sk_policy in use. While removing dst_entry garbage collection, commit 52df157f17e5 ("xfrm: take refcnt of dst when creating struct xfrm_dst bundle") changed xfrm_resolve_and_create_bundle so it returns an xdst with a refcount of 1 instead of 0. However, it did not delete the dst_hold performed by xfrm_lookup when a per-socket policy is in use. This means that when a socket policy is in use, dst entries returned by xfrm_lookup have a refcount of 2, and are not freed when no longer in use. Cc: Wei Wang Fixes: 52df157f17 ("xfrm: take refcnt of dst when creating struct xfrm_dst bundle") Tested: https://android-review.googlesource.com/417481 Tested: https://android-review.googlesource.com/418659 Tested: https://android-review.googlesource.com/424463 Tested: https://android-review.googlesource.com/452776 passes on net-next Signed-off-by: Lorenzo Colitti Acked-by: Wei Wang Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 6f5a0dad..69b16ee 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2226,7 +2226,6 @@ struct dst_entry *xfrm_lookup(struct net *net, struct dst_entry *dst_orig, goto no_transform; } - dst_hold(&xdst->u.dst); route = xdst->route; } } -- cgit v1.1 From 98cd249cf9d2c7e2322fbf20c454c019e141a28b Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 20 Jul 2017 06:58:25 +0200 Subject: c6x: defconfig: Cleanup from old Kconfig options Remove old, dead Kconfig options (in order appearing in this commit): - EXPERIMENTAL is gone since v3.9; - MISC_DEVICES: commit 7c5763b8453a ("drivers: misc: Remove MISC_DEVICES config option"); Signed-off-by: Krzysztof Kozlowski Signed-off-by: Mark Salter --- arch/c6x/configs/dsk6455_defconfig | 2 -- arch/c6x/configs/evmc6457_defconfig | 2 -- arch/c6x/configs/evmc6472_defconfig | 2 -- arch/c6x/configs/evmc6474_defconfig | 2 -- arch/c6x/configs/evmc6678_defconfig | 2 -- 5 files changed, 10 deletions(-) diff --git a/arch/c6x/configs/dsk6455_defconfig b/arch/c6x/configs/dsk6455_defconfig index 4663487..d764ea4 100644 --- a/arch/c6x/configs/dsk6455_defconfig +++ b/arch/c6x/configs/dsk6455_defconfig @@ -1,5 +1,4 @@ CONFIG_SOC_TMS320C6455=y -CONFIG_EXPERIMENTAL=y # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SYSVIPC=y CONFIG_SPARSE_IRQ=y @@ -25,7 +24,6 @@ CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=2 CONFIG_BLK_DEV_RAM_SIZE=17000 -CONFIG_MISC_DEVICES=y # CONFIG_INPUT is not set # CONFIG_SERIO is not set # CONFIG_VT is not set diff --git a/arch/c6x/configs/evmc6457_defconfig b/arch/c6x/configs/evmc6457_defconfig index bba40e1..05d0b4a 100644 --- a/arch/c6x/configs/evmc6457_defconfig +++ b/arch/c6x/configs/evmc6457_defconfig @@ -1,5 +1,4 @@ CONFIG_SOC_TMS320C6457=y -CONFIG_EXPERIMENTAL=y # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SYSVIPC=y CONFIG_SPARSE_IRQ=y @@ -26,7 +25,6 @@ CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=2 CONFIG_BLK_DEV_RAM_SIZE=17000 -CONFIG_MISC_DEVICES=y # CONFIG_INPUT is not set # CONFIG_SERIO is not set # CONFIG_VT is not set diff --git a/arch/c6x/configs/evmc6472_defconfig b/arch/c6x/configs/evmc6472_defconfig index 8c46155..8d81fcf 100644 --- a/arch/c6x/configs/evmc6472_defconfig +++ b/arch/c6x/configs/evmc6472_defconfig @@ -1,5 +1,4 @@ CONFIG_SOC_TMS320C6472=y -CONFIG_EXPERIMENTAL=y # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SYSVIPC=y CONFIG_SPARSE_IRQ=y @@ -27,7 +26,6 @@ CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=2 CONFIG_BLK_DEV_RAM_SIZE=17000 -CONFIG_MISC_DEVICES=y # CONFIG_INPUT is not set # CONFIG_SERIO is not set # CONFIG_VT is not set diff --git a/arch/c6x/configs/evmc6474_defconfig b/arch/c6x/configs/evmc6474_defconfig index 15533f6..8156a98 100644 --- a/arch/c6x/configs/evmc6474_defconfig +++ b/arch/c6x/configs/evmc6474_defconfig @@ -1,5 +1,4 @@ CONFIG_SOC_TMS320C6474=y -CONFIG_EXPERIMENTAL=y # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SYSVIPC=y CONFIG_SPARSE_IRQ=y @@ -27,7 +26,6 @@ CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=2 CONFIG_BLK_DEV_RAM_SIZE=17000 -CONFIG_MISC_DEVICES=y # CONFIG_INPUT is not set # CONFIG_SERIO is not set # CONFIG_VT is not set diff --git a/arch/c6x/configs/evmc6678_defconfig b/arch/c6x/configs/evmc6678_defconfig index 5f126d4..c4f433c 100644 --- a/arch/c6x/configs/evmc6678_defconfig +++ b/arch/c6x/configs/evmc6678_defconfig @@ -1,5 +1,4 @@ CONFIG_SOC_TMS320C6678=y -CONFIG_EXPERIMENTAL=y # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SYSVIPC=y CONFIG_SPARSE_IRQ=y @@ -27,7 +26,6 @@ CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=2 CONFIG_BLK_DEV_RAM_SIZE=17000 -CONFIG_MISC_DEVICES=y # CONFIG_INPUT is not set # CONFIG_SERIO is not set # CONFIG_VT is not set -- cgit v1.1 From 636d42117800db1a994726fcf017e3633db832a5 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Tue, 18 Jul 2017 16:42:43 -0500 Subject: c6x: Convert to using %pOF instead of full_name Now that we have a custom printf format specifier, convert users of full_name to use %pOF instead. This is preparation to remove storing of the full path string for each node. Signed-off-by: Rob Herring Cc: Mark Salter Cc: Aurelien Jacquiot Cc: linux-c6x-dev@linux-c6x.org Signed-off-by: Mark Salter --- arch/c6x/platforms/megamod-pic.c | 22 +++++++++++----------- arch/c6x/platforms/plldata.c | 4 ++-- arch/c6x/platforms/timer64.c | 8 ++++---- 3 files changed, 17 insertions(+), 17 deletions(-) diff --git a/arch/c6x/platforms/megamod-pic.c b/arch/c6x/platforms/megamod-pic.c index 43afc03..9519fa5 100644 --- a/arch/c6x/platforms/megamod-pic.c +++ b/arch/c6x/platforms/megamod-pic.c @@ -208,14 +208,14 @@ static struct megamod_pic * __init init_megamod_pic(struct device_node *np) pic = kzalloc(sizeof(struct megamod_pic), GFP_KERNEL); if (!pic) { - pr_err("%s: Could not alloc PIC structure.\n", np->full_name); + pr_err("%pOF: Could not alloc PIC structure.\n", np); return NULL; } pic->irqhost = irq_domain_add_linear(np, NR_COMBINERS * 32, &megamod_domain_ops, pic); if (!pic->irqhost) { - pr_err("%s: Could not alloc host.\n", np->full_name); + pr_err("%pOF: Could not alloc host.\n", np); goto error_free; } @@ -225,7 +225,7 @@ static struct megamod_pic * __init init_megamod_pic(struct device_node *np) pic->regs = of_iomap(np, 0); if (!pic->regs) { - pr_err("%s: Could not map registers.\n", np->full_name); + pr_err("%pOF: Could not map registers.\n", np); goto error_free; } @@ -253,8 +253,8 @@ static struct megamod_pic * __init init_megamod_pic(struct device_node *np) irq_data = irq_get_irq_data(irq); if (!irq_data) { - pr_err("%s: combiner-%d no irq_data for virq %d!\n", - np->full_name, i, irq); + pr_err("%pOF: combiner-%d no irq_data for virq %d!\n", + np, i, irq); continue; } @@ -265,16 +265,16 @@ static struct megamod_pic * __init init_megamod_pic(struct device_node *np) * of the core priority interrupts (4 - 15). */ if (hwirq < 4 || hwirq >= NR_PRIORITY_IRQS) { - pr_err("%s: combiner-%d core irq %ld out of range!\n", - np->full_name, i, hwirq); + pr_err("%pOF: combiner-%d core irq %ld out of range!\n", + np, i, hwirq); continue; } /* record the mapping */ mapping[hwirq - 4] = i; - pr_debug("%s: combiner-%d cascading to hwirq %ld\n", - np->full_name, i, hwirq); + pr_debug("%pOF: combiner-%d cascading to hwirq %ld\n", + np, i, hwirq); cascade_data[i].pic = pic; cascade_data[i].index = i; @@ -290,8 +290,8 @@ static struct megamod_pic * __init init_megamod_pic(struct device_node *np) /* Finally, set up the MUX registers */ for (i = 0; i < NR_MUX_OUTPUTS; i++) { if (mapping[i] != IRQ_UNMAPPED) { - pr_debug("%s: setting mux %d to priority %d\n", - np->full_name, mapping[i], i + 4); + pr_debug("%pOF: setting mux %d to priority %d\n", + np, mapping[i], i + 4); set_megamod_mux(pic, mapping[i], i); } } diff --git a/arch/c6x/platforms/plldata.c b/arch/c6x/platforms/plldata.c index 755359e..e8b6cc6 100644 --- a/arch/c6x/platforms/plldata.c +++ b/arch/c6x/platforms/plldata.c @@ -436,8 +436,8 @@ void __init c64x_setup_clocks(void) err = of_property_read_u32(node, "clock-frequency", &val); if (err || val == 0) { - pr_err("%s: no clock-frequency found! Using %dMHz\n", - node->full_name, (int)val / 1000000); + pr_err("%pOF: no clock-frequency found! Using %dMHz\n", + node, (int)val / 1000000); val = 25000000; } clkin1.rate = val; diff --git a/arch/c6x/platforms/timer64.c b/arch/c6x/platforms/timer64.c index 0bd0452..241a9a6 100644 --- a/arch/c6x/platforms/timer64.c +++ b/arch/c6x/platforms/timer64.c @@ -204,14 +204,14 @@ void __init timer64_init(void) timer = of_iomap(np, 0); if (!timer) { - pr_debug("%s: Cannot map timer registers.\n", np->full_name); + pr_debug("%pOF: Cannot map timer registers.\n", np); goto out; } - pr_debug("%s: Timer registers=%p.\n", np->full_name, timer); + pr_debug("%pOF: Timer registers=%p.\n", np, timer); cd->irq = irq_of_parse_and_map(np, 0); if (cd->irq == NO_IRQ) { - pr_debug("%s: Cannot find interrupt.\n", np->full_name); + pr_debug("%pOF: Cannot find interrupt.\n", np); iounmap(timer); goto out; } @@ -229,7 +229,7 @@ void __init timer64_init(void) dscr_set_devstate(timer64_devstate_id, DSCR_DEVSTATE_ENABLED); } - pr_debug("%s: Timer irq=%d.\n", np->full_name, cd->irq); + pr_debug("%pOF: Timer irq=%d.\n", np, cd->irq); clockevents_calc_mult_shift(cd, c6x_core_freq / TIMER_DIVISOR, 5); -- cgit v1.1 From 10a54d8196d11f6cc0db2f71249f93854cb9fe55 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Tue, 22 Aug 2017 10:37:29 +0300 Subject: iwlwifi: pcie: move rx workqueue initialization to iwl_trans_pcie_alloc() Work queues cannot be allocated when a mutex is held because the mutex may be in use and that would make it sleep. Doing so generates the following splat with 4.13+: [ 19.513298] ====================================================== [ 19.513429] WARNING: possible circular locking dependency detected [ 19.513557] 4.13.0-rc5+ #6 Not tainted [ 19.513638] ------------------------------------------------------ [ 19.513767] cpuhp/0/12 is trying to acquire lock: [ 19.513867] (&tz->lock){+.+.+.}, at: [] thermal_zone_get_temp+0x5b/0xb0 [ 19.514047] [ 19.514047] but task is already holding lock: [ 19.514166] (cpuhp_state){+.+.+.}, at: [] cpuhp_thread_fun+0x3a/0x210 [ 19.514338] [ 19.514338] which lock already depends on the new lock. This lock dependency already existed with previous kernel versions, but it was not detected until commit 49dfe2a67797 ("cpuhotplug: Link lock stacks for hotplug callbacks") was introduced. Reported-by: David Weinehall Reported-by: Jiri Kosina Signed-off-by: Luca Coelho Signed-off-by: Kalle Valo --- drivers/net/wireless/intel/iwlwifi/pcie/internal.h | 2 ++ drivers/net/wireless/intel/iwlwifi/pcie/rx.c | 10 +--------- drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 9 +++++++++ 3 files changed, 12 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index fa315d8..a1ea9ef 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -787,6 +787,8 @@ int iwl_pci_fw_enter_d0i3(struct iwl_trans *trans); void iwl_pcie_enable_rx_wake(struct iwl_trans *trans, bool enable); +void iwl_pcie_rx_allocator_work(struct work_struct *data); + /* common functions that are used by gen2 transport */ void iwl_pcie_apm_config(struct iwl_trans *trans); int iwl_pcie_prepare_card_hw(struct iwl_trans *trans); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 351c442..942736d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -597,7 +597,7 @@ static void iwl_pcie_rx_allocator_get(struct iwl_trans *trans, rxq->free_count += RX_CLAIM_REQ_ALLOC; } -static void iwl_pcie_rx_allocator_work(struct work_struct *data) +void iwl_pcie_rx_allocator_work(struct work_struct *data) { struct iwl_rb_allocator *rba_p = container_of(data, struct iwl_rb_allocator, rx_alloc); @@ -900,10 +900,6 @@ static int _iwl_pcie_rx_init(struct iwl_trans *trans) return err; } def_rxq = trans_pcie->rxq; - if (!rba->alloc_wq) - rba->alloc_wq = alloc_workqueue("rb_allocator", - WQ_HIGHPRI | WQ_UNBOUND, 1); - INIT_WORK(&rba->rx_alloc, iwl_pcie_rx_allocator_work); spin_lock(&rba->lock); atomic_set(&rba->req_pending, 0); @@ -1017,10 +1013,6 @@ void iwl_pcie_rx_free(struct iwl_trans *trans) } cancel_work_sync(&rba->rx_alloc); - if (rba->alloc_wq) { - destroy_workqueue(rba->alloc_wq); - rba->alloc_wq = NULL; - } iwl_pcie_free_rbs_pool(trans); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index f95eec5..3927bbf 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -1786,6 +1786,11 @@ void iwl_trans_pcie_free(struct iwl_trans *trans) iwl_pcie_tx_free(trans); iwl_pcie_rx_free(trans); + if (trans_pcie->rba.alloc_wq) { + destroy_workqueue(trans_pcie->rba.alloc_wq); + trans_pcie->rba.alloc_wq = NULL; + } + if (trans_pcie->msix_enabled) { for (i = 0; i < trans_pcie->alloc_vecs; i++) { irq_set_affinity_hint( @@ -3169,6 +3174,10 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, trans_pcie->inta_mask = CSR_INI_SET_MASK; } + trans_pcie->rba.alloc_wq = alloc_workqueue("rb_allocator", + WQ_HIGHPRI | WQ_UNBOUND, 1); + INIT_WORK(&trans_pcie->rba.rx_alloc, iwl_pcie_rx_allocator_work); + #ifdef CONFIG_IWLWIFI_PCIE_RTPM trans->runtime_pm_mode = IWL_PLAT_PM_MODE_D0I3; #else -- cgit v1.1 From a8f0f9e49956a74718874b800251455680085600 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Thu, 17 Aug 2017 16:37:25 -0400 Subject: ftrace: Check for null ret_stack on profile function graph entry function There's a small race when function graph shutsdown and the calling of the registered function graph entry callback. The callback must not reference the task's ret_stack without first checking that it is not NULL. Note, when a ret_stack is allocated for a task, it stays allocated until the task exits. The problem here, is that function_graph is shutdown, and a new task was created, which doesn't have its ret_stack allocated. But since some of the functions are still being traced, the callbacks can still be called. The normal function_graph code handles this, but starting with commit 8861dd303c ("ftrace: Access ret_stack->subtime only in the function profiler") the profiler code references the ret_stack on function entry, but doesn't check if it is NULL first. Link: https://bugzilla.kernel.org/show_bug.cgi?id=196611 Cc: stable@vger.kernel.org Fixes: 8861dd303c ("ftrace: Access ret_stack->subtime only in the function profiler") Reported-by: lilydjwg@gmail.com Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 02004ae9..96cea88 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -889,6 +889,10 @@ static int profile_graph_entry(struct ftrace_graph_ent *trace) function_profile_call(trace->func, 0, NULL, NULL); + /* If function graph is shutting down, ret_stack can be NULL */ + if (!current->ret_stack) + return 0; + if (index >= 0 && index < FTRACE_RETFUNC_DEPTH) current->ret_stack[index].subtime = 0; -- cgit v1.1 From 475bb3c69ab05df2a6ecef6acc2393703d134180 Mon Sep 17 00:00:00 2001 From: Chunyu Hu Date: Mon, 14 Aug 2017 18:18:17 +0800 Subject: tracing: Fix kmemleak in tracing_map_array_free() kmemleak reported the below leak when I was doing clear of the hist trigger. With this patch, the kmeamleak is gone. unreferenced object 0xffff94322b63d760 (size 32): comm "bash", pid 1522, jiffies 4403687962 (age 2442.311s) hex dump (first 32 bytes): 00 01 00 00 04 00 00 00 08 00 00 00 ff 00 00 00 ................ 10 00 00 00 00 00 00 00 80 a8 7a f2 31 94 ff ff ..........z.1... backtrace: [] kmemleak_alloc+0x4a/0xa0 [] kmem_cache_alloc_trace+0xca/0x1d0 [] tracing_map_array_alloc+0x26/0x140 [] kretprobe_trampoline+0x0/0x50 [] create_hist_data+0x535/0x750 [] event_hist_trigger_func+0x1f7/0x420 [] event_trigger_write+0xfd/0x1a0 [] __vfs_write+0x37/0x170 [] vfs_write+0xb2/0x1b0 [] SyS_write+0x55/0xc0 [] do_syscall_64+0x67/0x150 [] return_from_SYSCALL_64+0x0/0x6a [] 0xffffffffffffffff unreferenced object 0xffff9431f27aa880 (size 128): comm "bash", pid 1522, jiffies 4403687962 (age 2442.311s) hex dump (first 32 bytes): 00 00 8c 2a 32 94 ff ff 00 f0 8b 2a 32 94 ff ff ...*2......*2... 00 e0 8b 2a 32 94 ff ff 00 d0 8b 2a 32 94 ff ff ...*2......*2... backtrace: [] kmemleak_alloc+0x4a/0xa0 [] __kmalloc+0xe8/0x220 [] tracing_map_array_alloc+0xb1/0x140 [] kretprobe_trampoline+0x0/0x50 [] create_hist_data+0x535/0x750 [] event_hist_trigger_func+0x1f7/0x420 [] event_trigger_write+0xfd/0x1a0 [] __vfs_write+0x37/0x170 [] vfs_write+0xb2/0x1b0 [] SyS_write+0x55/0xc0 [] do_syscall_64+0x67/0x150 [] return_from_SYSCALL_64+0x0/0x6a [] 0xffffffffffffffff Link: http://lkml.kernel.org/r/1502705898-27571-1-git-send-email-chuhu@redhat.com Cc: stable@vger.kernel.org Fixes: 08d43a5fa063 ("tracing: Add lock-free tracing_map") Signed-off-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/tracing_map.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c index 0a689bb..305039b 100644 --- a/kernel/trace/tracing_map.c +++ b/kernel/trace/tracing_map.c @@ -221,16 +221,19 @@ void tracing_map_array_free(struct tracing_map_array *a) if (!a) return; - if (!a->pages) { - kfree(a); - return; - } + if (!a->pages) + goto free; for (i = 0; i < a->n_pages; i++) { if (!a->pages[i]) break; free_page((unsigned long)a->pages[i]); } + + kfree(a->pages); + + free: + kfree(a); } struct tracing_map_array *tracing_map_array_alloc(unsigned int n_elts, -- cgit v1.1 From 8b0db1a5bdfcee0dbfa89607672598ae203c9045 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Wed, 23 Aug 2017 12:46:27 -0400 Subject: tracing: Fix freeing of filter in create_filter() when set_str is false Performing the following task with kmemleak enabled: # cd /sys/kernel/tracing/events/irq/irq_handler_entry/ # echo 'enable_event:kmem:kmalloc:3 if irq >' > trigger # echo 'enable_event:kmem:kmalloc:3 if irq > 31' > trigger # echo scan > /sys/kernel/debug/kmemleak # cat /sys/kernel/debug/kmemleak unreferenced object 0xffff8800b9290308 (size 32): comm "bash", pid 1114, jiffies 4294848451 (age 141.139s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmemleak_alloc+0x4a/0xa0 [] kmem_cache_alloc_trace+0x158/0x290 [] create_filter_start.constprop.28+0x99/0x940 [] create_filter+0xa9/0x160 [] create_event_filter+0xc/0x10 [] set_trigger_filter+0xe5/0x210 [] event_enable_trigger_func+0x324/0x490 [] event_trigger_write+0x1a2/0x260 [] __vfs_write+0xd7/0x380 [] vfs_write+0x101/0x260 [] SyS_write+0xab/0x130 [] entry_SYSCALL_64_fastpath+0x1f/0xbe [] 0xffffffffffffffff The function create_filter() is passed a 'filterp' pointer that gets allocated, and if "set_str" is true, it is up to the caller to free it, even on error. The problem is that the pointer is not freed by create_filter() when set_str is false. This is a bug, and it is not up to the caller to free the filter on error if it doesn't care about the string. Link: http://lkml.kernel.org/r/1502705898-27571-2-git-send-email-chuhu@redhat.com Cc: stable@vger.kernel.org Fixes: 38b78eb85 ("tracing: Factorize filter creation") Reported-by: Chunyu Hu Tested-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_filter.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 59a411f..181e139 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1959,6 +1959,10 @@ static int create_filter(struct trace_event_call *call, if (err && set_str) append_filter_err(ps, filter); } + if (err && !set_str) { + free_event_filter(filter); + filter = NULL; + } create_filter_finish(ps); *filterp = filter; -- cgit v1.1 From ab6dd1beac7be3c17f8bf3d38bdf29ecb7293f1e Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 10 Aug 2017 10:22:24 +0800 Subject: netfilter: check for seqadj ext existence before adding it in nf_nat_setup_info Commit 4440a2ab3b9f ("netfilter: synproxy: Check oom when adding synproxy and seqadj ct extensions") wanted to drop the packet when it fails to add seqadj ext due to no memory by checking if nfct_seqadj_ext_add returns NULL. But that nfct_seqadj_ext_add returns NULL can also happen when seqadj ext already exists in a nf_conn. It will cause that userspace protocol doesn't work when both dnat and snat are configured. Li Shuang found this issue in the case: Topo: ftp client router ftp server 10.167.131.2 <-> 10.167.131.254 10.167.141.254 <-> 10.167.141.1 Rules: # iptables -t nat -A PREROUTING -i eth1 -p tcp -m tcp --dport 21 -j \ DNAT --to-destination 10.167.141.1 # iptables -t nat -A POSTROUTING -o eth2 -p tcp -m tcp --dport 21 -j \ SNAT --to-source 10.167.141.254 In router, when both dnat and snat are added, nf_nat_setup_info will be called twice. The packet can be dropped at the 2nd time for DNAT due to seqadj ext is already added at the 1st time for SNAT. This patch is to fix it by checking for seqadj ext existence before adding it, so that the packet will not be dropped if seqadj ext already exists. Note that as Florian mentioned, as a long term, we should review ext_add() behaviour, it's better to return a pointer to the existing ext instead. Fixes: 4440a2ab3b9f ("netfilter: synproxy: Check oom when adding synproxy and seqadj ct extensions") Reported-by: Li Shuang Acked-by: Florian Westphal Signed-off-by: Xin Long Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_nat_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index eb54178..b1d3740 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -441,7 +441,7 @@ nf_nat_setup_info(struct nf_conn *ct, else ct->status |= IPS_DST_NAT; - if (nfct_help(ct)) + if (nfct_help(ct) && !nfct_seqadj(ct)) if (!nfct_seqadj_ext_add(ct)) return NF_DROP; } -- cgit v1.1 From 50b4d485528d1dbe0bd249f2073140e3444f4a7b Mon Sep 17 00:00:00 2001 From: Benjamin Block Date: Thu, 24 Aug 2017 01:57:56 +0200 Subject: bsg-lib: fix kernel panic resulting from missing allocation of reply-buffer Since we split the scsi_request out of struct request bsg fails to provide a reply-buffer for the drivers. This was done via the pointer for sense-data, that is not preallocated anymore. Failing to allocate/assign it results in illegal dereferences because LLDs use this pointer unquestioned. An example panic on s390x, using the zFCP driver, looks like this (I had debugging on, otherwise NULL-pointer dereferences wouldn't even panic on s390x): Unable to handle kernel pointer dereference in virtual kernel address space Failing address: 6b6b6b6b6b6b6000 TEID: 6b6b6b6b6b6b6403 Fault in home space mode while using kernel ASCE. AS:0000000001590007 R3:0000000000000024 Oops: 0038 ilc:2 [#1] PREEMPT SMP DEBUG_PAGEALLOC Modules linked in: CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.12.0-bsg-regression+ #3 Hardware name: IBM 2964 N96 702 (z/VM 6.4.0) task: 0000000065cb0100 task.stack: 0000000065cb4000 Krnl PSW : 0704e00180000000 000003ff801e4156 (zfcp_fc_ct_els_job_handler+0x16/0x58 [zfcp]) R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:2 PM:0 RI:0 EA:3 Krnl GPRS: 0000000000000001 000000005fa9d0d0 000000005fa9d078 0000000000e16866 000003ff00000290 6b6b6b6b6b6b6b6b 0000000059f78f00 000000000000000f 00000000593a0958 00000000593a0958 0000000060d88800 000000005ddd4c38 0000000058b50100 07000000659cba08 000003ff801e8556 00000000659cb9a8 Krnl Code: 000003ff801e4146: e31020500004 lg %r1,80(%r2) 000003ff801e414c: 58402040 l %r4,64(%r2) #000003ff801e4150: e35020200004 lg %r5,32(%r2) >000003ff801e4156: 50405004 st %r4,4(%r5) 000003ff801e415a: e54c50080000 mvhi 8(%r5),0 000003ff801e4160: e33010280012 lt %r3,40(%r1) 000003ff801e4166: a718fffb lhi %r1,-5 000003ff801e416a: 1803 lr %r0,%r3 Call Trace: ([<000003ff801e8556>] zfcp_fsf_req_complete+0x726/0x768 [zfcp]) [<000003ff801ea82a>] zfcp_fsf_reqid_check+0x102/0x180 [zfcp] [<000003ff801eb980>] zfcp_qdio_int_resp+0x230/0x278 [zfcp] [<00000000009b91b6>] qdio_kick_handler+0x2ae/0x2c8 [<00000000009b9e3e>] __tiqdio_inbound_processing+0x406/0xc10 [<00000000001684c2>] tasklet_action+0x15a/0x1d8 [<0000000000bd28ec>] __do_softirq+0x3ec/0x848 [<00000000001675a4>] irq_exit+0x74/0xf8 [<000000000010dd6a>] do_IRQ+0xba/0xf0 [<0000000000bd19e8>] io_int_handler+0x104/0x2d4 [<00000000001033b6>] enabled_wait+0xb6/0x188 ([<000000000010339e>] enabled_wait+0x9e/0x188) [<000000000010396a>] arch_cpu_idle+0x32/0x50 [<0000000000bd0112>] default_idle_call+0x52/0x68 [<00000000001cd0fa>] do_idle+0x102/0x188 [<00000000001cd41e>] cpu_startup_entry+0x3e/0x48 [<0000000000118c64>] smp_start_secondary+0x11c/0x130 [<0000000000bd2016>] restart_int_handler+0x62/0x78 [<0000000000000000>] (null) INFO: lockdep is turned off. Last Breaking-Event-Address: [<000003ff801e41d6>] zfcp_fc_ct_job_handler+0x3e/0x48 [zfcp] Kernel panic - not syncing: Fatal exception in interrupt This patch moves bsg-lib to allocate and setup struct bsg_job ahead of time, including the allocation of a buffer for the reply-data. This means, struct bsg_job is not allocated separately anymore, but as part of struct request allocation - similar to struct scsi_cmd. Reflect this in the function names that used to handle creation/destruction of struct bsg_job. Reported-by: Steffen Maier Suggested-by: Christoph Hellwig Reviewed-by: Christoph Hellwig Signed-off-by: Benjamin Block Fixes: 82ed4db499b8 ("block: split scsi_request out of struct request") Cc: #4.11+ Signed-off-by: Jens Axboe --- block/bsg-lib.c | 74 +++++++++++++++++++++++++++++-------------------- include/linux/blkdev.h | 1 - include/linux/bsg-lib.h | 2 ++ 3 files changed, 46 insertions(+), 31 deletions(-) diff --git a/block/bsg-lib.c b/block/bsg-lib.c index c4513b2..dd56d74 100644 --- a/block/bsg-lib.c +++ b/block/bsg-lib.c @@ -29,26 +29,25 @@ #include /** - * bsg_destroy_job - routine to teardown/delete a bsg job + * bsg_teardown_job - routine to teardown a bsg job * @job: bsg_job that is to be torn down */ -static void bsg_destroy_job(struct kref *kref) +static void bsg_teardown_job(struct kref *kref) { struct bsg_job *job = container_of(kref, struct bsg_job, kref); struct request *rq = job->req; - blk_end_request_all(rq, BLK_STS_OK); - put_device(job->dev); /* release reference for the request */ kfree(job->request_payload.sg_list); kfree(job->reply_payload.sg_list); - kfree(job); + + blk_end_request_all(rq, BLK_STS_OK); } void bsg_job_put(struct bsg_job *job) { - kref_put(&job->kref, bsg_destroy_job); + kref_put(&job->kref, bsg_teardown_job); } EXPORT_SYMBOL_GPL(bsg_job_put); @@ -100,7 +99,7 @@ EXPORT_SYMBOL_GPL(bsg_job_done); */ static void bsg_softirq_done(struct request *rq) { - struct bsg_job *job = rq->special; + struct bsg_job *job = blk_mq_rq_to_pdu(rq); bsg_job_put(job); } @@ -122,33 +121,20 @@ static int bsg_map_buffer(struct bsg_buffer *buf, struct request *req) } /** - * bsg_create_job - create the bsg_job structure for the bsg request + * bsg_prepare_job - create the bsg_job structure for the bsg request * @dev: device that is being sent the bsg request * @req: BSG request that needs a job structure */ -static int bsg_create_job(struct device *dev, struct request *req) +static int bsg_prepare_job(struct device *dev, struct request *req) { struct request *rsp = req->next_rq; - struct request_queue *q = req->q; struct scsi_request *rq = scsi_req(req); - struct bsg_job *job; + struct bsg_job *job = blk_mq_rq_to_pdu(req); int ret; - BUG_ON(req->special); - - job = kzalloc(sizeof(struct bsg_job) + q->bsg_job_size, GFP_KERNEL); - if (!job) - return -ENOMEM; - - req->special = job; - job->req = req; - if (q->bsg_job_size) - job->dd_data = (void *)&job[1]; job->request = rq->cmd; job->request_len = rq->cmd_len; - job->reply = rq->sense; - job->reply_len = SCSI_SENSE_BUFFERSIZE; /* Size of sense buffer - * allocated */ + if (req->bio) { ret = bsg_map_buffer(&job->request_payload, req); if (ret) @@ -187,7 +173,6 @@ static void bsg_request_fn(struct request_queue *q) { struct device *dev = q->queuedata; struct request *req; - struct bsg_job *job; int ret; if (!get_device(dev)) @@ -199,7 +184,7 @@ static void bsg_request_fn(struct request_queue *q) break; spin_unlock_irq(q->queue_lock); - ret = bsg_create_job(dev, req); + ret = bsg_prepare_job(dev, req); if (ret) { scsi_req(req)->result = ret; blk_end_request_all(req, BLK_STS_OK); @@ -207,8 +192,7 @@ static void bsg_request_fn(struct request_queue *q) continue; } - job = req->special; - ret = q->bsg_job_fn(job); + ret = q->bsg_job_fn(blk_mq_rq_to_pdu(req)); spin_lock_irq(q->queue_lock); if (ret) break; @@ -219,6 +203,35 @@ static void bsg_request_fn(struct request_queue *q) spin_lock_irq(q->queue_lock); } +static int bsg_init_rq(struct request_queue *q, struct request *req, gfp_t gfp) +{ + struct bsg_job *job = blk_mq_rq_to_pdu(req); + struct scsi_request *sreq = &job->sreq; + + memset(job, 0, sizeof(*job)); + + scsi_req_init(sreq); + sreq->sense_len = SCSI_SENSE_BUFFERSIZE; + sreq->sense = kzalloc(sreq->sense_len, gfp); + if (!sreq->sense) + return -ENOMEM; + + job->req = req; + job->reply = sreq->sense; + job->reply_len = sreq->sense_len; + job->dd_data = job + 1; + + return 0; +} + +static void bsg_exit_rq(struct request_queue *q, struct request *req) +{ + struct bsg_job *job = blk_mq_rq_to_pdu(req); + struct scsi_request *sreq = &job->sreq; + + kfree(sreq->sense); +} + /** * bsg_setup_queue - Create and add the bsg hooks so we can receive requests * @dev: device to attach bsg device to @@ -235,7 +248,9 @@ struct request_queue *bsg_setup_queue(struct device *dev, char *name, q = blk_alloc_queue(GFP_KERNEL); if (!q) return ERR_PTR(-ENOMEM); - q->cmd_size = sizeof(struct scsi_request); + q->cmd_size = sizeof(struct bsg_job) + dd_job_size; + q->init_rq_fn = bsg_init_rq; + q->exit_rq_fn = bsg_exit_rq; q->request_fn = bsg_request_fn; ret = blk_init_allocated_queue(q); @@ -243,7 +258,6 @@ struct request_queue *bsg_setup_queue(struct device *dev, char *name, goto out_cleanup_queue; q->queuedata = dev; - q->bsg_job_size = dd_job_size; q->bsg_job_fn = job_fn; queue_flag_set_unlocked(QUEUE_FLAG_BIDI, q); queue_flag_set_unlocked(QUEUE_FLAG_SCSI_PASSTHROUGH, q); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 25f6a0c..2a5d52f 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -568,7 +568,6 @@ struct request_queue { #if defined(CONFIG_BLK_DEV_BSG) bsg_job_fn *bsg_job_fn; - int bsg_job_size; struct bsg_class_device bsg_dev; #endif diff --git a/include/linux/bsg-lib.h b/include/linux/bsg-lib.h index e34dde2..637a20c 100644 --- a/include/linux/bsg-lib.h +++ b/include/linux/bsg-lib.h @@ -24,6 +24,7 @@ #define _BLK_BSG_ #include +#include struct request; struct device; @@ -37,6 +38,7 @@ struct bsg_buffer { }; struct bsg_job { + struct scsi_request sreq; struct device *dev; struct request *req; -- cgit v1.1 From c26844eda9d4fdbd266660e3b3de2d0270e3a1ed Mon Sep 17 00:00:00 2001 From: andy zhou Date: Mon, 21 Aug 2017 12:38:53 -0700 Subject: netfilter: nf_tables: Fix nft limit burst handling Current implementation treats the burst configuration the same as rate configuration. This can cause the per packet cost to be lower than configured. In effect, this bug causes the token bucket to be refilled at a higher rate than what user has specified. This patch changes the implementation so that the token bucket size is controlled by "rate + burst", while maintain the token bucket refill rate the same as user specified. Fixes: 96518518cc41 ("netfilter: add nftables") Signed-off-by: Andy Zhou Acked-by: Joe Stringer Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_limit.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/net/netfilter/nft_limit.c b/net/netfilter/nft_limit.c index 18dd57a..14538b1 100644 --- a/net/netfilter/nft_limit.c +++ b/net/netfilter/nft_limit.c @@ -65,19 +65,23 @@ static int nft_limit_init(struct nft_limit *limit, limit->nsecs = unit * NSEC_PER_SEC; if (limit->rate == 0 || limit->nsecs < unit) return -EOVERFLOW; - limit->tokens = limit->tokens_max = limit->nsecs; - - if (tb[NFTA_LIMIT_BURST]) { - u64 rate; + if (tb[NFTA_LIMIT_BURST]) limit->burst = ntohl(nla_get_be32(tb[NFTA_LIMIT_BURST])); + else + limit->burst = 0; + + if (limit->rate + limit->burst < limit->rate) + return -EOVERFLOW; - rate = limit->rate + limit->burst; - if (rate < limit->rate) - return -EOVERFLOW; + /* The token bucket size limits the number of tokens can be + * accumulated. tokens_max specifies the bucket size. + * tokens_max = unit * (rate + burst) / rate. + */ + limit->tokens = div_u64(limit->nsecs * (limit->rate + limit->burst), + limit->rate); + limit->tokens_max = limit->tokens; - limit->rate = rate; - } if (tb[NFTA_LIMIT_FLAGS]) { u32 flags = ntohl(nla_get_be32(tb[NFTA_LIMIT_FLAGS])); @@ -95,9 +99,8 @@ static int nft_limit_dump(struct sk_buff *skb, const struct nft_limit *limit, { u32 flags = limit->invert ? NFT_LIMIT_F_INV : 0; u64 secs = div_u64(limit->nsecs, NSEC_PER_SEC); - u64 rate = limit->rate - limit->burst; - if (nla_put_be64(skb, NFTA_LIMIT_RATE, cpu_to_be64(rate), + if (nla_put_be64(skb, NFTA_LIMIT_RATE, cpu_to_be64(limit->rate), NFTA_LIMIT_PAD) || nla_put_be64(skb, NFTA_LIMIT_UNIT, cpu_to_be64(secs), NFTA_LIMIT_PAD) || -- cgit v1.1 From 64236e315955cc59e2b612e6a0e59579395530ae Mon Sep 17 00:00:00 2001 From: Cao jin Date: Tue, 22 Aug 2017 21:09:53 +0800 Subject: kbuild: update comments of Makefile.asm-generic Signed-off-by: Cao jin Signed-off-by: Masahiro Yamada --- scripts/Makefile.asm-generic | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/Makefile.asm-generic b/scripts/Makefile.asm-generic index 95f7d80..a6c8c17 100644 --- a/scripts/Makefile.asm-generic +++ b/scripts/Makefile.asm-generic @@ -1,9 +1,9 @@ # include/asm-generic contains a lot of files that are used # verbatim by several architectures. # -# This Makefile reads the file arch/$(SRCARCH)/include/asm/Kbuild +# This Makefile reads the file arch/$(SRCARCH)/include/$(src)/Kbuild # and for each file listed in this file with generic-y creates -# a small wrapper file in $(obj) (arch/$(SRCARCH)/include/generated/asm) +# a small wrapper file in $(obj) (arch/$(SRCARCH)/include/generated/$(src)) kbuild-file := $(srctree)/arch/$(SRCARCH)/include/$(src)/Kbuild -include $(kbuild-file) -- cgit v1.1 From 58efbc9f5463308c43d812fd0748a4dee44c8a16 Mon Sep 17 00:00:00 2001 From: Omar Sandoval Date: Tue, 22 Aug 2017 23:45:59 -0700 Subject: Btrfs: fix blk_status_t/errno confusion This fixes several instances of blk_status_t and bare errno ints being mixed up, some of which are real bugs. In the normal case, 0 matches BLK_STS_OK, so we don't observe any effects of the missing conversion, but in case of errors or passes through the repair/retry paths, the errors get mixed up. The changes were identified using 'sparse', we don't have reports of the buggy behaviour. Fixes: 4e4cbee93d56 ("block: switch bios to blk_status_t") Signed-off-by: Omar Sandoval Reviewed-by: Liu Bo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 4 ++-- fs/btrfs/inode.c | 70 +++++++++++++++++++++++++++++------------------------- fs/btrfs/raid56.c | 34 +++++++++++++------------- fs/btrfs/volumes.c | 10 ++++---- fs/btrfs/volumes.h | 6 ++--- 5 files changed, 64 insertions(+), 60 deletions(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 080e2eb..f45b61f 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3516,7 +3516,7 @@ static blk_status_t wait_dev_flush(struct btrfs_device *device) struct bio *bio = device->flush_bio; if (!device->flush_bio_sent) - return 0; + return BLK_STS_OK; device->flush_bio_sent = 0; wait_for_completion_io(&device->flush_wait); @@ -3563,7 +3563,7 @@ static int barrier_all_devices(struct btrfs_fs_info *info) continue; write_dev_flush(dev); - dev->last_flush_error = 0; + dev->last_flush_error = BLK_STS_OK; } /* wait for all the barriers */ diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 95c2120..24bcd5c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7924,11 +7924,12 @@ err: return ret; } -static inline int submit_dio_repair_bio(struct inode *inode, struct bio *bio, - int mirror_num) +static inline blk_status_t submit_dio_repair_bio(struct inode *inode, + struct bio *bio, + int mirror_num) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - int ret; + blk_status_t ret; BUG_ON(bio_op(bio) == REQ_OP_WRITE); @@ -7980,10 +7981,10 @@ static int btrfs_check_dio_repairable(struct inode *inode, return 1; } -static int dio_read_error(struct inode *inode, struct bio *failed_bio, - struct page *page, unsigned int pgoff, - u64 start, u64 end, int failed_mirror, - bio_end_io_t *repair_endio, void *repair_arg) +static blk_status_t dio_read_error(struct inode *inode, struct bio *failed_bio, + struct page *page, unsigned int pgoff, + u64 start, u64 end, int failed_mirror, + bio_end_io_t *repair_endio, void *repair_arg) { struct io_failure_record *failrec; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; @@ -7993,18 +7994,19 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, int read_mode = 0; int segs; int ret; + blk_status_t status; BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE); ret = btrfs_get_io_failure_record(inode, start, end, &failrec); if (ret) - return ret; + return errno_to_blk_status(ret); ret = btrfs_check_dio_repairable(inode, failed_bio, failrec, failed_mirror); if (!ret) { free_io_failure(failure_tree, io_tree, failrec); - return -EIO; + return BLK_STS_IOERR; } segs = bio_segments(failed_bio); @@ -8022,13 +8024,13 @@ static int dio_read_error(struct inode *inode, struct bio *failed_bio, "Repair DIO Read Error: submitting new dio read[%#x] to this_mirror=%d, in_validation=%d\n", read_mode, failrec->this_mirror, failrec->in_validation); - ret = submit_dio_repair_bio(inode, bio, failrec->this_mirror); - if (ret) { + status = submit_dio_repair_bio(inode, bio, failrec->this_mirror); + if (status) { free_io_failure(failure_tree, io_tree, failrec); bio_put(bio); } - return ret; + return status; } struct btrfs_retry_complete { @@ -8065,8 +8067,8 @@ end: bio_put(bio); } -static int __btrfs_correct_data_nocsum(struct inode *inode, - struct btrfs_io_bio *io_bio) +static blk_status_t __btrfs_correct_data_nocsum(struct inode *inode, + struct btrfs_io_bio *io_bio) { struct btrfs_fs_info *fs_info; struct bio_vec bvec; @@ -8076,8 +8078,8 @@ static int __btrfs_correct_data_nocsum(struct inode *inode, unsigned int pgoff; u32 sectorsize; int nr_sectors; - int ret; - int err = 0; + blk_status_t ret; + blk_status_t err = BLK_STS_OK; fs_info = BTRFS_I(inode)->root->fs_info; sectorsize = fs_info->sectorsize; @@ -8183,11 +8185,12 @@ static blk_status_t __btrfs_subio_endio_read(struct inode *inode, int csum_pos; bool uptodate = (err == 0); int ret; + blk_status_t status; fs_info = BTRFS_I(inode)->root->fs_info; sectorsize = fs_info->sectorsize; - err = 0; + err = BLK_STS_OK; start = io_bio->logical; done.inode = inode; io_bio->bio.bi_iter = io_bio->iter; @@ -8209,12 +8212,12 @@ try_again: done.start = start; init_completion(&done.done); - ret = dio_read_error(inode, &io_bio->bio, bvec.bv_page, - pgoff, start, start + sectorsize - 1, - io_bio->mirror_num, - btrfs_retry_endio, &done); - if (ret) { - err = errno_to_blk_status(ret); + status = dio_read_error(inode, &io_bio->bio, bvec.bv_page, + pgoff, start, start + sectorsize - 1, + io_bio->mirror_num, btrfs_retry_endio, + &done); + if (status) { + err = status; goto next; } @@ -8250,7 +8253,7 @@ static blk_status_t btrfs_subio_endio_read(struct inode *inode, if (unlikely(err)) return __btrfs_correct_data_nocsum(inode, io_bio); else - return 0; + return BLK_STS_OK; } else { return __btrfs_subio_endio_read(inode, io_bio, err); } @@ -8423,9 +8426,9 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode, return 0; } -static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, - u64 file_offset, int skip_sum, - int async_submit) +static inline blk_status_t +__btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, u64 file_offset, + int skip_sum, int async_submit) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_dio_private *dip = bio->bi_private; @@ -8488,6 +8491,7 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, int clone_offset = 0; int clone_len; int ret; + blk_status_t status; map_length = orig_bio->bi_iter.bi_size; submit_len = map_length; @@ -8537,9 +8541,9 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, */ atomic_inc(&dip->pending_bios); - ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, - async_submit); - if (ret) { + status = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, + async_submit); + if (status) { bio_put(bio); atomic_dec(&dip->pending_bios); goto out_err; @@ -8557,9 +8561,9 @@ static int btrfs_submit_direct_hook(struct btrfs_dio_private *dip, } while (submit_len > 0); submit: - ret = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, - async_submit); - if (!ret) + status = __btrfs_submit_dio_bio(bio, inode, file_offset, skip_sum, + async_submit); + if (!status) return 0; bio_put(bio); diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 2086383..2cf6ba4 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -905,7 +905,7 @@ static void raid_write_end_io(struct bio *bio) if (!atomic_dec_and_test(&rbio->stripes_pending)) return; - err = 0; + err = BLK_STS_OK; /* OK, we have read all the stripes we need to. */ max_errors = (rbio->operation == BTRFS_RBIO_PARITY_SCRUB) ? @@ -1324,7 +1324,7 @@ write_data: return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } /* @@ -1475,7 +1475,7 @@ static void raid_rmw_end_io(struct bio *bio) cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } static void async_rmw_stripe(struct btrfs_raid_bio *rbio) @@ -1579,7 +1579,7 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio) return 0; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); return -EIO; finish: @@ -1795,12 +1795,12 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio) void **pointers; int faila = -1, failb = -1; struct page *page; - int err; + blk_status_t err; int i; pointers = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS); if (!pointers) { - err = -ENOMEM; + err = BLK_STS_RESOURCE; goto cleanup_io; } @@ -1856,7 +1856,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio) * a bad data or Q stripe. * TODO, we should redo the xor here. */ - err = -EIO; + err = BLK_STS_IOERR; goto cleanup; } /* @@ -1882,7 +1882,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio) if (rbio->bbio->raid_map[failb] == RAID6_Q_STRIPE) { if (rbio->bbio->raid_map[faila] == RAID5_P_STRIPE) { - err = -EIO; + err = BLK_STS_IOERR; goto cleanup; } /* @@ -1954,13 +1954,13 @@ pstripe: } } - err = 0; + err = BLK_STS_OK; cleanup: kfree(pointers); cleanup_io: if (rbio->operation == BTRFS_RBIO_READ_REBUILD) { - if (err == 0) + if (err == BLK_STS_OK) cache_rbio_pages(rbio); else clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags); @@ -1968,7 +1968,7 @@ cleanup_io: rbio_orig_end_io(rbio, err); } else if (rbio->operation == BTRFS_RBIO_REBUILD_MISSING) { rbio_orig_end_io(rbio, err); - } else if (err == 0) { + } else if (err == BLK_STS_OK) { rbio->faila = -1; rbio->failb = -1; @@ -2005,7 +2005,7 @@ static void raid_recover_end_io(struct bio *bio) return; if (atomic_read(&rbio->error) > rbio->bbio->max_errors) - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); else __raid_recover_end_io(rbio); } @@ -2104,7 +2104,7 @@ out: cleanup: if (rbio->operation == BTRFS_RBIO_READ_REBUILD || rbio->operation == BTRFS_RBIO_REBUILD_MISSING) - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); return -EIO; } @@ -2431,7 +2431,7 @@ submit_write: nr_data = bio_list_size(&bio_list); if (!nr_data) { /* Every parity is right */ - rbio_orig_end_io(rbio, 0); + rbio_orig_end_io(rbio, BLK_STS_OK); return; } @@ -2451,7 +2451,7 @@ submit_write: return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } static inline int is_data_stripe(struct btrfs_raid_bio *rbio, int stripe) @@ -2519,7 +2519,7 @@ static void validate_rbio_for_parity_scrub(struct btrfs_raid_bio *rbio) return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); } /* @@ -2633,7 +2633,7 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio) return; cleanup: - rbio_orig_end_io(rbio, -EIO); + rbio_orig_end_io(rbio, BLK_STS_IOERR); return; finish: diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e8b9a26..bd679bc 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6212,8 +6212,8 @@ static void bbio_error(struct btrfs_bio *bbio, struct bio *bio, u64 logical) } } -int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, - int mirror_num, int async_submit) +blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, + int mirror_num, int async_submit) { struct btrfs_device *dev; struct bio *first_bio = bio; @@ -6233,7 +6233,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, &map_length, &bbio, mirror_num, 1); if (ret) { btrfs_bio_counter_dec(fs_info); - return ret; + return errno_to_blk_status(ret); } total_devs = bbio->num_stripes; @@ -6256,7 +6256,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, } btrfs_bio_counter_dec(fs_info); - return ret; + return errno_to_blk_status(ret); } if (map_length < length) { @@ -6283,7 +6283,7 @@ int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, dev_nr, async_submit); } btrfs_bio_counter_dec(fs_info); - return 0; + return BLK_STS_OK; } struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid, diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 6f45fd6..93277fc 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -74,7 +74,7 @@ struct btrfs_device { int missing; int can_discard; int is_tgtdev_for_dev_replace; - int last_flush_error; + blk_status_t last_flush_error; int flush_bio_sent; #ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED @@ -416,8 +416,8 @@ int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 type); void btrfs_mapping_init(struct btrfs_mapping_tree *tree); void btrfs_mapping_tree_free(struct btrfs_mapping_tree *tree); -int btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, - int mirror_num, int async_submit); +blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, + int mirror_num, int async_submit); int btrfs_open_devices(struct btrfs_fs_devices *fs_devices, fmode_t flags, void *holder); int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder, -- cgit v1.1 From 9ce76511b67be8fbcdff36b7e1662e3887bb7377 Mon Sep 17 00:00:00 2001 From: Tom Rini Date: Tue, 22 Aug 2017 21:51:46 -0400 Subject: ASoC: rt5677: Reintroduce I2C device IDs Not all devices with ACPI and this combination of sound devices will have the required information provided via ACPI. Reintroduce the I2C device ID to restore sound functionality on on the Chromebook 'Samus' model. [ More background note: the commit a36afb0ab648 ("ASoC: rt5677: Introduce proper table...") moved the i2c ID probed via ACPI ("RT5677CE:00") to a proper acpi_device_id table. Although the action itself is correct per se, the overseen issue is the reference id->driver_data at rt5677_i2c_probe() for retrieving the corresponding chip model for the given id. Since id=NULL is passed for ACPI matching case, we get an Oops now. We already have queued more fixes for 4.14 and they already address the issue, but they are bigger changes that aren't preferable for the late 4.13-rc stage. So, this patch just papers over the bug as a once-off quick fix for a particular ACPI matching. -- tiwai ] Fixes: a36afb0ab648 ("ASoC: rt5677: Introduce proper table for ACPI enumeration") Signed-off-by: Tom Rini Acked-by: Mark Brown Signed-off-by: Takashi Iwai --- sound/soc/codecs/rt5677.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/codecs/rt5677.c b/sound/soc/codecs/rt5677.c index 36e530a..6f62927 100644 --- a/sound/soc/codecs/rt5677.c +++ b/sound/soc/codecs/rt5677.c @@ -5021,6 +5021,7 @@ static const struct regmap_config rt5677_regmap = { static const struct i2c_device_id rt5677_i2c_id[] = { { "rt5677", RT5677 }, { "rt5676", RT5676 }, + { "RT5677CE:00", RT5677 }, { } }; MODULE_DEVICE_TABLE(i2c, rt5677_i2c_id); -- cgit v1.1 From 1c08c22c874ac88799cab1f78c40f46110274915 Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Thu, 24 Aug 2017 12:04:29 -0400 Subject: cpuset: Fix incorrect memory_pressure control file mapping The memory_pressure control file was incorrectly set up without a private value (0, by default). As a result, this control file was treated like memory_migrate on read. By adding back the FILE_MEMORY_PRESSURE private value, the correct memory pressure value will be returned. Signed-off-by: Waiman Long Signed-off-by: Tejun Heo Fixes: 7dbdb199d3bf ("cgroup: replace cftype->mode with CFTYPE_WORLD_WRITABLE") Cc: stable@vger.kernel.org # v4.4+ --- kernel/cgroup/cpuset.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index ca8376e..8362bac 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -1891,6 +1891,7 @@ static struct cftype files[] = { { .name = "memory_pressure", .read_u64 = cpuset_read_u64, + .private = FILE_MEMORY_PRESSURE, }, { -- cgit v1.1 From 79964dbaf662229253b281c42e82e2675a9d3b80 Mon Sep 17 00:00:00 2001 From: Maciej Purski Date: Mon, 21 Aug 2017 12:32:51 +0200 Subject: drm/bridge/sii8620: Fix memory corruption Function sii8620_mt_read_devcap_reg_recv() used to read array index from a wrong msg register, which caused writing out of array bounds. It led to writing on other fields of struct sii8620. Signed-off-by: Maciej Purski Fixes: e9c6da270 ("drm/bridge/sii8620: add reading device capability registers") Signed-off-by: Andrzej Hajda Link: https://patchwork.freedesktop.org/patch/msgid/1503311571-25819-1-git-send-email-m.purski@samsung.com --- drivers/gpu/drm/bridge/sil-sii8620.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/bridge/sil-sii8620.c b/drivers/gpu/drm/bridge/sil-sii8620.c index 2d51a22..5131bfb 100644 --- a/drivers/gpu/drm/bridge/sil-sii8620.c +++ b/drivers/gpu/drm/bridge/sil-sii8620.c @@ -597,9 +597,9 @@ static void sii8620_mt_read_devcap(struct sii8620 *ctx, bool xdevcap) static void sii8620_mt_read_devcap_reg_recv(struct sii8620 *ctx, struct sii8620_mt_msg *msg) { - u8 reg = msg->reg[0] & 0x7f; + u8 reg = msg->reg[1] & 0x7f; - if (msg->reg[0] & 0x80) + if (msg->reg[1] & 0x80) ctx->xdevcap[reg] = msg->ret; else ctx->devcap[reg] = msg->ret; -- cgit v1.1 From dadc0736f7be553a25ad34dc437ae379c5ab4a68 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 24 Aug 2017 09:02:49 -0700 Subject: virtio_net: be drop monitor friendly This change is needed to not fool drop monitor. (perf record ... -e skb:kfree_skb ) Packets were properly sent and are consumed after TX completion. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 98f17b0..b06169e 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1058,7 +1058,7 @@ static void free_old_xmit_skbs(struct send_queue *sq) bytes += skb->len; packets++; - dev_kfree_skb_any(skb); + dev_consume_skb_any(skb); } /* Avoid overhead when no packets have been processed -- cgit v1.1 From 498ca3c82a7b11e152a46c253f6b2087c929ce00 Mon Sep 17 00:00:00 2001 From: Noa Osherovich Date: Wed, 23 Aug 2017 08:35:40 +0300 Subject: IB/core: Avoid accessing non-allocated memory when inferring port type Commit 44c58487d51a ("IB/core: Define 'ib' and 'roce' rdma_ah_attr types") introduced the concept of type in ah_attr: * During ib_register_device, each port is checked for its type which is stored in ib_device's port_immutable array. * During uverbs' modify_qp, the type is inferred using the port number in ib_uverbs_qp_dest struct (address vector) by accessing the relevant port_immutable array and the type is passed on to providers. IB spec (version 1.3) enforces a valid port value only in Reset to Init. During Init to RTR, the address vector must be valid but port number is not mentioned as a field in the address vector, so its value is not validated, which leads to accesses to a non-allocated memory when inferring the port type. Save the real port number in ib_qp during modify to Init (when the comp_mask indicates that the port number is valid) and use this value to infer the port type. Avoid copying the address vector fields if the matching bit is not set in the attr_mask. Address vector can't be modified before the port, so no valid flow is affected. Fixes: 44c58487d51a ('IB/core: Define 'ib' and 'roce' rdma_ah_attr types') Signed-off-by: Noa Osherovich Reviewed-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/core/uverbs_cmd.c | 11 +++++++---- drivers/infiniband/core/verbs.c | 7 ++++++- include/rdma/ib_verbs.h | 1 + 3 files changed, 14 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 55822ae..739bd69 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1522,6 +1522,7 @@ static int create_qp(struct ib_uverbs_file *file, qp->qp_type = attr.qp_type; atomic_set(&qp->usecnt, 0); atomic_inc(&pd->usecnt); + qp->port = 0; if (attr.send_cq) atomic_inc(&attr.send_cq->usecnt); if (attr.recv_cq) @@ -1962,8 +1963,9 @@ static int modify_qp(struct ib_uverbs_file *file, attr->alt_timeout = cmd->base.alt_timeout; attr->rate_limit = cmd->rate_limit; - attr->ah_attr.type = rdma_ah_find_type(qp->device, - cmd->base.dest.port_num); + if (cmd->base.attr_mask & IB_QP_AV) + attr->ah_attr.type = rdma_ah_find_type(qp->device, + cmd->base.dest.port_num); if (cmd->base.dest.is_global) { rdma_ah_set_grh(&attr->ah_attr, NULL, cmd->base.dest.flow_label, @@ -1981,8 +1983,9 @@ static int modify_qp(struct ib_uverbs_file *file, rdma_ah_set_port_num(&attr->ah_attr, cmd->base.dest.port_num); - attr->alt_ah_attr.type = rdma_ah_find_type(qp->device, - cmd->base.dest.port_num); + if (cmd->base.attr_mask & IB_QP_ALT_PATH) + attr->alt_ah_attr.type = + rdma_ah_find_type(qp->device, cmd->base.dest.port_num); if (cmd->base.alt_dest.is_global) { rdma_ah_set_grh(&attr->alt_ah_attr, NULL, cmd->base.alt_dest.flow_label, diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 7f8fe44..b456e3c 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -838,6 +838,7 @@ struct ib_qp *ib_create_qp(struct ib_pd *pd, spin_lock_init(&qp->mr_lock); INIT_LIST_HEAD(&qp->rdma_mrs); INIT_LIST_HEAD(&qp->sig_mrs); + qp->port = 0; if (qp_init_attr->qp_type == IB_QPT_XRC_TGT) return ib_create_xrc_qp(qp, qp_init_attr); @@ -1297,7 +1298,11 @@ int ib_modify_qp_with_udata(struct ib_qp *qp, struct ib_qp_attr *attr, if (ret) return ret; } - return ib_security_modify_qp(qp, attr, attr_mask, udata); + ret = ib_security_modify_qp(qp, attr, attr_mask, udata); + if (!ret && (attr_mask & IB_QP_PORT)) + qp->port = attr->port_num; + + return ret; } EXPORT_SYMBOL(ib_modify_qp_with_udata); diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index b573243..88c32ab 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1683,6 +1683,7 @@ struct ib_qp { enum ib_qp_type qp_type; struct ib_rwq_ind_table *rwq_ind_tbl; struct ib_qp_security *qp_sec; + u8 port; }; struct ib_mr { -- cgit v1.1 From 1d31e9c09f41f9ffa53eaf8457b3e77a1d527f1e Mon Sep 17 00:00:00 2001 From: Majd Dibbiny Date: Wed, 23 Aug 2017 08:35:41 +0300 Subject: IB/mlx5: Fix Raw Packet QP event handler assignment In case we have SQ and RQ for Raw Packet QP, the SQ's event handler wasn't assigned. Fixing this by assigning event handler for each WQ after creation. [ 1877.145243] Call Trace: [ 1877.148644] [ 1877.150580] [] ? mlx5_rsc_event+0x105/0x210 [mlx5_core] [ 1877.159581] [] ? mlx5_cq_event+0x57/0xd0 [mlx5_core] [ 1877.167137] [] mlx5_eq_int+0x53e/0x6c0 [mlx5_core] [ 1877.174526] [] ? sched_clock+0x9/0x10 [ 1877.180753] [] handle_irq_event_percpu+0x3e/0x1e0 [ 1877.188014] [] handle_irq_event+0x3d/0x60 [ 1877.194567] [] handle_edge_irq+0x77/0x130 [ 1877.201129] [] handle_irq+0xbf/0x150 [ 1877.207244] [] ? atomic_notifier_call_chain+0x1a/0x20 [ 1877.214829] [] do_IRQ+0x4f/0xf0 [ 1877.220498] [] common_interrupt+0x6d/0x6d [ 1877.227025] [ 1877.228967] [] ? cpuidle_enter_state+0x52/0xc0 [ 1877.236990] [] cpuidle_idle_call+0xc5/0x200 [ 1877.243676] [] arch_cpu_idle+0xe/0x30 [ 1877.249831] [] cpu_startup_entry+0xf5/0x290 [ 1877.256513] [] start_secondary+0x265/0x27b [ 1877.263111] Code: Bad RIP value. [ 1877.267296] RIP [< (null)>] (null) [ 1877.273264] RSP [ 1877.277531] CR2: 0000000000000000 Fixes: 19098df2da78 ("IB/mlx5: Refactor mlx5_ib_qp to accommodate other QP types") Signed-off-by: Majd Dibbiny Reviewed-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/qp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 0889ff3..f58f8f5 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -1238,6 +1238,7 @@ static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, goto err_destroy_tis; sq->base.container_mibqp = qp; + sq->base.mqp.event = mlx5_ib_qp_event; } if (qp->rq.wqe_cnt) { -- cgit v1.1 From ec2558796d25e6024071b6bcb8e11392538d57bf Mon Sep 17 00:00:00 2001 From: Majd Dibbiny Date: Wed, 23 Aug 2017 08:35:42 +0300 Subject: IB/mlx5: Always return success for RoCE modify port CM layer calls ib_modify_port() regardless of the link layer. For the Ethernet ports, qkey violation and Port capabilities are meaningless. Therefore, always return success for ib_modify_port calls on the Ethernet ports. Cc: Selvin Xavier Signed-off-by: Majd Dibbiny Reviewed-by: Moni Shoua Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/main.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index a7f2e60..f7fcde1 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1085,6 +1085,12 @@ static int mlx5_ib_modify_port(struct ib_device *ibdev, u8 port, int mask, bool is_ib = (mlx5_ib_port_link_layer(ibdev, port) == IB_LINK_LAYER_INFINIBAND); + /* CM layer calls ib_modify_port() regardless of the link layer. For + * Ethernet ports, qkey violation and Port capabilities are meaningless. + */ + if (!is_ib) + return 0; + if (MLX5_CAP_GEN(dev->mdev, ib_virt) && is_ib) { change_mask = props->clr_port_cap_mask | props->set_port_cap_mask; value = ~props->clr_port_cap_mask | props->set_port_cap_mask; -- cgit v1.1 From 2fb44600fe784449404c6639de26af8361999ec7 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 18 Jul 2017 16:43:47 -0700 Subject: um: Fix check for _xstate for older hosts Commit 0a987645672e ("um: Allow building and running on older hosts") attempted to check for PTRACE_{GET,SET}REGSET under the premise that these ptrace(2) parameters were directly linked with the presence of the _xstate structure. After Richard's commit 61e8d462457f ("um: Correctly check for PTRACE_GETRESET/SETREGSET") which properly included linux/ptrace.h instead of asm/ptrace.h, we could get into the original build failure that I reported: arch/x86/um/user-offsets.c: In function 'foo': arch/x86/um/user-offsets.c:54: error: invalid application of 'sizeof' to incomplete type 'struct _xstate' On this particular host, we do have PTRACE_GETREGSET and PTRACE_SETREGSET defined in linux/ptrace.h, but not the structure _xstate that should be pulled from the following include chain: signal.h -> bits/sigcontext.h. Correctly fix this by checking for FP_XSTATE_MAGIC1 which is the correct way to see if struct _xstate is available or not on the host. Fixes: 61e8d462457f ("um: Correctly check for PTRACE_GETRESET/SETREGSET") Fixes: 0a987645672e ("um: Allow building and running on older hosts") Signed-off-by: Florian Fainelli Signed-off-by: Richard Weinberger --- arch/x86/um/user-offsets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/um/user-offsets.c b/arch/x86/um/user-offsets.c index ae4cd58..02250b2 100644 --- a/arch/x86/um/user-offsets.c +++ b/arch/x86/um/user-offsets.c @@ -50,7 +50,7 @@ void foo(void) DEFINE(HOST_GS, GS); DEFINE(HOST_ORIG_AX, ORIG_EAX); #else -#if defined(PTRACE_GETREGSET) && defined(PTRACE_SETREGSET) +#ifdef FP_XSTATE_MAGIC1 DEFINE(HOST_FP_SIZE, sizeof(struct _xstate) / sizeof(unsigned long)); #else DEFINE(HOST_FP_SIZE, sizeof(struct _fpstate) / sizeof(unsigned long)); -- cgit v1.1 From 311fc65c9fb9c966bca8e6f3ff8132ce57344ab9 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Thu, 24 Aug 2017 15:13:29 -0500 Subject: pty: Repair TIOCGPTPEER The implementation of TIOCGPTPEER has two issues. When /dev/ptmx (as opposed to /dev/pts/ptmx) is opened the wrong vfsmount is passed to dentry_open. Which results in the kernel displaying the wrong pathname for the peer. The second is simply by caching the vfsmount and dentry of the peer it leaves them open, in a way they were not previously Which because of the inreased reference counts can cause unnecessary behaviour differences resulting in regressions. To fix these move the ioctl into tty_io.c at a generic level allowing the ioctl to have access to the struct file on which the ioctl is being called. This allows the path of the slave to be derived when opening the slave through TIOCGPTPEER instead of requiring the path to the slave be cached. Thus removing the need for caching the path. A new function devpts_ptmx_path is factored out of devpts_acquire and used to implement a function devpts_mntget. The new function devpts_mntget takes a filp to perform the lookup on and fsi so that it can confirm that the superblock that is found by devpts_ptmx_path is the proper superblock. v2: Lots of fixes to make the code actually work v3: Suggestions by Linus - Removed the unnecessary initialization of filp in ptm_open_peer - Simplified devpts_ptmx_path as gotos are no longer required [ This is the fix for the issue that was reverted in commit 143c97cc6529, but this time without breaking 'pbuilder' due to increased reference counts - Linus ] Fixes: 54ebbfb16034 ("tty: add TIOCGPTPEER ioctl") Reported-by: Christian Brauner Reported-and-tested-by: Stefan Lippers-Hollmann Signed-off-by: "Eric W. Biederman" Signed-off-by: Linus Torvalds --- drivers/tty/pty.c | 64 ++++++++++++++++++++-------------------------- drivers/tty/tty_io.c | 3 +++ fs/devpts/inode.c | 65 +++++++++++++++++++++++++++++++++++------------ include/linux/devpts_fs.h | 10 ++++++++ 4 files changed, 89 insertions(+), 53 deletions(-) diff --git a/drivers/tty/pty.c b/drivers/tty/pty.c index 284749f..a6d5164 100644 --- a/drivers/tty/pty.c +++ b/drivers/tty/pty.c @@ -69,13 +69,8 @@ static void pty_close(struct tty_struct *tty, struct file *filp) #ifdef CONFIG_UNIX98_PTYS if (tty->driver == ptm_driver) { mutex_lock(&devpts_mutex); - if (tty->link->driver_data) { - struct path *path = tty->link->driver_data; - - devpts_pty_kill(path->dentry); - path_put(path); - kfree(path); - } + if (tty->link->driver_data) + devpts_pty_kill(tty->link->driver_data); mutex_unlock(&devpts_mutex); } #endif @@ -607,25 +602,24 @@ static inline void legacy_pty_init(void) { } static struct cdev ptmx_cdev; /** - * pty_open_peer - open the peer of a pty - * @tty: the peer of the pty being opened + * ptm_open_peer - open the peer of a pty + * @master: the open struct file of the ptmx device node + * @tty: the master of the pty being opened + * @flags: the flags for open * - * Open the cached dentry in tty->link, providing a safe way for userspace - * to get the slave end of a pty (where they have the master fd and cannot - * access or trust the mount namespace /dev/pts was mounted inside). + * Provide a race free way for userspace to open the slave end of a pty + * (where they have the master fd and cannot access or trust the mount + * namespace /dev/pts was mounted inside). */ -static struct file *pty_open_peer(struct tty_struct *tty, int flags) -{ - if (tty->driver->subtype != PTY_TYPE_MASTER) - return ERR_PTR(-EIO); - return dentry_open(tty->link->driver_data, flags, current_cred()); -} - -static int pty_get_peer(struct tty_struct *tty, int flags) +int ptm_open_peer(struct file *master, struct tty_struct *tty, int flags) { int fd = -1; - struct file *filp = NULL; + struct file *filp; int retval = -EINVAL; + struct path path; + + if (tty->driver != ptm_driver) + return -EIO; fd = get_unused_fd_flags(0); if (fd < 0) { @@ -633,7 +627,16 @@ static int pty_get_peer(struct tty_struct *tty, int flags) goto err; } - filp = pty_open_peer(tty, flags); + /* Compute the slave's path */ + path.mnt = devpts_mntget(master, tty->driver_data); + if (IS_ERR(path.mnt)) { + retval = PTR_ERR(path.mnt); + goto err_put; + } + path.dentry = tty->link->driver_data; + + filp = dentry_open(&path, flags, current_cred()); + mntput(path.mnt); if (IS_ERR(filp)) { retval = PTR_ERR(filp); goto err_put; @@ -662,8 +665,6 @@ static int pty_unix98_ioctl(struct tty_struct *tty, return pty_get_pktmode(tty, (int __user *)arg); case TIOCGPTN: /* Get PT Number */ return put_user(tty->index, (unsigned int __user *)arg); - case TIOCGPTPEER: /* Open the other end */ - return pty_get_peer(tty, (int) arg); case TIOCSIG: /* Send signal to other side of pty */ return pty_signal(tty, (int) arg); } @@ -791,7 +792,6 @@ static int ptmx_open(struct inode *inode, struct file *filp) { struct pts_fs_info *fsi; struct tty_struct *tty; - struct path *pts_path; struct dentry *dentry; int retval; int index; @@ -845,26 +845,16 @@ static int ptmx_open(struct inode *inode, struct file *filp) retval = PTR_ERR(dentry); goto err_release; } - /* We need to cache a fake path for TIOCGPTPEER. */ - pts_path = kmalloc(sizeof(struct path), GFP_KERNEL); - if (!pts_path) - goto err_release; - pts_path->mnt = filp->f_path.mnt; - pts_path->dentry = dentry; - path_get(pts_path); - tty->link->driver_data = pts_path; + tty->link->driver_data = dentry; retval = ptm_driver->ops->open(tty, filp); if (retval) - goto err_path_put; + goto err_release; tty_debug_hangup(tty, "opening (count=%d)\n", tty->count); tty_unlock(tty); return 0; -err_path_put: - path_put(pts_path); - kfree(pts_path); err_release: tty_unlock(tty); // This will also put-ref the fsi diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index 974b13d..10c4038 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -2518,6 +2518,9 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case TIOCSSERIAL: tty_warn_deprecated_flags(p); break; + case TIOCGPTPEER: + /* Special because the struct file is needed */ + return ptm_open_peer(file, tty, (int)arg); default: retval = tty_jobctrl_ioctl(tty, real_tty, file, cmd, arg); if (retval != -ENOIOCTLCMD) diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 108df2e..7eae33f 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -133,6 +133,50 @@ static inline struct pts_fs_info *DEVPTS_SB(struct super_block *sb) return sb->s_fs_info; } +static int devpts_ptmx_path(struct path *path) +{ + struct super_block *sb; + int err; + + /* Has the devpts filesystem already been found? */ + if (path->mnt->mnt_sb->s_magic == DEVPTS_SUPER_MAGIC) + return 0; + + /* Is a devpts filesystem at "pts" in the same directory? */ + err = path_pts(path); + if (err) + return err; + + /* Is the path the root of a devpts filesystem? */ + sb = path->mnt->mnt_sb; + if ((sb->s_magic != DEVPTS_SUPER_MAGIC) || + (path->mnt->mnt_root != sb->s_root)) + return -ENODEV; + + return 0; +} + +struct vfsmount *devpts_mntget(struct file *filp, struct pts_fs_info *fsi) +{ + struct path path; + int err; + + path = filp->f_path; + path_get(&path); + + err = devpts_ptmx_path(&path); + dput(path.dentry); + if (err) { + mntput(path.mnt); + path.mnt = ERR_PTR(err); + } + if (DEVPTS_SB(path.mnt->mnt_sb) != fsi) { + mntput(path.mnt); + path.mnt = ERR_PTR(-ENODEV); + } + return path.mnt; +} + struct pts_fs_info *devpts_acquire(struct file *filp) { struct pts_fs_info *result; @@ -143,27 +187,16 @@ struct pts_fs_info *devpts_acquire(struct file *filp) path = filp->f_path; path_get(&path); - /* Has the devpts filesystem already been found? */ - sb = path.mnt->mnt_sb; - if (sb->s_magic != DEVPTS_SUPER_MAGIC) { - /* Is a devpts filesystem at "pts" in the same directory? */ - err = path_pts(&path); - if (err) { - result = ERR_PTR(err); - goto out; - } - - /* Is the path the root of a devpts filesystem? */ - result = ERR_PTR(-ENODEV); - sb = path.mnt->mnt_sb; - if ((sb->s_magic != DEVPTS_SUPER_MAGIC) || - (path.mnt->mnt_root != sb->s_root)) - goto out; + err = devpts_ptmx_path(&path); + if (err) { + result = ERR_PTR(err); + goto out; } /* * pty code needs to hold extra references in case of last /dev/tty close */ + sb = path.mnt->mnt_sb; atomic_inc(&sb->s_active); result = DEVPTS_SB(sb); diff --git a/include/linux/devpts_fs.h b/include/linux/devpts_fs.h index 277ab9a..100cb43 100644 --- a/include/linux/devpts_fs.h +++ b/include/linux/devpts_fs.h @@ -19,6 +19,7 @@ struct pts_fs_info; +struct vfsmount *devpts_mntget(struct file *, struct pts_fs_info *); struct pts_fs_info *devpts_acquire(struct file *); void devpts_release(struct pts_fs_info *); @@ -32,6 +33,15 @@ void *devpts_get_priv(struct dentry *); /* unlink */ void devpts_pty_kill(struct dentry *); +/* in pty.c */ +int ptm_open_peer(struct file *master, struct tty_struct *tty, int flags); + +#else +static inline int +ptm_open_peer(struct file *master, struct tty_struct *tty, int flags) +{ + return -EIO; +} #endif -- cgit v1.1 From e58f95831e7468d25eb6e41f234842ecfe6f014f Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 23 Aug 2017 15:59:49 +0200 Subject: qlge: avoid memcpy buffer overflow gcc-8.0.0 (snapshot) points out that we copy a variable-length string into a fixed length field using memcpy() with the destination length, and that ends up copying whatever follows the string: inlined from 'ql_core_dump' at drivers/net/ethernet/qlogic/qlge/qlge_dbg.c:1106:2: drivers/net/ethernet/qlogic/qlge/qlge_dbg.c:708:2: error: 'memcpy' reading 15 bytes from a region of size 14 [-Werror=stringop-overflow=] memcpy(seg_hdr->description, desc, (sizeof(seg_hdr->description)) - 1); Changing it to use strncpy() will instead zero-pad the destination, which seems to be the right thing to do here. The bug is probably harmless, but it seems like a good idea to address it in stable kernels as well, if only for the purpose of building with gcc-8 without warnings. Fixes: a61f80261306 ("qlge: Add ethtool register dump function.") Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qlge/qlge_dbg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_dbg.c b/drivers/net/ethernet/qlogic/qlge/qlge_dbg.c index 28ea0af..e3223f2 100644 --- a/drivers/net/ethernet/qlogic/qlge/qlge_dbg.c +++ b/drivers/net/ethernet/qlogic/qlge/qlge_dbg.c @@ -724,7 +724,7 @@ static void ql_build_coredump_seg_header( seg_hdr->cookie = MPI_COREDUMP_COOKIE; seg_hdr->segNum = seg_number; seg_hdr->segSize = seg_size; - memcpy(seg_hdr->description, desc, (sizeof(seg_hdr->description)) - 1); + strncpy(seg_hdr->description, desc, (sizeof(seg_hdr->description)) - 1); } /* -- cgit v1.1 From 6c7e983b220f89e03286dc70a41c7ef3a8b409df Mon Sep 17 00:00:00 2001 From: Bob Peterson Date: Wed, 23 Aug 2017 10:43:02 -0400 Subject: tipc: Fix tipc_sk_reinit handling of -EAGAIN In 9dbbfb0ab6680c6a85609041011484e6658e7d3c function tipc_sk_reinit had additional logic added to loop in the event that function rhashtable_walk_next() returned -EAGAIN. No worries. However, if rhashtable_walk_start returns -EAGAIN, it does "continue", and therefore skips the call to rhashtable_walk_stop(). That has the effect of calling rcu_read_lock() without its paired call to rcu_read_unlock(). Since rcu_read_lock() may be nested, the problem may not be apparent for a while, especially since resize events may be rare. But the comments to rhashtable_walk_start() state: * ...Note that we take the RCU lock in all * cases including when we return an error. So you must always call * rhashtable_walk_stop to clean up. This patch replaces the continue with a goto and label to ensure a matching call to rhashtable_walk_stop(). Signed-off-by: Bob Peterson Acked-by: Herbert Xu Signed-off-by: David S. Miller --- net/tipc/socket.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 101e359..d50edd6 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -2255,8 +2255,8 @@ void tipc_sk_reinit(struct net *net) do { tsk = ERR_PTR(rhashtable_walk_start(&iter)); - if (tsk) - continue; + if (IS_ERR(tsk)) + goto walk_stop; while ((tsk = rhashtable_walk_next(&iter)) && !IS_ERR(tsk)) { spin_lock_bh(&tsk->sk.sk_lock.slock); @@ -2265,7 +2265,7 @@ void tipc_sk_reinit(struct net *net) msg_set_orignode(msg, tn->own_addr); spin_unlock_bh(&tsk->sk.sk_lock.slock); } - +walk_stop: rhashtable_walk_stop(&iter); } while (tsk == ERR_PTR(-EAGAIN)); } -- cgit v1.1 From fc788f64f1f3eb31e87d4f53bcf1ab76590d5838 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 18 Aug 2017 11:12:19 -0400 Subject: nfsd: Limit end of page list when decoding NFSv4 WRITE When processing an NFSv4 WRITE operation, argp->end should never point past the end of the data in the final page of the page list. Otherwise, nfsd4_decode_compound can walk into uninitialized memory. More critical, nfsd4_decode_write is failing to increment argp->pagelen when it increments argp->pagelist. This can cause later xdr decoders to assume more data is available than really is, which can cause server crashes on malformed requests. Signed-off-by: Chuck Lever Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4xdr.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 20fbcab..5f940d2 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -144,7 +144,7 @@ static void next_decode_page(struct nfsd4_compoundargs *argp) argp->p = page_address(argp->pagelist[0]); argp->pagelist++; if (argp->pagelen < PAGE_SIZE) { - argp->end = argp->p + (argp->pagelen>>2); + argp->end = argp->p + XDR_QUADLEN(argp->pagelen); argp->pagelen = 0; } else { argp->end = argp->p + (PAGE_SIZE>>2); @@ -1279,9 +1279,7 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) argp->pagelen -= pages * PAGE_SIZE; len -= pages * PAGE_SIZE; - argp->p = (__be32 *)page_address(argp->pagelist[0]); - argp->pagelist++; - argp->end = argp->p + XDR_QUADLEN(PAGE_SIZE); + next_decode_page(argp); } argp->p += XDR_QUADLEN(len); -- cgit v1.1 From eebe53e87f97975ee58a21693e44797608bf679c Mon Sep 17 00:00:00 2001 From: Vadim Lomovtsev Date: Mon, 21 Aug 2017 07:23:07 -0400 Subject: net: sunrpc: svcsock: fix NULL-pointer exception While running nfs/connectathon tests kernel NULL-pointer exception has been observed due to races in svcsock.c. Race is appear when kernel accepts connection by kernel_accept (which creates new socket) and start queuing ingress packets to new socket. This happens in ksoftirq context which could run concurrently on a different core while new socket setup is not done yet. The fix is to re-order socket user data init sequence and add write/read barrier calls to be sure that we got proper values for callback pointers before actually calling them. Test results: nfs/connectathon reports '0' failed tests for about 200+ iterations. Crash log: ---<-snip->--- [ 6708.638984] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 6708.647093] pgd = ffff0000094e0000 [ 6708.650497] [00000000] *pgd=0000010ffff90003, *pud=0000010ffff90003, *pmd=0000010ffff80003, *pte=0000000000000000 [ 6708.660761] Internal error: Oops: 86000005 [#1] SMP [ 6708.665630] Modules linked in: nfsv3 nfnetlink_queue nfnetlink_log nfnetlink rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache overlay xt_CONNSECMARK xt_SECMARK xt_conntrack iptable_security ip_tables ah4 xfrm4_mode_transport sctp tun binfmt_misc ext4 jbd2 mbcache loop tcp_diag udp_diag inet_diag rpcrdma ib_isert iscsi_target_mod ib_iser rdma_cm iw_cm libiscsi scsi_transport_iscsi ib_srpt target_core_mod ib_srp scsi_transport_srp ib_ipoib ib_ucm ib_uverbs ib_umad ib_cm ib_core nls_koi8_u nls_cp932 ts_kmp nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack vfat fat ghash_ce sha2_ce sha1_ce cavium_rng_vf i2c_thunderx sg thunderx_edac i2c_smbus edac_core cavium_rng nfsd auth_rpcgss nfs_acl lockd grace sunrpc xfs libcrc32c nicvf nicpf ast i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops [ 6708.736446] ttm drm i2c_core thunder_bgx thunder_xcv mdio_thunder mdio_cavium dm_mirror dm_region_hash dm_log dm_mod [last unloaded: stap_3c300909c5b3f46dcacd49aab3334af_87021] [ 6708.752275] CPU: 84 PID: 0 Comm: swapper/84 Tainted: G W OE 4.11.0-4.el7.aarch64 #1 [ 6708.760787] Hardware name: www.cavium.com CRB-2S/CRB-2S, BIOS 0.3 Mar 13 2017 [ 6708.767910] task: ffff810006842e80 task.stack: ffff81000689c000 [ 6708.773822] PC is at 0x0 [ 6708.776739] LR is at svc_data_ready+0x38/0x88 [sunrpc] [ 6708.781866] pc : [<0000000000000000>] lr : [] pstate: 60000145 [ 6708.789248] sp : ffff810ffbad3900 [ 6708.792551] x29: ffff810ffbad3900 x28: ffff000008c73d58 [ 6708.797853] x27: 0000000000000000 x26: ffff81000bbe1e00 [ 6708.803156] x25: 0000000000000020 x24: ffff800f7410bf28 [ 6708.808458] x23: ffff000008c63000 x22: ffff000008c63000 [ 6708.813760] x21: ffff800f7410bf28 x20: ffff81000bbe1e00 [ 6708.819063] x19: ffff810012412400 x18: 00000000d82a9df2 [ 6708.824365] x17: 0000000000000000 x16: 0000000000000000 [ 6708.829667] x15: 0000000000000000 x14: 0000000000000001 [ 6708.834969] x13: 0000000000000000 x12: 722e736f622e676e [ 6708.840271] x11: 00000000f814dd99 x10: 0000000000000000 [ 6708.845573] x9 : 7374687225000000 x8 : 0000000000000000 [ 6708.850875] x7 : 0000000000000000 x6 : 0000000000000000 [ 6708.856177] x5 : 0000000000000028 x4 : 0000000000000000 [ 6708.861479] x3 : 0000000000000000 x2 : 00000000e5000000 [ 6708.866781] x1 : 0000000000000000 x0 : ffff81000bbe1e00 [ 6708.872084] [ 6708.873565] Process swapper/84 (pid: 0, stack limit = 0xffff81000689c000) [ 6708.880341] Stack: (0xffff810ffbad3900 to 0xffff8100068a0000) [ 6708.886075] Call trace: [ 6708.888513] Exception stack(0xffff810ffbad3710 to 0xffff810ffbad3840) [ 6708.894942] 3700: ffff810012412400 0001000000000000 [ 6708.902759] 3720: ffff810ffbad3900 0000000000000000 0000000060000145 ffff800f79300000 [ 6708.910577] 3740: ffff000009274d00 00000000000003ea 0000000000000015 ffff000008c63000 [ 6708.918395] 3760: ffff810ffbad3830 ffff800f79300000 000000000000004d 0000000000000000 [ 6708.926212] 3780: ffff810ffbad3890 ffff0000080f88dc ffff800f79300000 000000000000004d [ 6708.934030] 37a0: ffff800f7930093c ffff000008c63000 0000000000000000 0000000000000140 [ 6708.941848] 37c0: ffff000008c2c000 0000000000040b00 ffff81000bbe1e00 0000000000000000 [ 6708.949665] 37e0: 00000000e5000000 0000000000000000 0000000000000000 0000000000000028 [ 6708.957483] 3800: 0000000000000000 0000000000000000 0000000000000000 7374687225000000 [ 6708.965300] 3820: 0000000000000000 00000000f814dd99 722e736f622e676e 0000000000000000 [ 6708.973117] [< (null)>] (null) [ 6708.977824] [] tcp_data_queue+0x754/0xc5c [ 6708.983386] [] tcp_rcv_established+0x1a0/0x67c [ 6708.989384] [] tcp_v4_do_rcv+0x15c/0x22c [ 6708.994858] [] tcp_v4_rcv+0xaf0/0xb58 [ 6709.000077] [] ip_local_deliver_finish+0x10c/0x254 [ 6709.006419] [] ip_local_deliver+0xf0/0xfc [ 6709.011980] [] ip_rcv_finish+0x208/0x3a4 [ 6709.017454] [] ip_rcv+0x2dc/0x3c8 [ 6709.022328] [] __netif_receive_skb_core+0x2f8/0xa0c [ 6709.028758] [] __netif_receive_skb+0x38/0x84 [ 6709.034580] [] netif_receive_skb_internal+0x68/0xdc [ 6709.041010] [] napi_gro_receive+0xcc/0x1a8 [ 6709.046690] [] nicvf_cq_intr_handler+0x59c/0x730 [nicvf] [ 6709.053559] [] nicvf_poll+0x38/0xb8 [nicvf] [ 6709.059295] [] net_rx_action+0x2f8/0x464 [ 6709.064771] [] __do_softirq+0x11c/0x308 [ 6709.070164] [] irq_exit+0x12c/0x174 [ 6709.075206] [] __handle_domain_irq+0x78/0xc4 [ 6709.081027] [] gic_handle_irq+0x94/0x190 [ 6709.086501] Exception stack(0xffff81000689fdf0 to 0xffff81000689ff20) [ 6709.092929] fde0: 0000810ff2ec0000 ffff000008c10000 [ 6709.100747] fe00: ffff000008c70ef4 0000000000000001 0000000000000000 ffff810ffbad9b18 [ 6709.108565] fe20: ffff810ffbad9c70 ffff8100169d3800 ffff810006843ab0 ffff81000689fe80 [ 6709.116382] fe40: 0000000000000bd0 0000ffffdf979cd0 183f5913da192500 0000ffff8a254ce4 [ 6709.124200] fe60: 0000ffff8a254b78 0000aaab10339808 0000000000000000 0000ffff8a0c2a50 [ 6709.132018] fe80: 0000ffffdf979b10 ffff000008d6d450 ffff000008c10000 ffff000008d6d000 [ 6709.139836] fea0: 0000000000000054 ffff000008cd3dbc 0000000000000000 0000000000000000 [ 6709.147653] fec0: 0000000000000000 0000000000000000 0000000000000000 ffff81000689ff20 [ 6709.155471] fee0: ffff000008085240 ffff81000689ff20 ffff000008085244 0000000060000145 [ 6709.163289] ff00: ffff81000689ff10 ffff00000813f1e4 ffffffffffffffff ffff00000813f238 [ 6709.171107] [] el1_irq+0xb4/0x140 [ 6709.175976] [] arch_cpu_idle+0x44/0x11c [ 6709.181368] [] default_idle_call+0x20/0x30 [ 6709.187020] [] do_idle+0x158/0x1e4 [ 6709.191973] [] cpu_startup_entry+0x2c/0x30 [ 6709.197624] [] secondary_start_kernel+0x13c/0x160 [ 6709.203878] [<0000000001bc71c4>] 0x1bc71c4 [ 6709.207967] Code: bad PC value [ 6709.211061] SMP: stopping secondary CPUs [ 6709.218830] Starting crashdump kernel... [ 6709.222749] Bye! ---<-snip>--- Signed-off-by: Vadim Lomovtsev Reviewed-by: Jeff Layton Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields --- net/sunrpc/svcsock.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 2b720fa..e185001 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -421,6 +421,9 @@ static void svc_data_ready(struct sock *sk) dprintk("svc: socket %p(inet %p), busy=%d\n", svsk, sk, test_bit(XPT_BUSY, &svsk->sk_xprt.xpt_flags)); + + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_odata(sk); if (!test_and_set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags)) svc_xprt_enqueue(&svsk->sk_xprt); @@ -437,6 +440,9 @@ static void svc_write_space(struct sock *sk) if (svsk) { dprintk("svc: socket %p(inet %p), write_space busy=%d\n", svsk, sk, test_bit(XPT_BUSY, &svsk->sk_xprt.xpt_flags)); + + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_owspace(sk); svc_xprt_enqueue(&svsk->sk_xprt); } @@ -760,8 +766,12 @@ static void svc_tcp_listen_data_ready(struct sock *sk) dprintk("svc: socket %p TCP (listen) state change %d\n", sk, sk->sk_state); - if (svsk) + if (svsk) { + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_odata(sk); + } + /* * This callback may called twice when a new connection * is established as a child socket inherits everything @@ -794,6 +804,8 @@ static void svc_tcp_state_change(struct sock *sk) if (!svsk) printk("svc: socket %p: no user data\n", sk); else { + /* Refer to svc_setup_socket() for details. */ + rmb(); svsk->sk_ostate(sk); if (sk->sk_state != TCP_ESTABLISHED) { set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags); @@ -1381,12 +1393,18 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, return ERR_PTR(err); } - inet->sk_user_data = svsk; svsk->sk_sock = sock; svsk->sk_sk = inet; svsk->sk_ostate = inet->sk_state_change; svsk->sk_odata = inet->sk_data_ready; svsk->sk_owspace = inet->sk_write_space; + /* + * This barrier is necessary in order to prevent race condition + * with svc_data_ready(), svc_listen_data_ready() and others + * when calling callbacks above. + */ + wmb(); + inet->sk_user_data = svsk; /* Initialize the socket */ if (sock->type == SOCK_DGRAM) -- cgit v1.1 From 4a646580f793d19717f7e034c8d473b509c27d49 Mon Sep 17 00:00:00 2001 From: Masaki Ota Date: Thu, 24 Aug 2017 15:44:36 -0700 Subject: Input: ALPS - fix two-finger scroll breakage in right side on ALPS touchpad Fixed the issue that two finger scroll does not work correctly on V8 protocol. The cause is that V8 protocol X-coordinate decode is wrong at SS4 PLUS device. I added SS4 PLUS X decode definition. Mote notes: the problem manifests itself by the commit e7348396c6d5 ("Input: ALPS - fix V8+ protocol handling (73 03 28)"), where a fix for the V8+ protocol was applied. Although the culprit must have been present beforehand, the two-finger scroll worked casually even with the wrongly reported values by some reason. It got broken by the commit above just because it changed x_max value, and this made libinput correctly figuring the MT events. Since the X coord is reported as falsely doubled, the events on the right-half side go outside the boundary, thus they are no longer handled. This resulted as a broken two-finger scroll. One finger event is decoded differently, and it didn't suffer from this problem. The problem was only about MT events. --tiwai Fixes: e7348396c6d5 ("Input: ALPS - fix V8+ protocol handling (73 03 28)") Signed-off-by: Masaki Ota Tested-by: Takashi Iwai Tested-by: Paul Donohue Cc: Signed-off-by: Takashi Iwai Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/alps.c | 41 +++++++++++++++++++++++++++++++---------- drivers/input/mouse/alps.h | 8 ++++++++ 2 files changed, 39 insertions(+), 10 deletions(-) diff --git a/drivers/input/mouse/alps.c b/drivers/input/mouse/alps.c index 262d105..850b00e 100644 --- a/drivers/input/mouse/alps.c +++ b/drivers/input/mouse/alps.c @@ -1215,14 +1215,24 @@ static int alps_decode_ss4_v2(struct alps_fields *f, case SS4_PACKET_ID_TWO: if (priv->flags & ALPS_BUTTONPAD) { - f->mt[0].x = SS4_BTL_MF_X_V2(p, 0); + if (IS_SS4PLUS_DEV(priv->dev_id)) { + f->mt[0].x = SS4_PLUS_BTL_MF_X_V2(p, 0); + f->mt[1].x = SS4_PLUS_BTL_MF_X_V2(p, 1); + } else { + f->mt[0].x = SS4_BTL_MF_X_V2(p, 0); + f->mt[1].x = SS4_BTL_MF_X_V2(p, 1); + } f->mt[0].y = SS4_BTL_MF_Y_V2(p, 0); - f->mt[1].x = SS4_BTL_MF_X_V2(p, 1); f->mt[1].y = SS4_BTL_MF_Y_V2(p, 1); } else { - f->mt[0].x = SS4_STD_MF_X_V2(p, 0); + if (IS_SS4PLUS_DEV(priv->dev_id)) { + f->mt[0].x = SS4_PLUS_STD_MF_X_V2(p, 0); + f->mt[1].x = SS4_PLUS_STD_MF_X_V2(p, 1); + } else { + f->mt[0].x = SS4_STD_MF_X_V2(p, 0); + f->mt[1].x = SS4_STD_MF_X_V2(p, 1); + } f->mt[0].y = SS4_STD_MF_Y_V2(p, 0); - f->mt[1].x = SS4_STD_MF_X_V2(p, 1); f->mt[1].y = SS4_STD_MF_Y_V2(p, 1); } f->pressure = SS4_MF_Z_V2(p, 0) ? 0x30 : 0; @@ -1239,16 +1249,27 @@ static int alps_decode_ss4_v2(struct alps_fields *f, case SS4_PACKET_ID_MULTI: if (priv->flags & ALPS_BUTTONPAD) { - f->mt[2].x = SS4_BTL_MF_X_V2(p, 0); + if (IS_SS4PLUS_DEV(priv->dev_id)) { + f->mt[0].x = SS4_PLUS_BTL_MF_X_V2(p, 0); + f->mt[1].x = SS4_PLUS_BTL_MF_X_V2(p, 1); + } else { + f->mt[2].x = SS4_BTL_MF_X_V2(p, 0); + f->mt[3].x = SS4_BTL_MF_X_V2(p, 1); + } + f->mt[2].y = SS4_BTL_MF_Y_V2(p, 0); - f->mt[3].x = SS4_BTL_MF_X_V2(p, 1); f->mt[3].y = SS4_BTL_MF_Y_V2(p, 1); no_data_x = SS4_MFPACKET_NO_AX_BL; no_data_y = SS4_MFPACKET_NO_AY_BL; } else { - f->mt[2].x = SS4_STD_MF_X_V2(p, 0); + if (IS_SS4PLUS_DEV(priv->dev_id)) { + f->mt[0].x = SS4_PLUS_STD_MF_X_V2(p, 0); + f->mt[1].x = SS4_PLUS_STD_MF_X_V2(p, 1); + } else { + f->mt[0].x = SS4_STD_MF_X_V2(p, 0); + f->mt[1].x = SS4_STD_MF_X_V2(p, 1); + } f->mt[2].y = SS4_STD_MF_Y_V2(p, 0); - f->mt[3].x = SS4_STD_MF_X_V2(p, 1); f->mt[3].y = SS4_STD_MF_Y_V2(p, 1); no_data_x = SS4_MFPACKET_NO_AX; no_data_y = SS4_MFPACKET_NO_AY; @@ -2541,8 +2562,8 @@ static int alps_set_defaults_ss4_v2(struct psmouse *psmouse, memset(otp, 0, sizeof(otp)); - if (alps_get_otp_values_ss4_v2(psmouse, 0, &otp[0][0]) || - alps_get_otp_values_ss4_v2(psmouse, 1, &otp[1][0])) + if (alps_get_otp_values_ss4_v2(psmouse, 1, &otp[1][0]) || + alps_get_otp_values_ss4_v2(psmouse, 0, &otp[0][0])) return -1; alps_update_device_area_ss4_v2(otp, priv); diff --git a/drivers/input/mouse/alps.h b/drivers/input/mouse/alps.h index ed2d687..c80a7c7 100644 --- a/drivers/input/mouse/alps.h +++ b/drivers/input/mouse/alps.h @@ -100,6 +100,10 @@ enum SS4_PACKET_ID { ((_b[1 + _i * 3] << 5) & 0x1F00) \ ) +#define SS4_PLUS_STD_MF_X_V2(_b, _i) (((_b[0 + (_i) * 3] << 4) & 0x0070) | \ + ((_b[1 + (_i) * 3] << 4) & 0x0F80) \ + ) + #define SS4_STD_MF_Y_V2(_b, _i) (((_b[1 + (_i) * 3] << 3) & 0x0010) | \ ((_b[2 + (_i) * 3] << 5) & 0x01E0) | \ ((_b[2 + (_i) * 3] << 4) & 0x0E00) \ @@ -109,6 +113,10 @@ enum SS4_PACKET_ID { ((_b[0 + (_i) * 3] >> 3) & 0x0010) \ ) +#define SS4_PLUS_BTL_MF_X_V2(_b, _i) (SS4_PLUS_STD_MF_X_V2(_b, _i) | \ + ((_b[0 + (_i) * 3] >> 4) & 0x0008) \ + ) + #define SS4_BTL_MF_Y_V2(_b, _i) (SS4_STD_MF_Y_V2(_b, _i) | \ ((_b[0 + (_i) * 3] >> 3) & 0x0008) \ ) -- cgit v1.1 From c45182eb967af11e9482168be5be41aa22e5d321 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 24 Aug 2017 15:20:41 -0700 Subject: net: systemport: Be drop monitor friendly Utilize dev_consume_skb_any(cb->skb) in bcm_sysport_free_cb() which is used when a TX packet is completed, as well as when the RX ring is cleaned on shutdown. None of these two cases are packet drops, so be drop monitor friendly. Suggested-by: Eric Dumazet Fixes: 80105befdb4b ("net: systemport: add Broadcom SYSTEMPORT Ethernet MAC driver") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcmsysport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index dc30527..e6add99 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -597,7 +597,7 @@ static int bcm_sysport_set_coalesce(struct net_device *dev, static void bcm_sysport_free_cb(struct bcm_sysport_cb *cb) { - dev_kfree_skb_any(cb->skb); + dev_consume_skb_any(cb->skb); cb->skb = NULL; dma_unmap_addr_set(cb, dma_addr, 0); } -- cgit v1.1 From 4e458debbb69af0cbde5bd6430d64519d5f59274 Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Thu, 24 Aug 2017 15:48:21 -0700 Subject: bpf: fix bpf_setsockopts return value This patch fixes a bug causing any sock operations to always return EINVAL. Fixes: a5192c52377e ("bpf: fix to bpf_setsockops"). Reported-by: Neal Cardwell Signed-off-by: Yuchung Cheng Acked-by: Neal Cardwell Acked-by: Craig Gallek Acked-by: Daniel Borkmann Acked-by: Lawrence Brakmo Signed-off-by: David S. Miller --- net/core/filter.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/core/filter.c b/net/core/filter.c index 6280a60..8eb81e5 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2872,7 +2872,6 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, ret = -EINVAL; } } - ret = -EINVAL; #endif } else { ret = -EINVAL; -- cgit v1.1 From d4fec855905fa8bd5fb1c59f73ad2d74a944876a Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 24 Aug 2017 15:56:29 -0700 Subject: net: bcmgenet: Be drop monitor friendly There are 3 spots where we call dev_kfree_skb() but we are actually just doing a normal SKB consumption: __bcmgenet_tx_reclaim() for normal TX reclamation, bcmgenet_alloc_rx_buffers() during the initial RX ring setup and bcmgenet_free_rx_buffers() during RX ring cleanup. Fixes: d6707bec5986 ("net: bcmgenet: rewrite bcmgenet_rx_refill()") Fixes: f48bed16a756 ("net: bcmgenet: Free skb after last Tx frag") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index a981c4e..fea3f9a 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -1360,7 +1360,7 @@ static unsigned int __bcmgenet_tx_reclaim(struct net_device *dev, if (skb) { pkts_compl++; bytes_compl += GENET_CB(skb)->bytes_sent; - dev_kfree_skb_any(skb); + dev_consume_skb_any(skb); } txbds_processed++; @@ -1875,7 +1875,7 @@ static int bcmgenet_alloc_rx_buffers(struct bcmgenet_priv *priv, cb = ring->cbs + i; skb = bcmgenet_rx_refill(priv, cb); if (skb) - dev_kfree_skb_any(skb); + dev_consume_skb_any(skb); if (!cb->skb) return -ENOMEM; } @@ -1894,7 +1894,7 @@ static void bcmgenet_free_rx_buffers(struct bcmgenet_priv *priv) skb = bcmgenet_free_rx_cb(&priv->pdev->dev, cb); if (skb) - dev_kfree_skb_any(skb); + dev_consume_skb_any(skb); } } -- cgit v1.1 From c2062ee3d9615828109ffe8089fbf69bed394d05 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 24 Aug 2017 16:01:13 -0700 Subject: net: systemport: Free DMA coherent descriptors on errors In case bcm_sysport_init_tx_ring() is not able to allocate ring->cbs, we would return with an error, and call bcm_sysport_fini_tx_ring() and it would see that ring->cbs is NULL and do nothing. This would leak the coherent DMA descriptor area, so we need to free it on error before returning. Reported-by: Eric Dumazet Fixes: 80105befdb4b ("net: systemport: add Broadcom SYSTEMPORT Ethernet MAC driver") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcmsysport.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index e6add99..c28fa5a 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -1346,6 +1346,8 @@ static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv, ring->cbs = kcalloc(size, sizeof(struct bcm_sysport_cb), GFP_KERNEL); if (!ring->cbs) { + dma_free_coherent(kdev, sizeof(struct dma_desc), + ring->desc_cpu, ring->desc_dma); netif_err(priv, hw, priv->netdev, "CB allocation failed\n"); return -ENOMEM; } -- cgit v1.1 From b974696da1cfc5aa0c29ed97dc8f6c239899e64b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Wed, 23 Aug 2017 09:03:04 +0200 Subject: mtd: nandsim: remove debugfs entries in error path MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The debugfs entries must be removed before an error is returned in the probe function. Otherwise another try to load the module fails and when the debugfs files are accessed without the module loaded, the kernel still tries to call a function in that module. Fixes: 5346c27c5fed ("mtd: nandsim: Introduce debugfs infrastructure") Signed-off-by: Uwe Kleine-König Reviewed-by: Richard Weinberger Acked-by: Boris Brezillon Signed-off-by: Brian Norris --- drivers/mtd/nand/nandsim.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c index 03a0d05..e4211c3 100644 --- a/drivers/mtd/nand/nandsim.c +++ b/drivers/mtd/nand/nandsim.c @@ -2373,6 +2373,7 @@ static int __init ns_init_module(void) return 0; err_exit: + nandsim_debugfs_remove(nand); free_nandsim(nand); nand_release(nsmtd); for (i = 0;i < ARRAY_SIZE(nand->partitions); ++i) -- cgit v1.1 From be3e83e3471cd0faff2c2d88fe9cfc73d9a9745a Mon Sep 17 00:00:00 2001 From: Boris Brezillon Date: Wed, 23 Aug 2017 20:45:01 +0200 Subject: mtd: nand: atmel: Relax tADL_min constraint Version 4 of the ONFI spec mandates that tADL be at least 400 nanoseconds, but, depending on the master clock rate, 400 ns may not fit in the tADL field of the SMC reg. We need to relax the check and accept the -ERANGE return code. Note that previous versions of the ONFI spec had a lower tADL_min (100 or 200 ns). It's not clear why this timing constraint got increased but it seems most NANDs are fine with values lower than 400ns, so we should be safe. Fixes: f9ce2eddf176 ("mtd: nand: atmel: Add ->setup_data_interface() hooks") Signed-off-by: Boris Brezillon Tested-by: Quentin Schulz Signed-off-by: Brian Norris --- drivers/mtd/nand/atmel/nand-controller.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/mtd/nand/atmel/nand-controller.c b/drivers/mtd/nand/atmel/nand-controller.c index 2c8baa0..ceec21b 100644 --- a/drivers/mtd/nand/atmel/nand-controller.c +++ b/drivers/mtd/nand/atmel/nand-controller.c @@ -1364,7 +1364,18 @@ static int atmel_smc_nand_prepare_smcconf(struct atmel_nand *nand, ret = atmel_smc_cs_conf_set_timing(smcconf, ATMEL_HSMC_TIMINGS_TADL_SHIFT, ncycles); - if (ret) + /* + * Version 4 of the ONFI spec mandates that tADL be at least 400 + * nanoseconds, but, depending on the master clock rate, 400 ns may not + * fit in the tADL field of the SMC reg. We need to relax the check and + * accept the -ERANGE return code. + * + * Note that previous versions of the ONFI spec had a lower tADL_min + * (100 or 200 ns). It's not clear why this timing constraint got + * increased but it seems most NANDs are fine with values lower than + * 400ns, so we should be safe. + */ + if (ret && ret != -ERANGE) return ret; ncycles = DIV_ROUND_UP(conf->timings.sdr.tAR_min, mckperiodps); -- cgit v1.1 From 551143d8d954fe398324a5caa276f518466c428b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 24 Aug 2017 21:12:28 -0700 Subject: net_sched: fix a refcount_t issue with noop_qdisc syzkaller reported a refcount_t warning [1] Issue here is that noop_qdisc refcnt was never really considered as a true refcount, since qdisc_destroy() found TCQ_F_BUILTIN set : if (qdisc->flags & TCQ_F_BUILTIN || !refcount_dec_and_test(&qdisc->refcnt))) return; Meaning that all atomic_inc() we did on noop_qdisc.refcnt were not really needed, but harmless until refcount_t came. To fix this problem, we simply need to not increment noop_qdisc.refcnt, since we never decrement it. [1] refcount_t: increment on 0; use-after-free. ------------[ cut here ]------------ WARNING: CPU: 0 PID: 21754 at lib/refcount.c:152 refcount_inc+0x47/0x50 lib/refcount.c:152 Kernel panic - not syncing: panic_on_warn set ... CPU: 0 PID: 21754 Comm: syz-executor7 Not tainted 4.13.0-rc6+ #20 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:16 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:52 panic+0x1e4/0x417 kernel/panic.c:180 __warn+0x1c4/0x1d9 kernel/panic.c:541 report_bug+0x211/0x2d0 lib/bug.c:183 fixup_bug+0x40/0x90 arch/x86/kernel/traps.c:190 do_trap_no_signal arch/x86/kernel/traps.c:224 [inline] do_trap+0x260/0x390 arch/x86/kernel/traps.c:273 do_error_trap+0x120/0x390 arch/x86/kernel/traps.c:310 do_invalid_op+0x1b/0x20 arch/x86/kernel/traps.c:323 invalid_op+0x1e/0x30 arch/x86/entry/entry_64.S:846 RIP: 0010:refcount_inc+0x47/0x50 lib/refcount.c:152 RSP: 0018:ffff8801c43477a0 EFLAGS: 00010282 RAX: 000000000000002b RBX: ffffffff86093c14 RCX: 0000000000000000 RDX: 000000000000002b RSI: ffffffff8159314e RDI: ffffed0038868ee8 RBP: ffff8801c43477a8 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffffffff86093ac0 R13: 0000000000000001 R14: ffff8801d0f3bac0 R15: dffffc0000000000 attach_default_qdiscs net/sched/sch_generic.c:792 [inline] dev_activate+0x7d3/0xaa0 net/sched/sch_generic.c:833 __dev_open+0x227/0x330 net/core/dev.c:1380 __dev_change_flags+0x695/0x990 net/core/dev.c:6726 dev_change_flags+0x88/0x140 net/core/dev.c:6792 dev_ifsioc+0x5a6/0x930 net/core/dev_ioctl.c:256 dev_ioctl+0x2bc/0xf90 net/core/dev_ioctl.c:554 sock_do_ioctl+0x94/0xb0 net/socket.c:968 sock_ioctl+0x2c2/0x440 net/socket.c:1058 vfs_ioctl fs/ioctl.c:45 [inline] do_vfs_ioctl+0x1b1/0x1520 fs/ioctl.c:685 SYSC_ioctl fs/ioctl.c:700 [inline] SyS_ioctl+0x8f/0xc0 fs/ioctl.c:691 Fixes: 7b9364050246 ("net, sched: convert Qdisc.refcnt from atomic_t to refcount_t") Signed-off-by: Eric Dumazet Reported-by: Dmitry Vyukov Cc: Reshetova, Elena Signed-off-by: David S. Miller --- include/net/sch_generic.h | 7 +++++++ net/sched/sch_api.c | 6 +++--- net/sched/sch_generic.c | 2 +- 3 files changed, 11 insertions(+), 4 deletions(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 67f815e..c1109cd 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -101,6 +101,13 @@ struct Qdisc { spinlock_t busylock ____cacheline_aligned_in_smp; }; +static inline void qdisc_refcount_inc(struct Qdisc *qdisc) +{ + if (qdisc->flags & TCQ_F_BUILTIN) + return; + refcount_inc(&qdisc->refcnt); +} + static inline bool qdisc_is_running(const struct Qdisc *qdisc) { return (raw_read_seqcount(&qdisc->running) & 1) ? true : false; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index a3fa144..4fb5a32 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -836,7 +836,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, old = dev_graft_qdisc(dev_queue, new); if (new && i > 0) - refcount_inc(&new->refcnt); + qdisc_refcount_inc(new); if (!ingress) qdisc_destroy(old); @@ -847,7 +847,7 @@ skip: notify_and_destroy(net, skb, n, classid, dev->qdisc, new); if (new && !new->ops->attach) - refcount_inc(&new->refcnt); + qdisc_refcount_inc(new); dev->qdisc = new ? : &noop_qdisc; if (new && new->ops->attach) @@ -1256,7 +1256,7 @@ replay: if (q == p || (p && check_loop(q, p, 0))) return -ELOOP; - refcount_inc(&q->refcnt); + qdisc_refcount_inc(q); goto graft; } else { if (!q) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 57ba406..4ba6da5 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -785,7 +785,7 @@ static void attach_default_qdiscs(struct net_device *dev) dev->priv_flags & IFF_NO_QUEUE) { netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL); dev->qdisc = txq->qdisc_sleeping; - refcount_inc(&dev->qdisc->refcnt); + qdisc_refcount_inc(dev->qdisc); } else { qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT); if (qdisc) { -- cgit v1.1 From 27163138b4d80e36f2006273d66b6c122d241f30 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Thu, 24 Aug 2017 16:31:22 +0200 Subject: tipc: perform skb_linearize() before parsing the inner header In tipc_rcv(), we linearize only the header and usually the packets are consumed as the nodes permit direct reception. However, if the skb contains tunnelled message due to fail over or synchronization we parse it in tipc_node_check_state() without performing linearization. This will cause link disturbances if the skb was non linear. In this commit, we perform linearization for the above messages. Signed-off-by: Parthasarathy Bhuvaragan Reviewed-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/node.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/tipc/node.c b/net/tipc/node.c index 9b4dcb6..b113a52 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1557,6 +1557,8 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b) /* Check/update node state before receiving */ if (unlikely(skb)) { + if (unlikely(skb_linearize(skb))) + goto discard; tipc_node_write_lock(n); if (tipc_node_check_state(n, skb, bearer_id, &xmitq)) { if (le->link) { -- cgit v1.1 From 60d1d93664a0bb3d5af722ed38c57ee165a45bf7 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Thu, 24 Aug 2017 16:31:23 +0200 Subject: tipc: reassign pointers after skb reallocation / linearization In tipc_msg_reverse(), we assign skb attributes to local pointers in stack at startup. This is followed by skb_linearize() and for cloned buffers we perform skb relocation using pskb_expand_head(). Both these methods may update the skb attributes and thus making the pointers incorrect. In this commit, we fix this error by ensuring that the pointers are re-assigned after any of these skb operations. Fixes: 29042e19f2c60 ("tipc: let function tipc_msg_reverse() expand header when needed") Signed-off-by: Parthasarathy Bhuvaragan Reviewed-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/msg.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/tipc/msg.c b/net/tipc/msg.c index dcd90e6..6ef379f 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -479,13 +479,14 @@ bool tipc_msg_make_bundle(struct sk_buff **skb, struct tipc_msg *msg, bool tipc_msg_reverse(u32 own_node, struct sk_buff **skb, int err) { struct sk_buff *_skb = *skb; - struct tipc_msg *hdr = buf_msg(_skb); + struct tipc_msg *hdr; struct tipc_msg ohdr; - int dlen = min_t(uint, msg_data_sz(hdr), MAX_FORWARD_SIZE); + int dlen; if (skb_linearize(_skb)) goto exit; hdr = buf_msg(_skb); + dlen = min_t(uint, msg_data_sz(hdr), MAX_FORWARD_SIZE); if (msg_dest_droppable(hdr)) goto exit; if (msg_errcode(hdr)) @@ -511,6 +512,8 @@ bool tipc_msg_reverse(u32 own_node, struct sk_buff **skb, int err) pskb_expand_head(_skb, BUF_HEADROOM, BUF_TAILROOM, GFP_ATOMIC)) goto exit; + /* reassign after skb header modifications */ + hdr = buf_msg(_skb); /* Now reverse the concerned fields */ msg_set_errcode(hdr, err); msg_set_non_seq(hdr, 0); -- cgit v1.1 From 991ca84daa001193066554fa49f3a934746317d6 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Thu, 24 Aug 2017 16:31:24 +0200 Subject: tipc: context imbalance at node read unlock If we fail to find a valid bearer in tipc_node_get_linkname(), node_read_unlock() is called without holding the node read lock. This commit fixes this error. Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/node.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/tipc/node.c b/net/tipc/node.c index b113a52..7dd2233 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1126,8 +1126,8 @@ int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 addr, strncpy(linkname, tipc_link_name(link), len); err = 0; } -exit: tipc_node_read_unlock(node); +exit: tipc_node_put(node); return err; } -- cgit v1.1 From 9b4e946ce14e20d7addbfb7d9139e604f9fda107 Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Thu, 24 Aug 2017 16:49:16 -0700 Subject: netvsc: fix deadlock betwen link status and removal There is a deadlock possible when canceling the link status delayed work queue. The removal process is run with RTNL held, and the link status callback is acquring RTNL. Resolve the issue by using trylock and rescheduling. If cancel is in process, that block it from happening. Fixes: 122a5f6410f4 ("staging: hv: use delayed_work for netvsc_send_garp()") Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- drivers/net/hyperv/netvsc_drv.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 0d78727..d91cbc6 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1269,7 +1269,12 @@ static void netvsc_link_change(struct work_struct *w) bool notify = false, reschedule = false; unsigned long flags, next_reconfig, delay; - rtnl_lock(); + /* if changes are happening, comeback later */ + if (!rtnl_trylock()) { + schedule_delayed_work(&ndev_ctx->dwork, LINKCHANGE_INT); + return; + } + net_device = rtnl_dereference(ndev_ctx->nvdev); if (!net_device) goto out_unlock; -- cgit v1.1 From 36ff0dd39f9b88ca83e1733b735e9f22b7be893b Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Fri, 25 Aug 2017 07:16:07 +0200 Subject: esp: Fix locking on page fragment allocation We allocate the page fragment for the ESP trailer inside a spinlock, but consume it outside of the lock. This is racy as some other cou could get the same page fragment then. Fix this by consuming the page fragment inside the lock too. Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible") Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible") Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 5 +++-- net/ipv6/esp6.c | 5 +++-- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index dbb31a9..a8ddb95 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -292,8 +292,6 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * kunmap_atomic(vaddr); - spin_unlock_bh(&x->lock); - nfrags = skb_shinfo(skb)->nr_frags; __skb_fill_page_desc(skb, nfrags, page, pfrag->offset, @@ -301,6 +299,9 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * skb_shinfo(skb)->nr_frags = ++nfrags; pfrag->offset = pfrag->offset + allocsize; + + spin_unlock_bh(&x->lock); + nfrags++; skb->len += tailen; diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 392def1..4e3fdc88 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -260,8 +260,6 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info kunmap_atomic(vaddr); - spin_unlock_bh(&x->lock); - nfrags = skb_shinfo(skb)->nr_frags; __skb_fill_page_desc(skb, nfrags, page, pfrag->offset, @@ -269,6 +267,9 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info skb_shinfo(skb)->nr_frags = ++nfrags; pfrag->offset = pfrag->offset + allocsize; + + spin_unlock_bh(&x->lock); + nfrags++; skb->len += tailen; -- cgit v1.1 From 54ffd790792898f05e215dce5aa593473e80e92f Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Fri, 25 Aug 2017 07:34:35 +0200 Subject: esp: Fix skb tailroom calculation We use skb_availroom to calculate the skb tailroom for the ESP trailer. skb_availroom calculates the tailroom and subtracts this value by reserved_tailroom. However reserved_tailroom is a union with the skb mark. This means that we subtract the tailroom by the skb mark if set. Fix this by using skb_tailroom instead. Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible") Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible") Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 2 +- net/ipv6/esp6.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index a8ddb95..df68963 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -258,7 +258,7 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * esp_output_udp_encap(x, skb, esp); if (!skb_cloned(skb)) { - if (tailen <= skb_availroom(skb)) { + if (tailen <= skb_tailroom(skb)) { nfrags = 1; trailer = skb; tail = skb_tail_pointer(trailer); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 4e3fdc88..ab64f36 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -226,7 +226,7 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info int tailen = esp->tailen; if (!skb_cloned(skb)) { - if (tailen <= skb_availroom(skb)) { + if (tailen <= skb_tailroom(skb)) { nfrags = 1; trailer = skb; tail = skb_tail_pointer(trailer); -- cgit v1.1 From c469268cd523245cc58255f6696e0c295485cb0b Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Thu, 24 Aug 2017 11:59:31 +0200 Subject: KVM: x86: block guest protection keys unless the host has them enabled If the host has protection keys disabled, we cannot read and write the guest PKRU---RDPKRU and WRPKRU fail with #GP(0) if CR4.PKE=0. Block the PKU cpuid bit in that case. This ensures that guest_CR4.PKE=1 implies host_CR4.PKE=1. Fixes: 1be0e61c1f255faaeab04a390e00c8b9b9042870 Cc: stable@vger.kernel.org Reviewed-by: David Hildenbrand Signed-off-by: Paolo Bonzini --- arch/x86/kvm/cpuid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 59ca2ee..19adbb4 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -469,7 +469,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, entry->ecx &= kvm_cpuid_7_0_ecx_x86_features; cpuid_mask(&entry->ecx, CPUID_7_ECX); /* PKU is not yet implemented for shadow paging. */ - if (!tdp_enabled) + if (!tdp_enabled || !boot_cpu_has(X86_FEATURE_OSPKE)) entry->ecx &= ~F(PKU); entry->edx &= kvm_cpuid_7_0_edx_x86_features; entry->edx &= get_scattered_cpuid_leaf(7, 0, CPUID_EDX); -- cgit v1.1 From b9dd21e104bcd45e124acfe978a79df71259e59b Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Wed, 23 Aug 2017 23:14:38 +0200 Subject: KVM: x86: simplify handling of PKRU Move it to struct kvm_arch_vcpu, replacing guest_pkru_valid with a simple comparison against the host value of the register. The write of PKRU in addition can be skipped if the guest has not enabled the feature. Once we do this, we need not test OSPKE in the host anymore, because guest_CR4.PKE=1 implies host_CR4.PKE=1. The static PKU test is kept to elide the code on older CPUs. Suggested-by: Yang Zhang Fixes: 1be0e61c1f255faaeab04a390e00c8b9b9042870 Cc: stable@vger.kernel.org Reviewed-by: David Hildenbrand Signed-off-by: Paolo Bonzini --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/kvm_cache_regs.h | 5 ----- arch/x86/kvm/mmu.h | 2 +- arch/x86/kvm/svm.c | 7 ------- arch/x86/kvm/vmx.c | 25 ++++++++----------------- 5 files changed, 10 insertions(+), 30 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 87ac4fb..f4d120a 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -492,6 +492,7 @@ struct kvm_vcpu_arch { unsigned long cr4; unsigned long cr4_guest_owned_bits; unsigned long cr8; + u32 pkru; u32 hflags; u64 efer; u64 apic_base; diff --git a/arch/x86/kvm/kvm_cache_regs.h b/arch/x86/kvm/kvm_cache_regs.h index 762cdf2..e1e89ee 100644 --- a/arch/x86/kvm/kvm_cache_regs.h +++ b/arch/x86/kvm/kvm_cache_regs.h @@ -84,11 +84,6 @@ static inline u64 kvm_read_edx_eax(struct kvm_vcpu *vcpu) | ((u64)(kvm_register_read(vcpu, VCPU_REGS_RDX) & -1u) << 32); } -static inline u32 kvm_read_pkru(struct kvm_vcpu *vcpu) -{ - return kvm_x86_ops->get_pkru(vcpu); -} - static inline void enter_guest_mode(struct kvm_vcpu *vcpu) { vcpu->arch.hflags |= HF_GUEST_MASK; diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index d7d248a..4b9a3ae 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -185,7 +185,7 @@ static inline u8 permission_fault(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, * index of the protection domain, so pte_pkey * 2 is * is the index of the first bit for the domain. */ - pkru_bits = (kvm_read_pkru(vcpu) >> (pte_pkey * 2)) & 3; + pkru_bits = (vcpu->arch.pkru >> (pte_pkey * 2)) & 3; /* clear present bit, replace PFEC.RSVD with ACC_USER_MASK. */ offset = (pfec & ~1) + diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 56ba053..af256b7 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1777,11 +1777,6 @@ static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) to_svm(vcpu)->vmcb->save.rflags = rflags; } -static u32 svm_get_pkru(struct kvm_vcpu *vcpu) -{ - return 0; -} - static void svm_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) { switch (reg) { @@ -5413,8 +5408,6 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = { .get_rflags = svm_get_rflags, .set_rflags = svm_set_rflags, - .get_pkru = svm_get_pkru, - .tlb_flush = svm_flush_tlb, .run = svm_vcpu_run, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 9b21b12..c6ef294 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -636,8 +636,6 @@ struct vcpu_vmx { u64 current_tsc_ratio; - bool guest_pkru_valid; - u32 guest_pkru; u32 host_pkru; /* @@ -2383,11 +2381,6 @@ static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) to_vmx(vcpu)->emulation_required = emulation_required(vcpu); } -static u32 vmx_get_pkru(struct kvm_vcpu *vcpu) -{ - return to_vmx(vcpu)->guest_pkru; -} - static u32 vmx_get_interrupt_shadow(struct kvm_vcpu *vcpu) { u32 interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO); @@ -9020,8 +9013,10 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) vmx_set_interrupt_shadow(vcpu, 0); - if (vmx->guest_pkru_valid) - __write_pkru(vmx->guest_pkru); + if (static_cpu_has(X86_FEATURE_PKU) && + kvm_read_cr4_bits(vcpu, X86_CR4_PKE) && + vcpu->arch.pkru != vmx->host_pkru) + __write_pkru(vcpu->arch.pkru); atomic_switch_perf_msrs(vmx); debugctlmsr = get_debugctlmsr(); @@ -9169,13 +9164,11 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) * back on host, so it is safe to read guest PKRU from current * XSAVE. */ - if (boot_cpu_has(X86_FEATURE_OSPKE)) { - vmx->guest_pkru = __read_pkru(); - if (vmx->guest_pkru != vmx->host_pkru) { - vmx->guest_pkru_valid = true; + if (static_cpu_has(X86_FEATURE_PKU) && + kvm_read_cr4_bits(vcpu, X86_CR4_PKE)) { + vcpu->arch.pkru = __read_pkru(); + if (vcpu->arch.pkru != vmx->host_pkru) __write_pkru(vmx->host_pkru); - } else - vmx->guest_pkru_valid = false; } /* @@ -11682,8 +11675,6 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = { .get_rflags = vmx_get_rflags, .set_rflags = vmx_set_rflags, - .get_pkru = vmx_get_pkru, - .tlb_flush = vmx_flush_tlb, .run = vmx_vcpu_run, -- cgit v1.1 From 38cfd5e3df9c4f88e76b547eee2087ee5c042ae2 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Wed, 23 Aug 2017 23:16:29 +0200 Subject: KVM, pkeys: do not use PKRU value in vcpu->arch.guest_fpu.state The host pkru is restored right after vcpu exit (commit 1be0e61), so KVM_GET_XSAVE will return the host PKRU value instead. Fix this by using the guest PKRU explicitly in fill_xsave and load_xsave. This part is based on a patch by Junkang Fu. The host PKRU data may also not match the value in vcpu->arch.guest_fpu.state, because it could have been changed by userspace since the last time it was saved, so skip loading it in kvm_load_guest_fpu. Reported-by: Junkang Fu Cc: Yang Zhang Fixes: 1be0e61c1f255faaeab04a390e00c8b9b9042870 Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini --- arch/x86/include/asm/fpu/internal.h | 6 +++--- arch/x86/kvm/x86.c | 17 ++++++++++++++--- 2 files changed, 17 insertions(+), 6 deletions(-) diff --git a/arch/x86/include/asm/fpu/internal.h b/arch/x86/include/asm/fpu/internal.h index 255645f..554cdb2 100644 --- a/arch/x86/include/asm/fpu/internal.h +++ b/arch/x86/include/asm/fpu/internal.h @@ -450,10 +450,10 @@ static inline int copy_fpregs_to_fpstate(struct fpu *fpu) return 0; } -static inline void __copy_kernel_to_fpregs(union fpregs_state *fpstate) +static inline void __copy_kernel_to_fpregs(union fpregs_state *fpstate, u64 mask) { if (use_xsave()) { - copy_kernel_to_xregs(&fpstate->xsave, -1); + copy_kernel_to_xregs(&fpstate->xsave, mask); } else { if (use_fxsr()) copy_kernel_to_fxregs(&fpstate->fxsave); @@ -477,7 +477,7 @@ static inline void copy_kernel_to_fpregs(union fpregs_state *fpstate) : : [addr] "m" (fpstate)); } - __copy_kernel_to_fpregs(fpstate); + __copy_kernel_to_fpregs(fpstate, -1); } extern int copy_fpstate_to_sigframe(void __user *buf, void __user *fp, int size); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index d734aa8..05a5e57 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3245,7 +3245,12 @@ static void fill_xsave(u8 *dest, struct kvm_vcpu *vcpu) u32 size, offset, ecx, edx; cpuid_count(XSTATE_CPUID, index, &size, &offset, &ecx, &edx); - memcpy(dest + offset, src, size); + if (feature == XFEATURE_MASK_PKRU) + memcpy(dest + offset, &vcpu->arch.pkru, + sizeof(vcpu->arch.pkru)); + else + memcpy(dest + offset, src, size); + } valid -= feature; @@ -3283,7 +3288,11 @@ static void load_xsave(struct kvm_vcpu *vcpu, u8 *src) u32 size, offset, ecx, edx; cpuid_count(XSTATE_CPUID, index, &size, &offset, &ecx, &edx); - memcpy(dest, src + offset, size); + if (feature == XFEATURE_MASK_PKRU) + memcpy(&vcpu->arch.pkru, src + offset, + sizeof(vcpu->arch.pkru)); + else + memcpy(dest, src + offset, size); } valid -= feature; @@ -7633,7 +7642,9 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) */ vcpu->guest_fpu_loaded = 1; __kernel_fpu_begin(); - __copy_kernel_to_fpregs(&vcpu->arch.guest_fpu.state); + /* PKRU is separately restored in kvm_x86_ops->run. */ + __copy_kernel_to_fpregs(&vcpu->arch.guest_fpu.state, + ~XFEATURE_MASK_PKRU); trace_kvm_fpu(1); } -- cgit v1.1 From ccd5b3235180eef3cfec337df1c8554ab151b5cc Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Thu, 24 Aug 2017 10:50:29 -0700 Subject: x86/mm: Fix use-after-free of ldt_struct The following commit: 39a0526fb3f7 ("x86/mm: Factor out LDT init from context init") renamed init_new_context() to init_new_context_ldt() and added a new init_new_context() which calls init_new_context_ldt(). However, the error code of init_new_context_ldt() was ignored. Consequently, if a memory allocation in alloc_ldt_struct() failed during a fork(), the ->context.ldt of the new task remained the same as that of the old task (due to the memcpy() in dup_mm()). ldt_struct's are not intended to be shared, so a use-after-free occurred after one task exited. Fix the bug by making init_new_context() pass through the error code of init_new_context_ldt(). This bug was found by syzkaller, which encountered the following splat: BUG: KASAN: use-after-free in free_ldt_struct.part.2+0x10a/0x150 arch/x86/kernel/ldt.c:116 Read of size 4 at addr ffff88006d2cb7c8 by task kworker/u9:0/3710 CPU: 1 PID: 3710 Comm: kworker/u9:0 Not tainted 4.13.0-rc4-next-20170811 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:16 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:52 print_address_description+0x73/0x250 mm/kasan/report.c:252 kasan_report_error mm/kasan/report.c:351 [inline] kasan_report+0x24e/0x340 mm/kasan/report.c:409 __asan_report_load4_noabort+0x14/0x20 mm/kasan/report.c:429 free_ldt_struct.part.2+0x10a/0x150 arch/x86/kernel/ldt.c:116 free_ldt_struct arch/x86/kernel/ldt.c:173 [inline] destroy_context_ldt+0x60/0x80 arch/x86/kernel/ldt.c:171 destroy_context arch/x86/include/asm/mmu_context.h:157 [inline] __mmdrop+0xe9/0x530 kernel/fork.c:889 mmdrop include/linux/sched/mm.h:42 [inline] exec_mmap fs/exec.c:1061 [inline] flush_old_exec+0x173c/0x1ff0 fs/exec.c:1291 load_elf_binary+0x81f/0x4ba0 fs/binfmt_elf.c:855 search_binary_handler+0x142/0x6b0 fs/exec.c:1652 exec_binprm fs/exec.c:1694 [inline] do_execveat_common.isra.33+0x1746/0x22e0 fs/exec.c:1816 do_execve+0x31/0x40 fs/exec.c:1860 call_usermodehelper_exec_async+0x457/0x8f0 kernel/umh.c:100 ret_from_fork+0x2a/0x40 arch/x86/entry/entry_64.S:431 Allocated by task 3700: save_stack_trace+0x16/0x20 arch/x86/kernel/stacktrace.c:59 save_stack+0x43/0xd0 mm/kasan/kasan.c:447 set_track mm/kasan/kasan.c:459 [inline] kasan_kmalloc+0xad/0xe0 mm/kasan/kasan.c:551 kmem_cache_alloc_trace+0x136/0x750 mm/slab.c:3627 kmalloc include/linux/slab.h:493 [inline] alloc_ldt_struct+0x52/0x140 arch/x86/kernel/ldt.c:67 write_ldt+0x7b7/0xab0 arch/x86/kernel/ldt.c:277 sys_modify_ldt+0x1ef/0x240 arch/x86/kernel/ldt.c:307 entry_SYSCALL_64_fastpath+0x1f/0xbe Freed by task 3700: save_stack_trace+0x16/0x20 arch/x86/kernel/stacktrace.c:59 save_stack+0x43/0xd0 mm/kasan/kasan.c:447 set_track mm/kasan/kasan.c:459 [inline] kasan_slab_free+0x71/0xc0 mm/kasan/kasan.c:524 __cache_free mm/slab.c:3503 [inline] kfree+0xca/0x250 mm/slab.c:3820 free_ldt_struct.part.2+0xdd/0x150 arch/x86/kernel/ldt.c:121 free_ldt_struct arch/x86/kernel/ldt.c:173 [inline] destroy_context_ldt+0x60/0x80 arch/x86/kernel/ldt.c:171 destroy_context arch/x86/include/asm/mmu_context.h:157 [inline] __mmdrop+0xe9/0x530 kernel/fork.c:889 mmdrop include/linux/sched/mm.h:42 [inline] __mmput kernel/fork.c:916 [inline] mmput+0x541/0x6e0 kernel/fork.c:927 copy_process.part.36+0x22e1/0x4af0 kernel/fork.c:1931 copy_process kernel/fork.c:1546 [inline] _do_fork+0x1ef/0xfb0 kernel/fork.c:2025 SYSC_clone kernel/fork.c:2135 [inline] SyS_clone+0x37/0x50 kernel/fork.c:2129 do_syscall_64+0x26c/0x8c0 arch/x86/entry/common.c:287 return_from_SYSCALL_64+0x0/0x7a Here is a C reproducer: #include #include #include #include #include #include #include static void *fork_thread(void *_arg) { fork(); } int main(void) { struct user_desc desc = { .entry_number = 8191 }; syscall(__NR_modify_ldt, 1, &desc, sizeof(desc)); for (;;) { if (fork() == 0) { pthread_t t; srand(getpid()); pthread_create(&t, NULL, fork_thread, NULL); usleep(rand() % 10000); syscall(__NR_exit_group, 0); } wait(NULL); } } Note: the reproducer takes advantage of the fact that alloc_ldt_struct() may use vmalloc() to allocate a large ->entries array, and after commit: 5d17a73a2ebe ("vmalloc: back off when the current task is killed") it is possible for userspace to fail a task's vmalloc() by sending a fatal signal, e.g. via exit_group(). It would be more difficult to reproduce this bug on kernels without that commit. This bug only affected kernels with CONFIG_MODIFY_LDT_SYSCALL=y. Signed-off-by: Eric Biggers Acked-by: Dave Hansen Cc: [v4.6+] Cc: Andrew Morton Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Christoph Hellwig Cc: Denys Vlasenko Cc: Dmitry Vyukov Cc: Linus Torvalds Cc: Michal Hocko Cc: Peter Zijlstra Cc: Rik van Riel Cc: Tetsuo Handa Cc: Thomas Gleixner Cc: linux-mm@kvack.org Fixes: 39a0526fb3f7 ("x86/mm: Factor out LDT init from context init") Link: http://lkml.kernel.org/r/20170824175029.76040-1-ebiggers3@gmail.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mmu_context.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h index 265c907..7a234be 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -140,9 +140,7 @@ static inline int init_new_context(struct task_struct *tsk, mm->context.execute_only_pkey = -1; } #endif - init_new_context_ldt(tsk, mm); - - return 0; + return init_new_context_ldt(tsk, mm); } static inline void destroy_context(struct mm_struct *mm) { -- cgit v1.1 From 64aee2a965cf2954a038b5522f11d2cd2f0f8f3e Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 22 Jun 2017 15:41:38 +0100 Subject: perf/core: Fix group {cpu,task} validation Regardless of which events form a group, it does not make sense for the events to target different tasks and/or CPUs, as this leaves the group inconsistent and impossible to schedule. The core perf code assumes that these are consistent across (successfully intialised) groups. Core perf code only verifies this when moving SW events into a HW context. Thus, we can violate this requirement for pure SW groups and pure HW groups, unless the relevant PMU driver happens to perform this verification itself. These mismatched groups subsequently wreak havoc elsewhere. For example, we handle watchpoints as SW events, and reserve watchpoint HW on a per-CPU basis at pmu::event_init() time to ensure that any event that is initialised is guaranteed to have a slot at pmu::add() time. However, the core code only checks the group leader's cpu filter (via event_filter_match()), and can thus install follower events onto CPUs violating thier (mismatched) CPU filters, potentially installing them into a CPU without sufficient reserved slots. This can be triggered with the below test case, resulting in warnings from arch backends. #define _GNU_SOURCE #include #include #include #include #include #include #include static int perf_event_open(struct perf_event_attr *attr, pid_t pid, int cpu, int group_fd, unsigned long flags) { return syscall(__NR_perf_event_open, attr, pid, cpu, group_fd, flags); } char watched_char; struct perf_event_attr wp_attr = { .type = PERF_TYPE_BREAKPOINT, .bp_type = HW_BREAKPOINT_RW, .bp_addr = (unsigned long)&watched_char, .bp_len = 1, .size = sizeof(wp_attr), }; int main(int argc, char *argv[]) { int leader, ret; cpu_set_t cpus; /* * Force use of CPU0 to ensure our CPU0-bound events get scheduled. */ CPU_ZERO(&cpus); CPU_SET(0, &cpus); ret = sched_setaffinity(0, sizeof(cpus), &cpus); if (ret) { printf("Unable to set cpu affinity\n"); return 1; } /* open leader event, bound to this task, CPU0 only */ leader = perf_event_open(&wp_attr, 0, 0, -1, 0); if (leader < 0) { printf("Couldn't open leader: %d\n", leader); return 1; } /* * Open a follower event that is bound to the same task, but a * different CPU. This means that the group should never be possible to * schedule. */ ret = perf_event_open(&wp_attr, 0, 1, leader, 0); if (ret < 0) { printf("Couldn't open mismatched follower: %d\n", ret); return 1; } else { printf("Opened leader/follower with mismastched CPUs\n"); } /* * Open as many independent events as we can, all bound to the same * task, CPU0 only. */ do { ret = perf_event_open(&wp_attr, 0, 0, -1, 0); } while (ret >= 0); /* * Force enable/disble all events to trigger the erronoeous * installation of the follower event. */ printf("Opened all events. Toggling..\n"); for (;;) { prctl(PR_TASK_PERF_EVENTS_DISABLE, 0, 0, 0, 0); prctl(PR_TASK_PERF_EVENTS_ENABLE, 0, 0, 0, 0); } return 0; } Fix this by validating this requirement regardless of whether we're moving events. Signed-off-by: Mark Rutland Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Zhou Chengming Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1498142498-15758-1-git-send-email-mark.rutland@arm.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 39 +++++++++++++++++++-------------------- 1 file changed, 19 insertions(+), 20 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index ee20d4c..3504125 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -10032,28 +10032,27 @@ SYSCALL_DEFINE5(perf_event_open, goto err_context; /* - * Do not allow to attach to a group in a different - * task or CPU context: + * Make sure we're both events for the same CPU; + * grouping events for different CPUs is broken; since + * you can never concurrently schedule them anyhow. */ - if (move_group) { - /* - * Make sure we're both on the same task, or both - * per-cpu events. - */ - if (group_leader->ctx->task != ctx->task) - goto err_context; + if (group_leader->cpu != event->cpu) + goto err_context; - /* - * Make sure we're both events for the same CPU; - * grouping events for different CPUs is broken; since - * you can never concurrently schedule them anyhow. - */ - if (group_leader->cpu != event->cpu) - goto err_context; - } else { - if (group_leader->ctx != ctx) - goto err_context; - } + /* + * Make sure we're both on the same task, or both + * per-CPU events. + */ + if (group_leader->ctx->task != ctx->task) + goto err_context; + + /* + * Do not allow to attach to a group in a different task + * or CPU context. If we're moving SW events, we'll fix + * this up later, so allow that. + */ + if (!move_group && group_leader->ctx != ctx) + goto err_context; /* * Only a group leader can be exclusive or pinned -- cgit v1.1 From 47c5310a8dbe7c2cb9f0083daa43ceed76c257fa Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 24 Aug 2017 19:14:47 +1000 Subject: KVM: PPC: Book3S: Fix race and leak in kvm_vm_ioctl_create_spapr_tce() Nixiaoming pointed out that there is a memory leak in kvm_vm_ioctl_create_spapr_tce() if the call to anon_inode_getfd() fails; the memory allocated for the kvmppc_spapr_tce_table struct is not freed, and nor are the pages allocated for the iommu tables. In addition, we have already incremented the process's count of locked memory pages, and this doesn't get restored on error. David Hildenbrand pointed out that there is a race in that the function checks early on that there is not already an entry in the stt->iommu_tables list with the same LIOBN, but an entry with the same LIOBN could get added between then and when the new entry is added to the list. This fixes all three problems. To simplify things, we now call anon_inode_getfd() before placing the new entry in the list. The check for an existing entry is done while holding the kvm->lock mutex, immediately before adding the new entry to the list. Finally, on failure we now call kvmppc_account_memlimit to decrement the process's count of locked memory pages. Reported-by: Nixiaoming Reported-by: David Hildenbrand Signed-off-by: Paul Mackerras Signed-off-by: Paolo Bonzini --- arch/powerpc/kvm/book3s_64_vio.c | 56 ++++++++++++++++++++++++---------------- 1 file changed, 34 insertions(+), 22 deletions(-) diff --git a/arch/powerpc/kvm/book3s_64_vio.c b/arch/powerpc/kvm/book3s_64_vio.c index a160c14..53766e2 100644 --- a/arch/powerpc/kvm/book3s_64_vio.c +++ b/arch/powerpc/kvm/book3s_64_vio.c @@ -294,32 +294,26 @@ long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm, struct kvm_create_spapr_tce_64 *args) { struct kvmppc_spapr_tce_table *stt = NULL; + struct kvmppc_spapr_tce_table *siter; unsigned long npages, size; int ret = -ENOMEM; int i; + int fd = -1; if (!args->size) return -EINVAL; - /* Check this LIOBN hasn't been previously allocated */ - list_for_each_entry(stt, &kvm->arch.spapr_tce_tables, list) { - if (stt->liobn == args->liobn) - return -EBUSY; - } - size = _ALIGN_UP(args->size, PAGE_SIZE >> 3); npages = kvmppc_tce_pages(size); ret = kvmppc_account_memlimit(kvmppc_stt_pages(npages), true); - if (ret) { - stt = NULL; - goto fail; - } + if (ret) + return ret; ret = -ENOMEM; stt = kzalloc(sizeof(*stt) + npages * sizeof(struct page *), GFP_KERNEL); if (!stt) - goto fail; + goto fail_acct; stt->liobn = args->liobn; stt->page_shift = args->page_shift; @@ -334,24 +328,42 @@ long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm, goto fail; } - kvm_get_kvm(kvm); + ret = fd = anon_inode_getfd("kvm-spapr-tce", &kvm_spapr_tce_fops, + stt, O_RDWR | O_CLOEXEC); + if (ret < 0) + goto fail; mutex_lock(&kvm->lock); - list_add_rcu(&stt->list, &kvm->arch.spapr_tce_tables); + + /* Check this LIOBN hasn't been previously allocated */ + ret = 0; + list_for_each_entry(siter, &kvm->arch.spapr_tce_tables, list) { + if (siter->liobn == args->liobn) { + ret = -EBUSY; + break; + } + } + + if (!ret) { + list_add_rcu(&stt->list, &kvm->arch.spapr_tce_tables); + kvm_get_kvm(kvm); + } mutex_unlock(&kvm->lock); - return anon_inode_getfd("kvm-spapr-tce", &kvm_spapr_tce_fops, - stt, O_RDWR | O_CLOEXEC); + if (!ret) + return fd; -fail: - if (stt) { - for (i = 0; i < npages; i++) - if (stt->pages[i]) - __free_page(stt->pages[i]); + put_unused_fd(fd); - kfree(stt); - } + fail: + for (i = 0; i < npages; i++) + if (stt->pages[i]) + __free_page(stt->pages[i]); + + kfree(stt); + fail_acct: + kvmppc_account_memlimit(kvmppc_stt_pages(npages), false); return ret; } -- cgit v1.1 From 22d538213ec4fa65b08b1edbf610066d8aab7bbb Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 18 Aug 2017 15:52:54 -0700 Subject: blk-mq-debugfs: Add names for recently added flags The symbolic constants QUEUE_FLAG_SCSI_PASSTHROUGH, QUEUE_FLAG_QUIESCED and REQ_NOWAIT are missing from blk-mq-debugfs.c. Add these to blk-mq-debugfs.c such that these appear as names in debugfs instead of as numbers. Reviewed-by: Omar Sandoval Signed-off-by: Bart Van Assche Cc: Hannes Reinecke Signed-off-by: Jens Axboe --- block/blk-mq-debugfs.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c index 9ebc294..4f927a5 100644 --- a/block/blk-mq-debugfs.c +++ b/block/blk-mq-debugfs.c @@ -75,6 +75,8 @@ static const char *const blk_queue_flag_name[] = { QUEUE_FLAG_NAME(STATS), QUEUE_FLAG_NAME(POLL_STATS), QUEUE_FLAG_NAME(REGISTERED), + QUEUE_FLAG_NAME(SCSI_PASSTHROUGH), + QUEUE_FLAG_NAME(QUIESCED), }; #undef QUEUE_FLAG_NAME @@ -265,6 +267,7 @@ static const char *const cmd_flag_name[] = { CMD_FLAG_NAME(RAHEAD), CMD_FLAG_NAME(BACKGROUND), CMD_FLAG_NAME(NOUNMAP), + CMD_FLAG_NAME(NOWAIT), }; #undef CMD_FLAG_NAME -- cgit v1.1 From 1046d304900cf9d4b2c730c6860b8e03cc704377 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Wed, 26 Jul 2017 15:32:23 +0100 Subject: virtio_blk: fix incorrect message when disk is resized The message printed on disk resize is incorrect. The following is printed when resizing to 2 GiB: $ truncate -s 1G test.img $ qemu -device virtio-blk-pci,logical_block_size=4096,... (qemu) block_resize drive1 2G virtio_blk virtio0: new size: 4194304 4096-byte logical blocks (17.2 GB/16.0 GiB) The virtio_blk capacity config field is in 512-byte sector units regardless of logical_block_size as per the VIRTIO specification. Therefore the message should read: virtio_blk virtio0: new size: 524288 4096-byte logical blocks (2.15 GB/2.0 GiB) Note that this only affects the printed message. Thankfully the actual block device has the correct size because the block layer expects capacity in sectors. Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- drivers/block/virtio_blk.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 1498b89..d3d5523 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -381,6 +381,7 @@ static void virtblk_config_changed_work(struct work_struct *work) struct request_queue *q = vblk->disk->queue; char cap_str_2[10], cap_str_10[10]; char *envp[] = { "RESIZE=1", NULL }; + unsigned long long nblocks; u64 capacity; /* Host must always specify the capacity. */ @@ -393,16 +394,19 @@ static void virtblk_config_changed_work(struct work_struct *work) capacity = (sector_t)-1; } - string_get_size(capacity, queue_logical_block_size(q), + nblocks = DIV_ROUND_UP_ULL(capacity, queue_logical_block_size(q) >> 9); + + string_get_size(nblocks, queue_logical_block_size(q), STRING_UNITS_2, cap_str_2, sizeof(cap_str_2)); - string_get_size(capacity, queue_logical_block_size(q), + string_get_size(nblocks, queue_logical_block_size(q), STRING_UNITS_10, cap_str_10, sizeof(cap_str_10)); dev_notice(&vdev->dev, - "new size: %llu %d-byte logical blocks (%s/%s)\n", - (unsigned long long)capacity, - queue_logical_block_size(q), - cap_str_10, cap_str_2); + "new size: %llu %d-byte logical blocks (%s/%s)\n", + nblocks, + queue_logical_block_size(q), + cap_str_10, + cap_str_2); set_capacity(vblk->disk, capacity); revalidate_disk(vblk->disk); -- cgit v1.1 From ba74b6f7fcc07355d087af6939712eed4a454821 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 24 Aug 2017 18:07:02 +0200 Subject: virtio_pci: fix cpu affinity support Commit 0b0f9dc5 ("Revert "virtio_pci: use shared interrupts for virtqueues"") removed the adjustment of the pre_vectors for the virtio MSI-X vector allocation which was added in commit fb5e31d9 ("virtio: allow drivers to request IRQ affinity when creating VQs"). This will lead to an incorrect assignment of MSI-X vectors, and potential deadlocks when offlining cpus. Signed-off-by: Christoph Hellwig Fixes: 0b0f9dc5 ("Revert "virtio_pci: use shared interrupts for virtqueues") Reported-by: YASUAKI ISHIMATSU Cc: stable@vger.kernel.org Signed-off-by: Michael S. Tsirkin --- drivers/virtio/virtio_pci_common.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/virtio/virtio_pci_common.c b/drivers/virtio/virtio_pci_common.c index 007a4f3..1c4797e 100644 --- a/drivers/virtio/virtio_pci_common.c +++ b/drivers/virtio/virtio_pci_common.c @@ -107,6 +107,7 @@ static int vp_request_msix_vectors(struct virtio_device *vdev, int nvectors, { struct virtio_pci_device *vp_dev = to_vp_device(vdev); const char *name = dev_name(&vp_dev->vdev.dev); + unsigned flags = PCI_IRQ_MSIX; unsigned i, v; int err = -ENOMEM; @@ -126,10 +127,13 @@ static int vp_request_msix_vectors(struct virtio_device *vdev, int nvectors, GFP_KERNEL)) goto error; + if (desc) { + flags |= PCI_IRQ_AFFINITY; + desc->pre_vectors++; /* virtio config vector */ + } + err = pci_alloc_irq_vectors_affinity(vp_dev->pci_dev, nvectors, - nvectors, PCI_IRQ_MSIX | - (desc ? PCI_IRQ_AFFINITY : 0), - desc); + nvectors, flags, desc); if (err < 0) goto error; vp_dev->msix_enabled = 1; -- cgit v1.1 From dc52d783d68c07a68743d42ef2f9a6f9a6d80fb1 Mon Sep 17 00:00:00 2001 From: Annie Li Date: Thu, 24 Aug 2017 17:25:59 -0400 Subject: xen-blkback: stop blkback thread of every queue in xen_blkif_disconnect MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In xen_blkif_disconnect, before checking inflight I/O, following code stops the blkback thread, if (ring->xenblkd) { kthread_stop(ring->xenblkd); wake_up(&ring->shutdown_wq); } If there is inflight I/O in any non-last queue, blkback returns -EBUSY directly, and above code would not be called to stop thread of remaining queue and processs them. When removing vbd device with lots of disk I/O load, some queues with inflight I/O still have blkback thread running even though the corresponding vbd device or guest is gone. And this could cause some problems, for example, if the backend device type is file, some loop devices and blkback thread always lingers there forever after guest is destroyed, and this causes failure of umounting repositories unless rebooting the dom0. This patch allows thread of every queue has the chance to get stopped. Otherwise, only thread of queue previous to(including) first busy one get stopped, blkthread of remaining queue will still run. So stop all threads properly and return -EBUSY if any queue has inflight I/O. Signed-off-by: Annie Li Reviewed-by: Herbert van den Bergh Reviewed-by: Bhavesh Davda Reviewed-by: Adnan Misherfi Acked-by: Roger Pau Monné Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/xenbus.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index 792da68..2adb859 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -244,6 +244,7 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif) { struct pending_req *req, *n; unsigned int j, r; + bool busy = false; for (r = 0; r < blkif->nr_rings; r++) { struct xen_blkif_ring *ring = &blkif->rings[r]; @@ -261,8 +262,10 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif) * don't have any discard_io or other_io requests. So, checking * for inflight IO is enough. */ - if (atomic_read(&ring->inflight) > 0) - return -EBUSY; + if (atomic_read(&ring->inflight) > 0) { + busy = true; + continue; + } if (ring->irq) { unbind_from_irqhandler(ring->irq, ring); @@ -300,6 +303,9 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif) WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages)); ring->active = false; } + if (busy) + return -EBUSY; + blkif->nr_ring_pages = 0; /* * blkif->rings was allocated in connect_ring, so we should free it in -- cgit v1.1 From 556b969a1cfe2686aae149137fa1dfcac0eefe54 Mon Sep 17 00:00:00 2001 From: Chen Yu Date: Fri, 25 Aug 2017 15:55:30 -0700 Subject: PM/hibernate: touch NMI watchdog when creating snapshot There is a problem that when counting the pages for creating the hibernation snapshot will take significant amount of time, especially on system with large memory. Since the counting job is performed with irq disabled, this might lead to NMI lockup. The following warning were found on a system with 1.5TB DRAM: Freezing user space processes ... (elapsed 0.002 seconds) done. OOM killer disabled. PM: Preallocating image memory... NMI watchdog: Watchdog detected hard LOCKUP on cpu 27 CPU: 27 PID: 3128 Comm: systemd-sleep Not tainted 4.13.0-0.rc2.git0.1.fc27.x86_64 #1 task: ffff9f01971ac000 task.stack: ffffb1a3f325c000 RIP: 0010:memory_bm_find_bit+0xf4/0x100 Call Trace: swsusp_set_page_free+0x2b/0x30 mark_free_pages+0x147/0x1c0 count_data_pages+0x41/0xa0 hibernate_preallocate_memory+0x80/0x450 hibernation_snapshot+0x58/0x410 hibernate+0x17c/0x310 state_store+0xdf/0xf0 kobj_attr_store+0xf/0x20 sysfs_kf_write+0x37/0x40 kernfs_fop_write+0x11c/0x1a0 __vfs_write+0x37/0x170 vfs_write+0xb1/0x1a0 SyS_write+0x55/0xc0 entry_SYSCALL_64_fastpath+0x1a/0xa5 ... done (allocated 6590003 pages) PM: Allocated 26360012 kbytes in 19.89 seconds (1325.28 MB/s) It has taken nearly 20 seconds(2.10GHz CPU) thus the NMI lockup was triggered. In case the timeout of the NMI watch dog has been set to 1 second, a safe interval should be 6590003/20 = 320k pages in theory. However there might also be some platforms running at a lower frequency, so feed the watchdog every 100k pages. [yu.c.chen@intel.com: simplification] Link: http://lkml.kernel.org/r/1503460079-29721-1-git-send-email-yu.c.chen@intel.com [yu.c.chen@intel.com: use interval of 128k instead of 100k to avoid modulus] Link: http://lkml.kernel.org/r/1503328098-5120-1-git-send-email-yu.c.chen@intel.com Signed-off-by: Chen Yu Reported-by: Jan Filipcewicz Suggested-by: Michal Hocko Reviewed-by: Michal Hocko Acked-by: Rafael J. Wysocki Cc: Mel Gorman Cc: Vlastimil Babka Cc: Len Brown Cc: Dan Williams Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 1bad301..7a58eb5 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -66,6 +66,7 @@ #include #include #include +#include #include #include @@ -2535,9 +2536,14 @@ void drain_all_pages(struct zone *zone) #ifdef CONFIG_HIBERNATION +/* + * Touch the watchdog for every WD_PAGE_COUNT pages. + */ +#define WD_PAGE_COUNT (128*1024) + void mark_free_pages(struct zone *zone) { - unsigned long pfn, max_zone_pfn; + unsigned long pfn, max_zone_pfn, page_count = WD_PAGE_COUNT; unsigned long flags; unsigned int order, t; struct page *page; @@ -2552,6 +2558,11 @@ void mark_free_pages(struct zone *zone) if (pfn_valid(pfn)) { page = pfn_to_page(pfn); + if (!--page_count) { + touch_nmi_watchdog(); + page_count = WD_PAGE_COUNT; + } + if (page_zone(page) != zone) continue; @@ -2565,8 +2576,13 @@ void mark_free_pages(struct zone *zone) unsigned long i; pfn = page_to_pfn(page); - for (i = 0; i < (1UL << order); i++) + for (i = 0; i < (1UL << order); i++) { + if (!--page_count) { + touch_nmi_watchdog(); + page_count = WD_PAGE_COUNT; + } swsusp_set_page_free(pfn_to_page(pfn + i)); + } } } spin_unlock_irqrestore(&zone->lock, flags); -- cgit v1.1 From 435c0b87d661da83771c30ed775f7c37eed193fb Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 25 Aug 2017 15:55:33 -0700 Subject: mm, shmem: fix handling /sys/kernel/mm/transparent_hugepage/shmem_enabled /sys/kernel/mm/transparent_hugepage/shmem_enabled controls if we want to allocate huge pages when allocate pages for private in-kernel shmem mount. Unfortunately, as Dan noticed, I've screwed it up and the only way to make kernel allocate huge page for the mount is to use "force" there. All other values will be effectively ignored. Link: http://lkml.kernel.org/r/20170822144254.66431-1-kirill.shutemov@linux.intel.com Fixes: 5a6e75f8110c ("shmem: prepare huge= mount option and sysfs knob") Signed-off-by: Kirill A. Shutemov Reported-by: Dan Carpenter Cc: stable [4.8+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 6540e59..fbcb3c9 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3967,7 +3967,7 @@ int __init shmem_init(void) } #ifdef CONFIG_TRANSPARENT_HUGE_PAGECACHE - if (has_transparent_hugepage() && shmem_huge < SHMEM_HUGE_DENY) + if (has_transparent_hugepage() && shmem_huge > SHMEM_HUGE_DENY) SHMEM_SB(shm_mnt->mnt_sb)->huge = shmem_huge; else shmem_huge = 0; /* just in case it was patched */ @@ -4028,7 +4028,7 @@ static ssize_t shmem_enabled_store(struct kobject *kobj, return -EINVAL; shmem_huge = huge; - if (shmem_huge < SHMEM_HUGE_DENY) + if (shmem_huge > SHMEM_HUGE_DENY) SHMEM_SB(shm_mnt->mnt_sb)->huge = shmem_huge; return count; } -- cgit v1.1 From fffa281b48a91ad6dac1a18c5907ece58fa3879b Mon Sep 17 00:00:00 2001 From: Ross Zwisler Date: Fri, 25 Aug 2017 15:55:36 -0700 Subject: dax: fix deadlock due to misaligned PMD faults In DAX there are two separate places where the 2MiB range of a PMD is defined. The first is in the page tables, where a PMD mapping inserted for a given address spans from (vmf->address & PMD_MASK) to ((vmf->address & PMD_MASK) + PMD_SIZE - 1). That is, from the 2MiB boundary below the address to the 2MiB boundary above the address. So, for example, a fault at address 3MiB (0x30 0000) falls within the PMD that ranges from 2MiB (0x20 0000) to 4MiB (0x40 0000). The second PMD range is in the mapping->page_tree, where a given file offset is covered by a radix tree entry that spans from one 2MiB aligned file offset to another 2MiB aligned file offset. So, for example, the file offset for 3MiB (pgoff 768) falls within the PMD range for the order 9 radix tree entry that ranges from 2MiB (pgoff 512) to 4MiB (pgoff 1024). This system works so long as the addresses and file offsets for a given mapping both have the same offsets relative to the start of each PMD. Consider the case where the starting address for a given file isn't 2MiB aligned - say our faulting address is 3 MiB (0x30 0000), but that corresponds to the beginning of our file (pgoff 0). Now all the PMDs in the mapping are misaligned so that the 2MiB range defined in the page tables never matches up with the 2MiB range defined in the radix tree. The current code notices this case for DAX faults to storage with the following test in dax_pmd_insert_mapping(): if (pfn_t_to_pfn(pfn) & PG_PMD_COLOUR) goto unlock_fallback; This test makes sure that the pfn we get from the driver is 2MiB aligned, and relies on the assumption that the 2MiB alignment of the pfn we get back from the driver matches the 2MiB alignment of the faulting address. However, faults to holes were not checked and we could hit the problem described above. This was reported in response to the NVML nvml/src/test/pmempool_sync TEST5: $ cd nvml/src/test/pmempool_sync $ make TEST5 You can grab NVML here: https://github.com/pmem/nvml/ The dmesg warning you see when you hit this error is: WARNING: CPU: 13 PID: 2900 at fs/dax.c:641 dax_insert_mapping_entry+0x2df/0x310 Where we notice in dax_insert_mapping_entry() that the radix tree entry we are about to replace doesn't match the locked entry that we had previously inserted into the tree. This happens because the initial insertion was done in grab_mapping_entry() using a pgoff calculated from the faulting address (vmf->address), and the replacement in dax_pmd_load_hole() => dax_insert_mapping_entry() is done using vmf->pgoff. In our failure case those two page offsets (one calculated from vmf->address, one using vmf->pgoff) point to different order 9 radix tree entries. This failure case can result in a deadlock because the radix tree unlock also happens on the pgoff calculated from vmf->address. This means that the locked radix tree entry that we swapped in to the tree in dax_insert_mapping_entry() using vmf->pgoff is never unlocked, so all future faults to that 2MiB range will block forever. Fix this by validating that the faulting address's PMD offset matches the PMD offset from the start of the file. This check is done at the very beginning of the fault and covers faults that would have mapped to storage as well as faults to holes. I left the COLOUR check in dax_pmd_insert_mapping() in place in case we ever hit the insanity condition where the alignment of the pfn we get from the driver doesn't match the alignment of the userspace address. Link: http://lkml.kernel.org/r/20170822222436.18926-1-ross.zwisler@linux.intel.com Signed-off-by: Ross Zwisler Reported-by: "Slusarz, Marcin" Reviewed-by: Jan Kara Cc: Alexander Viro Cc: Christoph Hellwig Cc: Dan Williams Cc: Dave Chinner Cc: Matthew Wilcox Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dax.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/fs/dax.c b/fs/dax.c index 306c2b6..865d42c 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -1383,6 +1383,16 @@ static int dax_iomap_pmd_fault(struct vm_fault *vmf, trace_dax_pmd_fault(inode, vmf, max_pgoff, 0); + /* + * Make sure that the faulting address's PMD offset (color) matches + * the PMD offset from the start of the file. This is necessary so + * that a PMD range in the page table overlaps exactly with a PMD + * range in the radix tree. + */ + if ((vmf->pgoff & PG_PMD_COLOUR) != + ((vmf->address >> PAGE_SHIFT) & PG_PMD_COLOUR)) + goto fallback; + /* Fall back to PTEs if we're going to COW */ if (write && !(vma->vm_flags & VM_SHARED)) goto fallback; -- cgit v1.1 From 263630e8d176d87308481ebdcd78ef9426739c6b Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Fri, 25 Aug 2017 15:55:39 -0700 Subject: mm/madvise.c: fix freeing of locked page with MADV_FREE If madvise(..., MADV_FREE) split a transparent hugepage, it called put_page() before unlock_page(). This was wrong because put_page() can free the page, e.g. if a concurrent madvise(..., MADV_DONTNEED) has removed it from the memory mapping. put_page() then rightfully complained about freeing a locked page. Fix this by moving the unlock_page() before put_page(). This bug was found by syzkaller, which encountered the following splat: BUG: Bad page state in process syzkaller412798 pfn:1bd800 page:ffffea0006f60000 count:0 mapcount:0 mapping: (null) index:0x20a00 flags: 0x200000000040019(locked|uptodate|dirty|swapbacked) raw: 0200000000040019 0000000000000000 0000000000020a00 00000000ffffffff raw: ffffea0006f60020 ffffea0006f60020 0000000000000000 0000000000000000 page dumped because: PAGE_FLAGS_CHECK_AT_FREE flag(s) set bad because of flags: 0x1(locked) Modules linked in: CPU: 1 PID: 3037 Comm: syzkaller412798 Not tainted 4.13.0-rc5+ #35 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:16 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:52 bad_page+0x230/0x2b0 mm/page_alloc.c:565 free_pages_check_bad+0x1f0/0x2e0 mm/page_alloc.c:943 free_pages_check mm/page_alloc.c:952 [inline] free_pages_prepare mm/page_alloc.c:1043 [inline] free_pcp_prepare mm/page_alloc.c:1068 [inline] free_hot_cold_page+0x8cf/0x12b0 mm/page_alloc.c:2584 __put_single_page mm/swap.c:79 [inline] __put_page+0xfb/0x160 mm/swap.c:113 put_page include/linux/mm.h:814 [inline] madvise_free_pte_range+0x137a/0x1ec0 mm/madvise.c:371 walk_pmd_range mm/pagewalk.c:50 [inline] walk_pud_range mm/pagewalk.c:108 [inline] walk_p4d_range mm/pagewalk.c:134 [inline] walk_pgd_range mm/pagewalk.c:160 [inline] __walk_page_range+0xc3a/0x1450 mm/pagewalk.c:249 walk_page_range+0x200/0x470 mm/pagewalk.c:326 madvise_free_page_range.isra.9+0x17d/0x230 mm/madvise.c:444 madvise_free_single_vma+0x353/0x580 mm/madvise.c:471 madvise_dontneed_free mm/madvise.c:555 [inline] madvise_vma mm/madvise.c:664 [inline] SYSC_madvise mm/madvise.c:832 [inline] SyS_madvise+0x7d3/0x13c0 mm/madvise.c:760 entry_SYSCALL_64_fastpath+0x1f/0xbe Here is a C reproducer: #define _GNU_SOURCE #include #include #include #define MADV_FREE 8 #define PAGE_SIZE 4096 static void *mapping; static const size_t mapping_size = 0x1000000; static void *madvise_thrproc(void *arg) { madvise(mapping, mapping_size, (long)arg); } int main(void) { pthread_t t[2]; for (;;) { mapping = mmap(NULL, mapping_size, PROT_WRITE, MAP_POPULATE|MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); munmap(mapping + mapping_size / 2, PAGE_SIZE); pthread_create(&t[0], 0, madvise_thrproc, (void*)MADV_DONTNEED); pthread_create(&t[1], 0, madvise_thrproc, (void*)MADV_FREE); pthread_join(t[0], NULL); pthread_join(t[1], NULL); munmap(mapping, mapping_size); } } Note: to see the splat, CONFIG_TRANSPARENT_HUGEPAGE=y and CONFIG_DEBUG_VM=y are needed. Google Bug Id: 64696096 Link: http://lkml.kernel.org/r/20170823205235.132061-1-ebiggers3@gmail.com Fixes: 854e9ed09ded ("mm: support madvise(MADV_FREE)") Signed-off-by: Eric Biggers Acked-by: David Rientjes Acked-by: Minchan Kim Acked-by: Michal Hocko Cc: Dmitry Vyukov Cc: Hugh Dickins Cc: Andrea Arcangeli Cc: [v4.5+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/madvise.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/madvise.c b/mm/madvise.c index 47d8d8a..23ed525 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -368,8 +368,8 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr, pte_offset_map_lock(mm, pmd, addr, &ptl); goto out; } - put_page(page); unlock_page(page); + put_page(page); pte = pte_offset_map_lock(mm, pmd, addr, &ptl); pte--; addr -= PAGE_SIZE; -- cgit v1.1 From 2b7e8665b4ff51c034c55df3cff76518d1a9ee3a Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Fri, 25 Aug 2017 15:55:43 -0700 Subject: fork: fix incorrect fput of ->exe_file causing use-after-free Commit 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable") made it possible to kill a forking task while it is waiting to acquire its ->mmap_sem for write, in dup_mmap(). However, it was overlooked that this introduced an new error path before a reference is taken on the mm_struct's ->exe_file. Since the ->exe_file of the new mm_struct was already set to the old ->exe_file by the memcpy() in dup_mm(), it was possible for the mmput() in the error path of dup_mm() to drop a reference to ->exe_file which was never taken. This caused the struct file to later be freed prematurely. Fix it by updating mm_init() to NULL out the ->exe_file, in the same place it clears other things like the list of mmaps. This bug was found by syzkaller. It can be reproduced using the following C program: #define _GNU_SOURCE #include #include #include #include #include #include static void *mmap_thread(void *_arg) { for (;;) { mmap(NULL, 0x1000000, PROT_READ, MAP_POPULATE|MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); } } static void *fork_thread(void *_arg) { usleep(rand() % 10000); fork(); } int main(void) { fork(); fork(); fork(); for (;;) { if (fork() == 0) { pthread_t t; pthread_create(&t, NULL, mmap_thread, NULL); pthread_create(&t, NULL, fork_thread, NULL); usleep(rand() % 10000); syscall(__NR_exit_group, 0); } wait(NULL); } } No special kernel config options are needed. It usually causes a NULL pointer dereference in __remove_shared_vm_struct() during exit, or in dup_mmap() (which is usually inlined into copy_process()) during fork. Both are due to a vm_area_struct's ->vm_file being used after it's already been freed. Google Bug Id: 64772007 Link: http://lkml.kernel.org/r/20170823211408.31198-1-ebiggers3@gmail.com Fixes: 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable") Signed-off-by: Eric Biggers Tested-by: Mark Rutland Acked-by: Michal Hocko Cc: Dmitry Vyukov Cc: Ingo Molnar Cc: Konstantin Khlebnikov Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Vlastimil Babka Cc: [v4.7+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/fork.c b/kernel/fork.c index e075b77..cbbea27 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -806,6 +806,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p, mm_init_cpumask(mm); mm_init_aio(mm); mm_init_owner(mm, p); + RCU_INIT_POINTER(mm->exe_file, NULL); mmu_notifier_mm_init(mm); init_tlb_flush_pending(mm); #if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS -- cgit v1.1 From 91b540f98872a206ea1c49e4aa6ea8eed0886644 Mon Sep 17 00:00:00 2001 From: Pavel Tatashin Date: Fri, 25 Aug 2017 15:55:46 -0700 Subject: mm/memblock.c: reversed logic in memblock_discard() In recently introduced memblock_discard() there is a reversed logic bug. Memory is freed of static array instead of dynamically allocated one. Link: http://lkml.kernel.org/r/1503511441-95478-2-git-send-email-pasha.tatashin@oracle.com Fixes: 3010f876500f ("mm: discard memblock data later") Signed-off-by: Pavel Tatashin Reported-by: Woody Suwalski Tested-by: Woody Suwalski Acked-by: Michal Hocko Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memblock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/memblock.c b/mm/memblock.c index bf14aea..9120578 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -299,7 +299,7 @@ void __init memblock_discard(void) __memblock_free_late(addr, size); } - if (memblock.memory.regions == memblock_memory_init_regions) { + if (memblock.memory.regions != memblock_memory_init_regions) { addr = __pa(memblock.memory.regions); size = PAGE_ALIGN(sizeof(struct memblock_region) * memblock.memory.max); -- cgit v1.1 From 8e1101d251647802d0a4ae19eb3d0e1453eaeff4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 25 Aug 2017 18:58:42 -0500 Subject: PCI/MSI: Don't warn when irq_create_affinity_masks() returns NULL irq_create_affinity_masks() can return NULL on non-SMP systems, when there are not enough "free" vectors available to spread, or if memory allocation for the CPU masks fails. Only the allocation failure is of interest, and even then the system will work just fine except for non-optimally spread vectors. Thus remove the warnings. Signed-off-by: Christoph Hellwig Signed-off-by: Bjorn Helgaas Acked-by: David S. Miller --- drivers/pci/msi.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 253d924..2225afc 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -538,12 +538,9 @@ msi_setup_entry(struct pci_dev *dev, int nvec, const struct irq_affinity *affd) struct msi_desc *entry; u16 control; - if (affd) { + if (affd) masks = irq_create_affinity_masks(nvec, affd); - if (!masks) - dev_err(&dev->dev, "can't allocate MSI affinity masks for %d vectors\n", - nvec); - } + /* MSI Entry Initialization */ entry = alloc_msi_entry(&dev->dev, nvec, masks); @@ -679,12 +676,8 @@ static int msix_setup_entries(struct pci_dev *dev, void __iomem *base, struct msi_desc *entry; int ret, i; - if (affd) { + if (affd) masks = irq_create_affinity_masks(nvec, affd); - if (!masks) - dev_err(&dev->dev, "can't allocate MSI-X affinity masks for %d vectors\n", - nvec); - } for (i = 0, curmsk = masks; i < nvec; i++) { entry = alloc_msi_entry(&dev->dev, 1, curmsk); -- cgit v1.1 From 3614364527daa870264f6dde77f02853cdecd02c Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Fri, 25 Aug 2017 09:05:42 +0200 Subject: ipv6: Fix may be used uninitialized warning in rt6_check rt_cookie might be used uninitialized, fix this by initializing it. Fixes: c5cff8561d2d ("ipv6: add rcu grace period before freeing fib6_node") Signed-off-by: Steffen Klassert Signed-off-by: David S. Miller --- net/ipv6/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a9d3564..48c8c92 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1289,7 +1289,7 @@ static void rt6_dst_from_metrics_check(struct rt6_info *rt) static struct dst_entry *rt6_check(struct rt6_info *rt, u32 cookie) { - u32 rt_cookie; + u32 rt_cookie = 0; if (!rt6_get_cookie_safe(rt, &rt_cookie) || rt_cookie != cookie) return NULL; -- cgit v1.1 From ebfa00c5745660fe7f0a91eea88d4dff658486c4 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 25 Aug 2017 13:10:12 +0200 Subject: tcp: fix refcnt leak with ebpf congestion control There are a few bugs around refcnt handling in the new BPF congestion control setsockopt: - The new ca is assigned to icsk->icsk_ca_ops even in the case where we cannot get a reference on it. This would lead to a use after free, since that ca is going away soon. - Changing the congestion control case doesn't release the refcnt on the previous ca. - In the reinit case, we first leak a reference on the old ca, then we call tcp_reinit_congestion_control on the ca that we have just assigned, leading to deinitializing the wrong ca (->release of the new ca on the old ca's data) and releasing the refcount on the ca that we actually want to use. This is visible by building (for example) BIC as a module and setting net.ipv4.tcp_congestion_control=bic, and using tcp_cong_kern.c from samples/bpf. This patch fixes the refcount issues, and moves reinit back into tcp core to avoid passing a ca pointer back to BPF. Fixes: 91b5b21c7c16 ("bpf: Add support for changing congestion control") Signed-off-by: Sabrina Dubroca Acked-by: Lawrence Brakmo Signed-off-by: David S. Miller --- include/net/tcp.h | 4 +--- net/core/filter.c | 7 ++----- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_cong.c | 19 ++++++++++++++----- 4 files changed, 18 insertions(+), 14 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index ada65e7..f642a39 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1004,9 +1004,7 @@ void tcp_get_default_congestion_control(char *name); void tcp_get_available_congestion_control(char *buf, size_t len); void tcp_get_allowed_congestion_control(char *buf, size_t len); int tcp_set_allowed_congestion_control(char *allowed); -int tcp_set_congestion_control(struct sock *sk, const char *name, bool load); -void tcp_reinit_congestion_control(struct sock *sk, - const struct tcp_congestion_ops *ca); +int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, bool reinit); u32 tcp_slow_start(struct tcp_sock *tp, u32 acked); void tcp_cong_avoid_ai(struct tcp_sock *tp, u32 w, u32 acked); diff --git a/net/core/filter.c b/net/core/filter.c index 8eb81e5..1699749 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2836,15 +2836,12 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, sk->sk_prot->setsockopt == tcp_setsockopt) { if (optname == TCP_CONGESTION) { char name[TCP_CA_NAME_MAX]; + bool reinit = bpf_sock->op > BPF_SOCK_OPS_NEEDS_ECN; strncpy(name, optval, min_t(long, optlen, TCP_CA_NAME_MAX-1)); name[TCP_CA_NAME_MAX-1] = 0; - ret = tcp_set_congestion_control(sk, name, false); - if (!ret && bpf_sock->op > BPF_SOCK_OPS_NEEDS_ECN) - /* replacing an existing ca */ - tcp_reinit_congestion_control(sk, - inet_csk(sk)->icsk_ca_ops); + ret = tcp_set_congestion_control(sk, name, false, reinit); } else { struct tcp_sock *tp = tcp_sk(sk); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 71ce33d..a3e91b5 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2481,7 +2481,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, name[val] = 0; lock_sock(sk); - err = tcp_set_congestion_control(sk, name, true); + err = tcp_set_congestion_control(sk, name, true, true); release_sock(sk); return err; } diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index fde983f..421ea1b 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -189,8 +189,8 @@ void tcp_init_congestion_control(struct sock *sk) INET_ECN_dontxmit(sk); } -void tcp_reinit_congestion_control(struct sock *sk, - const struct tcp_congestion_ops *ca) +static void tcp_reinit_congestion_control(struct sock *sk, + const struct tcp_congestion_ops *ca) { struct inet_connection_sock *icsk = inet_csk(sk); @@ -338,7 +338,7 @@ out: * tcp_reinit_congestion_control (if the current congestion control was * already initialized. */ -int tcp_set_congestion_control(struct sock *sk, const char *name, bool load) +int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, bool reinit) { struct inet_connection_sock *icsk = inet_csk(sk); const struct tcp_congestion_ops *ca; @@ -360,9 +360,18 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load) if (!ca) { err = -ENOENT; } else if (!load) { - icsk->icsk_ca_ops = ca; - if (!try_module_get(ca->owner)) + const struct tcp_congestion_ops *old_ca = icsk->icsk_ca_ops; + + if (try_module_get(ca->owner)) { + if (reinit) { + tcp_reinit_congestion_control(sk, ca); + } else { + icsk->icsk_ca_ops = ca; + module_put(old_ca->owner); + } + } else { err = -EBUSY; + } } else if (!((ca->flags & TCP_CONG_NON_RESTRICTED) || ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN))) { err = -EPERM; -- cgit v1.1 From 1089650d8837095f63e001bbf14d7b48043d67ad Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 24 Aug 2017 18:34:43 -0700 Subject: r8169: Do not increment tx_dropped in TX ring cleaning rtl8169_tx_clear_range() is responsible for cleaning up the TX ring during interface shutdown, incrementing tx_dropped for every SKB that we left at the time in the ring is misleading. Fixes: cac4b22f3d6a ("r8169: do not account fragments as packets") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index bd07a15..8a1bbd2 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -6863,7 +6863,6 @@ static void rtl8169_tx_clear_range(struct rtl8169_private *tp, u32 start, rtl8169_unmap_tx_skb(&tp->pci_dev->dev, tx_skb, tp->TxDescArray + entry); if (skb) { - tp->dev->stats.tx_dropped++; dev_kfree_skb_any(skb); tx_skb->skb = NULL; } -- cgit v1.1 From 7a4b813cb739ce598ffbad2e84d19d13fa23e25d Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 24 Aug 2017 18:34:44 -0700 Subject: r8169: Be drop monitor friendly rtl_tx() is the TX reclamation process whereas rtl8169_tx_clear_range() does the TX ring cleaning during shutdown, both of these functions should call dev_consume_skb_any() to be drop monitor friendly. Fixes: cac4b22f3d6a ("r8169: do not account fragments as packets") Fixes: eb781397904e ("r8169: Do not use dev_kfree_skb in xmit path") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index 8a1bbd2..e03fcf9 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -6863,7 +6863,7 @@ static void rtl8169_tx_clear_range(struct rtl8169_private *tp, u32 start, rtl8169_unmap_tx_skb(&tp->pci_dev->dev, tx_skb, tp->TxDescArray + entry); if (skb) { - dev_kfree_skb_any(skb); + dev_consume_skb_any(skb); tx_skb->skb = NULL; } } @@ -7318,7 +7318,7 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp) tp->tx_stats.packets++; tp->tx_stats.bytes += tx_skb->skb->len; u64_stats_update_end(&tp->tx_stats.syncp); - dev_kfree_skb_any(tx_skb->skb); + dev_consume_skb_any(tx_skb->skb); tx_skb->skb = NULL; } dirty_tx++; -- cgit v1.1 From 2207d182c14294d78b98142f0b6a16bea5e8b0fb Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Fri, 25 Aug 2017 07:35:51 +0200 Subject: net: sxgbe: check memory allocation failure Check memory allocation failure and return -ENOMEM in such a case, as already done few lines below for another memory allocation. Signed-off-by: Christophe JAILLET Signed-off-by: David S. Miller --- drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c index 73427e2..fbd00cb 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_platform.c @@ -47,6 +47,8 @@ static int sxgbe_probe_config_dt(struct platform_device *pdev, plat->mdio_bus_data = devm_kzalloc(&pdev->dev, sizeof(*plat->mdio_bus_data), GFP_KERNEL); + if (!plat->mdio_bus_data) + return -ENOMEM; dma_cfg = devm_kzalloc(&pdev->dev, sizeof(*dma_cfg), GFP_KERNEL); if (!dma_cfg) -- cgit v1.1 From 64f0f5d18a47c703c85576375cc010e83dac6a48 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 25 Aug 2017 14:31:01 +0200 Subject: udp6: set rx_dst_cookie on rx_dst updates Currently, in the udp6 code, the dst cookie is not initialized/updated concurrently with the RX dst used by early demux. As a result, the dst_check() in the early_demux path always fails, the rx dst cache is always invalidated, and we can't really leverage significant gain from the demux lookup. Fix it adding udp6 specific variant of sk_rx_dst_set() and use it to set the dst cookie when the dst entry is really changed. The issue is there since the introduction of early demux for ipv6. Fixes: 5425077d73e0 ("net: ipv6: Add early demux handler for UDP unicast") Acked-by: Hannes Frederic Sowa Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/net/udp.h | 2 +- net/ipv4/udp.c | 4 +++- net/ipv6/udp.c | 11 ++++++++++- 3 files changed, 14 insertions(+), 3 deletions(-) diff --git a/include/net/udp.h b/include/net/udp.h index 586de4b..626c2d8 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -260,7 +260,7 @@ static inline struct sk_buff *skb_recv_udp(struct sock *sk, unsigned int flags, } void udp_v4_early_demux(struct sk_buff *skb); -void udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst); +bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst); int udp_get_port(struct sock *sk, unsigned short snum, int (*saddr_cmp)(const struct sock *, const struct sock *)); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index cd1d044..a6dc48d 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1929,14 +1929,16 @@ drop: /* For TCP sockets, sk_rx_dst is protected by socket lock * For UDP, we use xchg() to guard against concurrent changes. */ -void udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) +bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) { struct dst_entry *old; if (dst_hold_safe(dst)) { old = xchg(&sk->sk_rx_dst, dst); dst_release(old); + return old != dst; } + return false; } EXPORT_SYMBOL(udp_sk_rx_dst_set); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 20039c8..d688622 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -768,6 +768,15 @@ start_lookup: return 0; } +static void udp6_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) +{ + if (udp_sk_rx_dst_set(sk, dst)) { + const struct rt6_info *rt = (const struct rt6_info *)dst; + + inet6_sk(sk)->rx_dst_cookie = rt6_get_cookie(rt); + } +} + int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, int proto) { @@ -817,7 +826,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, int ret; if (unlikely(sk->sk_rx_dst != dst)) - udp_sk_rx_dst_set(sk, dst); + udp6_sk_rx_dst_set(sk, dst); ret = udpv6_queue_rcv_skb(sk, skb); sock_put(sk); -- cgit v1.1 From 0cc3b0ec23ce4c69e1e890ed2b8d2fa932b14aad Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 27 Aug 2017 12:12:25 -0700 Subject: Clarify (and fix) MAX_LFS_FILESIZE macros We have a MAX_LFS_FILESIZE macro that is meant to be filled in by filesystems (and other IO targets) that know they are 64-bit clean and don't have any 32-bit limits in their IO path. It turns out that our 32-bit value for that limit was bogus. On 32-bit, the VM layer is limited by the page cache to only 32-bit index values, but our logic for that was confusing and actually wrong. We used to define that value to (((loff_t)PAGE_SIZE << (BITS_PER_LONG-1))-1) which is actually odd in several ways: it limits the index to 31 bits, and then it limits files so that they can't have data in that last byte of a page that has the highest 31-bit index (ie page index 0x7fffffff). Neither of those limitations make sense. The index is actually the full 32 bit unsigned value, and we can use that whole full page. So the maximum size of the file would logically be "PAGE_SIZE << BITS_PER_LONG". However, we do wan tto avoid the maximum index, because we have code that iterates over the page indexes, and we don't want that code to overflow. So the maximum size of a file on a 32-bit host should actually be one page less than the full 32-bit index. So the actual limit is ULONG_MAX << PAGE_SHIFT. That means that we will not actually be using the page of that last index (ULONG_MAX), but we can grow a file up to that limit. The wrong value of MAX_LFS_FILESIZE actually caused problems for Doug Nazar, who was still using a 32-bit host, but with a 9.7TB 2 x RAID5 volume. It turns out that our old MAX_LFS_FILESIZE was 8TiB (well, one byte less), but the actual true VM limit is one page less than 16TiB. This was invisible until commit c2a9737f45e2 ("vfs,mm: fix a dead loop in truncate_inode_pages_range()"), which started applying that MAX_LFS_FILESIZE limit to block devices too. NOTE! On 64-bit, the page index isn't a limiter at all, and the limit is actually just the offset type itself (loff_t), which is signed. But for clarity, on 64-bit, just use the maximum signed value, and don't make people have to count the number of 'f' characters in the hex constant. So just use LLONG_MAX for the 64-bit case. That was what the value had been before too, just written out as a hex constant. Fixes: c2a9737f45e2 ("vfs,mm: fix a dead loop in truncate_inode_pages_range()") Reported-and-tested-by: Doug Nazar Cc: Andreas Dilger Cc: Mark Fasheh Cc: Joel Becker Cc: Dave Kleikamp Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- include/linux/fs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/fs.h b/include/linux/fs.h index 6e1fd5d..cbfe127 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -907,9 +907,9 @@ static inline struct file *get_file(struct file *f) /* Page cache limit. The filesystems should put that into their s_maxbytes limits, otherwise bad things can happen in VM. */ #if BITS_PER_LONG==32 -#define MAX_LFS_FILESIZE (((loff_t)PAGE_SIZE << (BITS_PER_LONG-1))-1) +#define MAX_LFS_FILESIZE ((loff_t)ULONG_MAX << PAGE_SHIFT) #elif BITS_PER_LONG==64 -#define MAX_LFS_FILESIZE ((loff_t)0x7fffffffffffffffLL) +#define MAX_LFS_FILESIZE ((loff_t)LLONG_MAX) #endif #define FL_POSIX 1 -- cgit v1.1 From 3510ca20ece0150af6b10c77a74ff1b5c198e3e2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 27 Aug 2017 13:55:12 -0700 Subject: Minor page waitqueue cleanups Tim Chen and Kan Liang have been battling a customer load that shows extremely long page wakeup lists. The cause seems to be constant NUMA migration of a hot page that is shared across a lot of threads, but the actual root cause for the exact behavior has not been found. Tim has a patch that batches the wait list traversal at wakeup time, so that we at least don't get long uninterruptible cases where we traverse and wake up thousands of processes and get nasty latency spikes. That is likely 4.14 material, but we're still discussing the page waitqueue specific parts of it. In the meantime, I've tried to look at making the page wait queues less expensive, and failing miserably. If you have thousands of threads waiting for the same page, it will be painful. We'll need to try to figure out the NUMA balancing issue some day, in addition to avoiding the excessive spinlock hold times. That said, having tried to rewrite the page wait queues, I can at least fix up some of the braindamage in the current situation. In particular: (a) we don't want to continue walking the page wait list if the bit we're waiting for already got set again (which seems to be one of the patterns of the bad load). That makes no progress and just causes pointless cache pollution chasing the pointers. (b) we don't want to put the non-locking waiters always on the front of the queue, and the locking waiters always on the back. Not only is that unfair, it means that we wake up thousands of reading threads that will just end up being blocked by the writer later anyway. Also add a comment about the layout of 'struct wait_page_key' - there is an external user of it in the cachefiles code that means that it has to match the layout of 'struct wait_bit_key' in the two first members. It so happens to match, because 'struct page *' and 'unsigned long *' end up having the same values simply because the page flags are the first member in struct page. Cc: Tim Chen Cc: Kan Liang Cc: Mel Gorman Cc: Christopher Lameter Cc: Andi Kleen Cc: Davidlohr Bueso Cc: Peter Zijlstra Signed-off-by: Linus Torvalds --- kernel/sched/wait.c | 7 ++++--- mm/filemap.c | 11 ++++++----- 2 files changed, 10 insertions(+), 8 deletions(-) diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 17f11c6..d6afed6 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -70,9 +70,10 @@ static void __wake_up_common(struct wait_queue_head *wq_head, unsigned int mode, list_for_each_entry_safe(curr, next, &wq_head->head, entry) { unsigned flags = curr->flags; - - if (curr->func(curr, mode, wake_flags, key) && - (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) + int ret = curr->func(curr, mode, wake_flags, key); + if (ret < 0) + break; + if (ret && (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) break; } } diff --git a/mm/filemap.c b/mm/filemap.c index a497024..baba290 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -885,6 +885,7 @@ void __init pagecache_init(void) page_writeback_init(); } +/* This has the same layout as wait_bit_key - see fs/cachefiles/rdwr.c */ struct wait_page_key { struct page *page; int bit_nr; @@ -909,8 +910,10 @@ static int wake_page_function(wait_queue_entry_t *wait, unsigned mode, int sync, if (wait_page->bit_nr != key->bit_nr) return 0; + + /* Stop walking if it's locked */ if (test_bit(key->bit_nr, &key->page->flags)) - return 0; + return -1; return autoremove_wake_function(wait, mode, sync, key); } @@ -964,6 +967,7 @@ static inline int wait_on_page_bit_common(wait_queue_head_t *q, int ret = 0; init_wait(wait); + wait->flags = lock ? WQ_FLAG_EXCLUSIVE : 0; wait->func = wake_page_function; wait_page.page = page; wait_page.bit_nr = bit_nr; @@ -972,10 +976,7 @@ static inline int wait_on_page_bit_common(wait_queue_head_t *q, spin_lock_irq(&q->lock); if (likely(list_empty(&wait->entry))) { - if (lock) - __add_wait_queue_entry_tail_exclusive(q, wait); - else - __add_wait_queue(q, wait); + __add_wait_queue_entry_tail(q, wait); SetPageWaiters(page); } -- cgit v1.1 From a8b169afbf06a678437632709caac98e16f99263 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 27 Aug 2017 16:25:09 -0700 Subject: Avoid page waitqueue race leaving possible page locker waiting The "lock_page_killable()" function waits for exclusive access to the page lock bit using the WQ_FLAG_EXCLUSIVE bit in the waitqueue entry set. That means that if it gets woken up, other waiters may have been skipped. That, in turn, means that if it sees the page being unlocked, it *must* take that lock and return success, even if a lethal signal is also pending. So instead of checking for lethal signals first, we need to check for them after we've checked the actual bit that we were waiting for. Even if that might then delay the killing of the process. This matches the order of the old "wait_on_bit_lock()" infrastructure that the page locking used to use (and is still used in a few other areas). Note that if we still return an error after having unsuccessfully tried to acquire the page lock, that is ok: that means that some other thread was able to get ahead of us and lock the page, and when that other thread then unlocks the page, the wakeup event will be repeated. So any other pending waiters will now get properly woken up. Fixes: 62906027091f ("mm: add PageWaiters indicating tasks are waiting for a page bit") Cc: Nick Piggin Cc: Peter Zijlstra Cc: Mel Gorman Cc: Jan Kara Cc: Davidlohr Bueso Cc: Andi Kleen Signed-off-by: Linus Torvalds --- mm/filemap.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index baba290..0b41c8c 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -986,10 +986,6 @@ static inline int wait_on_page_bit_common(wait_queue_head_t *q, if (likely(test_bit(bit_nr, &page->flags))) { io_schedule(); - if (unlikely(signal_pending_state(state, current))) { - ret = -EINTR; - break; - } } if (lock) { @@ -999,6 +995,11 @@ static inline int wait_on_page_bit_common(wait_queue_head_t *q, if (!test_bit(bit_nr, &page->flags)) break; } + + if (unlikely(signal_pending_state(state, current))) { + ret = -EINTR; + break; + } } finish_wait(q, wait); -- cgit v1.1 From cc4a41fe5541a73019a864883297bd5043aa6d98 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 27 Aug 2017 17:20:40 -0700 Subject: Linux 4.13-rc7 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index dda88e7..8db6be7 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 4 PATCHLEVEL = 13 SUBLEVEL = 0 -EXTRAVERSION = -rc6 +EXTRAVERSION = -rc7 NAME = Fearless Coyote # *DOCUMENTATION* -- cgit v1.1 From 5fe0d4bd8f86d19f7f24c1ae5a9b6e6a5a52e51a Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Sat, 26 Aug 2017 17:08:57 +0200 Subject: xfrm_user: fix info leak in copy_user_offload() The memory reserved to dump the xfrm offload state includes padding bytes of struct xfrm_user_offload added by the compiler for alignment. Add an explicit memset(0) before filling the buffer to avoid the heap info leak. Cc: Steffen Klassert Fixes: d77e38e612a0 ("xfrm: Add an IPsec hardware offloading API") Signed-off-by: Mathias Krause Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 2be4c6a..3259555 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -796,7 +796,7 @@ static int copy_user_offload(struct xfrm_state_offload *xso, struct sk_buff *skb return -EMSGSIZE; xuo = nla_data(attr); - + memset(xuo, 0, sizeof(*xuo)); xuo->ifindex = xso->dev->ifindex; xuo->flags = xso->flags; -- cgit v1.1 From 50329c8a340c9dea60d837645fcf13fc36bfb84d Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Sat, 26 Aug 2017 17:08:58 +0200 Subject: xfrm_user: fix info leak in xfrm_notify_sa() The memory reserved to dump the ID of the xfrm state includes a padding byte in struct xfrm_usersa_id added by the compiler for alignment. To prevent the heap info leak, memset(0) the whole struct before filling it. Cc: Herbert Xu Fixes: 0603eac0d6b7 ("[IPSEC]: Add XFRMA_SA/XFRMA_POLICY for delete notification") Signed-off-by: Mathias Krause Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 3259555..c33516e 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2715,6 +2715,7 @@ static int xfrm_notify_sa(struct xfrm_state *x, const struct km_event *c) struct nlattr *attr; id = nlmsg_data(nlh); + memset(id, 0, sizeof(*id)); memcpy(&id->daddr, &x->id.daddr, sizeof(id->daddr)); id->spi = x->id.spi; id->family = x->props.family; -- cgit v1.1 From e3e5fc1698ae35ac60d075b477e84accb96e2652 Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Sat, 26 Aug 2017 17:08:59 +0200 Subject: xfrm_user: fix info leak in build_expire() The memory reserved to dump the expired xfrm state includes padding bytes in struct xfrm_user_expire added by the compiler for alignment. To prevent the heap info leak, memset(0) the remainder of the struct. Initializing the whole structure isn't needed as copy_to_user_state() already takes care of clearing the padding bytes within the 'state' member. Signed-off-by: Mathias Krause Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index c33516e..2cbdc81 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2578,6 +2578,8 @@ static int build_expire(struct sk_buff *skb, struct xfrm_state *x, const struct ue = nlmsg_data(nlh); copy_to_user_state(x, &ue->state); ue->hard = (c->data.hard != 0) ? 1 : 0; + /* clear the padding bytes */ + memset(&ue->hard + 1, 0, sizeof(*ue) - offsetofend(typeof(*ue), hard)); err = xfrm_mark_put(skb, &x->mark); if (err) -- cgit v1.1 From 931e79d7a7ddee4709c56b39de169a36804589a1 Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Sat, 26 Aug 2017 17:09:00 +0200 Subject: xfrm_user: fix info leak in build_aevent() The memory reserved to dump the ID of the xfrm state includes a padding byte in struct xfrm_usersa_id added by the compiler for alignment. To prevent the heap info leak, memset(0) the sa_id before filling it. Cc: Jamal Hadi Salim Fixes: d51d081d6504 ("[IPSEC]: Sync series - user") Signed-off-by: Mathias Krause Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 2cbdc81..9391ced 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1869,6 +1869,7 @@ static int build_aevent(struct sk_buff *skb, struct xfrm_state *x, const struct return -EMSGSIZE; id = nlmsg_data(nlh); + memset(&id->sa_id, 0, sizeof(id->sa_id)); memcpy(&id->sa_id.daddr, &x->id.daddr, sizeof(x->id.daddr)); id->sa_id.spi = x->id.spi; id->sa_id.family = x->props.family; -- cgit v1.1 From 54385bf75cc6451f29f3a149582584d5015d2c98 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 9 Aug 2017 11:32:10 -0700 Subject: dm: fix the second dec_pending() argument in __split_and_process_bio() Detected by sparse. Fixes: 4e4cbee93d56 ("block: switch bios to blk_status_t") Signed-off-by: Bart Van Assche Reviewed-by: Christoph Hellwig Tested-by: Laurence Oberman Signed-off-by: Mike Snitzer --- drivers/md/dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 2edbcc2..8298670 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1523,7 +1523,7 @@ static void __split_and_process_bio(struct mapped_device *md, } /* drop the extra reference count */ - dec_pending(ci.io, error); + dec_pending(ci.io, errno_to_blk_status(error)); } /*----------------------------------------------------------------- * CRUD END -- cgit v1.1 From 68515cc72191f5c5d86f6ef7091dc8ba714d96e4 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 9 Aug 2017 11:32:15 -0700 Subject: dm mpath: retry BLK_STS_RESOURCE errors Retry requests instead of failing them if an out-of-memory error occurs or the block driver below dm-mpath is busy. This restores the v4.12 behavior of noretry_error(), namely that -ENOMEM results in a retry. Fixes: 2a842acab109 ("block: introduce new block status code type") Signed-off-by: Bart Van Assche Reviewed-by: Christoph Hellwig Signed-off-by: Mike Snitzer --- drivers/md/dm-mpath.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 0e8ab5b..3486c82 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -1458,7 +1458,6 @@ static int noretry_error(blk_status_t error) case BLK_STS_TARGET: case BLK_STS_NEXUS: case BLK_STS_MEDIUM: - case BLK_STS_RESOURCE: return 1; } -- cgit v1.1 From 604407890ecf624c2fb41013c82b22aade59b455 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 9 Aug 2017 11:32:11 -0700 Subject: dm: fix printk() rate limiting code Using the same rate limiting state for different kinds of messages is wrong because this can cause a high frequency message to suppress a report of a low frequency message. Hence use a unique rate limiting state per message type. Fixes: 71a16736a15e ("dm: use local printk ratelimit") Cc: stable@vger.kernel.org Signed-off-by: Bart Van Assche Signed-off-by: Mike Snitzer --- drivers/md/dm.c | 10 ---------- include/linux/device-mapper.h | 41 ++++++++++++----------------------------- 2 files changed, 12 insertions(+), 39 deletions(-) diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 8298670..d669fdd 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -27,16 +27,6 @@ #define DM_MSG_PREFIX "core" -#ifdef CONFIG_PRINTK -/* - * ratelimit state to be used in DMXXX_LIMIT(). - */ -DEFINE_RATELIMIT_STATE(dm_ratelimit_state, - DEFAULT_RATELIMIT_INTERVAL, - DEFAULT_RATELIMIT_BURST); -EXPORT_SYMBOL(dm_ratelimit_state); -#endif - /* * Cookies are numeric values sent with CHANGE and REMOVE * uevents while resuming, removing or renaming the device. diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 1473455..4f2b3b2 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -549,46 +549,29 @@ void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size); *---------------------------------------------------------------*/ #define DM_NAME "device-mapper" -#ifdef CONFIG_PRINTK -extern struct ratelimit_state dm_ratelimit_state; - -#define dm_ratelimit() __ratelimit(&dm_ratelimit_state) -#else -#define dm_ratelimit() 0 -#endif +#define DM_RATELIMIT(pr_func, fmt, ...) \ +do { \ + static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL, \ + DEFAULT_RATELIMIT_BURST); \ + \ + if (__ratelimit(&rs)) \ + pr_func(DM_FMT(fmt), ##__VA_ARGS__); \ +} while (0) #define DM_FMT(fmt) DM_NAME ": " DM_MSG_PREFIX ": " fmt "\n" #define DMCRIT(fmt, ...) pr_crit(DM_FMT(fmt), ##__VA_ARGS__) #define DMERR(fmt, ...) pr_err(DM_FMT(fmt), ##__VA_ARGS__) -#define DMERR_LIMIT(fmt, ...) \ -do { \ - if (dm_ratelimit()) \ - DMERR(fmt, ##__VA_ARGS__); \ -} while (0) - +#define DMERR_LIMIT(fmt, ...) DM_RATELIMIT(pr_err, fmt, ##__VA_ARGS__) #define DMWARN(fmt, ...) pr_warn(DM_FMT(fmt), ##__VA_ARGS__) -#define DMWARN_LIMIT(fmt, ...) \ -do { \ - if (dm_ratelimit()) \ - DMWARN(fmt, ##__VA_ARGS__); \ -} while (0) - +#define DMWARN_LIMIT(fmt, ...) DM_RATELIMIT(pr_warn, fmt, ##__VA_ARGS__) #define DMINFO(fmt, ...) pr_info(DM_FMT(fmt), ##__VA_ARGS__) -#define DMINFO_LIMIT(fmt, ...) \ -do { \ - if (dm_ratelimit()) \ - DMINFO(fmt, ##__VA_ARGS__); \ -} while (0) +#define DMINFO_LIMIT(fmt, ...) DM_RATELIMIT(pr_info, fmt, ##__VA_ARGS__) #ifdef CONFIG_DM_DEBUG #define DMDEBUG(fmt, ...) printk(KERN_DEBUG DM_FMT(fmt), ##__VA_ARGS__) -#define DMDEBUG_LIMIT(fmt, ...) \ -do { \ - if (dm_ratelimit()) \ - DMDEBUG(fmt, ##__VA_ARGS__); \ -} while (0) +#define DMDEBUG_LIMIT(fmt, ...) DM_RATELIMIT(pr_debug, fmt, ##__VA_ARGS__) #else #define DMDEBUG(fmt, ...) no_printk(fmt, ##__VA_ARGS__) #define DMDEBUG_LIMIT(fmt, ...) no_printk(fmt, ##__VA_ARGS__) -- cgit v1.1 From 1c23484c355ec360ca2f37914f8a4802c6baeead Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 9 Aug 2017 11:32:12 -0700 Subject: dm mpath: do not lock up a CPU with requeuing activity When using the block layer in single queue mode, get_request() returns ERR_PTR(-EAGAIN) if the queue is dying and the REQ_NOWAIT flag has been passed to get_request(). Avoid that the kernel reports soft lockup complaints in this case due to continuous requeuing activity. Fixes: 7083abbbf ("dm mpath: avoid that path removal can trigger an infinite loop") Cc: stable@vger.kernel.org Signed-off-by: Bart Van Assche Tested-by: Laurence Oberman Reviewed-by: Christoph Hellwig Signed-off-by: Mike Snitzer --- drivers/md/dm-mpath.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 3486c82..d24e4b0 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -504,7 +504,6 @@ static int multipath_clone_and_map(struct dm_target *ti, struct request *rq, if (queue_dying) { atomic_inc(&m->pg_init_in_progress); activate_or_offline_path(pgpath); - return DM_MAPIO_REQUEUE; } return DM_MAPIO_DELAY_REQUEUE; } -- cgit v1.1 From 7a14724f54bf9889fcb1a9f1d4aa4e1d2e969d93 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Mon, 28 Aug 2017 08:33:20 -0700 Subject: libnvdimm: clean up command definitions Remove the command payloads that do not have an associated libnvdimm ioctl. I.e. remove the payloads that would only ever be carried in the ND_CMD_CALL envelope. This prevents userspace from growing unnecessary dependencies on this kernel header when userspace already has everything it needs to craft and send these commands. Cc: Jerry Hoemann Reported-by: Yasunori Goto Signed-off-by: Dan Williams --- include/uapi/linux/ndctl.h | 37 ------------------------------------- 1 file changed, 37 deletions(-) diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h index 6d3c542..3f03567 100644 --- a/include/uapi/linux/ndctl.h +++ b/include/uapi/linux/ndctl.h @@ -145,43 +145,6 @@ struct nd_cmd_clear_error { __u64 cleared; } __packed; -struct nd_cmd_trans_spa { - __u64 spa; - __u32 status; - __u8 flags; - __u8 _reserved[3]; - __u64 trans_length; - __u32 num_nvdimms; - struct nd_nvdimm_device { - __u32 nfit_device_handle; - __u32 _reserved; - __u64 dpa; - } __packed devices[0]; - -} __packed; - -struct nd_cmd_ars_err_inj { - __u64 err_inj_spa_range_base; - __u64 err_inj_spa_range_length; - __u8 err_inj_options; - __u32 status; -} __packed; - -struct nd_cmd_ars_err_inj_clr { - __u64 err_inj_clr_spa_range_base; - __u64 err_inj_clr_spa_range_length; - __u32 status; -} __packed; - -struct nd_cmd_ars_err_inj_stat { - __u32 status; - __u32 inj_err_rec_count; - struct nd_error_stat_query_record { - __u64 err_inj_stat_spa_range_base; - __u64 err_inj_stat_spa_range_length; - } __packed record[0]; -} __packed; - enum { ND_CMD_IMPLEMENTED = 0, -- cgit v1.1 From 35f0b6a779b8b7a98faefd7c1c660b4dac9a5c26 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Aug 2017 08:28:08 +0200 Subject: libata: quirk read log on no-name M.2 SSD Ido reported that reading the log page on his systems fails, so quirk it as it won't support ZBC or security protocols. Signed-off-by: Christoph Hellwig Reported-by: Ido Schimmel Tested-by: Ido Schimmel Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 4 ++++ include/linux/libata.h | 1 + 2 files changed, 5 insertions(+) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index fa7dd43..697f5f8 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2079,6 +2079,8 @@ unsigned int ata_read_log_page(struct ata_device *dev, u8 log, */ if (ap_flags & ATA_FLAG_NO_LOG_PAGE) return AC_ERR_DEV; + if (dev->horkage & ATA_HORKAGE_NO_LOG_PAGE) + return AC_ERR_DEV; retry: ata_tf_init(dev, &tf); @@ -4578,6 +4580,8 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "WDC WD3000JD-*", NULL, ATA_HORKAGE_WD_BROKEN_LPM }, { "WDC WD3200JD-*", NULL, ATA_HORKAGE_WD_BROKEN_LPM }, + + { "M.2 (S42) 3ME3", NULL, ATA_HORKAGE_NO_LOG_PAGE }, /* End Marker */ { } }; diff --git a/include/linux/libata.h b/include/linux/libata.h index 931c32f..9e927ae 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -438,6 +438,7 @@ enum { ATA_HORKAGE_NO_DMA_LOG = (1 << 23), /* don't use DMA for log read */ ATA_HORKAGE_NOTRIM = (1 << 24), /* don't use TRIM */ ATA_HORKAGE_MAX_SEC_1024 = (1 << 25), /* Limit max sects to 1024 */ + ATA_HORKAGE_NO_LOG_PAGE = (1 << 26), /* Doesn't like Get Log Page */ /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- cgit v1.1 From 3f9db52dc87b003a1732f3e03f7f5fc8701ef4ad Mon Sep 17 00:00:00 2001 From: Anthony Martin Date: Mon, 28 Aug 2017 10:26:12 -0700 Subject: Input: synaptics - fix device info appearing different on reconnect User-modified input settings no longer survive a suspend/resume cycle. Starting with 4.12, the touchpad is reinitialized on every reconnect because the hardware appears to be different. This can be reproduced by running the following as root: echo -n reconnect >/sys/devices/platform/i8042/serio1/drvctl A line like the following will show up in dmesg: [30378.295794] psmouse serio1: synaptics: hardware appears to be different: id(149271-149271), model(114865-114865), caps(d047b3-d047b1), ext(b40000-b40000). Note the single bit difference in caps: bit 1 (SYN_CAP_MULTIFINGER). This happens because we modify our stored copy of the device info capabilities when we enable advanced gesture mode but this change is not reflected in the actual hardware capabilities. It worked in the past because synaptics_query_hardware used to modify the stored synaptics_device_info struct instead of filling in a new one, as it does now. Fix it by no longer faking the SYN_CAP_MULTIFINGER bit when setting advanced gesture mode. This necessitated a small refactoring. Fixes: 6c53694fb222 ("Input: synaptics - split device info into a separate structure") Signed-off-by: Anthony Martin Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 35 ++++++++++++++++++++--------------- 1 file changed, 20 insertions(+), 15 deletions(-) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 16c3046..5af0b7d 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -535,16 +535,17 @@ static void synaptics_apply_quirks(struct psmouse *psmouse, } } +static bool synaptics_has_agm(struct synaptics_data *priv) +{ + return (SYN_CAP_ADV_GESTURE(priv->info.ext_cap_0c) || + SYN_CAP_IMAGE_SENSOR(priv->info.ext_cap_0c)); +} + static int synaptics_set_advanced_gesture_mode(struct psmouse *psmouse) { static u8 param = 0xc8; - struct synaptics_data *priv = psmouse->private; int error; - if (!(SYN_CAP_ADV_GESTURE(priv->info.ext_cap_0c) || - SYN_CAP_IMAGE_SENSOR(priv->info.ext_cap_0c))) - return 0; - error = psmouse_sliced_command(psmouse, SYN_QUE_MODEL); if (error) return error; @@ -553,9 +554,6 @@ static int synaptics_set_advanced_gesture_mode(struct psmouse *psmouse) if (error) return error; - /* Advanced gesture mode also sends multi finger data */ - priv->info.capabilities |= BIT(1); - return 0; } @@ -578,7 +576,7 @@ static int synaptics_set_mode(struct psmouse *psmouse) if (error) return error; - if (priv->absolute_mode) { + if (priv->absolute_mode && synaptics_has_agm(priv)) { error = synaptics_set_advanced_gesture_mode(psmouse); if (error) { psmouse_err(psmouse, @@ -766,9 +764,7 @@ static int synaptics_parse_hw_state(const u8 buf[], ((buf[0] & 0x04) >> 1) | ((buf[3] & 0x04) >> 2)); - if ((SYN_CAP_ADV_GESTURE(priv->info.ext_cap_0c) || - SYN_CAP_IMAGE_SENSOR(priv->info.ext_cap_0c)) && - hw->w == 2) { + if (synaptics_has_agm(priv) && hw->w == 2) { synaptics_parse_agm(buf, priv, hw); return 1; } @@ -1033,6 +1029,15 @@ static void synaptics_image_sensor_process(struct psmouse *psmouse, synaptics_report_mt_data(psmouse, sgm, num_fingers); } +static bool synaptics_has_multifinger(struct synaptics_data *priv) +{ + if (SYN_CAP_MULTIFINGER(priv->info.capabilities)) + return true; + + /* Advanced gesture mode also sends multi finger data */ + return synaptics_has_agm(priv); +} + /* * called for each full received packet from the touchpad */ @@ -1079,7 +1084,7 @@ static void synaptics_process_packet(struct psmouse *psmouse) if (SYN_CAP_EXTENDED(info->capabilities)) { switch (hw.w) { case 0 ... 1: - if (SYN_CAP_MULTIFINGER(info->capabilities)) + if (synaptics_has_multifinger(priv)) num_fingers = hw.w + 2; break; case 2: @@ -1123,7 +1128,7 @@ static void synaptics_process_packet(struct psmouse *psmouse) input_report_abs(dev, ABS_TOOL_WIDTH, finger_width); input_report_key(dev, BTN_TOOL_FINGER, num_fingers == 1); - if (SYN_CAP_MULTIFINGER(info->capabilities)) { + if (synaptics_has_multifinger(priv)) { input_report_key(dev, BTN_TOOL_DOUBLETAP, num_fingers == 2); input_report_key(dev, BTN_TOOL_TRIPLETAP, num_fingers == 3); } @@ -1283,7 +1288,7 @@ static void set_input_params(struct psmouse *psmouse, __set_bit(BTN_TOUCH, dev->keybit); __set_bit(BTN_TOOL_FINGER, dev->keybit); - if (SYN_CAP_MULTIFINGER(info->capabilities)) { + if (synaptics_has_multifinger(priv)) { __set_bit(BTN_TOOL_DOUBLETAP, dev->keybit); __set_bit(BTN_TOOL_TRIPLETAP, dev->keybit); } -- cgit v1.1 From 1e22391e8fbec9c3709bad82b997b108d1c6228b Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Fri, 25 Aug 2017 15:04:32 +0200 Subject: net: missing call of trace_napi_poll in busy_poll_stop Noticed that busy_poll_stop() also invoke the drivers napi->poll() function pointer, but didn't have an associated call to trace_napi_poll() like all other call sites. Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- net/core/dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/dev.c b/net/core/dev.c index ce15a06..818dfa6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5289,6 +5289,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock) * Ideally, a new ndo_busy_poll_stop() could avoid another round. */ rc = napi->poll(napi, BUSY_POLL_BUDGET); + trace_napi_poll(napi, rc, BUSY_POLL_BUDGET); netpoll_poll_unlock(have_poll_lock); if (rc == BUSY_POLL_BUDGET) __napi_schedule(napi); -- cgit v1.1 From 3b638f0f0b94fc41f419033f2a02c49d52f3fec9 Mon Sep 17 00:00:00 2001 From: Aleksander Morgado Date: Fri, 25 Aug 2017 15:39:16 +0200 Subject: cdc_ncm: flag the u-blox TOBY-L4 as wwan The u-blox TOBY-L4 is a LTE Advanced (Cat 6) module with HSPA+ and 2G fallback. Unlike the TOBY-L2, this module has one single USB layout and exposes several TTYs for control and a NCM interface for data. Connecting this module may be done just by activating the desired PDP context with 'AT+CGACT=1,' and then running DHCP on the NCM interface. Signed-off-by: Aleksander Morgado Signed-off-by: David S. Miller --- drivers/net/usb/cdc_ncm.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index 8f572b9..9c80e80 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -1758,6 +1758,13 @@ static const struct usb_device_id cdc_devs[] = { .driver_info = (unsigned long)&wwan_noarp_info, }, + /* u-blox TOBY-L4 */ + { USB_DEVICE_AND_INTERFACE_INFO(0x1546, 0x1010, + USB_CLASS_COMM, + USB_CDC_SUBCLASS_NCM, USB_CDC_PROTO_NONE), + .driver_info = (unsigned long)&wwan_info, + }, + /* Generic CDC-NCM devices */ { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_NCM, USB_CDC_PROTO_NONE), -- cgit v1.1 From 4c22868264516fe0c42817a87f37efb44254e7a9 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Fri, 25 Aug 2017 16:14:17 +0200 Subject: net: mvpp2: fix the mac address used when using PPv2.2 The mac address is only retrieved from h/w when using PPv2.1. Otherwise the variable holding it is still checked and used if it contains a valid value. As the variable isn't initialized to an invalid mac address value, we end up with random mac addresses which can be the same for all the ports handled by this PPv2 driver. Fixes this by initializing the h/w mac address variable to {0}, which is an invalid mac address value. This way the random assignation fallback is called and all ports end up with their own addresses. Signed-off-by: Antoine Tenart Fixes: 2697582144dd ("net: mvpp2: handle misc PPv2.1/PPv2.2 differences") Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/mvpp2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c index 48d21c1..4d598ca 100644 --- a/drivers/net/ethernet/marvell/mvpp2.c +++ b/drivers/net/ethernet/marvell/mvpp2.c @@ -6504,7 +6504,7 @@ static int mvpp2_port_probe(struct platform_device *pdev, struct resource *res; const char *dt_mac_addr; const char *mac_from; - char hw_mac_addr[ETH_ALEN]; + char hw_mac_addr[ETH_ALEN] = {0}; u32 id; int features; int phy_mode; -- cgit v1.1 From 9ee369a405c57613d7c83a3967780c3e30c52ecc Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 25 Aug 2017 16:22:17 +0200 Subject: l2tp: initialise session's refcount before making it reachable Sessions must be fully initialised before calling l2tp_session_add_to_tunnel(). Otherwise, there's a short time frame where partially initialised sessions can be accessed by external users. Fixes: dbdbc73b4478 ("l2tp: fix duplicate session creation") Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index b0c2d4a..f363669 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1844,6 +1844,8 @@ struct l2tp_session *l2tp_session_create(int priv_size, struct l2tp_tunnel *tunn l2tp_session_set_header_len(session, tunnel->version); + refcount_set(&session->ref_count, 1); + err = l2tp_session_add_to_tunnel(tunnel, session); if (err) { kfree(session); @@ -1851,10 +1853,6 @@ struct l2tp_session *l2tp_session_create(int priv_size, struct l2tp_tunnel *tunn return ERR_PTR(err); } - /* Bump the reference count. The session context is deleted - * only when this drops to zero. - */ - refcount_set(&session->ref_count, 1); l2tp_tunnel_inc_refcount(tunnel); /* Ensure tunnel socket isn't deleted */ -- cgit v1.1 From 54652eb12c1b72e9602d09cb2821d5760939190f Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 25 Aug 2017 16:51:40 +0200 Subject: l2tp: hold tunnel while looking up sessions in l2tp_netlink l2tp_tunnel_find() doesn't take a reference on the returned tunnel. Therefore, it's unsafe to use it because the returned tunnel can go away on us anytime. Fix this by defining l2tp_tunnel_get(), which works like l2tp_tunnel_find(), but takes a reference on the returned tunnel. Caller then has to drop this reference using l2tp_tunnel_dec_refcount(). As l2tp_tunnel_dec_refcount() needs to be moved to l2tp_core.h, let's simplify the patch and not move the L2TP_REFCNT_DEBUG part. This code has been broken (not even compiling) in May 2012 by commit a4ca44fa578c ("net: l2tp: Standardize logging styles") and fixed more than two years later by commit 29abe2fda54f ("l2tp: fix missing line continuation"). So it doesn't appear to be used by anyone. Same thing for l2tp_tunnel_free(); instead of moving it to l2tp_core.h, let's just simplify things and call kfree_rcu() directly in l2tp_tunnel_dec_refcount(). Extra assertions and debugging code provided by l2tp_tunnel_free() didn't help catching any of the reference counting and socket handling issues found while working on this series. Fixes: 309795f4bec2 ("l2tp: Add netlink control API for L2TP") Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 66 ++++++++++++++++--------------------------------- net/l2tp/l2tp_core.h | 13 ++++++++++ net/l2tp/l2tp_netlink.c | 6 +++-- 3 files changed, 38 insertions(+), 47 deletions(-) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index f363669..90165a6 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -113,7 +113,6 @@ struct l2tp_net { spinlock_t l2tp_session_hlist_lock; }; -static void l2tp_tunnel_free(struct l2tp_tunnel *tunnel); static inline struct l2tp_tunnel *l2tp_tunnel(struct sock *sk) { @@ -127,39 +126,6 @@ static inline struct l2tp_net *l2tp_pernet(const struct net *net) return net_generic(net, l2tp_net_id); } -/* Tunnel reference counts. Incremented per session that is added to - * the tunnel. - */ -static inline void l2tp_tunnel_inc_refcount_1(struct l2tp_tunnel *tunnel) -{ - refcount_inc(&tunnel->ref_count); -} - -static inline void l2tp_tunnel_dec_refcount_1(struct l2tp_tunnel *tunnel) -{ - if (refcount_dec_and_test(&tunnel->ref_count)) - l2tp_tunnel_free(tunnel); -} -#ifdef L2TP_REFCNT_DEBUG -#define l2tp_tunnel_inc_refcount(_t) \ -do { \ - pr_debug("l2tp_tunnel_inc_refcount: %s:%d %s: cnt=%d\n", \ - __func__, __LINE__, (_t)->name, \ - refcount_read(&_t->ref_count)); \ - l2tp_tunnel_inc_refcount_1(_t); \ -} while (0) -#define l2tp_tunnel_dec_refcount(_t) \ -do { \ - pr_debug("l2tp_tunnel_dec_refcount: %s:%d %s: cnt=%d\n", \ - __func__, __LINE__, (_t)->name, \ - refcount_read(&_t->ref_count)); \ - l2tp_tunnel_dec_refcount_1(_t); \ -} while (0) -#else -#define l2tp_tunnel_inc_refcount(t) l2tp_tunnel_inc_refcount_1(t) -#define l2tp_tunnel_dec_refcount(t) l2tp_tunnel_dec_refcount_1(t) -#endif - /* Session hash global list for L2TPv3. * The session_id SHOULD be random according to RFC3931, but several * L2TP implementations use incrementing session_ids. So we do a real @@ -229,6 +195,27 @@ l2tp_session_id_hash(struct l2tp_tunnel *tunnel, u32 session_id) return &tunnel->session_hlist[hash_32(session_id, L2TP_HASH_BITS)]; } +/* Lookup a tunnel. A new reference is held on the returned tunnel. */ +struct l2tp_tunnel *l2tp_tunnel_get(const struct net *net, u32 tunnel_id) +{ + const struct l2tp_net *pn = l2tp_pernet(net); + struct l2tp_tunnel *tunnel; + + rcu_read_lock_bh(); + list_for_each_entry_rcu(tunnel, &pn->l2tp_tunnel_list, list) { + if (tunnel->tunnel_id == tunnel_id) { + l2tp_tunnel_inc_refcount(tunnel); + rcu_read_unlock_bh(); + + return tunnel; + } + } + rcu_read_unlock_bh(); + + return NULL; +} +EXPORT_SYMBOL_GPL(l2tp_tunnel_get); + /* Lookup a session. A new reference is held on the returned session. * Optionally calls session->ref() too if do_ref is true. */ @@ -1348,17 +1335,6 @@ static void l2tp_udp_encap_destroy(struct sock *sk) } } -/* Really kill the tunnel. - * Come here only when all sessions have been cleared from the tunnel. - */ -static void l2tp_tunnel_free(struct l2tp_tunnel *tunnel) -{ - BUG_ON(refcount_read(&tunnel->ref_count) != 0); - BUG_ON(tunnel->sock != NULL); - l2tp_info(tunnel, L2TP_MSG_CONTROL, "%s: free...\n", tunnel->name); - kfree_rcu(tunnel, rcu); -} - /* Workqueue tunnel deletion function */ static void l2tp_tunnel_del_work(struct work_struct *work) { diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h index cdb6e33..9101297 100644 --- a/net/l2tp/l2tp_core.h +++ b/net/l2tp/l2tp_core.h @@ -231,6 +231,8 @@ out: return tunnel; } +struct l2tp_tunnel *l2tp_tunnel_get(const struct net *net, u32 tunnel_id); + struct l2tp_session *l2tp_session_get(const struct net *net, struct l2tp_tunnel *tunnel, u32 session_id, bool do_ref); @@ -269,6 +271,17 @@ int l2tp_nl_register_ops(enum l2tp_pwtype pw_type, void l2tp_nl_unregister_ops(enum l2tp_pwtype pw_type); int l2tp_ioctl(struct sock *sk, int cmd, unsigned long arg); +static inline void l2tp_tunnel_inc_refcount(struct l2tp_tunnel *tunnel) +{ + refcount_inc(&tunnel->ref_count); +} + +static inline void l2tp_tunnel_dec_refcount(struct l2tp_tunnel *tunnel) +{ + if (refcount_dec_and_test(&tunnel->ref_count)) + kfree_rcu(tunnel, rcu); +} + /* Session reference counts. Incremented when code obtains a reference * to a session. */ diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index 12cfcd0..27ee94b 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -65,10 +65,12 @@ static struct l2tp_session *l2tp_nl_session_get(struct genl_info *info, (info->attrs[L2TP_ATTR_CONN_ID])) { tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); session_id = nla_get_u32(info->attrs[L2TP_ATTR_SESSION_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel) + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (tunnel) { session = l2tp_session_get(net, tunnel, session_id, do_ref); + l2tp_tunnel_dec_refcount(tunnel); + } } return session; -- cgit v1.1 From bb0a32ce4389e17e47e198d2cddaf141561581ad Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 25 Aug 2017 16:51:42 +0200 Subject: l2tp: hold tunnel while processing genl delete command l2tp_nl_cmd_tunnel_delete() needs to take a reference on the tunnel, to prevent it from being concurrently freed by l2tp_tunnel_destruct(). Fixes: 309795f4bec2 ("l2tp: Add netlink control API for L2TP") Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_netlink.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index 27ee94b..8089665 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -273,8 +273,8 @@ static int l2tp_nl_cmd_tunnel_delete(struct sk_buff *skb, struct genl_info *info } tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel == NULL) { + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (!tunnel) { ret = -ENODEV; goto out; } @@ -284,6 +284,8 @@ static int l2tp_nl_cmd_tunnel_delete(struct sk_buff *skb, struct genl_info *info (void) l2tp_tunnel_delete(tunnel); + l2tp_tunnel_dec_refcount(tunnel); + out: return ret; } -- cgit v1.1 From 8c0e421525c9eb50d68e8f633f703ca31680b746 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 25 Aug 2017 16:51:42 +0200 Subject: l2tp: hold tunnel while handling genl tunnel updates We need to make sure the tunnel is not going to be destroyed by l2tp_tunnel_destruct() concurrently. Fixes: 309795f4bec2 ("l2tp: Add netlink control API for L2TP") Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_netlink.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index 8089665..d61e75b 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -303,8 +303,8 @@ static int l2tp_nl_cmd_tunnel_modify(struct sk_buff *skb, struct genl_info *info } tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel == NULL) { + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (!tunnel) { ret = -ENODEV; goto out; } @@ -315,6 +315,8 @@ static int l2tp_nl_cmd_tunnel_modify(struct sk_buff *skb, struct genl_info *info ret = l2tp_tunnel_notify(&l2tp_nl_family, info, tunnel, L2TP_CMD_TUNNEL_MODIFY); + l2tp_tunnel_dec_refcount(tunnel); + out: return ret; } -- cgit v1.1 From 4e4b21da3acc68a7ea55f850cacc13706b7480e9 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 25 Aug 2017 16:51:43 +0200 Subject: l2tp: hold tunnel while handling genl TUNNEL_GET commands Use l2tp_tunnel_get() instead of l2tp_tunnel_find() so that we get a reference on the tunnel, preventing l2tp_tunnel_destruct() from freeing it from under us. Also move l2tp_tunnel_get() below nlmsg_new() so that we only take the reference when needed. Fixes: 309795f4bec2 ("l2tp: Add netlink control API for L2TP") Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_netlink.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index d61e75b..ae5170e 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -444,34 +444,37 @@ static int l2tp_nl_cmd_tunnel_get(struct sk_buff *skb, struct genl_info *info) if (!info->attrs[L2TP_ATTR_CONN_ID]) { ret = -EINVAL; - goto out; + goto err; } tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); - if (tunnel == NULL) { - ret = -ENODEV; - goto out; - } - msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) { ret = -ENOMEM; - goto out; + goto err; + } + + tunnel = l2tp_tunnel_get(net, tunnel_id); + if (!tunnel) { + ret = -ENODEV; + goto err_nlmsg; } ret = l2tp_nl_tunnel_send(msg, info->snd_portid, info->snd_seq, NLM_F_ACK, tunnel, L2TP_CMD_TUNNEL_GET); if (ret < 0) - goto err_out; + goto err_nlmsg_tunnel; + + l2tp_tunnel_dec_refcount(tunnel); return genlmsg_unicast(net, msg, info->snd_portid); -err_out: +err_nlmsg_tunnel: + l2tp_tunnel_dec_refcount(tunnel); +err_nlmsg: nlmsg_free(msg); - -out: +err: return ret; } -- cgit v1.1 From e702c1204eb57788ef189c839c8c779368267d70 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 25 Aug 2017 16:51:46 +0200 Subject: l2tp: hold tunnel used while creating sessions with netlink Use l2tp_tunnel_get() to retrieve tunnel, so that it can't go away on us. Otherwise l2tp_tunnel_destruct() might release the last reference count concurrently, thus freeing the tunnel while we're using it. Fixes: 309795f4bec2 ("l2tp: Add netlink control API for L2TP") Signed-off-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_netlink.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index ae5170e..57427d4 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -518,8 +518,9 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf ret = -EINVAL; goto out; } + tunnel_id = nla_get_u32(info->attrs[L2TP_ATTR_CONN_ID]); - tunnel = l2tp_tunnel_find(net, tunnel_id); + tunnel = l2tp_tunnel_get(net, tunnel_id); if (!tunnel) { ret = -ENODEV; goto out; @@ -527,24 +528,24 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf if (!info->attrs[L2TP_ATTR_SESSION_ID]) { ret = -EINVAL; - goto out; + goto out_tunnel; } session_id = nla_get_u32(info->attrs[L2TP_ATTR_SESSION_ID]); if (!info->attrs[L2TP_ATTR_PEER_SESSION_ID]) { ret = -EINVAL; - goto out; + goto out_tunnel; } peer_session_id = nla_get_u32(info->attrs[L2TP_ATTR_PEER_SESSION_ID]); if (!info->attrs[L2TP_ATTR_PW_TYPE]) { ret = -EINVAL; - goto out; + goto out_tunnel; } cfg.pw_type = nla_get_u16(info->attrs[L2TP_ATTR_PW_TYPE]); if (cfg.pw_type >= __L2TP_PWTYPE_MAX) { ret = -EINVAL; - goto out; + goto out_tunnel; } if (tunnel->version > 2) { @@ -566,7 +567,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf u16 len = nla_len(info->attrs[L2TP_ATTR_COOKIE]); if (len > 8) { ret = -EINVAL; - goto out; + goto out_tunnel; } cfg.cookie_len = len; memcpy(&cfg.cookie[0], nla_data(info->attrs[L2TP_ATTR_COOKIE]), len); @@ -575,7 +576,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf u16 len = nla_len(info->attrs[L2TP_ATTR_PEER_COOKIE]); if (len > 8) { ret = -EINVAL; - goto out; + goto out_tunnel; } cfg.peer_cookie_len = len; memcpy(&cfg.peer_cookie[0], nla_data(info->attrs[L2TP_ATTR_PEER_COOKIE]), len); @@ -618,7 +619,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf if ((l2tp_nl_cmd_ops[cfg.pw_type] == NULL) || (l2tp_nl_cmd_ops[cfg.pw_type]->session_create == NULL)) { ret = -EPROTONOSUPPORT; - goto out; + goto out_tunnel; } /* Check that pseudowire-specific params are present */ @@ -628,7 +629,7 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf case L2TP_PWTYPE_ETH_VLAN: if (!info->attrs[L2TP_ATTR_VLAN_ID]) { ret = -EINVAL; - goto out; + goto out_tunnel; } break; case L2TP_PWTYPE_ETH: @@ -656,6 +657,8 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf } } +out_tunnel: + l2tp_tunnel_dec_refcount(tunnel); out: return ret; } -- cgit v1.1 From a7cd39e0c7805a93eaa4256370bcd48c506d46c1 Mon Sep 17 00:00:00 2001 From: Pieter Jansen van Vuuren Date: Fri, 25 Aug 2017 19:31:01 +0200 Subject: nfp: fix unchecked flow dissector use Previously flow dissectors were referenced without first checking that they are in use and correctly populated by TC. This patch fixes this by checking each flow dissector key before referencing them. Fixes: 5571e8c9f241 ("nfp: extend flower matching capabilities") Signed-off-by: Pieter Jansen van Vuuren Reviewed-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/flower/match.c | 133 +++++++++++---------- .../net/ethernet/netronome/nfp/flower/offload.c | 41 ++++--- 2 files changed, 93 insertions(+), 81 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c index 0e08404..b365110 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/match.c +++ b/drivers/net/ethernet/netronome/nfp/flower/match.c @@ -45,6 +45,7 @@ nfp_flower_compile_meta_tci(struct nfp_flower_meta_two *frame, struct flow_dissector_key_vlan *flow_vlan; u16 tmp_tci; + memset(frame, 0, sizeof(struct nfp_flower_meta_two)); /* Populate the metadata frame. */ frame->nfp_flow_key_layer = key_type; frame->mask_id = ~0; @@ -54,21 +55,20 @@ nfp_flower_compile_meta_tci(struct nfp_flower_meta_two *frame, return; } - flow_vlan = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_VLAN, - flow->key); - - /* Populate the tci field. */ - if (!flow_vlan->vlan_id) { - tmp_tci = 0; - } else { - tmp_tci = FIELD_PREP(NFP_FLOWER_MASK_VLAN_PRIO, - flow_vlan->vlan_priority) | - FIELD_PREP(NFP_FLOWER_MASK_VLAN_VID, - flow_vlan->vlan_id) | - NFP_FLOWER_MASK_VLAN_CFI; + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_VLAN)) { + flow_vlan = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_VLAN, + flow->key); + /* Populate the tci field. */ + if (flow_vlan->vlan_id) { + tmp_tci = FIELD_PREP(NFP_FLOWER_MASK_VLAN_PRIO, + flow_vlan->vlan_priority) | + FIELD_PREP(NFP_FLOWER_MASK_VLAN_VID, + flow_vlan->vlan_id) | + NFP_FLOWER_MASK_VLAN_CFI; + frame->tci = cpu_to_be16(tmp_tci); + } } - frame->tci = cpu_to_be16(tmp_tci); } static void @@ -99,17 +99,18 @@ nfp_flower_compile_mac(struct nfp_flower_mac_mpls *frame, bool mask_version) { struct fl_flow_key *target = mask_version ? flow->mask : flow->key; - struct flow_dissector_key_eth_addrs *flow_mac; - - flow_mac = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_ETH_ADDRS, - target); + struct flow_dissector_key_eth_addrs *addr; memset(frame, 0, sizeof(struct nfp_flower_mac_mpls)); - /* Populate mac frame. */ - ether_addr_copy(frame->mac_dst, &flow_mac->dst[0]); - ether_addr_copy(frame->mac_src, &flow_mac->src[0]); + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { + addr = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_ETH_ADDRS, + target); + /* Populate mac frame. */ + ether_addr_copy(frame->mac_dst, &addr->dst[0]); + ether_addr_copy(frame->mac_src, &addr->src[0]); + } if (mask_version) frame->mpls_lse = cpu_to_be32(~0); @@ -121,14 +122,17 @@ nfp_flower_compile_tport(struct nfp_flower_tp_ports *frame, bool mask_version) { struct fl_flow_key *target = mask_version ? flow->mask : flow->key; - struct flow_dissector_key_ports *flow_tp; + struct flow_dissector_key_ports *tp; - flow_tp = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_PORTS, - target); + memset(frame, 0, sizeof(struct nfp_flower_tp_ports)); - frame->port_src = flow_tp->src; - frame->port_dst = flow_tp->dst; + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_PORTS)) { + tp = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_PORTS, + target); + frame->port_src = tp->src; + frame->port_dst = tp->dst; + } } static void @@ -137,25 +141,27 @@ nfp_flower_compile_ipv4(struct nfp_flower_ipv4 *frame, bool mask_version) { struct fl_flow_key *target = mask_version ? flow->mask : flow->key; - struct flow_dissector_key_ipv4_addrs *flow_ipv4; - struct flow_dissector_key_basic *flow_basic; - - flow_ipv4 = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_IPV4_ADDRS, - target); + struct flow_dissector_key_ipv4_addrs *addr; + struct flow_dissector_key_basic *basic; - flow_basic = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_BASIC, - target); - - /* Populate IPv4 frame. */ - frame->reserved = 0; - frame->ipv4_src = flow_ipv4->src; - frame->ipv4_dst = flow_ipv4->dst; - frame->proto = flow_basic->ip_proto; /* Wildcard TOS/TTL for now. */ - frame->tos = 0; - frame->ttl = 0; + memset(frame, 0, sizeof(struct nfp_flower_ipv4)); + + if (dissector_uses_key(flow->dissector, + FLOW_DISSECTOR_KEY_IPV4_ADDRS)) { + addr = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_IPV4_ADDRS, + target); + frame->ipv4_src = addr->src; + frame->ipv4_dst = addr->dst; + } + + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_BASIC)) { + basic = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_BASIC, + target); + frame->proto = basic->ip_proto; + } } static void @@ -164,26 +170,27 @@ nfp_flower_compile_ipv6(struct nfp_flower_ipv6 *frame, bool mask_version) { struct fl_flow_key *target = mask_version ? flow->mask : flow->key; - struct flow_dissector_key_ipv6_addrs *flow_ipv6; - struct flow_dissector_key_basic *flow_basic; - - flow_ipv6 = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_IPV6_ADDRS, - target); - - flow_basic = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_BASIC, - target); + struct flow_dissector_key_ipv6_addrs *addr; + struct flow_dissector_key_basic *basic; - /* Populate IPv6 frame. */ - frame->reserved = 0; - frame->ipv6_src = flow_ipv6->src; - frame->ipv6_dst = flow_ipv6->dst; - frame->proto = flow_basic->ip_proto; /* Wildcard LABEL/TOS/TTL for now. */ - frame->ipv6_flow_label_exthdr = 0; - frame->tos = 0; - frame->ttl = 0; + memset(frame, 0, sizeof(struct nfp_flower_ipv6)); + + if (dissector_uses_key(flow->dissector, + FLOW_DISSECTOR_KEY_IPV6_ADDRS)) { + addr = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_IPV6_ADDRS, + target); + frame->ipv6_src = addr->src; + frame->ipv6_dst = addr->dst; + } + + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_BASIC)) { + basic = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_BASIC, + target); + frame->proto = basic->ip_proto; + } } int nfp_flower_compile_flow_match(struct tc_cls_flower_offload *flow, diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c index 4ad10bd..6c8ecc2 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/offload.c +++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c @@ -105,35 +105,40 @@ static int nfp_flower_calculate_key_layers(struct nfp_fl_key_ls *ret_key_ls, struct tc_cls_flower_offload *flow) { - struct flow_dissector_key_control *mask_enc_ctl; - struct flow_dissector_key_basic *mask_basic; - struct flow_dissector_key_basic *key_basic; + struct flow_dissector_key_basic *mask_basic = NULL; + struct flow_dissector_key_basic *key_basic = NULL; u32 key_layer_two; u8 key_layer; int key_size; - mask_enc_ctl = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_ENC_CONTROL, - flow->mask); + if (dissector_uses_key(flow->dissector, + FLOW_DISSECTOR_KEY_ENC_CONTROL)) { + struct flow_dissector_key_control *mask_enc_ctl = + skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_ENC_CONTROL, + flow->mask); + /* We are expecting a tunnel. For now we ignore offloading. */ + if (mask_enc_ctl->addr_type) + return -EOPNOTSUPP; + } + + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_BASIC)) { + mask_basic = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_BASIC, + flow->mask); - mask_basic = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_BASIC, - flow->mask); + key_basic = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_BASIC, + flow->key); + } - key_basic = skb_flow_dissector_target(flow->dissector, - FLOW_DISSECTOR_KEY_BASIC, - flow->key); key_layer_two = 0; key_layer = NFP_FLOWER_LAYER_PORT | NFP_FLOWER_LAYER_MAC; key_size = sizeof(struct nfp_flower_meta_one) + sizeof(struct nfp_flower_in_port) + sizeof(struct nfp_flower_mac_mpls); - /* We are expecting a tunnel. For now we ignore offloading. */ - if (mask_enc_ctl->addr_type) - return -EOPNOTSUPP; - - if (mask_basic->n_proto) { + if (mask_basic && mask_basic->n_proto) { /* Ethernet type is present in the key. */ switch (key_basic->n_proto) { case cpu_to_be16(ETH_P_IP): @@ -166,7 +171,7 @@ nfp_flower_calculate_key_layers(struct nfp_fl_key_ls *ret_key_ls, } } - if (mask_basic->ip_proto) { + if (mask_basic && mask_basic->ip_proto) { /* Ethernet type is present in the key. */ switch (key_basic->ip_proto) { case IPPROTO_TCP: -- cgit v1.1 From 74af5975108f54f9443952c0b4d52487031a7569 Mon Sep 17 00:00:00 2001 From: Pieter Jansen van Vuuren Date: Fri, 25 Aug 2017 19:31:02 +0200 Subject: nfp: fix supported key layers calculation Previously when calculating the supported key layers MPLS, IPv4/6 TTL and TOS were not considered. This patch checks that the TTL and TOS fields are masked out before offloading. Additionally this patch checks that MPLS packets are correctly handled, by not offloading them. Fixes: af9d842c1354 ("nfp: extend flower add flow offload") Signed-off-by: Pieter Jansen van Vuuren Reviewed-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/flower/offload.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c index 6c8ecc2..74a96d6 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/offload.c +++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c @@ -107,6 +107,7 @@ nfp_flower_calculate_key_layers(struct nfp_fl_key_ls *ret_key_ls, { struct flow_dissector_key_basic *mask_basic = NULL; struct flow_dissector_key_basic *key_basic = NULL; + struct flow_dissector_key_ip *mask_ip = NULL; u32 key_layer_two; u8 key_layer; int key_size; @@ -132,6 +133,11 @@ nfp_flower_calculate_key_layers(struct nfp_fl_key_ls *ret_key_ls, flow->key); } + if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_IP)) + mask_ip = skb_flow_dissector_target(flow->dissector, + FLOW_DISSECTOR_KEY_IP, + flow->mask); + key_layer_two = 0; key_layer = NFP_FLOWER_LAYER_PORT | NFP_FLOWER_LAYER_MAC; key_size = sizeof(struct nfp_flower_meta_one) + @@ -142,11 +148,19 @@ nfp_flower_calculate_key_layers(struct nfp_fl_key_ls *ret_key_ls, /* Ethernet type is present in the key. */ switch (key_basic->n_proto) { case cpu_to_be16(ETH_P_IP): + if (mask_ip && mask_ip->tos) + return -EOPNOTSUPP; + if (mask_ip && mask_ip->ttl) + return -EOPNOTSUPP; key_layer |= NFP_FLOWER_LAYER_IPV4; key_size += sizeof(struct nfp_flower_ipv4); break; case cpu_to_be16(ETH_P_IPV6): + if (mask_ip && mask_ip->tos) + return -EOPNOTSUPP; + if (mask_ip && mask_ip->ttl) + return -EOPNOTSUPP; key_layer |= NFP_FLOWER_LAYER_IPV6; key_size += sizeof(struct nfp_flower_ipv6); break; @@ -157,6 +171,11 @@ nfp_flower_calculate_key_layers(struct nfp_fl_key_ls *ret_key_ls, case cpu_to_be16(ETH_P_ARP): return -EOPNOTSUPP; + /* Currently we do not offload MPLS. */ + case cpu_to_be16(ETH_P_MPLS_UC): + case cpu_to_be16(ETH_P_MPLS_MC): + return -EOPNOTSUPP; + /* Will be included in layer 2. */ case cpu_to_be16(ETH_P_8021Q): break; -- cgit v1.1 From 6afd33e4384060e692705912337b184c1e159aff Mon Sep 17 00:00:00 2001 From: Pieter Jansen van Vuuren Date: Fri, 25 Aug 2017 19:31:03 +0200 Subject: nfp: remove incorrect mask check for vlan matching Previously the vlan tci field was incorrectly exact matched. This patch fixes this by using the flow dissector to populate the vlan tci field. Fixes: 5571e8c9f241 ("nfp: extend flower matching capabilities") Signed-off-by: Pieter Jansen van Vuuren Reviewed-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/flower/match.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c index b365110..d25b503 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/match.c +++ b/drivers/net/ethernet/netronome/nfp/flower/match.c @@ -42,6 +42,7 @@ nfp_flower_compile_meta_tci(struct nfp_flower_meta_two *frame, struct tc_cls_flower_offload *flow, u8 key_type, bool mask_version) { + struct fl_flow_key *target = mask_version ? flow->mask : flow->key; struct flow_dissector_key_vlan *flow_vlan; u16 tmp_tci; @@ -50,15 +51,10 @@ nfp_flower_compile_meta_tci(struct nfp_flower_meta_two *frame, frame->nfp_flow_key_layer = key_type; frame->mask_id = ~0; - if (mask_version) { - frame->tci = cpu_to_be16(~0); - return; - } - if (dissector_uses_key(flow->dissector, FLOW_DISSECTOR_KEY_VLAN)) { flow_vlan = skb_flow_dissector_target(flow->dissector, FLOW_DISSECTOR_KEY_VLAN, - flow->key); + target); /* Populate the tci field. */ if (flow_vlan->vlan_id) { tmp_tci = FIELD_PREP(NFP_FLOWER_MASK_VLAN_PRIO, -- cgit v1.1 From ad4540cc5aa3dccb8e1e12458d57f8c40fae5a1c Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Fri, 25 Aug 2017 21:12:17 +0200 Subject: net: stmmac: sun8i: Remove the compatibles Since the bindings have been controversial, and we follow the DT stable ABI rule, we shouldn't let a driver with a DT binding that might change slip through in a stable release. Remove the compatibles to make sure the driver will not probe and no-one will start using the binding currently implemented. This commit will obviously need to be reverted in due time. Signed-off-by: Maxime Ripard Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c index fffd6d5..39c2122 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c @@ -979,14 +979,6 @@ static int sun8i_dwmac_probe(struct platform_device *pdev) } static const struct of_device_id sun8i_dwmac_match[] = { - { .compatible = "allwinner,sun8i-h3-emac", - .data = &emac_variant_h3 }, - { .compatible = "allwinner,sun8i-v3s-emac", - .data = &emac_variant_v3s }, - { .compatible = "allwinner,sun8i-a83t-emac", - .data = &emac_variant_a83t }, - { .compatible = "allwinner,sun50i-a64-emac", - .data = &emac_variant_a64 }, { } }; MODULE_DEVICE_TABLE(of, sun8i_dwmac_match); -- cgit v1.1 From 0f3086868e8889a823a6e0f3d299102aa895d947 Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Fri, 25 Aug 2017 22:48:48 +0200 Subject: cxgb4: Fix stack out-of-bounds read due to wrong size to t4_record_mbox() Passing commands for logging to t4_record_mbox() with size MBOX_LEN, when the actual command size is actually smaller, causes out-of-bounds stack accesses in t4_record_mbox() while copying command words here: for (i = 0; i < size / 8; i++) entry->cmd[i] = be64_to_cpu(cmd[i]); Up to 48 bytes from the stack are then leaked to debugfs. This happens whenever we send (and log) commands described by structs fw_sched_cmd (32 bytes leaked), fw_vi_rxmode_cmd (48), fw_hello_cmd (48), fw_bye_cmd (48), fw_initialize_cmd (48), fw_reset_cmd (48), fw_pfvf_cmd (32), fw_eq_eth_cmd (16), fw_eq_ctrl_cmd (32), fw_eq_ofld_cmd (32), fw_acl_mac_cmd(16), fw_rss_glb_config_cmd(32), fw_rss_vi_config_cmd(32), fw_devlog_cmd(32), fw_vi_enable_cmd(48), fw_port_cmd(32), fw_sched_cmd(32), fw_devlog_cmd(32). The cxgb4vf driver got this right instead. When we call t4_record_mbox() to log a command reply, a MBOX_LEN size can be used though, as get_mbox_rpl() will fill cmd_rpl up completely. Fixes: 7f080c3f2ff0 ("cxgb4: Add support to enable logging of firmware mailbox commands") Signed-off-by: Stefano Brivio Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index 82bf7aa..0293b41 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -369,12 +369,12 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd, list_del(&entry.list); spin_unlock(&adap->mbox_lock); ret = (v == MBOX_OWNER_FW) ? -EBUSY : -ETIMEDOUT; - t4_record_mbox(adap, cmd, MBOX_LEN, access, ret); + t4_record_mbox(adap, cmd, size, access, ret); return ret; } /* Copy in the new mailbox command and send it on its way ... */ - t4_record_mbox(adap, cmd, MBOX_LEN, access, 0); + t4_record_mbox(adap, cmd, size, access, 0); for (i = 0; i < size; i += 8) t4_write_reg64(adap, data_reg + i, be64_to_cpu(*p++)); @@ -426,7 +426,7 @@ int t4_wr_mbox_meat_timeout(struct adapter *adap, int mbox, const void *cmd, } ret = (pcie_fw & PCIE_FW_ERR_F) ? -ENXIO : -ETIMEDOUT; - t4_record_mbox(adap, cmd, MBOX_LEN, access, ret); + t4_record_mbox(adap, cmd, size, access, ret); dev_err(adap->pdev_dev, "command %#x in mailbox %d timed out\n", *(const u8 *)cmd, mbox); t4_report_fw_error(adap); -- cgit v1.1 From 4e587ea71bf924f7dac621f1351653bd41e446cb Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 25 Aug 2017 15:03:10 -0700 Subject: ipv6: fix sparse warning on rt6i_node Commit c5cff8561d2d adds rcu grace period before freeing fib6_node. This generates a new sparse warning on rt->rt6i_node related code: net/ipv6/route.c:1394:30: error: incompatible types in comparison expression (different address spaces) ./include/net/ip6_fib.h:187:14: error: incompatible types in comparison expression (different address spaces) This commit adds "__rcu" tag for rt6i_node and makes sure corresponding rcu API is used for it. After this fix, sparse no longer generates the above warning. Fixes: c5cff8561d2d ("ipv6: add rcu grace period before freeing fib6_node") Signed-off-by: Wei Wang Acked-by: Eric Dumazet Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 2 +- net/ipv6/addrconf.c | 2 +- net/ipv6/ip6_fib.c | 11 +++++++---- net/ipv6/route.c | 3 ++- 4 files changed, 11 insertions(+), 7 deletions(-) diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index e9c59db..af509f8 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -105,7 +105,7 @@ struct rt6_info { * the same cache line. */ struct fib6_table *rt6i_table; - struct fib6_node *rt6i_node; + struct fib6_node __rcu *rt6i_node; struct in6_addr rt6i_gateway; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 3c46e95..936e9ab 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5556,7 +5556,7 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp) * our DAD process, so we don't need * to do it again */ - if (!(ifp->rt->rt6i_node)) + if (!rcu_access_pointer(ifp->rt->rt6i_node)) ip6_ins_rt(ifp->rt); if (ifp->idev->cnf.forwarding) addrconf_join_anycast(ifp); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index a5ebf86..10b4b1f 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -889,7 +889,7 @@ add: rt->dst.rt6_next = iter; *ins = rt; - rt->rt6i_node = fn; + rcu_assign_pointer(rt->rt6i_node, fn); atomic_inc(&rt->rt6i_ref); if (!info->skip_notify) inet6_rt_notify(RTM_NEWROUTE, rt, info, nlflags); @@ -915,7 +915,7 @@ add: return err; *ins = rt; - rt->rt6i_node = fn; + rcu_assign_pointer(rt->rt6i_node, fn); rt->dst.rt6_next = iter->dst.rt6_next; atomic_inc(&rt->rt6i_ref); if (!info->skip_notify) @@ -1480,8 +1480,9 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, int fib6_del(struct rt6_info *rt, struct nl_info *info) { + struct fib6_node *fn = rcu_dereference_protected(rt->rt6i_node, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); struct net *net = info->nl_net; - struct fib6_node *fn = rt->rt6i_node; struct rt6_info **rtp; #if RT6_DEBUG >= 2 @@ -1670,7 +1671,9 @@ static int fib6_clean_node(struct fib6_walker *w) if (res) { #if RT6_DEBUG >= 2 pr_debug("%s: del failed: rt=%p@%p err=%d\n", - __func__, rt, rt->rt6i_node, res); + __func__, rt, + rcu_access_pointer(rt->rt6i_node), + res); #endif continue; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 48c8c92..86fb241 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1383,7 +1383,8 @@ static void rt6_do_update_pmtu(struct rt6_info *rt, u32 mtu) static bool rt6_cache_allowed_for_pmtu(const struct rt6_info *rt) { return !(rt->rt6i_flags & RTF_CACHE) && - (rt->rt6i_flags & RTF_PCPU || rt->rt6i_node); + (rt->rt6i_flags & RTF_PCPU || + rcu_access_pointer(rt->rt6i_node)); } static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk, -- cgit v1.1 From 1e2ea8ad37be25a7cdcc974945935829d534d5d3 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 26 Aug 2017 20:10:10 +0800 Subject: ipv6: set dst.obsolete when a cached route has expired Now it doesn't check for the cached route expiration in ipv6's dst_ops->check(), because it trusts dst_gc that would clean the cached route up when it's expired. The problem is in dst_gc, it would clean the cached route only when it's refcount is 1. If some other module (like xfrm) keeps holding it and the module only release it when dst_ops->check() fails. But without checking for the cached route expiration, .check() may always return true. Meanwhile, without releasing the cached route, dst_gc couldn't del it. It will cause this cached route never to expire. This patch is to set dst.obsolete with DST_OBSOLETE_KILL in .gc when it's expired, and check obsolete != DST_OBSOLETE_FORCE_CHK in .check. Note that this is even needed when ipv6 dst_gc timer is removed one day. It would set dst.obsolete in .redirect and .update_pmtu instead, and check for cached route expiration when getting it, just like what ipv4 route does. Reported-by: Jianlin Shi Signed-off-by: Xin Long Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- net/ipv6/ip6_fib.c | 4 +++- net/ipv6/route.c | 3 ++- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 10b4b1f..e1c85bb 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1795,8 +1795,10 @@ static int fib6_age(struct rt6_info *rt, void *arg) } gc_args->more++; } else if (rt->rt6i_flags & RTF_CACHE) { + if (time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) + rt->dst.obsolete = DST_OBSOLETE_KILL; if (atomic_read(&rt->dst.__refcnt) == 1 && - time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) { + rt->dst.obsolete == DST_OBSOLETE_KILL) { RT6_TRACE("aging clone %p\n", rt); return -1; } else if (rt->rt6i_flags & RTF_GATEWAY) { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 86fb241..2d0e779 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -440,7 +440,8 @@ static bool rt6_check_expired(const struct rt6_info *rt) if (time_after(jiffies, rt->dst.expires)) return true; } else if (rt->dst.from) { - return rt6_check_expired((struct rt6_info *) rt->dst.from); + return rt->dst.obsolete != DST_OBSOLETE_FORCE_CHK || + rt6_check_expired((struct rt6_info *)rt->dst.from); } return false; } -- cgit v1.1 From 79de3cbe9a974e03a02b71da80da9ee0eb15a2d0 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Wed, 23 Aug 2017 22:37:00 +0200 Subject: fs/select: Fix memory corruption in compat_get_fd_set() Commit 464d62421cb8 ("select: switch compat_{get,put}_fd_set() to compat_{get,put}_bitmap()") changed the calculation on how many bytes need to be zeroed when userspace handed over a NULL pointer for a fdset array in the select syscall. The calculation was changed in compat_get_fd_set() wrongly from memset(fdset, 0, ((nr + 1) & ~1)*sizeof(compat_ulong_t)); to memset(fdset, 0, ALIGN(nr, BITS_PER_LONG)); The ALIGN(nr, BITS_PER_LONG) calculates the number of _bits_ which need to be zeroed in the target fdset array (rounded up to the next full bits for an unsigned long). But the memset() call expects the number of _bytes_ to be zeroed. This leads to clearing more memory than wanted (on the stack area or even at kmalloc()ed memory areas) and to random kernel crashes as we have seen them on the parisc platform. The correct change should have been memset(fdset, 0, (ALIGN(nr, BITS_PER_LONG) / BITS_PER_LONG) * BYTES_PER_LONG); which is the same as can be archieved with a call to zero_fd_set(nr, fdset). Fixes: 464d62421cb8 ("select: switch compat_{get,put}_fd_set() to compat_{get,put}_bitmap()" Acked-by:: Al Viro Signed-off-by: Helge Deller Signed-off-by: Linus Torvalds --- fs/select.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/fs/select.c b/fs/select.c index 9d5f15e..c6362e3 100644 --- a/fs/select.c +++ b/fs/select.c @@ -1164,11 +1164,7 @@ int compat_get_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, if (ufdset) { return compat_get_bitmap(fdset, ufdset, nr); } else { - /* Tricky, must clear full unsigned long in the - * kernel fdset at the end, ALIGN makes sure that - * actually happens. - */ - memset(fdset, 0, ALIGN(nr, BITS_PER_LONG)); + zero_fd_set(nr, fdset); return 0; } } -- cgit v1.1 From e8206d2baa41a4c7cf4590929f8819f8309b564d Mon Sep 17 00:00:00 2001 From: Alexey Brodkin Date: Mon, 28 Aug 2017 15:03:58 -0700 Subject: ARCv2: SMP: Mask only private-per-core IRQ lines on boot at core intc Recent commit a8ec3ee861b6 "arc: Mask individual IRQ lines during core INTC init" breaks interrupt handling on ARCv2 SMP systems. That commit masked all interrupts at onset, as some controllers on some boards (customer as well as internal), would assert interrutps early before any handlers were installed. For SMP systems, the masking was done at each cpu's core-intc. Later, when the IRQ was actually requested, it was unmasked, but only on the requesting cpu. For "common" interrupts, which were wired up from the 2nd level IDU intc, this was as issue as they needed to be enabled on ALL the cpus (given that IDU IRQs are by default served Round Robin across cpus) So fix that by NOT masking "common" interrupts at core-intc, but instead at the 2nd level IDU intc (latter already being done in idu_of_init()) Fixes: a8ec3ee861b6 ("arc: Mask individual IRQ lines during core INTC init") Signed-off-by: Alexey Brodkin [vgupta: reworked changelog, removed the extraneous idu_irq_mask_raw()] Signed-off-by: Vineet Gupta Signed-off-by: Linus Torvalds --- arch/arc/kernel/intc-arcv2.c | 11 +++++++++-- arch/arc/kernel/intc-compact.c | 2 +- 2 files changed, 10 insertions(+), 3 deletions(-) diff --git a/arch/arc/kernel/intc-arcv2.c b/arch/arc/kernel/intc-arcv2.c index cf90714..067ea36 100644 --- a/arch/arc/kernel/intc-arcv2.c +++ b/arch/arc/kernel/intc-arcv2.c @@ -75,13 +75,20 @@ void arc_init_IRQ(void) * Set a default priority for all available interrupts to prevent * switching of register banks if Fast IRQ and multiple register banks * are supported by CPU. - * Also disable all IRQ lines so faulty external hardware won't + * Also disable private-per-core IRQ lines so faulty external HW won't * trigger interrupt that kernel is not ready to handle. */ for (i = NR_EXCEPTIONS; i < irq_bcr.irqs + NR_EXCEPTIONS; i++) { write_aux_reg(AUX_IRQ_SELECT, i); write_aux_reg(AUX_IRQ_PRIORITY, ARCV2_IRQ_DEF_PRIO); - write_aux_reg(AUX_IRQ_ENABLE, 0); + + /* + * Only mask cpu private IRQs here. + * "common" interrupts are masked at IDU, otherwise it would + * need to be unmasked at each cpu, with IPIs + */ + if (i < FIRST_EXT_IRQ) + write_aux_reg(AUX_IRQ_ENABLE, 0); } /* setup status32, don't enable intr yet as kernel doesn't want */ diff --git a/arch/arc/kernel/intc-compact.c b/arch/arc/kernel/intc-compact.c index cef3880..47b421f 100644 --- a/arch/arc/kernel/intc-compact.c +++ b/arch/arc/kernel/intc-compact.c @@ -27,7 +27,7 @@ */ void arc_init_IRQ(void) { - int level_mask = 0, i; + unsigned int level_mask = 0, i; /* Is timer high priority Interrupt (Level2 in ARCompact jargon) */ level_mask |= IS_ENABLED(CONFIG_ARC_COMPACT_IRQ_LEVELS) << TIMER0_IRQ; -- cgit v1.1 From b339752d054fb32863418452dff350a1086885b1 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 28 Aug 2017 14:51:27 -0700 Subject: cpumask: fix spurious cpumask_of_node() on non-NUMA multi-node configs When !NUMA, cpumask_of_node(@node) equals cpu_online_mask regardless of @node. The assumption seems that if !NUMA, there shouldn't be more than one node and thus reporting cpu_online_mask regardless of @node is correct. However, that assumption was broken years ago to support DISCONTIGMEM and whether a system has multiple nodes or not is separately controlled by NEED_MULTIPLE_NODES. This means that, on a system with !NUMA && NEED_MULTIPLE_NODES, cpumask_of_node() will report cpu_online_mask for all possible nodes, indicating that the CPUs are associated with multiple nodes which is an impossible configuration. This bug has been around forever but doesn't look like it has caused any noticeable symptoms. However, it triggers a WARN recently added to workqueue to verify NUMA affinity configuration. Fix it by reporting empty cpumask on non-zero nodes if !NUMA. Signed-off-by: Tejun Heo Reported-and-tested-by: Geert Uytterhoeven Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds --- include/asm-generic/topology.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/include/asm-generic/topology.h b/include/asm-generic/topology.h index fc824e2..5d2add1 100644 --- a/include/asm-generic/topology.h +++ b/include/asm-generic/topology.h @@ -48,7 +48,11 @@ #define parent_node(node) ((void)(node),0) #endif #ifndef cpumask_of_node -#define cpumask_of_node(node) ((void)node, cpu_online_mask) + #ifdef CONFIG_NEED_MULTIPLE_NODES + #define cpumask_of_node(node) ((node) == 0 ? cpu_online_mask : cpu_none_mask) + #else + #define cpumask_of_node(node) ((void)node, cpu_online_mask) + #endif #endif #ifndef pcibus_to_node #define pcibus_to_node(bus) ((void)(bus), -1) -- cgit v1.1 From ef9a5a62c63456cbba1beef2fc1372fce105fbbc Mon Sep 17 00:00:00 2001 From: Roopa Prabhu Date: Sat, 26 Aug 2017 21:13:48 -0700 Subject: bridge: check for null fdb->dst before notifying switchdev drivers current switchdev drivers dont seem to support offloading fdb entries pointing to the bridge device which have fdb->dst not set to any port. This patch adds a NULL fdb->dst check in the switchdev notifier code. This patch fixes the below NULL ptr dereference: $bridge fdb add 00:02:00:00:00:33 dev br0 self [ 69.953374] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 [ 69.954044] IP: br_switchdev_fdb_notify+0x29/0x80 [ 69.954044] PGD 66527067 [ 69.954044] P4D 66527067 [ 69.954044] PUD 7899c067 [ 69.954044] PMD 0 [ 69.954044] [ 69.954044] Oops: 0000 [#1] SMP [ 69.954044] Modules linked in: [ 69.954044] CPU: 1 PID: 3074 Comm: bridge Not tainted 4.13.0-rc6+ #1 [ 69.954044] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5.1-0-g8936dbb-20141113_115728-nilsson.home.kraxel.org 04/01/2014 [ 69.954044] task: ffff88007b827140 task.stack: ffffc90001564000 [ 69.954044] RIP: 0010:br_switchdev_fdb_notify+0x29/0x80 [ 69.954044] RSP: 0018:ffffc90001567918 EFLAGS: 00010246 [ 69.954044] RAX: 0000000000000000 RBX: ffff8800795e0880 RCX: 00000000000000c0 [ 69.954044] RDX: ffffc90001567920 RSI: 000000000000001c RDI: ffff8800795d0600 [ 69.954044] RBP: ffffc90001567938 R08: ffff8800795d0600 R09: 0000000000000000 [ 69.954044] R10: ffffc90001567a88 R11: ffff88007b849400 R12: ffff8800795e0880 [ 69.954044] R13: ffff8800795d0600 R14: ffffffff81ef8880 R15: 000000000000001c [ 69.954044] FS: 00007f93d3085700(0000) GS:ffff88007fd00000(0000) knlGS:0000000000000000 [ 69.954044] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 69.954044] CR2: 0000000000000008 CR3: 0000000066551000 CR4: 00000000000006e0 [ 69.954044] Call Trace: [ 69.954044] fdb_notify+0x3f/0xf0 [ 69.954044] __br_fdb_add.isra.12+0x1a7/0x370 [ 69.954044] br_fdb_add+0x178/0x280 [ 69.954044] rtnl_fdb_add+0x10a/0x200 [ 69.954044] rtnetlink_rcv_msg+0x1b4/0x240 [ 69.954044] ? skb_free_head+0x21/0x40 [ 69.954044] ? rtnl_calcit.isra.18+0xf0/0xf0 [ 69.954044] netlink_rcv_skb+0xed/0x120 [ 69.954044] rtnetlink_rcv+0x15/0x20 [ 69.954044] netlink_unicast+0x180/0x200 [ 69.954044] netlink_sendmsg+0x291/0x370 [ 69.954044] ___sys_sendmsg+0x180/0x2e0 [ 69.954044] ? filemap_map_pages+0x2db/0x370 [ 69.954044] ? do_wp_page+0x11d/0x420 [ 69.954044] ? __handle_mm_fault+0x794/0xd80 [ 69.954044] ? vma_link+0xcb/0xd0 [ 69.954044] __sys_sendmsg+0x4c/0x90 [ 69.954044] SyS_sendmsg+0x12/0x20 [ 69.954044] do_syscall_64+0x63/0xe0 [ 69.954044] entry_SYSCALL64_slow_path+0x25/0x25 [ 69.954044] RIP: 0033:0x7f93d2bad690 [ 69.954044] RSP: 002b:00007ffc7217a638 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 69.954044] RAX: ffffffffffffffda RBX: 00007ffc72182eac RCX: 00007f93d2bad690 [ 69.954044] RDX: 0000000000000000 RSI: 00007ffc7217a670 RDI: 0000000000000003 [ 69.954044] RBP: 0000000059a1f7f8 R08: 0000000000000006 R09: 000000000000000a [ 69.954044] R10: 00007ffc7217a400 R11: 0000000000000246 R12: 00007ffc7217a670 [ 69.954044] R13: 00007ffc72182a98 R14: 00000000006114c0 R15: 00007ffc72182aa0 [ 69.954044] Code: 1f 00 66 66 66 66 90 55 48 89 e5 48 83 ec 20 f6 47 20 04 74 0a 83 fe 1c 74 09 83 fe 1d 74 2c c9 66 90 c3 48 8b 47 10 48 8d 55 e8 <48> 8b 70 08 0f b7 47 1e 48 83 c7 18 48 89 7d f0 bf 03 00 00 00 [ 69.954044] RIP: br_switchdev_fdb_notify+0x29/0x80 RSP: ffffc90001567918 [ 69.954044] CR2: 0000000000000008 [ 69.954044] ---[ end trace 03e9eec4a82c238b ]--- Fixes: 6b26b51b1d13 ("net: bridge: Add support for notifying devices about FDB add/del") Signed-off-by: Roopa Prabhu Signed-off-by: David S. Miller --- net/bridge/br_switchdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index 181a44d..f6b1c7d 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -115,7 +115,7 @@ br_switchdev_fdb_call_notifiers(bool adding, const unsigned char *mac, void br_switchdev_fdb_notify(const struct net_bridge_fdb_entry *fdb, int type) { - if (!fdb->added_by_user) + if (!fdb->added_by_user || !fdb->dst) return; switch (type) { -- cgit v1.1 From 9c3a815f471a84811cf8021cf64aae3b8081dfde Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 28 Aug 2017 16:45:40 -0700 Subject: page waitqueue: always add new entries at the end Commit 3510ca20ece0 ("Minor page waitqueue cleanups") made the page queue code always add new waiters to the back of the queue, which helps upcoming patches to batch the wakeups for some horrid loads where the wait queues grow to thousands of entries. However, I forgot about the nasrt add_page_wait_queue() special case code that is only used by the cachefiles code. That one still continued to add the new wait queue entries at the beginning of the list. Fix it, because any sane batched wakeup will require that we don't suddenly start getting new entries at the beginning of the list that we already handled in a previous batch. [ The current code always does the whole list while holding the lock, so wait queue ordering doesn't matter for correctness, but even then it's better to add later entries at the end from a fairness standpoint ] Signed-off-by: Linus Torvalds --- mm/filemap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/filemap.c b/mm/filemap.c index 0b41c8c..65b4b6e 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1041,7 +1041,7 @@ void add_page_wait_queue(struct page *page, wait_queue_entry_t *waiter) unsigned long flags; spin_lock_irqsave(&q->lock, flags); - __add_wait_queue(q, waiter); + __add_wait_queue_entry_tail(q, waiter); SetPageWaiters(page); spin_unlock_irqrestore(&q->lock, flags); } -- cgit v1.1 From c7848399ec7612c2fa4dc0c2eeb9e0b89d00afef Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 28 Aug 2017 17:10:51 -0700 Subject: net: dsa: Don't dereference dst->cpu_dp->netdev If we do not have a master network device attached dst->cpu_dp will be NULL and accessing cpu_dp->netdev will create a trace similar to the one below. The correct check is on dst->cpu_dp period. [ 1.004650] DSA: switch 0 0 parsed [ 1.008078] Unable to handle kernel NULL pointer dereference at virtual address 00000010 [ 1.016195] pgd = c0003000 [ 1.018918] [00000010] *pgd=80000000004003, *pmd=00000000 [ 1.024349] Internal error: Oops: 206 [#1] SMP ARM [ 1.029157] Modules linked in: [ 1.032228] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.13.0-rc6-00071-g45b45afab9bd-dirty #7 [ 1.040772] Hardware name: Broadcom STB (Flattened Device Tree) [ 1.046704] task: ee08f840 task.stack: ee090000 [ 1.051258] PC is at dsa_register_switch+0x5e0/0x9dc [ 1.056234] LR is at dsa_register_switch+0x5d0/0x9dc [ 1.061211] pc : [] lr : [] psr: 60000213 [ 1.067491] sp : ee091d88 ip : 00000000 fp : 0000000c [ 1.072728] r10: 00000000 r9 : 00000001 r8 : ee208010 [ 1.077965] r7 : ee2b57b0 r6 : ee2b5780 r5 : 00000000 r4 : ee208e0c [ 1.084506] r3 : 00000000 r2 : 00040d00 r1 : 2d1b2000 r0 : 00000016 [ 1.091050] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user [ 1.098199] Control: 32c5387d Table: 00003000 DAC: fffffffd [ 1.103957] Process swapper/0 (pid: 1, stack limit = 0xee090210) Reported-by: Dan Carpenter Fixes: 6d3c8c0dd88a ("net: dsa: Remove master_netdev and use dst->cpu_dp->netdev") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/dsa2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index c442051..20bc9c5 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -577,7 +577,7 @@ static int dsa_dst_parse(struct dsa_switch_tree *dst) return err; } - if (!dst->cpu_dp->netdev) { + if (!dst->cpu_dp) { pr_warn("Tree has no master device\n"); return -EINVAL; } -- cgit v1.1 From e8f11db956aa09c1618051a7aaf367d6810d8d8c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 29 Aug 2017 14:42:06 +0200 Subject: libata: check for trusted computing in IDENTIFY DEVICE data ATA-8 and later mirrors the TRUSTED COMPUTING SUPPORTED bit in word 48 of the IDENTIFY DEVICE data. Check this before issuing a READ LOG PAGE command to avoid issues with buggy devices. The only downside is that we can't support Security Send / Receive for a device with an older revision due to the conflicting use of this field in earlier specifications. tj: The reason we need this is because some devices which don't support READ LOG PAGE lock up after getting issued that command. Signed-off-by: Christoph Hellwig Tested-by: David Ahern Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 3 +++ include/linux/ata.h | 10 +++++++++- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 697f5f8..ca57b03 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2413,6 +2413,9 @@ static void ata_dev_config_trusted(struct ata_device *dev) u64 trusted_cap; unsigned int err; + if (!ata_id_has_trusted(dev->id)) + return; + if (!ata_identify_page_supported(dev, ATA_LOG_SECURITY)) { ata_dev_warn(dev, "Security Log not supported\n"); diff --git a/include/linux/ata.h b/include/linux/ata.h index e65ae4b..c7a3538 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -60,7 +60,8 @@ enum { ATA_ID_FW_REV = 23, ATA_ID_PROD = 27, ATA_ID_MAX_MULTSECT = 47, - ATA_ID_DWORD_IO = 48, + ATA_ID_DWORD_IO = 48, /* before ATA-8 */ + ATA_ID_TRUSTED = 48, /* ATA-8 and later */ ATA_ID_CAPABILITY = 49, ATA_ID_OLD_PIO_MODES = 51, ATA_ID_OLD_DMA_MODES = 52, @@ -889,6 +890,13 @@ static inline bool ata_id_has_dword_io(const u16 *id) return id[ATA_ID_DWORD_IO] & (1 << 0); } +static inline bool ata_id_has_trusted(const u16 *id) +{ + if (ata_id_major_version(id) <= 7) + return false; + return id[ATA_ID_TRUSTED] & (1 << 0); +} + static inline bool ata_id_has_unload(const u16 *id) { if (ata_id_major_version(id) >= 7 && -- cgit v1.1 From 2aca392398384f8a2f339f4c661a1699f4f2e2eb Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 29 Aug 2017 08:36:58 -0700 Subject: Revert "libata: quirk read log on no-name M.2 SSD" This reverts commit 35f0b6a779b8b7a98faefd7c1c660b4dac9a5c26. We now conditionalize issuing of READ LOG PAGE on the TRUSTED COMPUTING SUPPORTED bit in the identity data and this shouldn't be necessary. Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 4 ---- include/linux/libata.h | 1 - 2 files changed, 5 deletions(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index ca57b03..1945a8e 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2079,8 +2079,6 @@ unsigned int ata_read_log_page(struct ata_device *dev, u8 log, */ if (ap_flags & ATA_FLAG_NO_LOG_PAGE) return AC_ERR_DEV; - if (dev->horkage & ATA_HORKAGE_NO_LOG_PAGE) - return AC_ERR_DEV; retry: ata_tf_init(dev, &tf); @@ -4583,8 +4581,6 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { { "WDC WD3000JD-*", NULL, ATA_HORKAGE_WD_BROKEN_LPM }, { "WDC WD3200JD-*", NULL, ATA_HORKAGE_WD_BROKEN_LPM }, - - { "M.2 (S42) 3ME3", NULL, ATA_HORKAGE_NO_LOG_PAGE }, /* End Marker */ { } }; diff --git a/include/linux/libata.h b/include/linux/libata.h index 9e927ae..931c32f 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -438,7 +438,6 @@ enum { ATA_HORKAGE_NO_DMA_LOG = (1 << 23), /* don't use DMA for log read */ ATA_HORKAGE_NOTRIM = (1 << 24), /* don't use TRIM */ ATA_HORKAGE_MAX_SEC_1024 = (1 << 25), /* Limit max sects to 1024 */ - ATA_HORKAGE_NO_LOG_PAGE = (1 << 26), /* Doesn't like Get Log Page */ /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- cgit v1.1 From 785373b4c38719f4af6775845df6be1dfaea120f Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 29 Aug 2017 09:11:06 -0700 Subject: Revert "rmap: do not call mmu_notifier_invalidate_page() under ptl" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit aac2fea94f7a3df8ad1eeb477eb2643f81fd5393. It turns out that that patch was complete and utter garbage, and broke KVM, resulting in odd oopses. Quoting Andrea Arcangeli: "The aforementioned commit has 3 bugs. 1) mmu_notifier_invalidate_range cannot be used in replacement of mmu_notifier_invalidate_range_start/end. For KVM mmu_notifier_invalidate_range is a noop and rightfully so. A MMU notifier implementation has to implement either ->invalidate_range method or the invalidate_range_start/end methods, not both. And if you implement invalidate_range_start/end like KVM is forced to do, calling mmu_notifier_invalidate_range in common code is a noop for KVM. For those MMU notifiers that can get away only implementing ->invalidate_range, the ->invalidate_range is implicitly called by mmu_notifier_invalidate_range_end(). And only those secondary MMUs that share the same pagetable with the primary MMU (like AMD iommuv2) can get away only implementing ->invalidate_range. So all cases (THP on/off) are broken right now. To fix this is enough to replace mmu_notifier_invalidate_range with mmu_notifier_invalidate_range_start;mmu_notifier_invalidate_range_end. Either that or call multiple mmu_notifier_invalidate_page like before. 2) address + (1UL << compound_order(page) is buggy, it should be PAGE_SIZE << compound_order(page), it's bytes not pages, 2M not 512. 3) The whole invalidate_range thing was an attempt to call a single invalidate while walking multiple 4k ptes that maps the same THP (after a pmd virtual split without physical compound page THP split). It's unclear if the rmap_walk will always provide an address that is 2M aligned as parameter to try_to_unmap_one, in presence of THP. I think it needs also an address &= (PAGE_SIZE << compound_order(page)) - 1 to be safe" In general, we should stop making excuses for horrible MMU notifier users. It's much more important that the core VM is sane and safe, than letting MMU notifiers sleep. So if some MMU notifier is sleeping under a spinlock, we need to fix the notifier, not try to make excuses for that garbage in the core VM. Reported-and-tested-by: Bernhard Held Reported-and-tested-by: Adam Borowski Cc: Andrea Arcangeli Cc: Radim Krčmář Cc: Wanpeng Li Cc: Paolo Bonzini Cc: Takashi Iwai Cc: Nadav Amit Cc: Mike Galbraith Cc: Kirill A. Shutemov Cc: Jérôme Glisse Cc: axie Cc: Andrew Morton Signed-off-by: Linus Torvalds --- mm/rmap.c | 52 ++++++++++++++++++++++------------------------------ 1 file changed, 22 insertions(+), 30 deletions(-) diff --git a/mm/rmap.c b/mm/rmap.c index c1286d4..c8993c6 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -888,10 +888,10 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, .flags = PVMW_SYNC, }; int *cleaned = arg; - bool invalidation_needed = false; while (page_vma_mapped_walk(&pvmw)) { int ret = 0; + address = pvmw.address; if (pvmw.pte) { pte_t entry; pte_t *pte = pvmw.pte; @@ -899,11 +899,11 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, if (!pte_dirty(*pte) && !pte_write(*pte)) continue; - flush_cache_page(vma, pvmw.address, pte_pfn(*pte)); - entry = ptep_clear_flush(vma, pvmw.address, pte); + flush_cache_page(vma, address, pte_pfn(*pte)); + entry = ptep_clear_flush(vma, address, pte); entry = pte_wrprotect(entry); entry = pte_mkclean(entry); - set_pte_at(vma->vm_mm, pvmw.address, pte, entry); + set_pte_at(vma->vm_mm, address, pte, entry); ret = 1; } else { #ifdef CONFIG_TRANSPARENT_HUGE_PAGECACHE @@ -913,11 +913,11 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, if (!pmd_dirty(*pmd) && !pmd_write(*pmd)) continue; - flush_cache_page(vma, pvmw.address, page_to_pfn(page)); - entry = pmdp_huge_clear_flush(vma, pvmw.address, pmd); + flush_cache_page(vma, address, page_to_pfn(page)); + entry = pmdp_huge_clear_flush(vma, address, pmd); entry = pmd_wrprotect(entry); entry = pmd_mkclean(entry); - set_pmd_at(vma->vm_mm, pvmw.address, pmd, entry); + set_pmd_at(vma->vm_mm, address, pmd, entry); ret = 1; #else /* unexpected pmd-mapped page? */ @@ -926,16 +926,11 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, } if (ret) { + mmu_notifier_invalidate_page(vma->vm_mm, address); (*cleaned)++; - invalidation_needed = true; } } - if (invalidation_needed) { - mmu_notifier_invalidate_range(vma->vm_mm, address, - address + (1UL << compound_order(page))); - } - return true; } @@ -1328,7 +1323,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, }; pte_t pteval; struct page *subpage; - bool ret = true, invalidation_needed = false; + bool ret = true; enum ttu_flags flags = (enum ttu_flags)arg; /* munlock has nothing to gain from examining un-locked vmas */ @@ -1368,9 +1363,11 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, VM_BUG_ON_PAGE(!pvmw.pte, page); subpage = page - page_to_pfn(page) + pte_pfn(*pvmw.pte); + address = pvmw.address; + if (!(flags & TTU_IGNORE_ACCESS)) { - if (ptep_clear_flush_young_notify(vma, pvmw.address, + if (ptep_clear_flush_young_notify(vma, address, pvmw.pte)) { ret = false; page_vma_mapped_walk_done(&pvmw); @@ -1379,7 +1376,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, } /* Nuke the page table entry. */ - flush_cache_page(vma, pvmw.address, pte_pfn(*pvmw.pte)); + flush_cache_page(vma, address, pte_pfn(*pvmw.pte)); if (should_defer_flush(mm, flags)) { /* * We clear the PTE but do not flush so potentially @@ -1389,12 +1386,11 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, * transition on a cached TLB entry is written through * and traps if the PTE is unmapped. */ - pteval = ptep_get_and_clear(mm, pvmw.address, - pvmw.pte); + pteval = ptep_get_and_clear(mm, address, pvmw.pte); set_tlb_ubc_flush_pending(mm, pte_dirty(pteval)); } else { - pteval = ptep_clear_flush(vma, pvmw.address, pvmw.pte); + pteval = ptep_clear_flush(vma, address, pvmw.pte); } /* Move the dirty bit to the page. Now the pte is gone. */ @@ -1409,12 +1405,12 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, if (PageHuge(page)) { int nr = 1 << compound_order(page); hugetlb_count_sub(nr, mm); - set_huge_swap_pte_at(mm, pvmw.address, + set_huge_swap_pte_at(mm, address, pvmw.pte, pteval, vma_mmu_pagesize(vma)); } else { dec_mm_counter(mm, mm_counter(page)); - set_pte_at(mm, pvmw.address, pvmw.pte, pteval); + set_pte_at(mm, address, pvmw.pte, pteval); } } else if (pte_unused(pteval)) { @@ -1438,7 +1434,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, swp_pte = swp_entry_to_pte(entry); if (pte_soft_dirty(pteval)) swp_pte = pte_swp_mksoft_dirty(swp_pte); - set_pte_at(mm, pvmw.address, pvmw.pte, swp_pte); + set_pte_at(mm, address, pvmw.pte, swp_pte); } else if (PageAnon(page)) { swp_entry_t entry = { .val = page_private(subpage) }; pte_t swp_pte; @@ -1464,7 +1460,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, * If the page was redirtied, it cannot be * discarded. Remap the page to page table. */ - set_pte_at(mm, pvmw.address, pvmw.pte, pteval); + set_pte_at(mm, address, pvmw.pte, pteval); SetPageSwapBacked(page); ret = false; page_vma_mapped_walk_done(&pvmw); @@ -1472,7 +1468,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, } if (swap_duplicate(entry) < 0) { - set_pte_at(mm, pvmw.address, pvmw.pte, pteval); + set_pte_at(mm, address, pvmw.pte, pteval); ret = false; page_vma_mapped_walk_done(&pvmw); break; @@ -1488,18 +1484,14 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, swp_pte = swp_entry_to_pte(entry); if (pte_soft_dirty(pteval)) swp_pte = pte_swp_mksoft_dirty(swp_pte); - set_pte_at(mm, pvmw.address, pvmw.pte, swp_pte); + set_pte_at(mm, address, pvmw.pte, swp_pte); } else dec_mm_counter(mm, mm_counter_file(page)); discard: page_remove_rmap(subpage, PageHuge(page)); put_page(page); - invalidation_needed = true; + mmu_notifier_invalidate_page(mm, address); } - - if (invalidation_needed) - mmu_notifier_invalidate_range(mm, address, - address + (1UL << compound_order(page))); return ret; } -- cgit v1.1 From e8d411d2980723b8f8ba8e4dd78b694c5fd9ea3e Mon Sep 17 00:00:00 2001 From: Xin Long Date: Mon, 28 Aug 2017 10:45:01 +0800 Subject: ipv6: do not set sk_destruct in IPV6_ADDRFORM sockopt ChunYu found a kernel warn_on during syzkaller fuzzing: [40226.038539] WARNING: CPU: 5 PID: 23720 at net/ipv4/af_inet.c:152 inet_sock_destruct+0x78d/0x9a0 [40226.144849] Call Trace: [40226.147590] [40226.149859] dump_stack+0xe2/0x186 [40226.176546] __warn+0x1a4/0x1e0 [40226.180066] warn_slowpath_null+0x31/0x40 [40226.184555] inet_sock_destruct+0x78d/0x9a0 [40226.246355] __sk_destruct+0xfa/0x8c0 [40226.290612] rcu_process_callbacks+0xaa0/0x18a0 [40226.336816] __do_softirq+0x241/0x75e [40226.367758] irq_exit+0x1f6/0x220 [40226.371458] smp_apic_timer_interrupt+0x7b/0xa0 [40226.376507] apic_timer_interrupt+0x93/0xa0 The warn_on happned when sk->sk_rmem_alloc wasn't 0 in inet_sock_destruct. As after commit f970bd9e3a06 ("udp: implement memory accounting helpers"), udp has changed to use udp_destruct_sock as sk_destruct where it would udp_rmem_release all rmem. But IPV6_ADDRFORM sockopt sets sk_destruct with inet_sock_destruct after changing family to PF_INET. If rmem is not 0 at that time, and there is no place to release rmem before calling inet_sock_destruct, the warn_on will be triggered. This patch is to fix it by not setting sk_destruct in IPV6_ADDRFORM sockopt any more. As IPV6_ADDRFORM sockopt only works for tcp and udp. TCP sock has already set it's sk_destruct with inet_sock_destruct and UDP has set with udp_destruct_sock since they're created. Fixes: f970bd9e3a06 ("udp: implement memory accounting helpers") Reported-by: ChunYu Wang Signed-off-by: Xin Long Acked-by: Paolo Abeni Signed-off-by: David S. Miller --- net/ipv6/ipv6_sockglue.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 02d795f..a5e466d 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -242,7 +242,6 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, pktopt = xchg(&np->pktoptions, NULL); kfree_skb(pktopt); - sk->sk_destruct = inet_sock_destruct; /* * ... and add it to the refcnt debug socks count * in the new family. -acme -- cgit v1.1 From e42faf553c8d7bdd4735cc6b9a9e969aa985d078 Mon Sep 17 00:00:00 2001 From: Matt Turner Date: Mon, 24 Oct 2016 22:07:41 -0700 Subject: alpha: Fix build error without CONFIG_VGA_HOSE. pci_vga_hose is #defined to 0 in include/asm/vga.h if CONFIG_VGA_HOSE is not set. Signed-off-by: Matt Turner --- arch/alpha/kernel/core_marvel.c | 2 ++ arch/alpha/kernel/core_titan.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/arch/alpha/kernel/core_marvel.c b/arch/alpha/kernel/core_marvel.c index d5f0580..db72356 100644 --- a/arch/alpha/kernel/core_marvel.c +++ b/arch/alpha/kernel/core_marvel.c @@ -369,6 +369,7 @@ marvel_io7_present(gct6_node *node) static void __init marvel_find_console_vga_hose(void) { +#ifdef CONFIG_VGA_HOSE u64 *pu64 = (u64 *)((u64)hwrpb + hwrpb->ctbt_offset); if (pu64[7] == 3) { /* TERM_TYPE == graphics */ @@ -402,6 +403,7 @@ marvel_find_console_vga_hose(void) pci_vga_hose = hose; } } +#endif } gct6_search_struct gct_wanted_node_list[] = { diff --git a/arch/alpha/kernel/core_titan.c b/arch/alpha/kernel/core_titan.c index 219bf27..b532d92 100644 --- a/arch/alpha/kernel/core_titan.c +++ b/arch/alpha/kernel/core_titan.c @@ -461,6 +461,7 @@ titan_ioremap(unsigned long addr, unsigned long size) unsigned long *ptes; unsigned long pfn; +#ifdef CONFIG_VGA_HOSE /* * Adjust the address and hose, if necessary. */ @@ -468,6 +469,7 @@ titan_ioremap(unsigned long addr, unsigned long size) h = pci_vga_hose->index; addr += pci_vga_hose->mem_space->start; } +#endif /* * Find the hose. -- cgit v1.1 From a720830613eaa25eb5bc9b76705a88a36296709a Mon Sep 17 00:00:00 2001 From: Richard Henderson Date: Fri, 23 Jun 2017 14:19:59 -0700 Subject: alpha: Update for new syscalls Signed-off-by: Richard Henderson Signed-off-by: Matt Turner --- arch/alpha/include/asm/unistd.h | 2 +- arch/alpha/include/uapi/asm/unistd.h | 14 ++++++++++++++ arch/alpha/kernel/systbls.S | 9 +++++++++ 3 files changed, 24 insertions(+), 1 deletion(-) diff --git a/arch/alpha/include/asm/unistd.h b/arch/alpha/include/asm/unistd.h index b37153e..db7fc0f 100644 --- a/arch/alpha/include/asm/unistd.h +++ b/arch/alpha/include/asm/unistd.h @@ -3,7 +3,7 @@ #include -#define NR_SYSCALLS 514 +#define NR_SYSCALLS 523 #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_STAT64 diff --git a/arch/alpha/include/uapi/asm/unistd.h b/arch/alpha/include/uapi/asm/unistd.h index aa33bf5..a2945fe 100644 --- a/arch/alpha/include/uapi/asm/unistd.h +++ b/arch/alpha/include/uapi/asm/unistd.h @@ -475,5 +475,19 @@ #define __NR_getrandom 511 #define __NR_memfd_create 512 #define __NR_execveat 513 +#define __NR_seccomp 514 +#define __NR_bpf 515 +#define __NR_userfaultfd 516 +#define __NR_membarrier 517 +#define __NR_mlock2 518 +#define __NR_copy_file_range 519 +#define __NR_preadv2 520 +#define __NR_pwritev2 521 +#define __NR_statx 522 + +/* Alpha doesn't have protection keys. */ +#define __IGNORE_pkey_mprotect +#define __IGNORE_pkey_alloc +#define __IGNORE_pkey_free #endif /* _UAPI_ALPHA_UNISTD_H */ diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S index 9b62e3f..5b4514a 100644 --- a/arch/alpha/kernel/systbls.S +++ b/arch/alpha/kernel/systbls.S @@ -532,6 +532,15 @@ sys_call_table: .quad sys_getrandom .quad sys_memfd_create .quad sys_execveat + .quad sys_seccomp + .quad sys_bpf /* 515 */ + .quad sys_userfaultfd + .quad sys_membarrier + .quad sys_mlock2 + .quad sys_copy_file_range + .quad sys_preadv2 /* 520 */ + .quad sys_pwritev2 + .quad sys_statx .size sys_call_table, . - sys_call_table .type sys_call_table, @object -- cgit v1.1 From 4758ce82e66711b1a4557577e30a5f9b88d4a4b5 Mon Sep 17 00:00:00 2001 From: Richard Henderson Date: Fri, 23 Jun 2017 14:20:00 -0700 Subject: alpha: Package string routines together There are direct branches between {str*cpy,str*cat} and stx*cpy. Ensure the branches are within range by merging these objects. Signed-off-by: Richard Henderson Signed-off-by: Matt Turner --- arch/alpha/lib/Makefile | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/arch/alpha/lib/Makefile b/arch/alpha/lib/Makefile index 7083434..a808159 100644 --- a/arch/alpha/lib/Makefile +++ b/arch/alpha/lib/Makefile @@ -20,12 +20,8 @@ lib-y = __divqu.o __remqu.o __divlu.o __remlu.o \ checksum.o \ csum_partial_copy.o \ $(ev67-y)strlen.o \ - $(ev67-y)strcat.o \ - strcpy.o \ - $(ev67-y)strncat.o \ - strncpy.o \ - $(ev6-y)stxcpy.o \ - $(ev6-y)stxncpy.o \ + stycpy.o \ + styncpy.o \ $(ev67-y)strchr.o \ $(ev67-y)strrchr.o \ $(ev6-y)memchr.o \ @@ -49,3 +45,17 @@ AFLAGS___remlu.o = -DREM -DINTSIZE $(addprefix $(obj)/,__divqu.o __remqu.o __divlu.o __remlu.o): \ $(src)/$(ev6-y)divide.S FORCE $(call if_changed_rule,as_o_S) + +# There are direct branches between {str*cpy,str*cat} and stx*cpy. +# Ensure the branches are within range by merging these objects. + +LDFLAGS_stycpy.o := -r +LDFLAGS_styncpy.o := -r + +$(obj)/stycpy.o: $(obj)/strcpy.o $(obj)/$(ev67-y)strcat.o \ + $(obj)/$(ev6-y)stxcpy.o FORCE + $(call if_changed,ld) + +$(obj)/styncpy.o: $(obj)/strncpy.o $(obj)/$(ev67-y)strncat.o \ + $(obj)/$(ev6-y)stxncpy.o FORCE + $(call if_changed,ld) -- cgit v1.1 From 4606f68faff1b16734f9379937fc49ae289d4c6e Mon Sep 17 00:00:00 2001 From: Richard Henderson Date: Fri, 23 Jun 2017 14:20:01 -0700 Subject: alpha: Fix typo in ev6-copy_user.S Patch 8525023121de4848b5f0a7d867ffeadbc477774d introduced a typo. That said, the identity AND insns added by that patch are more clearly written as MOV. At the same time, re-schedule the ev6 version so that the first dispatch can execute in parallel. Signed-off-by: Richard Henderson Signed-off-by: Matt Turner --- arch/alpha/lib/copy_user.S | 2 +- arch/alpha/lib/ev6-copy_user.S | 7 ++++--- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/alpha/lib/copy_user.S b/arch/alpha/lib/copy_user.S index 159f1b7..c277a1a 100644 --- a/arch/alpha/lib/copy_user.S +++ b/arch/alpha/lib/copy_user.S @@ -34,7 +34,7 @@ .ent __copy_user __copy_user: .prologue 0 - and $18,$18,$0 + mov $18,$0 and $16,7,$3 beq $0,$35 beq $3,$36 diff --git a/arch/alpha/lib/ev6-copy_user.S b/arch/alpha/lib/ev6-copy_user.S index 35e6710..954ca03 100644 --- a/arch/alpha/lib/ev6-copy_user.S +++ b/arch/alpha/lib/ev6-copy_user.S @@ -45,9 +45,10 @@ # Pipeline info: Slotting & Comments __copy_user: .prologue 0 - andq $18, $18, $0 - subq $18, 32, $1 # .. E .. .. : Is this going to be a small copy? - beq $0, $zerolength # U .. .. .. : U L U L + mov $18, $0 # .. .. .. E + subq $18, 32, $1 # .. .. E. .. : Is this going to be a small copy? + nop # .. E .. .. + beq $18, $zerolength # U .. .. .. : U L U L and $16,7,$3 # .. .. .. E : is leading dest misalignment ble $1, $onebyteloop # .. .. U .. : 1st branch : small amount of data -- cgit v1.1 From 4f61e0781b701f5eb4f0e186d361c4104659001e Mon Sep 17 00:00:00 2001 From: Michael Cree Date: Sat, 24 Jun 2017 13:37:02 +1200 Subject: alpha: support R_ALPHA_REFLONG relocations for module loading Since commit 71810db27c1c853b33 (modversions: treat symbol CRCs as 32 bit quantities) R_ALPHA_REFLONG relocations can be required to load modules. This implements it. Tested-by: Bob Tracy Reviewed-by: Richard Henderson Signed-off-by: Michael Cree Signed-off-by: Matt Turner --- arch/alpha/kernel/module.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/alpha/kernel/module.c b/arch/alpha/kernel/module.c index 936bc8f..47632fa 100644 --- a/arch/alpha/kernel/module.c +++ b/arch/alpha/kernel/module.c @@ -181,6 +181,9 @@ apply_relocate_add(Elf64_Shdr *sechdrs, const char *strtab, switch (r_type) { case R_ALPHA_NONE: break; + case R_ALPHA_REFLONG: + *(u32 *)location = value; + break; case R_ALPHA_REFQUAD: /* BUG() can produce misaligned relocations. */ ((u32 *)location)[0] = value; -- cgit v1.1 From 69f0678239b9f45652e1914efa20a397b3acac7a Mon Sep 17 00:00:00 2001 From: Matt Turner Date: Thu, 24 Aug 2017 08:59:36 -0700 Subject: alpha: Fix section mismatches Signed-off-by: Matt Turner --- arch/alpha/kernel/core_marvel.c | 4 ++-- arch/alpha/kernel/smp.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/alpha/kernel/core_marvel.c b/arch/alpha/kernel/core_marvel.c index db72356..03ff832 100644 --- a/arch/alpha/kernel/core_marvel.c +++ b/arch/alpha/kernel/core_marvel.c @@ -351,7 +351,7 @@ marvel_init_io7(struct io7 *io7) } } -void +void __init marvel_io7_present(gct6_node *node) { int pe; @@ -406,7 +406,7 @@ marvel_find_console_vga_hose(void) #endif } -gct6_search_struct gct_wanted_node_list[] = { +gct6_search_struct gct_wanted_node_list[] __initdata = { { GCT_TYPE_HOSE, GCT_SUBTYPE_IO_PORT_MODULE, marvel_io7_present }, { 0, 0, NULL } }; diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c index 9fc5604..f6726a7 100644 --- a/arch/alpha/kernel/smp.c +++ b/arch/alpha/kernel/smp.c @@ -115,7 +115,7 @@ wait_boot_cpu_to_stop(int cpuid) /* * Where secondaries begin a life of C. */ -void +void __init smp_callin(void) { int cpuid = hard_smp_processor_id(); -- cgit v1.1 From 7817cedce04325bed7b618386a245fb461233746 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Fri, 31 Jul 2015 19:32:39 -0700 Subject: alpha: Define ioremap_wc Commit 3cc2dac5be3f ("drivers/video/fbdev/atyfb: Replace MTRR UC hole with strong UC") introduces calls to ioremap_wc and ioremap_uc. This causes build failures with alpha:allmodconfig. Map the missing functions to ioremap_nocache. Fixes: 3cc2dac5be3f ("drivers/video/fbdev/atyfb: Replace MTRR UC hole with strong UC") Cc: Paul Gortmaker Cc: Luis R. Rodriguez Signed-off-by: Guenter Roeck Signed-off-by: Matt Turner --- arch/alpha/include/asm/io.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/alpha/include/asm/io.h b/arch/alpha/include/asm/io.h index ff40491..4d61d2a 100644 --- a/arch/alpha/include/asm/io.h +++ b/arch/alpha/include/asm/io.h @@ -299,6 +299,7 @@ static inline void __iomem * ioremap_nocache(unsigned long offset, return ioremap(offset, size); } +#define ioremap_wc ioremap_nocache #define ioremap_uc ioremap_nocache static inline void iounmap(volatile void __iomem *addr) -- cgit v1.1 From cec80d82142ab25c71eee24b529cfeaf17c43062 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 1 Oct 2015 01:35:55 +0100 Subject: alpha: uapi: Add support for __SANE_USERSPACE_TYPES__ This fixes compiler errors in perf such as: tests/attr.c: In function 'store_event': tests/attr.c:66:27: error: format '%llu' expects argument of type 'long long unsigned int', but argument 6 has type '__u64 {aka long unsigned int}' [-Werror=format=] snprintf(path, PATH_MAX, "%s/event-%d-%llu-%d", dir, ^ Signed-off-by: Ben Hutchings Tested-by: Michael Cree Cc: stable@vger.kernel.org Signed-off-by: Matt Turner --- arch/alpha/include/asm/types.h | 2 +- arch/alpha/include/uapi/asm/types.h | 12 +++++++++++- 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/arch/alpha/include/asm/types.h b/arch/alpha/include/asm/types.h index 4cb4b6d..0bc66e1 100644 --- a/arch/alpha/include/asm/types.h +++ b/arch/alpha/include/asm/types.h @@ -1,6 +1,6 @@ #ifndef _ALPHA_TYPES_H #define _ALPHA_TYPES_H -#include +#include #endif /* _ALPHA_TYPES_H */ diff --git a/arch/alpha/include/uapi/asm/types.h b/arch/alpha/include/uapi/asm/types.h index 9fd3cd4..8d1024d 100644 --- a/arch/alpha/include/uapi/asm/types.h +++ b/arch/alpha/include/uapi/asm/types.h @@ -9,8 +9,18 @@ * need to be careful to avoid a name clashes. */ -#ifndef __KERNEL__ +/* + * This is here because we used to use l64 for alpha + * and we don't want to impact user mode with our change to ll64 + * in the kernel. + * + * However, some user programs are fine with this. They can + * flag __SANE_USERSPACE_TYPES__ to get int-ll64.h here. + */ +#if !defined(__SANE_USERSPACE_TYPES__) && !defined(__KERNEL__) #include +#else +#include #endif #endif /* _UAPI_ALPHA_TYPES_H */ -- cgit v1.1 From 021aba761f2a6c12158afb9993524c300c01fae2 Mon Sep 17 00:00:00 2001 From: Sinclair Yeh Date: Tue, 29 Aug 2017 18:55:09 +0200 Subject: drm/vmwgfx: Fix F26 Wayland screen update issue vmwgfx currently cannot support non-blocking commit because when vmw_*_crtc_page_flip is called, drm_atomic_nonblocking_commit() schedules the update on a thread. This means vmw_*_crtc_page_flip cannot rely on the new surface being bound before the subsequent dirty and flush operations happen. Cc: # 4.12.x Signed-off-by: Sinclair Yeh Reviewed-by: Thomas Hellstrom Reviewed-by: Charmaine Lee --- drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 61e06f0..625ba24 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -1567,10 +1567,34 @@ vmw_kms_atomic_check_modeset(struct drm_device *dev, } +/** + * vmw_kms_atomic_commit - Perform an atomic state commit + * + * @dev: DRM device + * @state: the driver state object + * @nonblock: Whether nonblocking behaviour is requested + * + * This is a simple wrapper around drm_atomic_helper_commit() for + * us to clear the nonblocking value. + * + * Nonblocking commits currently cause synchronization issues + * for vmwgfx. + * + * RETURNS + * Zero for success or negative error code on failure. + */ +int vmw_kms_atomic_commit(struct drm_device *dev, + struct drm_atomic_state *state, + bool nonblock) +{ + return drm_atomic_helper_commit(dev, state, false); +} + + static const struct drm_mode_config_funcs vmw_kms_funcs = { .fb_create = vmw_kms_fb_create, .atomic_check = vmw_kms_atomic_check_modeset, - .atomic_commit = drm_atomic_helper_commit, + .atomic_commit = vmw_kms_atomic_commit, }; static int vmw_kms_generic_present(struct vmw_private *dev_priv, -- cgit v1.1 From d55c60eba0ef44ec21831ce26db300763eafd865 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Mon, 28 Aug 2017 17:57:02 +0200 Subject: tipc: permit bond slave as bearer For a bond slave device as a tipc bearer, the dev represents the bond interface and orig_dev represents the slave in tipc_l2_rcv_msg(). Since we decode the tipc_ptr from bonding device (dev), we fail to find the bearer and thus tipc links are not established. In this commit, we register the tipc protocol callback per device and look for tipc bearer from both the devices. Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- net/tipc/bearer.c | 26 +++++++++++--------------- net/tipc/bearer.h | 2 ++ 2 files changed, 13 insertions(+), 15 deletions(-) diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 767e053..89cd061 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -65,6 +65,8 @@ static struct tipc_bearer *bearer_get(struct net *net, int bearer_id) } static void bearer_disable(struct net *net, struct tipc_bearer *b); +static int tipc_l2_rcv_msg(struct sk_buff *skb, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev); /** * tipc_media_find - locates specified media object by name @@ -428,6 +430,10 @@ int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b, /* Associate TIPC bearer with L2 bearer */ rcu_assign_pointer(b->media_ptr, dev); + b->pt.dev = dev; + b->pt.type = htons(ETH_P_TIPC); + b->pt.func = tipc_l2_rcv_msg; + dev_add_pack(&b->pt); memset(&b->bcast_addr, 0, sizeof(b->bcast_addr)); memcpy(b->bcast_addr.value, dev->broadcast, b->media->hwaddr_len); b->bcast_addr.media_id = b->media->type_id; @@ -447,6 +453,7 @@ void tipc_disable_l2_media(struct tipc_bearer *b) struct net_device *dev; dev = (struct net_device *)rtnl_dereference(b->media_ptr); + dev_remove_pack(&b->pt); RCU_INIT_POINTER(dev->tipc_ptr, NULL); synchronize_net(); dev_put(dev); @@ -594,11 +601,12 @@ static int tipc_l2_rcv_msg(struct sk_buff *skb, struct net_device *dev, struct tipc_bearer *b; rcu_read_lock(); - b = rcu_dereference_rtnl(dev->tipc_ptr); + b = rcu_dereference_rtnl(dev->tipc_ptr) ?: + rcu_dereference_rtnl(orig_dev->tipc_ptr); if (likely(b && test_bit(0, &b->up) && (skb->pkt_type <= PACKET_MULTICAST))) { skb->next = NULL; - tipc_rcv(dev_net(dev), skb, b); + tipc_rcv(dev_net(b->pt.dev), skb, b); rcu_read_unlock(); return NET_RX_SUCCESS; } @@ -659,11 +667,6 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, return NOTIFY_OK; } -static struct packet_type tipc_packet_type __read_mostly = { - .type = htons(ETH_P_TIPC), - .func = tipc_l2_rcv_msg, -}; - static struct notifier_block notifier = { .notifier_call = tipc_l2_device_event, .priority = 0, @@ -671,19 +674,12 @@ static struct notifier_block notifier = { int tipc_bearer_setup(void) { - int err; - - err = register_netdevice_notifier(¬ifier); - if (err) - return err; - dev_add_pack(&tipc_packet_type); - return 0; + return register_netdevice_notifier(¬ifier); } void tipc_bearer_cleanup(void) { unregister_netdevice_notifier(¬ifier); - dev_remove_pack(&tipc_packet_type); } void tipc_bearer_stop(struct net *net) diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 635c908..e07a55a 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -131,6 +131,7 @@ struct tipc_media { * @name: bearer name (format = media:interface) * @media: ptr to media structure associated with bearer * @bcast_addr: media address used in broadcasting + * @pt: packet type for bearer * @rcu: rcu struct for tipc_bearer * @priority: default link priority for bearer * @window: default window size for bearer @@ -151,6 +152,7 @@ struct tipc_bearer { char name[TIPC_MAX_BEARER_NAME]; struct tipc_media *media; struct tipc_media_addr bcast_addr; + struct packet_type pt; struct rcu_head rcu; u32 priority; u32 window; -- cgit v1.1 From edbd58be15a957f6a760c4a514cd475217eb97fd Mon Sep 17 00:00:00 2001 From: Benjamin Poirier Date: Mon, 28 Aug 2017 14:29:41 -0400 Subject: packet: Don't write vnet header beyond end of buffer ... which may happen with certain values of tp_reserve and maclen. Fixes: 58d19b19cd99 ("packet: vnet_hdr support for tpacket_rcv") Signed-off-by: Benjamin Poirier Cc: Willem de Bruijn Acked-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/packet/af_packet.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 008a45c..1c61af9 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2191,6 +2191,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct timespec ts; __u32 ts_status; bool is_drop_n_account = false; + bool do_vnet = false; /* struct tpacket{2,3}_hdr is aligned to a multiple of TPACKET_ALIGNMENT. * We may add members to them until current aligned size without forcing @@ -2241,8 +2242,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, netoff = TPACKET_ALIGN(po->tp_hdrlen + (maclen < 16 ? 16 : maclen)) + po->tp_reserve; - if (po->has_vnet_hdr) + if (po->has_vnet_hdr) { netoff += sizeof(struct virtio_net_hdr); + do_vnet = true; + } macoff = netoff - maclen; } if (po->tp_version <= TPACKET_V2) { @@ -2259,8 +2262,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, skb_set_owner_r(copy_skb, sk); } snaplen = po->rx_ring.frame_size - macoff; - if ((int)snaplen < 0) + if ((int)snaplen < 0) { snaplen = 0; + do_vnet = false; + } } } else if (unlikely(macoff + snaplen > GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len)) { @@ -2273,6 +2278,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, if (unlikely((int)snaplen < 0)) { snaplen = 0; macoff = GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len; + do_vnet = false; } } spin_lock(&sk->sk_receive_queue.lock); @@ -2298,7 +2304,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, } spin_unlock(&sk->sk_receive_queue.lock); - if (po->has_vnet_hdr) { + if (do_vnet) { if (virtio_net_hdr_from_skb(skb, h.raw + macoff - sizeof(struct virtio_net_hdr), vio_le(), true)) { -- cgit v1.1 From 278175aba363dcc5b0978abe16fa39dcdca67ffb Mon Sep 17 00:00:00 2001 From: Pavel Belous Date: Mon, 28 Aug 2017 21:52:08 +0300 Subject: net:ethernet:aquantia: Extra spinlocks removed. This patch removes datapath spinlocks which does not perform any useful work. Fixes: 6e70637f9f1e ("net: ethernet: aquantia: Add ring support code") Signed-off-by: Pavel Belous Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 42 +++++++---------------- drivers/net/ethernet/aquantia/atlantic/aq_ring.c | 1 - drivers/net/ethernet/aquantia/atlantic/aq_utils.h | 1 - drivers/net/ethernet/aquantia/atlantic/aq_vec.c | 11 ++---- 4 files changed, 14 insertions(+), 41 deletions(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 9ee1c50..08b7275 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -597,14 +597,11 @@ exit: } int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb) -__releases(&ring->lock) -__acquires(&ring->lock) { struct aq_ring_s *ring = NULL; unsigned int frags = 0U; unsigned int vec = skb->queue_mapping % self->aq_nic_cfg.vecs; unsigned int tc = 0U; - unsigned int trys = AQ_CFG_LOCK_TRYS; int err = NETDEV_TX_OK; bool is_nic_in_bad_state; @@ -628,36 +625,21 @@ __acquires(&ring->lock) goto err_exit; } - do { - if (spin_trylock(&ring->header.lock)) { - frags = aq_nic_map_skb(self, skb, ring); - - if (likely(frags)) { - err = self->aq_hw_ops.hw_ring_tx_xmit( - self->aq_hw, - ring, frags); - if (err >= 0) { - if (aq_ring_avail_dx(ring) < - AQ_CFG_SKB_FRAGS_MAX + 1) - aq_nic_ndev_queue_stop( - self, - ring->idx); - - ++ring->stats.tx.packets; - ring->stats.tx.bytes += skb->len; - } - } else { - err = NETDEV_TX_BUSY; - } + frags = aq_nic_map_skb(self, skb, ring); - spin_unlock(&ring->header.lock); - break; - } - } while (--trys); + if (likely(frags)) { + err = self->aq_hw_ops.hw_ring_tx_xmit(self->aq_hw, + ring, + frags); + if (err >= 0) { + if (aq_ring_avail_dx(ring) < AQ_CFG_SKB_FRAGS_MAX + 1) + aq_nic_ndev_queue_stop(self, ring->idx); - if (!trys) { + ++ring->stats.tx.packets; + ring->stats.tx.bytes += skb->len; + } + } else { err = NETDEV_TX_BUSY; - goto err_exit; } err_exit: diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c index 9a08179..ec5579f 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c @@ -101,7 +101,6 @@ int aq_ring_init(struct aq_ring_s *self) self->hw_head = 0; self->sw_head = 0; self->sw_tail = 0; - spin_lock_init(&self->header.lock); return 0; } diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_utils.h b/drivers/net/ethernet/aquantia/atlantic/aq_utils.h index f6012b3..e12bcdf 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_utils.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_utils.h @@ -17,7 +17,6 @@ #define AQ_DIMOF(_ARY_) ARRAY_SIZE(_ARY_) struct aq_obj_s { - spinlock_t lock; /* spinlock for nic/rings processing */ atomic_t flags; }; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_vec.c b/drivers/net/ethernet/aquantia/atlantic/aq_vec.c index ad5b4d4d..fee446a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_vec.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_vec.c @@ -34,8 +34,6 @@ struct aq_vec_s { #define AQ_VEC_RX_ID 1 static int aq_vec_poll(struct napi_struct *napi, int budget) -__releases(&self->lock) -__acquires(&self->lock) { struct aq_vec_s *self = container_of(napi, struct aq_vec_s, napi); struct aq_ring_s *ring = NULL; @@ -47,7 +45,7 @@ __acquires(&self->lock) if (!self) { err = -EINVAL; - } else if (spin_trylock(&self->header.lock)) { + } else { for (i = 0U, ring = self->ring[0]; self->tx_rings > i; ++i, ring = self->ring[i]) { if (self->aq_hw_ops->hw_ring_tx_head_update) { @@ -105,11 +103,8 @@ __acquires(&self->lock) self->aq_hw_ops->hw_irq_enable(self->aq_hw, 1U << self->aq_ring_param.vec_idx); } - -err_exit: - spin_unlock(&self->header.lock); } - +err_exit: return work_done; } @@ -185,8 +180,6 @@ int aq_vec_init(struct aq_vec_s *self, struct aq_hw_ops *aq_hw_ops, self->aq_hw_ops = aq_hw_ops; self->aq_hw = aq_hw; - spin_lock_init(&self->header.lock); - for (i = 0U, ring = self->ring[0]; self->tx_rings > i; ++i, ring = self->ring[i]) { err = aq_ring_init(&ring[AQ_VEC_TX_ID]); -- cgit v1.1 From 64fc7953ffd9424726988dd04945c28141ee41af Mon Sep 17 00:00:00 2001 From: Pavel Belous Date: Mon, 28 Aug 2017 21:52:09 +0300 Subject: net:ethernet:aquantia: Fix for number of RSS queues. The number of RSS queues should be not more than numbers of CPU. Its does not make sense to increase perfomance, and also cause problems on some motherboards. Fixes: 94f6c9e4cdf6 ("net: ethernet: aquantia: Support for NIC-specific code") Signed-off-by: Pavel Belous Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 08b7275..d6d8e70 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -103,6 +103,8 @@ int aq_nic_cfg_start(struct aq_nic_s *self) else cfg->vecs = 1U; + cfg->num_rss_queues = min(cfg->vecs, AQ_CFG_NUM_RSS_QUEUES_DEF); + cfg->irq_type = aq_pci_func_get_irq_type(self->aq_pci_func); if ((cfg->irq_type == AQ_HW_IRQ_LEGACY) || -- cgit v1.1 From 0a402e7b9725611069dad4c873d1516f8c805f38 Mon Sep 17 00:00:00 2001 From: Pavel Belous Date: Mon, 28 Aug 2017 21:52:10 +0300 Subject: net:ethernet:aquantia: Workaround for HW checksum bug. The hardware has the HW Checksum Offload bug when small TCP patckets (with length <= 60 bytes) has wrong "checksum valid" bit. The solution is - ignore checksum valid bit for small packets (with length <= 60 bytes) and mark this as CHECKSUM_NONE to allow network stack recalculate checksum itself. Fixes: ccf9a5ed14be ("net: ethernet: aquantia: Atlantic A0 and B0 specific functions.") Signed-off-by: Pavel Belous Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c | 6 ++++++ drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c index faeb493..c5a02df 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c @@ -629,6 +629,12 @@ static int hw_atl_a0_hw_ring_rx_receive(struct aq_hw_s *self, buff->is_udp_cso = (is_err & 0x10U) ? 0 : 1; else if (0x0U == (pkt_type & 0x1CU)) buff->is_tcp_cso = (is_err & 0x10U) ? 0 : 1; + + /* Checksum offload workaround for small packets */ + if (rxd_wb->pkt_len <= 60) { + buff->is_ip_cso = 0U; + buff->is_cso_err = 0U; + } } is_err &= ~0x18U; diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 1bceb73..21784cc 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -645,6 +645,12 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, buff->is_udp_cso = buff->is_cso_err ? 0U : 1U; else if (0x0U == (pkt_type & 0x1CU)) buff->is_tcp_cso = buff->is_cso_err ? 0U : 1U; + + /* Checksum offload workaround for small packets */ + if (rxd_wb->pkt_len <= 60) { + buff->is_ip_cso = 0U; + buff->is_cso_err = 0U; + } } is_err &= ~0x18U; -- cgit v1.1 From bd8ed4415ff8584ccdd1f61c8d7279dc1f9e623e Mon Sep 17 00:00:00 2001 From: Pavel Belous Date: Mon, 28 Aug 2017 21:52:11 +0300 Subject: net:ethernet:aquantia: Fix for incorrect speed index. The driver choose the optimal interrupt throttling settings depends of current link speed. Due this bug link_status field from aq_hw is never updated and as result always used same interrupt throttling values. Fixes: 3d2ff7eebe26 ("net: ethernet: aquantia: Atlantic hardware abstraction layer") Signed-off-by: Pavel Belous Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_hw.h | 3 +-- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 31 ++++++++++------------ .../aquantia/atlantic/hw_atl/hw_atl_utils.c | 4 +-- .../aquantia/atlantic/hw_atl/hw_atl_utils.h | 3 +-- 4 files changed, 18 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index fce0fd3..bf9b3f0 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -105,8 +105,7 @@ struct aq_hw_ops { int (*hw_set_mac_address)(struct aq_hw_s *self, u8 *mac_addr); - int (*hw_get_link_status)(struct aq_hw_s *self, - struct aq_hw_link_status_s *link_status); + int (*hw_get_link_status)(struct aq_hw_s *self); int (*hw_set_link_speed)(struct aq_hw_s *self, u32 speed); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index d6d8e70..dce17a5 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -125,33 +125,30 @@ static void aq_nic_service_timer_cb(unsigned long param) struct net_device *ndev = aq_nic_get_ndev(self); int err = 0; unsigned int i = 0U; - struct aq_hw_link_status_s link_status; struct aq_ring_stats_rx_s stats_rx; struct aq_ring_stats_tx_s stats_tx; if (aq_utils_obj_test(&self->header.flags, AQ_NIC_FLAGS_IS_NOT_READY)) goto err_exit; - err = self->aq_hw_ops.hw_get_link_status(self->aq_hw, &link_status); + err = self->aq_hw_ops.hw_get_link_status(self->aq_hw); if (err < 0) goto err_exit; - self->aq_hw_ops.hw_interrupt_moderation_set(self->aq_hw, - self->aq_nic_cfg.is_interrupt_moderation); - - if (memcmp(&link_status, &self->link_status, sizeof(link_status))) { - if (link_status.mbps) { - aq_utils_obj_set(&self->header.flags, - AQ_NIC_FLAG_STARTED); - aq_utils_obj_clear(&self->header.flags, - AQ_NIC_LINK_DOWN); - netif_carrier_on(self->ndev); - } else { - netif_carrier_off(self->ndev); - aq_utils_obj_set(&self->header.flags, AQ_NIC_LINK_DOWN); - } + self->link_status = self->aq_hw->aq_link_status; - self->link_status = link_status; + self->aq_hw_ops.hw_interrupt_moderation_set(self->aq_hw, + self->aq_nic_cfg.is_interrupt_moderation); + + if (self->link_status.mbps) { + aq_utils_obj_set(&self->header.flags, + AQ_NIC_FLAG_STARTED); + aq_utils_obj_clear(&self->header.flags, + AQ_NIC_LINK_DOWN); + netif_carrier_on(self->ndev); + } else { + netif_carrier_off(self->ndev); + aq_utils_obj_set(&self->header.flags, AQ_NIC_LINK_DOWN); } memset(&stats_rx, 0U, sizeof(struct aq_ring_stats_rx_s)); diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c index 8d6d8f5..7a1332e 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c @@ -313,11 +313,11 @@ void hw_atl_utils_mpi_set(struct aq_hw_s *self, err_exit:; } -int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self, - struct aq_hw_link_status_s *link_status) +int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self) { u32 cp0x036C = aq_hw_read_reg(self, HW_ATL_MPI_STATE_ADR); u32 link_speed_mask = cp0x036C >> HW_ATL_MPI_SPEED_SHIFT; + struct aq_hw_link_status_s *link_status = &self->aq_link_status; if (!link_speed_mask) { link_status->mbps = 0U; diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h index a66aee5..e0360a6 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h @@ -180,8 +180,7 @@ void hw_atl_utils_mpi_set(struct aq_hw_s *self, int hw_atl_utils_mpi_set_speed(struct aq_hw_s *self, u32 speed, enum hal_atl_utils_fw_state_e state); -int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self, - struct aq_hw_link_status_s *link_status); +int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self); int hw_atl_utils_get_mac_permanent(struct aq_hw_s *self, struct aq_hw_caps_s *aq_hw_caps, -- cgit v1.1 From b21f502f84be082fb63cca8e7ab6eb8f7ee88024 Mon Sep 17 00:00:00 2001 From: Igor Russkikh Date: Mon, 28 Aug 2017 21:52:12 +0300 Subject: net:ethernet:aquantia: Fix for multicast filter handling. Since the HW supports up to 32 multicast filters we should track count of multicast filters to avoid overflow. If we attempt to add >32 multicast filter - just set NETIF_ALLMULTI flag instead. Fixes: 94f6c9e4cdf6 ("net: ethernet: aquantia: Support for NIC-specific code") Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index dce17a5..6ac9e26 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -669,11 +669,26 @@ int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev) netdev_for_each_mc_addr(ha, ndev) { ether_addr_copy(self->mc_list.ar[i++], ha->addr); ++self->mc_list.count; + + if (i >= AQ_CFG_MULTICAST_ADDRESS_MAX) + break; } - return self->aq_hw_ops.hw_multicast_list_set(self->aq_hw, + if (i >= AQ_CFG_MULTICAST_ADDRESS_MAX) { + /* Number of filters is too big: atlantic does not support this. + * Force all multi filter to support this. + * With this we disable all UC filters and setup "all pass" + * multicast mask + */ + self->packet_filter |= IFF_ALLMULTI; + self->aq_hw->aq_nic_cfg->mc_list_count = 0; + return self->aq_hw_ops.hw_packet_filter_set(self->aq_hw, + self->packet_filter); + } else { + return self->aq_hw_ops.hw_multicast_list_set(self->aq_hw, self->mc_list.ar, self->mc_list.count); + } } int aq_nic_set_mtu(struct aq_nic_s *self, int new_mtu) -- cgit v1.1 From 6d3f58e09f0c1457aa5a6c60e5da08786dd8a18e Mon Sep 17 00:00:00 2001 From: Pavel Belous Date: Mon, 28 Aug 2017 21:52:13 +0300 Subject: net:ethernet:aquantia: Show info message if bad firmware version detected. We should inform user about wrong firmware version by printing message in dmesg. Fixes: 3d2ff7eebe26 ("net: ethernet: aquantia: Atlantic hardware abstraction layer") Signed-off-by: Pavel Belous Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c index 7a1332e..4f5ec9a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c @@ -141,6 +141,12 @@ static int hw_atl_utils_init_ucp(struct aq_hw_s *self, err = hw_atl_utils_ver_match(aq_hw_caps->fw_ver_expected, aq_hw_read_reg(self, 0x18U)); + + if (err < 0) + pr_err("%s: Bad FW version detected: expected=%x, actual=%x\n", + AQ_CFG_DRV_NAME, + aq_hw_caps->fw_ver_expected, + aq_hw_read_reg(self, 0x18U)); return err; } -- cgit v1.1 From e7562597b46d099a78d45c781ee8d1a7a93c53f1 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 29 Aug 2017 22:15:16 +0300 Subject: nfp: double free on error in probe Both the nfp_net_pf_app_start() and the nfp_net_pci_probe() functions call nfp_net_pf_app_stop_ctrl(pf) so there is a double free. The free should be done from the probe function because it's allocated there so I have removed the call from nfp_net_pf_app_start(). Fixes: 02082701b974 ("nfp: create control vNICs and wire up rx/tx") Signed-off-by: Dan Carpenter Acked-by: Jakub Kicinski Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_main.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c index 1aca4e5..34b9853 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_main.c @@ -458,7 +458,7 @@ static int nfp_net_pf_app_start(struct nfp_pf *pf) err = nfp_app_start(pf->app, pf->ctrl_vnic); if (err) - goto err_ctrl_stop; + return err; if (pf->num_vfs) { err = nfp_app_sriov_enable(pf->app, pf->num_vfs); @@ -470,8 +470,6 @@ static int nfp_net_pf_app_start(struct nfp_pf *pf) err_app_stop: nfp_app_stop(pf->app); -err_ctrl_stop: - nfp_net_pf_app_stop_ctrl(pf); return err; } -- cgit v1.1 From 183db481279437590f75a8a0479d512e5dd597de Mon Sep 17 00:00:00 2001 From: Quan Nguyen Date: Tue, 29 Aug 2017 15:43:12 -0700 Subject: drivers: net: xgene: Correct probe sequence handling The phy is connected at early stage of probe but not properly disconnected if error occurs. This patch fixes the issue. Also changing the return type of xgene_enet_check_phy_handle(), since this function always returns success. Signed-off-by: Quan Nguyen Signed-off-by: Iyappan Subramanian Signed-off-by: David S. Miller --- drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 27 ++++++++++++------------ 1 file changed, 13 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 1d307f2..6e253d9 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -1661,21 +1661,21 @@ static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata) return 0; } -static int xgene_enet_check_phy_handle(struct xgene_enet_pdata *pdata) +static void xgene_enet_check_phy_handle(struct xgene_enet_pdata *pdata) { int ret; if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) - return 0; + return; if (!IS_ENABLED(CONFIG_MDIO_XGENE)) - return 0; + return; ret = xgene_enet_phy_connect(pdata->ndev); if (!ret) pdata->mdio_driver = true; - return 0; + return; } static void xgene_enet_gpiod_get(struct xgene_enet_pdata *pdata) @@ -1779,10 +1779,6 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) if (ret) return ret; - ret = xgene_enet_check_phy_handle(pdata); - if (ret) - return ret; - xgene_enet_gpiod_get(pdata); pdata->clk = devm_clk_get(&pdev->dev, NULL); @@ -2097,9 +2093,11 @@ static int xgene_enet_probe(struct platform_device *pdev) goto err; } + xgene_enet_check_phy_handle(pdata); + ret = xgene_enet_init_hw(pdata); if (ret) - goto err; + goto err2; link_state = pdata->mac_ops->link_state; if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) { @@ -2117,29 +2115,30 @@ static int xgene_enet_probe(struct platform_device *pdev) spin_lock_init(&pdata->stats_lock); ret = xgene_extd_stats_init(pdata); if (ret) - goto err2; + goto err1; xgene_enet_napi_add(pdata); ret = register_netdev(ndev); if (ret) { netdev_err(ndev, "Failed to register netdev\n"); - goto err2; + goto err1; } return 0; -err2: +err1: /* * If necessary, free_netdev() will call netif_napi_del() and undo * the effects of xgene_enet_napi_add()'s calls to netif_napi_add(). */ + xgene_enet_delete_desc_rings(pdata); + +err2: if (pdata->mdio_driver) xgene_enet_phy_disconnect(pdata); else if (phy_interface_mode_is_rgmii(pdata->phy_mode)) xgene_enet_mdio_remove(pdata); -err1: - xgene_enet_delete_desc_rings(pdata); err: free_netdev(ndev); return ret; -- cgit v1.1 From a7c17d8ae4918afe527c141cc20ed34dfe18e7db Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Fri, 25 Aug 2017 15:43:45 +0300 Subject: mmc: block: Fix block status codes Commit 2a842acab109 ("block: introduce new block status code type") changed the error type but not in patches merged through the mmc tree, like commit 0493f6fe5bde ("mmc: block: Move boot partition locking into a driver op"). Fix one error code that is incorrect and also use BLK_STS_OK in preference to 0. Fixes: 17ece345a042 ("Merge tag 'mmc-v4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc") Signed-off-by: Adrian Hunter Signed-off-by: Ulf Hansson --- drivers/mmc/core/block.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c index 80d1ec6..8bd7aba 100644 --- a/drivers/mmc/core/block.c +++ b/drivers/mmc/core/block.c @@ -1213,7 +1213,7 @@ static void mmc_blk_issue_drv_op(struct mmc_queue *mq, struct request *req) break; } mq_rq->drv_op_result = ret; - blk_end_request_all(req, ret); + blk_end_request_all(req, ret ? BLK_STS_IOERR : BLK_STS_OK); } static void mmc_blk_issue_discard_rq(struct mmc_queue *mq, struct request *req) @@ -1718,9 +1718,9 @@ static bool mmc_blk_rw_cmd_err(struct mmc_blk_data *md, struct mmc_card *card, if (err) req_pending = old_req_pending; else - req_pending = blk_end_request(req, 0, blocks << 9); + req_pending = blk_end_request(req, BLK_STS_OK, blocks << 9); } else { - req_pending = blk_end_request(req, 0, brq->data.bytes_xfered); + req_pending = blk_end_request(req, BLK_STS_OK, brq->data.bytes_xfered); } return req_pending; } -- cgit v1.1 From 99c14fc360dbbb583a03ab985551b12b5c5ca4f1 Mon Sep 17 00:00:00 2001 From: Zhoujie Wu Date: Mon, 21 Aug 2017 11:02:09 -0700 Subject: mmc: sdhci-xenon: add set_power callback Xenon sdh controller requests proper SD bus voltage select bits programmed even with vmmc power supply. Any reserved value(100b-000b) programmed in this field will lead to controller ignore SD bus power bit and keep its value at zero. Add set_power callback to handle this. Signed-off-by: Zhoujie Wu Acked-by: Adrian Hunter Tested-by: Gregory CLEMENT Fixes: 3a3748dba881 ("mmc: sdhci-xenon: Add Marvell Xenon SDHC core functionality") Cc: Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-xenon.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/drivers/mmc/host/sdhci-xenon.c b/drivers/mmc/host/sdhci-xenon.c index bc1781b..c580af0 100644 --- a/drivers/mmc/host/sdhci-xenon.c +++ b/drivers/mmc/host/sdhci-xenon.c @@ -210,8 +210,27 @@ static void xenon_set_uhs_signaling(struct sdhci_host *host, sdhci_writew(host, ctrl_2, SDHCI_HOST_CONTROL2); } +static void xenon_set_power(struct sdhci_host *host, unsigned char mode, + unsigned short vdd) +{ + struct mmc_host *mmc = host->mmc; + u8 pwr = host->pwr; + + sdhci_set_power_noreg(host, mode, vdd); + + if (host->pwr == pwr) + return; + + if (host->pwr == 0) + vdd = 0; + + if (!IS_ERR(mmc->supply.vmmc)) + mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd); +} + static const struct sdhci_ops sdhci_xenon_ops = { .set_clock = sdhci_set_clock, + .set_power = xenon_set_power, .set_bus_width = sdhci_set_bus_width, .reset = xenon_reset, .set_uhs_signaling = xenon_set_uhs_signaling, -- cgit v1.1 From b925a2dc165e5ec2330ca1256704faef8ed96913 Mon Sep 17 00:00:00 2001 From: Max Gurtovoy Date: Mon, 28 Aug 2017 12:52:27 +0300 Subject: nvme-rdma: default MR page size to 4k Due to various page sizes in the system (IOMMU/device/kernel), we set the fabrics controller page size to 4k and block layer boundaries accordinglly. In architectures that uses different kernel page size we'll have a mismatch to the MR page size that may cause a mapping error. Update the MR page size to correspond to the core ctrl settings. Signed-off-by: Max Gurtovoy Reviewed-by: Sagi Grimberg Signed-off-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index da04df1..a03299d 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -920,7 +920,11 @@ static int nvme_rdma_map_sg_fr(struct nvme_rdma_queue *queue, struct nvme_keyed_sgl_desc *sg = &c->common.dptr.ksgl; int nr; - nr = ib_map_mr_sg(req->mr, req->sg_table.sgl, count, NULL, PAGE_SIZE); + /* + * Align the MR to a 4K page size to match the ctrl page size and + * the block virtual boundary. + */ + nr = ib_map_mr_sg(req->mr, req->sg_table.sgl, count, NULL, SZ_4K); if (nr < count) { if (nr < 0) return nr; @@ -1583,7 +1587,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl) goto out_cleanup_queue; ctrl->ctrl.max_hw_sectors = - (ctrl->max_fr_pages - 1) << (PAGE_SHIFT - 9); + (ctrl->max_fr_pages - 1) << (ilog2(SZ_4K) - 9); error = nvme_init_identify(&ctrl->ctrl); if (error) -- cgit v1.1 From 4033f35d174af4804a79fd5731d9e6be976f9f28 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 28 Aug 2017 10:47:18 +0200 Subject: nvme-pci: use dma memory for the host memory buffer descriptors The NVMe 1.3 specification says in section 5.21.1.13: "After a successful completion of a Set Features enabling the host memory buffer, the host shall not write to the associated host memory region, buffer size, or descriptor list until the host memory buffer has been disabled." While this doesn't state that the descriptor list must remain accessible to the device it certainly implies it must remaing readable by the device. So switch to a dma coherent allocation for the descriptor list just to be safe - it's not like the cost for it matters compared to the actual memory buffers. Signed-off-by: Christoph Hellwig Reviewed-by: Keith Busch Reviewed-by: Sagi Grimberg Reviewed-by: Johannes Thumshirn Fixes: 87ad72a59a38 ("nvme-pci: implement host memory buffer support") --- drivers/nvme/host/pci.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 925467b..ea892e7 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -109,6 +109,7 @@ struct nvme_dev { /* host memory buffer support: */ u64 host_mem_size; u32 nr_host_mem_descs; + dma_addr_t host_mem_descs_dma; struct nvme_host_mem_buf_desc *host_mem_descs; void **host_mem_desc_bufs; }; @@ -1565,16 +1566,10 @@ static inline void nvme_release_cmb(struct nvme_dev *dev) static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits) { - size_t len = dev->nr_host_mem_descs * sizeof(*dev->host_mem_descs); + u64 dma_addr = dev->host_mem_descs_dma; struct nvme_command c; - u64 dma_addr; int ret; - dma_addr = dma_map_single(dev->dev, dev->host_mem_descs, len, - DMA_TO_DEVICE); - if (dma_mapping_error(dev->dev, dma_addr)) - return -ENOMEM; - memset(&c, 0, sizeof(c)); c.features.opcode = nvme_admin_set_features; c.features.fid = cpu_to_le32(NVME_FEAT_HOST_MEM_BUF); @@ -1591,7 +1586,6 @@ static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits) "failed to set host mem (err %d, flags %#x).\n", ret, bits); } - dma_unmap_single(dev->dev, dma_addr, len, DMA_TO_DEVICE); return ret; } @@ -1609,7 +1603,9 @@ static void nvme_free_host_mem(struct nvme_dev *dev) kfree(dev->host_mem_desc_bufs); dev->host_mem_desc_bufs = NULL; - kfree(dev->host_mem_descs); + dma_free_coherent(dev->dev, + dev->nr_host_mem_descs * sizeof(*dev->host_mem_descs), + dev->host_mem_descs, dev->host_mem_descs_dma); dev->host_mem_descs = NULL; } @@ -1617,6 +1613,7 @@ static int nvme_alloc_host_mem(struct nvme_dev *dev, u64 min, u64 preferred) { struct nvme_host_mem_buf_desc *descs; u32 chunk_size, max_entries, len; + dma_addr_t descs_dma; int i = 0; void **bufs; u64 size = 0, tmp; @@ -1627,7 +1624,8 @@ retry: tmp = (preferred + chunk_size - 1); do_div(tmp, chunk_size); max_entries = tmp; - descs = kcalloc(max_entries, sizeof(*descs), GFP_KERNEL); + descs = dma_zalloc_coherent(dev->dev, max_entries * sizeof(*descs), + &descs_dma, GFP_KERNEL); if (!descs) goto out; @@ -1661,6 +1659,7 @@ retry: dev->nr_host_mem_descs = i; dev->host_mem_size = size; dev->host_mem_descs = descs; + dev->host_mem_descs_dma = descs_dma; dev->host_mem_desc_bufs = bufs; return 0; @@ -1674,7 +1673,8 @@ out_free_bufs: kfree(bufs); out_free_descs: - kfree(descs); + dma_free_coherent(dev->dev, max_entries * sizeof(*descs), descs, + descs_dma); out: /* try a smaller chunk size if we failed early */ if (chunk_size >= PAGE_SIZE * 2 && (i == 0 || size < min)) { -- cgit v1.1 From 223694b9ae8bfba99f3528d49d07a740af6ff95a Mon Sep 17 00:00:00 2001 From: Changpeng Liu Date: Thu, 31 Aug 2017 11:22:49 +0800 Subject: nvme: fix the definition of the doorbell buffer config support bit NVMe 1.3 specification defines the Optional Admin Command Support feature flags, bit 8 set to '1' then the controller supports the Doorbell Buffer Config command. Bit 7 is used for Virtualization Mangement command. Signed-off-by: Changpeng Liu Reviewed-by: Sagi Grimberg Reviewed-by: Max Gurtovoy Reviewed-by: Johannes Thumshirn Signed-off-by: Christoph Hellwig Fixes: f9f38e33 ("nvme: improve performance for virtual NVMe devices") Cc: stable@vger.kernel.org --- include/linux/nvme.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 25d8225..8efff88 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -254,7 +254,7 @@ enum { NVME_CTRL_VWC_PRESENT = 1 << 0, NVME_CTRL_OACS_SEC_SUPP = 1 << 0, NVME_CTRL_OACS_DIRECTIVES = 1 << 5, - NVME_CTRL_OACS_DBBUF_SUPP = 1 << 7, + NVME_CTRL_OACS_DBBUF_SUPP = 1 << 8, }; struct nvme_lbaf { -- cgit v1.1 From bcab3a6e64a9647e022b46182c9687a8c3aa9d11 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 30 Aug 2017 14:50:07 +0200 Subject: ALSA: pcm: Fix power lock unbalance via OSS emulation PCM OSS emulation issues the drain ioctl without power lock. It used to work in the earlier kernels as the power lock was taken inside snd_pcm_drain() itself. But since 68b4acd32249 ("ALSA: pcm: Apply power lock globally to common ioctls"), the power lock is taken outside the function. Due to that change, the call via OSS emulation leads to the unbalanced power lock, thus it deadlocks. As a quick fix, just take the power lock before snd_pcm_drain() call for OSS emulation path. A better cleanup will follow later. Fixes: 68b4acd32249 ("ALSA: pcm: Apply power lock globally to common ioctls") Reported-and-tested-by: Markus Trippelsdorf Signed-off-by: Takashi Iwai --- sound/core/pcm_native.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index 22995cb..cf0433f 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -3064,6 +3064,7 @@ int snd_pcm_kernel_ioctl(struct snd_pcm_substream *substream, { snd_pcm_uframes_t *frames = arg; snd_pcm_sframes_t result; + int err; switch (cmd) { case SNDRV_PCM_IOCTL_FORWARD: @@ -3083,7 +3084,10 @@ int snd_pcm_kernel_ioctl(struct snd_pcm_substream *substream, case SNDRV_PCM_IOCTL_START: return snd_pcm_start_lock_irq(substream); case SNDRV_PCM_IOCTL_DRAIN: - return snd_pcm_drain(substream, NULL); + snd_power_lock(substream->pcm->card); + err = snd_pcm_drain(substream, NULL); + snd_power_unlock(substream->pcm->card); + return err; case SNDRV_PCM_IOCTL_DROP: return snd_pcm_drop(substream); case SNDRV_PCM_IOCTL_DELAY: -- cgit v1.1 From f0e82d737d1b727e93adec16cf7132391249356a Mon Sep 17 00:00:00 2001 From: Sekhar Nori Date: Wed, 30 Aug 2017 13:37:13 +0530 Subject: net: ti: cpsw-common: dont print error if ti_cm_get_macid() fails It is quite common for ti_cm_get_macid() to fail on some of the platforms it is invoked on. They include any platform where mac address is not part of SoC register space. On these platforms, mac address is read and populated in device-tree by bootloader. An example is TI DA850. Downgrade the severity of message to "information", so it does not spam logs when 'quiet' boot is desired. Signed-off-by: Sekhar Nori Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/cpsw-common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/cpsw-common.c b/drivers/net/ethernet/ti/cpsw-common.c index 56ba411..38d1cc5 100644 --- a/drivers/net/ethernet/ti/cpsw-common.c +++ b/drivers/net/ethernet/ti/cpsw-common.c @@ -96,7 +96,7 @@ int ti_cm_get_macid(struct device *dev, int slave, u8 *mac_addr) if (of_machine_is_compatible("ti,dra7")) return davinci_emac_3517_get_macid(dev, 0x514, slave, mac_addr); - dev_err(dev, "incompatible machine/device type for reading mac address\n"); + dev_info(dev, "incompatible machine/device type for reading mac address\n"); return -ENOENT; } EXPORT_SYMBOL_GPL(ti_cm_get_macid); -- cgit v1.1 From 1b4a56cd7f5c2f835ffae2a8204fb95597269b10 Mon Sep 17 00:00:00 2001 From: Lucas Stach Date: Tue, 29 Aug 2017 17:51:22 +0200 Subject: ASoC: simple_card_utils: fix fallback when "label" property isn't present snd_soc_of_parse_card_name() doesn't return an error if the requested property isn't present, but silently fails to fill the card name. This can not be changed, as it is a backwards compatibility measure itself. We can not rely on the return value of this function alone, but must check if the card name has been filled sucessfully when deciding to skip the fallback path, which is in place for existing users. Fixes: dedfaa1eedc7 (ASoC: simple-card-utils: enable "label" on asoc_simple_card_parse_card_name) Signed-off-by: Lucas Stach Acked-by: Kuninori Morimoto Signed-off-by: Mark Brown --- sound/soc/generic/simple-card-utils.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/generic/simple-card-utils.c b/sound/soc/generic/simple-card-utils.c index 26d64fa..0442099 100644 --- a/sound/soc/generic/simple-card-utils.c +++ b/sound/soc/generic/simple-card-utils.c @@ -132,7 +132,7 @@ int asoc_simple_card_parse_card_name(struct snd_soc_card *card, /* Parse the card name from DT */ ret = snd_soc_of_parse_card_name(card, "label"); - if (ret < 0) { + if (ret < 0 || !card->name) { char prop[128]; snprintf(prop, sizeof(prop), "%sname", prefix); -- cgit v1.1 From 33c52b6718d2a6cb414440c98560818910d896dc Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Thu, 13 Jul 2017 13:45:11 -0500 Subject: net/mlx5e: Check for qos capability in dcbnl_initialize qos capability is the master capability bit that determines if the DCBX is supported for the PCI function. If this bit is off, driver cannot run any dcbx code. Fixes: e207b7e99176 ("net/mlx5e: ConnectX-4 firmware support for DCBX") Signed-off-by: Huy Nguyen Reviewed-by: Parav Pandit Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 2eb54d3..810b510 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -754,6 +754,9 @@ void mlx5e_dcbnl_initialize(struct mlx5e_priv *priv) { struct mlx5e_dcbx *dcbx = &priv->dcbx; + if (!MLX5_CAP_GEN(priv->mdev, qos)) + return; + if (MLX5_CAP_GEN(priv->mdev, dcbx)) mlx5e_dcbnl_query_dcbx_mode(priv, &dcbx->mode); -- cgit v1.1 From 9e10bf1d349787f373484d835efe2dbb5f9c5614 Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Mon, 10 Jul 2017 14:00:23 -0500 Subject: net/mlx5e: Fix DCB_CAP_ATTR_DCBX capability for DCBNL getcap. Current code doesn't report DCB_CAP_DCBX_HOST capability when query through getcap. User space lldptool expects capability to have HOST mode set when it wants to configure DCBX CEE mode. In absence of HOST mode capability, lldptool fails to switch to CEE mode. This fix returns DCB_CAP_DCBX_HOST capability when port's DCBX controlled mode is under software control. Fixes: 3a6a931dfb8e ("net/mlx5e: Support DCBX CEE API") Signed-off-by: Huy Nguyen Reviewed-by: Parav Pandit Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 1 + drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 21 ++++++++++++--------- 2 files changed, 13 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 0039b47..2f26fb3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -263,6 +263,7 @@ struct mlx5e_dcbx { /* The only setting that cannot be read from FW */ u8 tc_tsa[IEEE_8021QAZ_MAX_TCS]; + u8 cap; }; #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 810b510..c1d384f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -288,13 +288,8 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, static u8 mlx5e_dcbnl_getdcbx(struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); - struct mlx5e_dcbx *dcbx = &priv->dcbx; - u8 mode = DCB_CAP_DCBX_VER_IEEE | DCB_CAP_DCBX_VER_CEE; - - if (dcbx->mode == MLX5E_DCBX_PARAM_VER_OPER_HOST) - mode |= DCB_CAP_DCBX_HOST; - return mode; + return priv->dcbx.cap; } static u8 mlx5e_dcbnl_setdcbx(struct net_device *dev, u8 mode) @@ -312,6 +307,7 @@ static u8 mlx5e_dcbnl_setdcbx(struct net_device *dev, u8 mode) /* set dcbx to fw controlled */ if (!mlx5e_dcbnl_set_dcbx_mode(priv, MLX5E_DCBX_PARAM_VER_OPER_AUTO)) { dcbx->mode = MLX5E_DCBX_PARAM_VER_OPER_AUTO; + dcbx->cap &= ~DCB_CAP_DCBX_HOST; return 0; } @@ -324,6 +320,8 @@ static u8 mlx5e_dcbnl_setdcbx(struct net_device *dev, u8 mode) if (mlx5e_dcbnl_switch_to_host_mode(netdev_priv(dev))) return 1; + dcbx->cap = mode; + return 0; } @@ -628,9 +626,9 @@ static u8 mlx5e_dcbnl_getcap(struct net_device *netdev, *cap = false; break; case DCB_CAP_ATTR_DCBX: - *cap = (DCB_CAP_DCBX_LLD_MANAGED | - DCB_CAP_DCBX_VER_CEE | - DCB_CAP_DCBX_STATIC); + *cap = priv->dcbx.cap | + DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_VER_IEEE; break; default: *cap = 0; @@ -760,5 +758,10 @@ void mlx5e_dcbnl_initialize(struct mlx5e_priv *priv) if (MLX5_CAP_GEN(priv->mdev, dcbx)) mlx5e_dcbnl_query_dcbx_mode(priv, &dcbx->mode); + priv->dcbx.cap = DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_VER_IEEE; + if (priv->dcbx.mode == MLX5E_DCBX_PARAM_VER_OPER_HOST) + priv->dcbx.cap |= DCB_CAP_DCBX_HOST; + mlx5e_ets_init(priv); } -- cgit v1.1 From 672d0880b7798a917bcc622308f25a0fbb991dab Mon Sep 17 00:00:00 2001 From: Noa Osherovich Date: Sun, 30 Jul 2017 13:55:48 +0300 Subject: net/mlx5: Fix arm SRQ command for ISSI version 0 Support for ISSI version 0 was recently broken as the arm_srq_cmd command, which is used only for ISSI version 0, was given the opcode for ISSI version 1 instead of ISSI version 0. Change arm_srq_cmd to use the correct command opcode for ISSI version 0. Fixes: af1ba291c5e4 ('{net, IB}/mlx5: Refactor internal SRQ API') Signed-off-by: Noa Osherovich Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/srq.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/srq.c b/drivers/net/ethernet/mellanox/mlx5/core/srq.c index f774de6..520f638 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/srq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/srq.c @@ -201,13 +201,13 @@ static int destroy_srq_cmd(struct mlx5_core_dev *dev, static int arm_srq_cmd(struct mlx5_core_dev *dev, struct mlx5_core_srq *srq, u16 lwm, int is_srq) { - /* arm_srq structs missing using identical xrc ones */ - u32 srq_in[MLX5_ST_SZ_DW(arm_xrc_srq_in)] = {0}; - u32 srq_out[MLX5_ST_SZ_DW(arm_xrc_srq_out)] = {0}; + u32 srq_in[MLX5_ST_SZ_DW(arm_rq_in)] = {0}; + u32 srq_out[MLX5_ST_SZ_DW(arm_rq_out)] = {0}; - MLX5_SET(arm_xrc_srq_in, srq_in, opcode, MLX5_CMD_OP_ARM_XRC_SRQ); - MLX5_SET(arm_xrc_srq_in, srq_in, xrc_srqn, srq->srqn); - MLX5_SET(arm_xrc_srq_in, srq_in, lwm, lwm); + MLX5_SET(arm_rq_in, srq_in, opcode, MLX5_CMD_OP_ARM_RQ); + MLX5_SET(arm_rq_in, srq_in, op_mod, MLX5_ARM_RQ_IN_OP_MOD_SRQ); + MLX5_SET(arm_rq_in, srq_in, srq_number, srq->srqn); + MLX5_SET(arm_rq_in, srq_in, lwm, lwm); return mlx5_cmd_exec(dev, srq_in, sizeof(srq_in), srq_out, sizeof(srq_out)); -- cgit v1.1 From b3cb5388499c5e219324bfe7da2e46cbad82bfcf Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Tue, 8 Aug 2017 13:17:00 -0500 Subject: net/mlx5: Skip mlx5_unload_one if mlx5_load_one fails There is an issue where the firmware fails during mlx5_load_one, the health_care timer detects the issue and schedules a health_care call. Then the mlx5_load_one detects the issue, cleans up and quits. Then the health_care starts and calls mlx5_unload_one to clean up the resources that no longer exist and causes kernel panic. The root cause is that the bit MLX5_INTERFACE_STATE_DOWN is not set after mlx5_load_one fails. The solution is removing the bit MLX5_INTERFACE_STATE_DOWN and quit mlx5_unload_one if the bit MLX5_INTERFACE_STATE_UP is not set. The bit MLX5_INTERFACE_STATE_DOWN is redundant and we can use MLX5_INTERFACE_STATE_UP instead. Fixes: 5fc7197d3a25 ("net/mlx5: Add pci shutdown callback") Signed-off-by: Huy Nguyen Reviewed-by: Daniel Jurgens Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 4 +--- include/linux/mlx5/driver.h | 5 ++--- 2 files changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index c065132..4cdb414 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1186,7 +1186,6 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv, } } - clear_bit(MLX5_INTERFACE_STATE_DOWN, &dev->intf_state); set_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state); out: mutex_unlock(&dev->intf_state_mutex); @@ -1261,7 +1260,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv, mlx5_drain_health_recovery(dev); mutex_lock(&dev->intf_state_mutex); - if (test_bit(MLX5_INTERFACE_STATE_DOWN, &dev->intf_state)) { + if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) { dev_warn(&dev->pdev->dev, "%s: interface is down, NOP\n", __func__); if (cleanup) @@ -1270,7 +1269,6 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv, } clear_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state); - set_bit(MLX5_INTERFACE_STATE_DOWN, &dev->intf_state); if (mlx5_device_registered(dev)) mlx5_detach_device(dev); diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index df6ce59..918f5e6 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -673,9 +673,8 @@ enum mlx5_device_state { }; enum mlx5_interface_state { - MLX5_INTERFACE_STATE_DOWN = BIT(0), - MLX5_INTERFACE_STATE_UP = BIT(1), - MLX5_INTERFACE_STATE_SHUTDOWN = BIT(2), + MLX5_INTERFACE_STATE_UP = BIT(0), + MLX5_INTERFACE_STATE_SHUTDOWN = BIT(1), }; enum mlx5_pci_status { -- cgit v1.1 From 10a8d00707082955b177164d4b4e758ffcbd4017 Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Wed, 9 Aug 2017 10:03:40 -0500 Subject: net/mlx5: Remove the flag MLX5_INTERFACE_STATE_SHUTDOWN MLX5_INTERFACE_STATE_SHUTDOWN is not used in the code. Fixes: 5fc7197d3a25 ("net/mlx5: Add pci shutdown callback") Signed-off-by: Huy Nguyen Reviewed-by: Daniel Jurgens Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 -- include/linux/mlx5/driver.h | 1 - 2 files changed, 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 4cdb414..1688582 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1563,8 +1563,6 @@ static void shutdown(struct pci_dev *pdev) int err; dev_info(&pdev->dev, "Shutdown was called\n"); - /* Notify mlx5 clients that the kernel is being shut down */ - set_bit(MLX5_INTERFACE_STATE_SHUTDOWN, &dev->intf_state); err = mlx5_try_fast_unload(dev); if (err) mlx5_unload_one(dev, priv, false); diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 918f5e6..205d82d 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -674,7 +674,6 @@ enum mlx5_device_state { enum mlx5_interface_state { MLX5_INTERFACE_STATE_UP = BIT(0), - MLX5_INTERFACE_STATE_SHUTDOWN = BIT(1), }; enum mlx5_pci_status { -- cgit v1.1 From 0556ce72ab16156af6c94cdc7964e4310acc97c0 Mon Sep 17 00:00:00 2001 From: Eran Ben Elisha Date: Wed, 16 Aug 2017 14:37:11 +0300 Subject: net/mlx5e: Fix dangling page pointer on DMA mapping error Function mlx5e_dealloc_rx_wqe is using page pointer value as an indication to valid DMA mapping. In case that the mapping failed, we released the page but kept the dangling pointer. Store the page pointer only after the DMA mapping passed to avoid invalid page DMA unmap. Fixes: bc77b240b3c5 ("net/mlx5e: Add fragmented memory support for RX multi packet WQE") Signed-off-by: Eran Ben Elisha Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 325b2c8..7344433 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -222,13 +222,13 @@ static inline int mlx5e_page_alloc_mapped(struct mlx5e_rq *rq, if (unlikely(!page)) return -ENOMEM; - dma_info->page = page; dma_info->addr = dma_map_page(rq->pdev, page, 0, RQ_PAGE_SIZE(rq), rq->buff.map_dir); if (unlikely(dma_mapping_error(rq->pdev, dma_info->addr))) { put_page(page); return -ENOMEM; } + dma_info->page = page; return 0; } -- cgit v1.1 From 5a8e12678c767ccf8bb16d6237569e4a707d655b Mon Sep 17 00:00:00 2001 From: Inbar Karmy Date: Mon, 14 Aug 2017 16:12:16 +0300 Subject: net/mlx5e: Don't override user RSS upon set channels Currently, increasing the number of combined channels is changing the RSS spread to use the new created channels. Prevent the RSS spread change in case the user explicitly declare it, to avoid overriding user configuration. Tested: when RSS default: # ethtool -L ens8 combined 4 RSS spread will change and point to 4 channels. # ethtool -X ens8 equal 4 # ethtool -L ens8 combined 6 RSS will not change after increasing the number of the channels. Fixes: 8bf368620486 ('ethtool: ensure channel counts are within bounds during SCHANNELS') Signed-off-by: Inbar Karmy Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 917fade..f559401 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -641,8 +641,10 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv, new_channels.params = priv->channels.params; new_channels.params.num_channels = count; - mlx5e_build_default_indir_rqt(priv->mdev, new_channels.params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, count); + if (!netif_is_rxfh_configured(priv->netdev)) + mlx5e_build_default_indir_rqt(priv->mdev, + new_channels.params.indirection_rqt, + MLX5E_INDIR_RQT_SIZE, count); if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { priv->channels.params = new_channels.params; -- cgit v1.1 From 08820528c9d3ff0d0eda047d7ef5ecac2da1ef6c Mon Sep 17 00:00:00 2001 From: Paul Blakey Date: Tue, 22 Aug 2017 13:51:56 +0300 Subject: net/mlx5e: Properly resolve TC offloaded ipv6 vxlan tunnel source address Currently if vxlan tunnel ipv6 src isn't supplied the driver fails to resolve it as part of the route lookup. The resulting encap header is left with a zeroed out ipv6 src address so the packets are sent with this src ip. Use an appropriate route lookup API that also resolves the source ipv6 address if it's not supplied. Fixes: ce99f6b97fcd ('net/mlx5e: Support SRIOV TC encapsulation offloads for IPv6 tunnels') Signed-off-by: Paul Blakey Reviewed-by: Or Gerlitz Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 3c536f5..7f282e8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -1443,12 +1443,10 @@ static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv, struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; int ret; - dst = ip6_route_output(dev_net(mirred_dev), NULL, fl6); - ret = dst->error; - if (ret) { - dst_release(dst); + ret = ipv6_stub->ipv6_dst_lookup(dev_net(mirred_dev), NULL, &dst, + fl6); + if (ret < 0) return ret; - } *out_ttl = ip6_dst_hoplimit(dst); -- cgit v1.1 From 191220396db840822fc818edf03c49f0c02eb237 Mon Sep 17 00:00:00 2001 From: Shahar Klein Date: Tue, 1 Aug 2017 15:29:55 +0300 Subject: net/mlx5: E-Switch, Unload the representors in the correct order When changing from switchdev to legacy mode, all the representor port devices (uplink nic and reps) are cleaned up. Part of this cleaning process is removing the neigh entries and the hash table containing them. However, a representor neigh entry might be linked to the uplink port hash table and if the uplink nic is cleaned first the cleaning of the representor will end up in null deref. Fix that by unloading the representors in the opposite order of load. Fixes: cb67b832921c ("net/mlx5e: Introduce SRIOV VF representors") Signed-off-by: Shahar Klein Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 95b6402..5bc0593 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -815,7 +815,7 @@ void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports) struct mlx5_eswitch_rep *rep; int vport; - for (vport = 0; vport < nvports; vport++) { + for (vport = nvports - 1; vport >= 0; vport--) { rep = &esw->offloads.vport_reps[vport]; if (!rep->valid) continue; -- cgit v1.1 From 6aace17e64f4aa1c49802c46bd10688968b3787f Mon Sep 17 00:00:00 2001 From: Moshe Shemesh Date: Tue, 8 Aug 2017 15:56:37 +0300 Subject: net/mlx5e: Fix inline header size for small packets Fix inline header size, make sure it is not greater than skb len. This bug effects small packets, for example L2 packets with size < 18. Fixes: ae76715d153e ("net/mlx5e: Check the minimum inline header mode before xmit") Signed-off-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index aaa0f4e..31353e5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -128,10 +128,10 @@ static inline int mlx5e_skb_l3_header_offset(struct sk_buff *skb) return mlx5e_skb_l2_header_offset(skb); } -static inline unsigned int mlx5e_calc_min_inline(enum mlx5_inline_modes mode, - struct sk_buff *skb) +static inline u16 mlx5e_calc_min_inline(enum mlx5_inline_modes mode, + struct sk_buff *skb) { - int hlen; + u16 hlen; switch (mode) { case MLX5_INLINE_MODE_NONE: @@ -140,19 +140,22 @@ static inline unsigned int mlx5e_calc_min_inline(enum mlx5_inline_modes mode, hlen = eth_get_headlen(skb->data, skb_headlen(skb)); if (hlen == ETH_HLEN && !skb_vlan_tag_present(skb)) hlen += VLAN_HLEN; - return hlen; + break; case MLX5_INLINE_MODE_IP: /* When transport header is set to zero, it means no transport * header. When transport header is set to 0xff's, it means * transport header wasn't set. */ - if (skb_transport_offset(skb)) - return mlx5e_skb_l3_header_offset(skb); + if (skb_transport_offset(skb)) { + hlen = mlx5e_skb_l3_header_offset(skb); + break; + } /* fall through */ case MLX5_INLINE_MODE_L2: default: - return mlx5e_skb_l2_header_offset(skb); + hlen = mlx5e_skb_l2_header_offset(skb); } + return min_t(u16, hlen, skb->len); } static inline void mlx5e_tx_skb_pull_inline(unsigned char **skb_data, -- cgit v1.1 From 1213ad28f9595a08e3877248bbba1a25c40225d6 Mon Sep 17 00:00:00 2001 From: Tal Gilboa Date: Mon, 28 Aug 2017 18:45:08 +0300 Subject: net/mlx5e: Fix CQ moderation mode not set properly cq_period_mode assignment was mistakenly removed so it was always set to "0", which is EQE based moderation, regardless of the device CAPs and requested value in ethtool. Fixes: 6a9764efb255 ("net/mlx5e: Isolate open_channels from priv->params") Signed-off-by: Tal Gilboa Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 57f31fa..6ad7f07 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1969,6 +1969,7 @@ static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, } mlx5e_build_common_cq_param(priv, param); + param->cq_period_mode = params->rx_cq_period_mode; } static void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, -- cgit v1.1 From 9e37b1784f2be9397a903307574ee565bbadfd75 Mon Sep 17 00:00:00 2001 From: Pavel Shilovsky Date: Thu, 24 Aug 2017 15:16:40 -0700 Subject: CIFS: Fix maximum SMB2 header size Currently the maximum size of SMB2/3 header is set incorrectly which leads to hanging of directory listing operations on encrypted SMB3 connections. Fix this by setting the maximum size to 170 bytes that is calculated as RFC1002 length field size (4) + transform header size (52) + SMB2 header size (64) + create response size (56). Cc: Signed-off-by: Pavel Shilovsky Signed-off-by: Steve French Acked-by: Sachin Prabhu --- fs/cifs/smb2pdu.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h index 18700fd..2826882 100644 --- a/fs/cifs/smb2pdu.h +++ b/fs/cifs/smb2pdu.h @@ -84,8 +84,8 @@ #define NUMBER_OF_SMB2_COMMANDS 0x0013 -/* BB FIXME - analyze following length BB */ -#define MAX_SMB2_HDR_SIZE 0x78 /* 4 len + 64 hdr + (2*24 wct) + 2 bct + 2 pad */ +/* 4 len + 52 transform hdr + 64 hdr + 56 create rsp */ +#define MAX_SMB2_HDR_SIZE 0x00b0 #define SMB2_PROTO_NUMBER cpu_to_le32(0x424d53fe) #define SMB2_TRANSFORM_PROTO_NUM cpu_to_le32(0x424d53fd) -- cgit v1.1 From 6e3c1529c39e92ed64ca41d53abadabbaa1d5393 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 27 Aug 2017 16:56:08 -0500 Subject: CIFS: remove endian related sparse warning Recent patch had an endian warning ie cifs: return ENAMETOOLONG for overlong names in cifs_open()/cifs_lookup() Signed-off-by: Steve French CC: Ronnie Sahlberg CC: Stable Acked-by: Pavel Shilovsky --- fs/cifs/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 569d3fb..e702d48 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -205,7 +205,7 @@ check_name(struct dentry *direntry, struct cifs_tcon *tcon) int i; if (unlikely(direntry->d_name.len > - tcon->fsAttrInfo.MaxPathNameComponentLength)) + le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength))) return -ENAMETOOLONG; if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS)) { -- cgit v1.1 From 88c2ace69dbef696edba77712882af03879abc9c Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:48:57 +0300 Subject: sch_htb: fix crash on init failure The commit below added a call to the ->destroy() callback for all qdiscs which failed in their ->init(), but some were not prepared for such change and can't handle partially initialized qdisc. HTB is one of them and if any error occurs before the qdisc watchdog timer and qdisc work are initialized then we can hit either a null ptr deref (timer->base) when canceling in ->destroy or lockdep error info about trying to register a non-static key and a stack dump. So to fix these two move the watchdog timer and workqueue init before anything that can err out. To reproduce userspace needs to send broken htb qdisc create request, tested with a modified tc (q_htb.c). Trace log: [ 2710.897602] BUG: unable to handle kernel NULL pointer dereference at (null) [ 2710.897977] IP: hrtimer_active+0x17/0x8a [ 2710.898174] PGD 58fab067 [ 2710.898175] P4D 58fab067 [ 2710.898353] PUD 586c0067 [ 2710.898531] PMD 0 [ 2710.898710] [ 2710.899045] Oops: 0000 [#1] SMP [ 2710.899232] Modules linked in: [ 2710.899419] CPU: 1 PID: 950 Comm: tc Not tainted 4.13.0-rc6+ #54 [ 2710.899646] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 2710.900035] task: ffff880059ed2700 task.stack: ffff88005ad4c000 [ 2710.900262] RIP: 0010:hrtimer_active+0x17/0x8a [ 2710.900467] RSP: 0018:ffff88005ad4f960 EFLAGS: 00010246 [ 2710.900684] RAX: 0000000000000000 RBX: ffff88003701e298 RCX: 0000000000000000 [ 2710.900933] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88003701e298 [ 2710.901177] RBP: ffff88005ad4f980 R08: 0000000000000001 R09: 0000000000000001 [ 2710.901419] R10: ffff88005ad4f800 R11: 0000000000000400 R12: 0000000000000000 [ 2710.901663] R13: ffff88003701e298 R14: ffffffff822a4540 R15: ffff88005ad4fac0 [ 2710.901907] FS: 00007f2f5e90f740(0000) GS:ffff88005d880000(0000) knlGS:0000000000000000 [ 2710.902277] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 2710.902500] CR2: 0000000000000000 CR3: 0000000058ca3000 CR4: 00000000000406e0 [ 2710.902744] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 2710.902977] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 2710.903180] Call Trace: [ 2710.903332] hrtimer_try_to_cancel+0x1a/0x93 [ 2710.903504] hrtimer_cancel+0x15/0x20 [ 2710.903667] qdisc_watchdog_cancel+0x12/0x14 [ 2710.903866] htb_destroy+0x2e/0xf7 [ 2710.904097] qdisc_create+0x377/0x3fd [ 2710.904330] tc_modify_qdisc+0x4d2/0x4fd [ 2710.904511] rtnetlink_rcv_msg+0x188/0x197 [ 2710.904682] ? rcu_read_unlock+0x3e/0x5f [ 2710.904849] ? rtnl_newlink+0x729/0x729 [ 2710.905017] netlink_rcv_skb+0x6c/0xce [ 2710.905183] rtnetlink_rcv+0x23/0x2a [ 2710.905345] netlink_unicast+0x103/0x181 [ 2710.905511] netlink_sendmsg+0x326/0x337 [ 2710.905679] sock_sendmsg_nosec+0x14/0x3f [ 2710.905847] sock_sendmsg+0x29/0x2e [ 2710.906010] ___sys_sendmsg+0x209/0x28b [ 2710.906176] ? do_raw_spin_unlock+0xcd/0xf8 [ 2710.906346] ? _raw_spin_unlock+0x27/0x31 [ 2710.906514] ? __handle_mm_fault+0x651/0xdb1 [ 2710.906685] ? check_chain_key+0xb0/0xfd [ 2710.906855] __sys_sendmsg+0x45/0x63 [ 2710.907018] ? __sys_sendmsg+0x45/0x63 [ 2710.907185] SyS_sendmsg+0x19/0x1b [ 2710.907344] entry_SYSCALL_64_fastpath+0x23/0xc2 Note that probably this bug goes further back because the default qdisc handling always calls ->destroy on init failure too. Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Fixes: 0fbbeb1ba43b ("[PKT_SCHED]: Fix missing qdisc_destroy() in qdisc_create_dflt()") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_htb.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 5d65ec5..5bf5177 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1017,6 +1017,9 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt) int err; int i; + qdisc_watchdog_init(&q->watchdog, sch); + INIT_WORK(&q->work, htb_work_func); + if (!opt) return -EINVAL; @@ -1041,8 +1044,6 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt) for (i = 0; i < TC_HTB_NUMPRIO; i++) INIT_LIST_HEAD(q->drops + i); - qdisc_watchdog_init(&q->watchdog, sch); - INIT_WORK(&q->work, htb_work_func); qdisc_skb_head_init(&q->direct_queue); if (tb[TCA_HTB_DIRECT_QLEN]) -- cgit v1.1 From e89d469e3be3ed3d7124a803211a463ff83d0964 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:48:58 +0300 Subject: sch_multiq: fix double free on init failure The below commit added a call to ->destroy() on init failure, but multiq still frees ->queues on error in init, but ->queues is also freed by ->destroy() thus we get double free and corrupted memory. Very easy to reproduce (eth0 not multiqueue): $ tc qdisc add dev eth0 root multiq RTNETLINK answers: Operation not supported $ ip l add dumdum type dummy (crash) Trace log: [ 3929.467747] general protection fault: 0000 [#1] SMP [ 3929.468083] Modules linked in: [ 3929.468302] CPU: 3 PID: 967 Comm: ip Not tainted 4.13.0-rc6+ #56 [ 3929.468625] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 3929.469124] task: ffff88003716a700 task.stack: ffff88005872c000 [ 3929.469449] RIP: 0010:__kmalloc_track_caller+0x117/0x1be [ 3929.469746] RSP: 0018:ffff88005872f6a0 EFLAGS: 00010246 [ 3929.470042] RAX: 00000000000002de RBX: 0000000058a59000 RCX: 00000000000002df [ 3929.470406] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff821f7020 [ 3929.470770] RBP: ffff88005872f6e8 R08: 000000000001f010 R09: 0000000000000000 [ 3929.471133] R10: ffff88005872f730 R11: 0000000000008cdd R12: ff006d75646d7564 [ 3929.471496] R13: 00000000014000c0 R14: ffff88005b403c00 R15: ffff88005b403c00 [ 3929.471869] FS: 00007f0b70480740(0000) GS:ffff88005d980000(0000) knlGS:0000000000000000 [ 3929.472286] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 3929.472677] CR2: 00007ffcee4f3000 CR3: 0000000059d45000 CR4: 00000000000406e0 [ 3929.473209] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 3929.474109] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 3929.474873] Call Trace: [ 3929.475337] ? kstrdup_const+0x23/0x25 [ 3929.475863] kstrdup+0x2e/0x4b [ 3929.476338] kstrdup_const+0x23/0x25 [ 3929.478084] __kernfs_new_node+0x28/0xbc [ 3929.478478] kernfs_new_node+0x35/0x55 [ 3929.478929] kernfs_create_link+0x23/0x76 [ 3929.479478] sysfs_do_create_link_sd.isra.2+0x85/0xd7 [ 3929.480096] sysfs_create_link+0x33/0x35 [ 3929.480649] device_add+0x200/0x589 [ 3929.481184] netdev_register_kobject+0x7c/0x12f [ 3929.481711] register_netdevice+0x373/0x471 [ 3929.482174] rtnl_newlink+0x614/0x729 [ 3929.482610] ? rtnl_newlink+0x17f/0x729 [ 3929.483080] rtnetlink_rcv_msg+0x188/0x197 [ 3929.483533] ? rcu_read_unlock+0x3e/0x5f [ 3929.483984] ? rtnl_newlink+0x729/0x729 [ 3929.484420] netlink_rcv_skb+0x6c/0xce [ 3929.484858] rtnetlink_rcv+0x23/0x2a [ 3929.485291] netlink_unicast+0x103/0x181 [ 3929.485735] netlink_sendmsg+0x326/0x337 [ 3929.486181] sock_sendmsg_nosec+0x14/0x3f [ 3929.486614] sock_sendmsg+0x29/0x2e [ 3929.486973] ___sys_sendmsg+0x209/0x28b [ 3929.487340] ? do_raw_spin_unlock+0xcd/0xf8 [ 3929.487719] ? _raw_spin_unlock+0x27/0x31 [ 3929.488092] ? __handle_mm_fault+0x651/0xdb1 [ 3929.488471] ? check_chain_key+0xb0/0xfd [ 3929.488847] __sys_sendmsg+0x45/0x63 [ 3929.489206] ? __sys_sendmsg+0x45/0x63 [ 3929.489576] SyS_sendmsg+0x19/0x1b [ 3929.489901] entry_SYSCALL_64_fastpath+0x23/0xc2 [ 3929.490172] RIP: 0033:0x7f0b6fb93690 [ 3929.490423] RSP: 002b:00007ffcee4ed588 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 3929.490881] RAX: ffffffffffffffda RBX: ffffffff810d278c RCX: 00007f0b6fb93690 [ 3929.491198] RDX: 0000000000000000 RSI: 00007ffcee4ed5d0 RDI: 0000000000000003 [ 3929.491521] RBP: ffff88005872ff98 R08: 0000000000000001 R09: 0000000000000000 [ 3929.491801] R10: 00007ffcee4ed350 R11: 0000000000000246 R12: 0000000000000002 [ 3929.492075] R13: 000000000066f1a0 R14: 00007ffcee4f5680 R15: 0000000000000000 [ 3929.492352] ? trace_hardirqs_off_caller+0xa7/0xcf [ 3929.492590] Code: 8b 45 c0 48 8b 45 b8 74 17 48 8b 4d c8 83 ca ff 44 89 ee 4c 89 f7 e8 83 ca ff ff 49 89 c4 eb 49 49 63 56 20 48 8d 48 01 4d 8b 06 <49> 8b 1c 14 48 89 c2 4c 89 e0 65 49 0f c7 08 0f 94 c0 83 f0 01 [ 3929.493335] RIP: __kmalloc_track_caller+0x117/0x1be RSP: ffff88005872f6a0 Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Fixes: f07d1501292b ("multiq: Further multiqueue cleanup") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_multiq.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index f143b7b..9c454f5 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -257,12 +257,7 @@ static int multiq_init(struct Qdisc *sch, struct nlattr *opt) for (i = 0; i < q->max_bands; i++) q->queues[i] = &noop_qdisc; - err = multiq_tune(sch, opt); - - if (err) - kfree(q->queues); - - return err; + return multiq_tune(sch, opt); } static int multiq_dump(struct Qdisc *sch, struct sk_buff *skb) -- cgit v1.1 From 32db864d33c21fd70a217ba53cb7224889354ffb Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:48:59 +0300 Subject: sch_hhf: fix null pointer dereference on init failure If sch_hhf fails in its ->init() function (either due to wrong user-space arguments as below or memory alloc failure of hh_flows) it will do a null pointer deref of q->hh_flows in its ->destroy() function. To reproduce the crash: $ tc qdisc add dev eth0 root hhf quantum 2000000 non_hh_weight 10000000 Crash log: [ 690.654882] BUG: unable to handle kernel NULL pointer dereference at (null) [ 690.655565] IP: hhf_destroy+0x48/0xbc [ 690.655944] PGD 37345067 [ 690.655948] P4D 37345067 [ 690.656252] PUD 58402067 [ 690.656554] PMD 0 [ 690.656857] [ 690.657362] Oops: 0000 [#1] SMP [ 690.657696] Modules linked in: [ 690.658032] CPU: 3 PID: 920 Comm: tc Not tainted 4.13.0-rc6+ #57 [ 690.658525] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 690.659255] task: ffff880058578000 task.stack: ffff88005acbc000 [ 690.659747] RIP: 0010:hhf_destroy+0x48/0xbc [ 690.660146] RSP: 0018:ffff88005acbf9e0 EFLAGS: 00010246 [ 690.660601] RAX: 0000000000000000 RBX: 0000000000000020 RCX: 0000000000000000 [ 690.661155] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffffffff821f63f0 [ 690.661710] RBP: ffff88005acbfa08 R08: ffffffff81b10a90 R09: 0000000000000000 [ 690.662267] R10: 00000000f42b7019 R11: ffff880058578000 R12: 00000000ffffffea [ 690.662820] R13: ffff8800372f6400 R14: 0000000000000000 R15: 0000000000000000 [ 690.663769] FS: 00007f8ae5e8b740(0000) GS:ffff88005d980000(0000) knlGS:0000000000000000 [ 690.667069] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 690.667965] CR2: 0000000000000000 CR3: 0000000058523000 CR4: 00000000000406e0 [ 690.668918] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 690.669945] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 690.671003] Call Trace: [ 690.671743] qdisc_create+0x377/0x3fd [ 690.672534] tc_modify_qdisc+0x4d2/0x4fd [ 690.673324] rtnetlink_rcv_msg+0x188/0x197 [ 690.674204] ? rcu_read_unlock+0x3e/0x5f [ 690.675091] ? rtnl_newlink+0x729/0x729 [ 690.675877] netlink_rcv_skb+0x6c/0xce [ 690.676648] rtnetlink_rcv+0x23/0x2a [ 690.677405] netlink_unicast+0x103/0x181 [ 690.678179] netlink_sendmsg+0x326/0x337 [ 690.678958] sock_sendmsg_nosec+0x14/0x3f [ 690.679743] sock_sendmsg+0x29/0x2e [ 690.680506] ___sys_sendmsg+0x209/0x28b [ 690.681283] ? __handle_mm_fault+0xc7d/0xdb1 [ 690.681915] ? check_chain_key+0xb0/0xfd [ 690.682449] __sys_sendmsg+0x45/0x63 [ 690.682954] ? __sys_sendmsg+0x45/0x63 [ 690.683471] SyS_sendmsg+0x19/0x1b [ 690.683974] entry_SYSCALL_64_fastpath+0x23/0xc2 [ 690.684516] RIP: 0033:0x7f8ae529d690 [ 690.685016] RSP: 002b:00007fff26d2d6b8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 690.685931] RAX: ffffffffffffffda RBX: ffffffff810d278c RCX: 00007f8ae529d690 [ 690.686573] RDX: 0000000000000000 RSI: 00007fff26d2d700 RDI: 0000000000000003 [ 690.687047] RBP: ffff88005acbff98 R08: 0000000000000001 R09: 0000000000000000 [ 690.687519] R10: 00007fff26d2d480 R11: 0000000000000246 R12: 0000000000000002 [ 690.687996] R13: 0000000001258070 R14: 0000000000000001 R15: 0000000000000000 [ 690.688475] ? trace_hardirqs_off_caller+0xa7/0xcf [ 690.688887] Code: 00 00 e8 2a 02 ae ff 49 8b bc 1d 60 02 00 00 48 83 c3 08 e8 19 02 ae ff 48 83 fb 20 75 dc 45 31 f6 4d 89 f7 4d 03 bd 20 02 00 00 <49> 8b 07 49 39 c7 75 24 49 83 c6 10 49 81 fe 00 40 00 00 75 e1 [ 690.690200] RIP: hhf_destroy+0x48/0xbc RSP: ffff88005acbf9e0 [ 690.690636] CR2: 0000000000000000 Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Fixes: 10239edf86f1 ("net-qdisc-hhf: Heavy-Hitter Filter (HHF) qdisc") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_hhf.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index 51d3ba6..73a53c0 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -477,6 +477,9 @@ static void hhf_destroy(struct Qdisc *sch) kvfree(q->hhf_valid_bits[i]); } + if (!q->hh_flows) + return; + for (i = 0; i < HH_FLOWS_CNT; i++) { struct hh_flow_state *flow, *next; struct list_head *head = &q->hh_flows[i]; -- cgit v1.1 From 3bdac362a2f89ed3e148fa6f38c5f5d858f50b1a Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:49:00 +0300 Subject: sch_hfsc: fix null pointer deref and double free on init failure Depending on where ->init fails we can get a null pointer deref due to uninitialized hires timer (watchdog) or a double free of the qdisc hash because it is already freed by ->destroy(). Fixes: 8d5537387505 ("net/sched/hfsc: allocate tcf block for hfsc root class") Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_hfsc.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index fd15200..11ab8da 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1418,6 +1418,8 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) struct tc_hfsc_qopt *qopt; int err; + qdisc_watchdog_init(&q->watchdog, sch); + if (opt == NULL || nla_len(opt) < sizeof(*qopt)) return -EINVAL; qopt = nla_data(opt); @@ -1430,7 +1432,7 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) err = tcf_block_get(&q->root.block, &q->root.filter_list); if (err) - goto err_tcf; + return err; q->root.cl_common.classid = sch->handle; q->root.refcnt = 1; @@ -1448,13 +1450,7 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) qdisc_class_hash_insert(&q->clhash, &q->root.cl_common); qdisc_class_hash_grow(sch, &q->clhash); - qdisc_watchdog_init(&q->watchdog, sch); - return 0; - -err_tcf: - qdisc_class_hash_destroy(&q->clhash); - return err; } static int -- cgit v1.1 From 3501d059921246ff617b43e86250a719c140bd97 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:49:01 +0300 Subject: sch_cbq: fix null pointer dereferences on init failure CBQ can fail on ->init by wrong nl attributes or simply for missing any, f.e. if it's set as a default qdisc then TCA_OPTIONS (opt) will be NULL when it is activated. The first thing init does is parse opt but it will dereference a null pointer if used as a default qdisc, also since init failure at default qdisc invokes ->reset() which cancels all timers then we'll also dereference two more null pointers (timer->base) as they were never initialized. To reproduce: $ sysctl net.core.default_qdisc=cbq $ ip l set ethX up Crash log of the first null ptr deref: [44727.907454] BUG: unable to handle kernel NULL pointer dereference at (null) [44727.907600] IP: cbq_init+0x27/0x205 [44727.907676] PGD 59ff4067 [44727.907677] P4D 59ff4067 [44727.907742] PUD 59c70067 [44727.907807] PMD 0 [44727.907873] [44727.907982] Oops: 0000 [#1] SMP [44727.908054] Modules linked in: [44727.908126] CPU: 1 PID: 21312 Comm: ip Not tainted 4.13.0-rc6+ #60 [44727.908235] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [44727.908477] task: ffff88005ad42700 task.stack: ffff880037214000 [44727.908672] RIP: 0010:cbq_init+0x27/0x205 [44727.908838] RSP: 0018:ffff8800372175f0 EFLAGS: 00010286 [44727.909018] RAX: ffffffff816c3852 RBX: ffff880058c53800 RCX: 0000000000000000 [44727.909222] RDX: 0000000000000004 RSI: 0000000000000000 RDI: ffff8800372175f8 [44727.909427] RBP: ffff880037217650 R08: ffffffff81b0f380 R09: 0000000000000000 [44727.909631] R10: ffff880037217660 R11: 0000000000000020 R12: ffffffff822a44c0 [44727.909835] R13: ffff880058b92000 R14: 00000000ffffffff R15: 0000000000000001 [44727.910040] FS: 00007ff8bc583740(0000) GS:ffff88005d880000(0000) knlGS:0000000000000000 [44727.910339] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [44727.910525] CR2: 0000000000000000 CR3: 00000000371e5000 CR4: 00000000000406e0 [44727.910731] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [44727.910936] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [44727.911141] Call Trace: [44727.911291] ? lockdep_init_map+0xb6/0x1ba [44727.911461] ? qdisc_alloc+0x14e/0x187 [44727.911626] qdisc_create_dflt+0x7a/0x94 [44727.911794] ? dev_activate+0x129/0x129 [44727.911959] attach_one_default_qdisc+0x36/0x63 [44727.912132] netdev_for_each_tx_queue+0x3d/0x48 [44727.912305] dev_activate+0x4b/0x129 [44727.912468] __dev_open+0xe7/0x104 [44727.912631] __dev_change_flags+0xc6/0x15c [44727.912799] dev_change_flags+0x25/0x59 [44727.912966] do_setlink+0x30c/0xb3f [44727.913129] ? check_chain_key+0xb0/0xfd [44727.913294] ? check_chain_key+0xb0/0xfd [44727.913463] rtnl_newlink+0x3a4/0x729 [44727.913626] ? rtnl_newlink+0x117/0x729 [44727.913801] ? ns_capable_common+0xd/0xb1 [44727.913968] ? ns_capable+0x13/0x15 [44727.914131] rtnetlink_rcv_msg+0x188/0x197 [44727.914300] ? rcu_read_unlock+0x3e/0x5f [44727.914465] ? rtnl_newlink+0x729/0x729 [44727.914630] netlink_rcv_skb+0x6c/0xce [44727.914796] rtnetlink_rcv+0x23/0x2a [44727.914956] netlink_unicast+0x103/0x181 [44727.915122] netlink_sendmsg+0x326/0x337 [44727.915291] sock_sendmsg_nosec+0x14/0x3f [44727.915459] sock_sendmsg+0x29/0x2e [44727.915619] ___sys_sendmsg+0x209/0x28b [44727.915784] ? do_raw_spin_unlock+0xcd/0xf8 [44727.915954] ? _raw_spin_unlock+0x27/0x31 [44727.916121] ? __handle_mm_fault+0x651/0xdb1 [44727.916290] ? check_chain_key+0xb0/0xfd [44727.916461] __sys_sendmsg+0x45/0x63 [44727.916626] ? __sys_sendmsg+0x45/0x63 [44727.916792] SyS_sendmsg+0x19/0x1b [44727.916950] entry_SYSCALL_64_fastpath+0x23/0xc2 [44727.917125] RIP: 0033:0x7ff8bbc96690 [44727.917286] RSP: 002b:00007ffc360991e8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [44727.917579] RAX: ffffffffffffffda RBX: ffffffff810d278c RCX: 00007ff8bbc96690 [44727.917783] RDX: 0000000000000000 RSI: 00007ffc36099230 RDI: 0000000000000003 [44727.917987] RBP: ffff880037217f98 R08: 0000000000000001 R09: 0000000000000003 [44727.918190] R10: 00007ffc36098fb0 R11: 0000000000000246 R12: 0000000000000006 [44727.918393] R13: 000000000066f1a0 R14: 00007ffc360a12e0 R15: 0000000000000000 [44727.918597] ? trace_hardirqs_off_caller+0xa7/0xcf [44727.918774] Code: 41 5f 5d c3 66 66 66 66 90 55 48 8d 56 04 45 31 c9 49 c7 c0 80 f3 b0 81 48 89 e5 41 55 41 54 53 48 89 fb 48 8d 7d a8 48 83 ec 48 <0f> b7 0e be 07 00 00 00 83 e9 04 e8 e6 f7 d8 ff 85 c0 0f 88 bb [44727.919332] RIP: cbq_init+0x27/0x205 RSP: ffff8800372175f0 [44727.919516] CR2: 0000000000000000 Fixes: 0fbbeb1ba43b ("[PKT_SCHED]: Fix missing qdisc_destroy() in qdisc_create_dflt()") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_cbq.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 780db43..156c8a3 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1139,6 +1139,13 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) struct tc_ratespec *r; int err; + qdisc_watchdog_init(&q->watchdog, sch); + hrtimer_init(&q->delay_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); + q->delay_timer.function = cbq_undelay; + + if (!opt) + return -EINVAL; + err = nla_parse_nested(tb, TCA_CBQ_MAX, opt, cbq_policy, NULL); if (err < 0) return err; @@ -1177,9 +1184,6 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) q->link.avpkt = q->link.allot/2; q->link.minidle = -0x7FFFFFFF; - qdisc_watchdog_init(&q->watchdog, sch); - hrtimer_init(&q->delay_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); - q->delay_timer.function = cbq_undelay; q->toplevel = TC_CBQ_MAXLEVEL; q->now = psched_get_time(); -- cgit v1.1 From 30c31d746d0eb458ae327f522bc8e4c44cbea0f0 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:49:02 +0300 Subject: sch_fq_codel: avoid double free on init failure It is very unlikely to happen but the backlogs memory allocation could fail and will free q->flows, but then ->destroy() will free q->flows too. For correctness remove the first free and let ->destroy clean up. Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_fq_codel.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 337f2d6..2c0c05f 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -491,10 +491,8 @@ static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt) if (!q->flows) return -ENOMEM; q->backlogs = kvzalloc(q->flows_cnt * sizeof(u32), GFP_KERNEL); - if (!q->backlogs) { - kvfree(q->flows); + if (!q->backlogs) return -ENOMEM; - } for (i = 0; i < q->flows_cnt; i++) { struct fq_codel_flow *flow = q->flows + i; -- cgit v1.1 From 634576a1844dba15bc5e6fc61d72f37e13a21615 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:49:03 +0300 Subject: sch_netem: avoid null pointer deref on init failure netem can fail in ->init due to missing options (either not supplied by user-space or used as a default qdisc) causing a timer->base null pointer deref in its ->destroy() and ->reset() callbacks. Reproduce: $ sysctl net.core.default_qdisc=netem $ ip l set ethX up Crash log: [ 1814.846943] BUG: unable to handle kernel NULL pointer dereference at (null) [ 1814.847181] IP: hrtimer_active+0x17/0x8a [ 1814.847270] PGD 59c34067 [ 1814.847271] P4D 59c34067 [ 1814.847337] PUD 37374067 [ 1814.847403] PMD 0 [ 1814.847468] [ 1814.847582] Oops: 0000 [#1] SMP [ 1814.847655] Modules linked in: sch_netem(O) sch_fq_codel(O) [ 1814.847761] CPU: 3 PID: 1573 Comm: ip Tainted: G O 4.13.0-rc6+ #62 [ 1814.847884] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 1814.848043] task: ffff88003723a700 task.stack: ffff88005adc8000 [ 1814.848235] RIP: 0010:hrtimer_active+0x17/0x8a [ 1814.848407] RSP: 0018:ffff88005adcb590 EFLAGS: 00010246 [ 1814.848590] RAX: 0000000000000000 RBX: ffff880058e359d8 RCX: 0000000000000000 [ 1814.848793] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff880058e359d8 [ 1814.848998] RBP: ffff88005adcb5b0 R08: 00000000014080c0 R09: 00000000ffffffff [ 1814.849204] R10: ffff88005adcb660 R11: 0000000000000020 R12: 0000000000000000 [ 1814.849410] R13: ffff880058e359d8 R14: 00000000ffffffff R15: 0000000000000001 [ 1814.849616] FS: 00007f733bbca740(0000) GS:ffff88005d980000(0000) knlGS:0000000000000000 [ 1814.849919] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1814.850107] CR2: 0000000000000000 CR3: 0000000059f0d000 CR4: 00000000000406e0 [ 1814.850313] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 1814.850518] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 1814.850723] Call Trace: [ 1814.850875] hrtimer_try_to_cancel+0x1a/0x93 [ 1814.851047] hrtimer_cancel+0x15/0x20 [ 1814.851211] qdisc_watchdog_cancel+0x12/0x14 [ 1814.851383] netem_reset+0xe6/0xed [sch_netem] [ 1814.851561] qdisc_destroy+0x8b/0xe5 [ 1814.851723] qdisc_create_dflt+0x86/0x94 [ 1814.851890] ? dev_activate+0x129/0x129 [ 1814.852057] attach_one_default_qdisc+0x36/0x63 [ 1814.852232] netdev_for_each_tx_queue+0x3d/0x48 [ 1814.852406] dev_activate+0x4b/0x129 [ 1814.852569] __dev_open+0xe7/0x104 [ 1814.852730] __dev_change_flags+0xc6/0x15c [ 1814.852899] dev_change_flags+0x25/0x59 [ 1814.853064] do_setlink+0x30c/0xb3f [ 1814.853228] ? check_chain_key+0xb0/0xfd [ 1814.853396] ? check_chain_key+0xb0/0xfd [ 1814.853565] rtnl_newlink+0x3a4/0x729 [ 1814.853728] ? rtnl_newlink+0x117/0x729 [ 1814.853905] ? ns_capable_common+0xd/0xb1 [ 1814.854072] ? ns_capable+0x13/0x15 [ 1814.854234] rtnetlink_rcv_msg+0x188/0x197 [ 1814.854404] ? rcu_read_unlock+0x3e/0x5f [ 1814.854572] ? rtnl_newlink+0x729/0x729 [ 1814.854737] netlink_rcv_skb+0x6c/0xce [ 1814.854902] rtnetlink_rcv+0x23/0x2a [ 1814.855064] netlink_unicast+0x103/0x181 [ 1814.855230] netlink_sendmsg+0x326/0x337 [ 1814.855398] sock_sendmsg_nosec+0x14/0x3f [ 1814.855584] sock_sendmsg+0x29/0x2e [ 1814.855747] ___sys_sendmsg+0x209/0x28b [ 1814.855912] ? do_raw_spin_unlock+0xcd/0xf8 [ 1814.856082] ? _raw_spin_unlock+0x27/0x31 [ 1814.856251] ? __handle_mm_fault+0x651/0xdb1 [ 1814.856421] ? check_chain_key+0xb0/0xfd [ 1814.856592] __sys_sendmsg+0x45/0x63 [ 1814.856755] ? __sys_sendmsg+0x45/0x63 [ 1814.856923] SyS_sendmsg+0x19/0x1b [ 1814.857083] entry_SYSCALL_64_fastpath+0x23/0xc2 [ 1814.857256] RIP: 0033:0x7f733b2dd690 [ 1814.857419] RSP: 002b:00007ffe1d3387d8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 1814.858238] RAX: ffffffffffffffda RBX: ffffffff810d278c RCX: 00007f733b2dd690 [ 1814.858445] RDX: 0000000000000000 RSI: 00007ffe1d338820 RDI: 0000000000000003 [ 1814.858651] RBP: ffff88005adcbf98 R08: 0000000000000001 R09: 0000000000000003 [ 1814.858856] R10: 00007ffe1d3385a0 R11: 0000000000000246 R12: 0000000000000002 [ 1814.859060] R13: 000000000066f1a0 R14: 00007ffe1d3408d0 R15: 0000000000000000 [ 1814.859267] ? trace_hardirqs_off_caller+0xa7/0xcf [ 1814.859446] Code: 10 55 48 89 c7 48 89 e5 e8 45 a1 fb ff 31 c0 5d c3 31 c0 c3 66 66 66 66 90 55 48 89 e5 41 56 41 55 41 54 53 49 89 fd 49 8b 45 30 <4c> 8b 20 41 8b 5c 24 38 31 c9 31 d2 48 c7 c7 50 8e 1d 82 41 89 [ 1814.860022] RIP: hrtimer_active+0x17/0x8a RSP: ffff88005adcb590 [ 1814.860214] CR2: 0000000000000000 Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Fixes: 0fbbeb1ba43b ("[PKT_SCHED]: Fix missing qdisc_destroy() in qdisc_create_dflt()") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_netem.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 1b3dd61..14d1724 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -933,11 +933,11 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt) struct netem_sched_data *q = qdisc_priv(sch); int ret; + qdisc_watchdog_init(&q->watchdog, sch); + if (!opt) return -EINVAL; - qdisc_watchdog_init(&q->watchdog, sch); - q->loss_model = CLG_RANDOM; ret = netem_change(sch, opt); if (ret) -- cgit v1.1 From e232657661242cc8a169595160847b3e66aa7056 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:49:04 +0300 Subject: sch_sfq: fix null pointer dereference on init failure Currently only a memory allocation failure can lead to this, so let's initialize the timer first. Fixes: 6529eaba33f0 ("net: sched: introduce tcf block infractructure") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_sfq.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 82469ef..fc69fc5 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -716,13 +716,13 @@ static int sfq_init(struct Qdisc *sch, struct nlattr *opt) int i; int err; + setup_deferrable_timer(&q->perturb_timer, sfq_perturbation, + (unsigned long)sch); + err = tcf_block_get(&q->block, &q->filter_list); if (err) return err; - setup_deferrable_timer(&q->perturb_timer, sfq_perturbation, - (unsigned long)sch); - for (i = 0; i < SFQ_MAX_DEPTH + 1; i++) { q->dep[i].next = i + SFQ_MAX_FLOWS; q->dep[i].prev = i + SFQ_MAX_FLOWS; -- cgit v1.1 From c2d6511e6a4f1f3673d711569c00c3849549e9b0 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 30 Aug 2017 12:49:05 +0300 Subject: sch_tbf: fix two null pointer dereferences on init failure sch_tbf calls qdisc_watchdog_cancel() in both its ->reset and ->destroy callbacks but it may fail before the timer is initialized due to missing options (either not supplied by user-space or set as a default qdisc), also q->qdisc is used by ->reset and ->destroy so we need it initialized. Reproduce: $ sysctl net.core.default_qdisc=tbf $ ip l set ethX up Crash log: [ 959.160172] BUG: unable to handle kernel NULL pointer dereference at 0000000000000018 [ 959.160323] IP: qdisc_reset+0xa/0x5c [ 959.160400] PGD 59cdb067 [ 959.160401] P4D 59cdb067 [ 959.160466] PUD 59ccb067 [ 959.160532] PMD 0 [ 959.160597] [ 959.160706] Oops: 0000 [#1] SMP [ 959.160778] Modules linked in: sch_tbf sch_sfb sch_prio sch_netem [ 959.160891] CPU: 2 PID: 1562 Comm: ip Not tainted 4.13.0-rc6+ #62 [ 959.160998] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 959.161157] task: ffff880059c9a700 task.stack: ffff8800376d0000 [ 959.161263] RIP: 0010:qdisc_reset+0xa/0x5c [ 959.161347] RSP: 0018:ffff8800376d3610 EFLAGS: 00010286 [ 959.161531] RAX: ffffffffa001b1dd RBX: ffff8800373a2800 RCX: 0000000000000000 [ 959.161733] RDX: ffffffff8215f160 RSI: ffffffff8215f160 RDI: 0000000000000000 [ 959.161939] RBP: ffff8800376d3618 R08: 00000000014080c0 R09: 00000000ffffffff [ 959.162141] R10: ffff8800376d3578 R11: 0000000000000020 R12: ffffffffa001d2c0 [ 959.162343] R13: ffff880037538000 R14: 00000000ffffffff R15: 0000000000000001 [ 959.162546] FS: 00007fcc5126b740(0000) GS:ffff88005d900000(0000) knlGS:0000000000000000 [ 959.162844] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 959.163030] CR2: 0000000000000018 CR3: 000000005abc4000 CR4: 00000000000406e0 [ 959.163233] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 959.163436] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 959.163638] Call Trace: [ 959.163788] tbf_reset+0x19/0x64 [sch_tbf] [ 959.163957] qdisc_destroy+0x8b/0xe5 [ 959.164119] qdisc_create_dflt+0x86/0x94 [ 959.164284] ? dev_activate+0x129/0x129 [ 959.164449] attach_one_default_qdisc+0x36/0x63 [ 959.164623] netdev_for_each_tx_queue+0x3d/0x48 [ 959.164795] dev_activate+0x4b/0x129 [ 959.164957] __dev_open+0xe7/0x104 [ 959.165118] __dev_change_flags+0xc6/0x15c [ 959.165287] dev_change_flags+0x25/0x59 [ 959.165451] do_setlink+0x30c/0xb3f [ 959.165613] ? check_chain_key+0xb0/0xfd [ 959.165782] rtnl_newlink+0x3a4/0x729 [ 959.165947] ? rtnl_newlink+0x117/0x729 [ 959.166121] ? ns_capable_common+0xd/0xb1 [ 959.166288] ? ns_capable+0x13/0x15 [ 959.166450] rtnetlink_rcv_msg+0x188/0x197 [ 959.166617] ? rcu_read_unlock+0x3e/0x5f [ 959.166783] ? rtnl_newlink+0x729/0x729 [ 959.166948] netlink_rcv_skb+0x6c/0xce [ 959.167113] rtnetlink_rcv+0x23/0x2a [ 959.167273] netlink_unicast+0x103/0x181 [ 959.167439] netlink_sendmsg+0x326/0x337 [ 959.167607] sock_sendmsg_nosec+0x14/0x3f [ 959.167772] sock_sendmsg+0x29/0x2e [ 959.167932] ___sys_sendmsg+0x209/0x28b [ 959.168098] ? do_raw_spin_unlock+0xcd/0xf8 [ 959.168267] ? _raw_spin_unlock+0x27/0x31 [ 959.168432] ? __handle_mm_fault+0x651/0xdb1 [ 959.168602] ? check_chain_key+0xb0/0xfd [ 959.168773] __sys_sendmsg+0x45/0x63 [ 959.168934] ? __sys_sendmsg+0x45/0x63 [ 959.169100] SyS_sendmsg+0x19/0x1b [ 959.169260] entry_SYSCALL_64_fastpath+0x23/0xc2 [ 959.169432] RIP: 0033:0x7fcc5097e690 [ 959.169592] RSP: 002b:00007ffd0d5c7b48 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 959.169887] RAX: ffffffffffffffda RBX: ffffffff810d278c RCX: 00007fcc5097e690 [ 959.170089] RDX: 0000000000000000 RSI: 00007ffd0d5c7b90 RDI: 0000000000000003 [ 959.170292] RBP: ffff8800376d3f98 R08: 0000000000000001 R09: 0000000000000003 [ 959.170494] R10: 00007ffd0d5c7910 R11: 0000000000000246 R12: 0000000000000006 [ 959.170697] R13: 000000000066f1a0 R14: 00007ffd0d5cfc40 R15: 0000000000000000 [ 959.170900] ? trace_hardirqs_off_caller+0xa7/0xcf [ 959.171076] Code: 00 41 c7 84 24 14 01 00 00 00 00 00 00 41 c7 84 24 98 00 00 00 00 00 00 00 41 5c 41 5d 41 5e 5d c3 66 66 66 66 90 55 48 89 e5 53 <48> 8b 47 18 48 89 fb 48 8b 40 48 48 85 c0 74 02 ff d0 48 8b bb [ 959.171637] RIP: qdisc_reset+0xa/0x5c RSP: ffff8800376d3610 [ 959.171821] CR2: 0000000000000018 Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Fixes: 0fbbeb1ba43b ("[PKT_SCHED]: Fix missing qdisc_destroy() in qdisc_create_dflt()") Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/sched/sch_tbf.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index b2e4b6a..493270f 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -425,12 +425,13 @@ static int tbf_init(struct Qdisc *sch, struct nlattr *opt) { struct tbf_sched_data *q = qdisc_priv(sch); + qdisc_watchdog_init(&q->watchdog, sch); + q->qdisc = &noop_qdisc; + if (opt == NULL) return -EINVAL; q->t_c = ktime_get_ns(); - qdisc_watchdog_init(&q->watchdog, sch); - q->qdisc = &noop_qdisc; return tbf_change(sch, opt); } -- cgit v1.1 From 351050ecd6523374b370341cc29fe61e2201556b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 30 Aug 2017 09:29:31 -0700 Subject: kcm: do not attach PF_KCM sockets to avoid deadlock syzkaller had no problem to trigger a deadlock, attaching a KCM socket to another one (or itself). (original syzkaller report was a very confusing lockdep splat during a sendmsg()) It seems KCM claims to only support TCP, but no enforcement is done, so we might need to add additional checks. Fixes: ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module") Signed-off-by: Eric Dumazet Reported-by: Dmitry Vyukov Acked-by: Tom Herbert Signed-off-by: David S. Miller --- net/kcm/kcmsock.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c index da49191..4abf628 100644 --- a/net/kcm/kcmsock.c +++ b/net/kcm/kcmsock.c @@ -1383,6 +1383,10 @@ static int kcm_attach(struct socket *sock, struct socket *csock, if (!csk) return -EINVAL; + /* We must prevent loops or risk deadlock ! */ + if (csk->sk_family == PF_KCM) + return -EOPNOTSUPP; + psock = kmem_cache_zalloc(kcm_psockp, GFP_KERNEL); if (!psock) return -ENOMEM; -- cgit v1.1 From df191632f814357ee4d646421662d866028b569d Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 30 Aug 2017 12:39:33 -0700 Subject: net: dsa: bcm_sf2: Fix number of CFP entries for BCM7278 BCM7278 has only 128 entries while BCM7445 has the full 256 entries set, fix that. Fixes: 7318166cacad ("net: dsa: bcm_sf2: Add support for ethtool::rxnfc") Signed-off-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller --- drivers/net/dsa/bcm_sf2.c | 4 ++++ drivers/net/dsa/bcm_sf2.h | 1 + drivers/net/dsa/bcm_sf2_cfp.c | 8 ++++---- 3 files changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 648f91b..9b6ce7c 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -1048,6 +1048,7 @@ struct bcm_sf2_of_data { u32 type; const u16 *reg_offsets; unsigned int core_reg_align; + unsigned int num_cfp_rules; }; /* Register offsets for the SWITCH_REG_* block */ @@ -1071,6 +1072,7 @@ static const struct bcm_sf2_of_data bcm_sf2_7445_data = { .type = BCM7445_DEVICE_ID, .core_reg_align = 0, .reg_offsets = bcm_sf2_7445_reg_offsets, + .num_cfp_rules = 256, }; static const u16 bcm_sf2_7278_reg_offsets[] = { @@ -1093,6 +1095,7 @@ static const struct bcm_sf2_of_data bcm_sf2_7278_data = { .type = BCM7278_DEVICE_ID, .core_reg_align = 1, .reg_offsets = bcm_sf2_7278_reg_offsets, + .num_cfp_rules = 128, }; static const struct of_device_id bcm_sf2_of_match[] = { @@ -1149,6 +1152,7 @@ static int bcm_sf2_sw_probe(struct platform_device *pdev) priv->type = data->type; priv->reg_offsets = data->reg_offsets; priv->core_reg_align = data->core_reg_align; + priv->num_cfp_rules = data->num_cfp_rules; /* Auto-detection using standard registers will not work, so * provide an indication of what kind of device we are for diff --git a/drivers/net/dsa/bcm_sf2.h b/drivers/net/dsa/bcm_sf2.h index 7d3030e..7f9125e 100644 --- a/drivers/net/dsa/bcm_sf2.h +++ b/drivers/net/dsa/bcm_sf2.h @@ -72,6 +72,7 @@ struct bcm_sf2_priv { u32 type; const u16 *reg_offsets; unsigned int core_reg_align; + unsigned int num_cfp_rules; /* spinlock protecting access to the indirect registers */ spinlock_t indir_lock; diff --git a/drivers/net/dsa/bcm_sf2_cfp.c b/drivers/net/dsa/bcm_sf2_cfp.c index 2fb32d6..8a1da7e 100644 --- a/drivers/net/dsa/bcm_sf2_cfp.c +++ b/drivers/net/dsa/bcm_sf2_cfp.c @@ -98,7 +98,7 @@ static inline void bcm_sf2_cfp_rule_addr_set(struct bcm_sf2_priv *priv, { u32 reg; - WARN_ON(addr >= CFP_NUM_RULES); + WARN_ON(addr >= priv->num_cfp_rules); reg = core_readl(priv, CORE_CFP_ACC); reg &= ~(XCESS_ADDR_MASK << XCESS_ADDR_SHIFT); @@ -109,7 +109,7 @@ static inline void bcm_sf2_cfp_rule_addr_set(struct bcm_sf2_priv *priv, static inline unsigned int bcm_sf2_cfp_rule_size(struct bcm_sf2_priv *priv) { /* Entry #0 is reserved */ - return CFP_NUM_RULES - 1; + return priv->num_cfp_rules - 1; } static int bcm_sf2_cfp_rule_set(struct dsa_switch *ds, int port, @@ -523,7 +523,7 @@ static int bcm_sf2_cfp_rule_get_all(struct bcm_sf2_priv *priv, if (!(reg & OP_STR_DONE)) break; - } while (index < CFP_NUM_RULES); + } while (index < priv->num_cfp_rules); /* Put the TCAM size here */ nfc->data = bcm_sf2_cfp_rule_size(priv); @@ -544,7 +544,7 @@ int bcm_sf2_get_rxnfc(struct dsa_switch *ds, int port, case ETHTOOL_GRXCLSRLCNT: /* Subtract the default, unusable rule */ nfc->rule_cnt = bitmap_weight(priv->cfp.used, - CFP_NUM_RULES) - 1; + priv->num_cfp_rules) - 1; /* We support specifying rule locations */ nfc->data |= RX_CLS_LOC_SPECIAL; break; -- cgit v1.1 From ebc8254aeae34226d0bc8fda309fd9790d4dccfe Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 30 Aug 2017 17:49:29 -0700 Subject: Revert "net: phy: Correctly process PHY_HALTED in phy_stop_machine()" This reverts commit 7ad813f208533cebfcc32d3d7474dc1677d1b09a ("net: phy: Correctly process PHY_HALTED in phy_stop_machine()") because it is creating the possibility for a NULL pointer dereference. David Daney provide the following call trace and diagram of events: When ndo_stop() is called we call: phy_disconnect() +---> phy_stop_interrupts() implies: phydev->irq = PHY_POLL; +---> phy_stop_machine() | +---> phy_state_machine() | +----> queue_delayed_work(): Work queued. +--->phy_detach() implies: phydev->attached_dev = NULL; Now at a later time the queued work does: phy_state_machine() +---->netif_carrier_off(phydev->attached_dev): Oh no! It is NULL: CPU 12 Unable to handle kernel paging request at virtual address 0000000000000048, epc == ffffffff80de37ec, ra == ffffffff80c7c Oops[#1]: CPU: 12 PID: 1502 Comm: kworker/12:1 Not tainted 4.9.43-Cavium-Octeon+ #1 Workqueue: events_power_efficient phy_state_machine task: 80000004021ed100 task.stack: 8000000409d70000 $ 0 : 0000000000000000 ffffffff84720060 0000000000000048 0000000000000004 $ 4 : 0000000000000000 0000000000000001 0000000000000004 0000000000000000 $ 8 : 0000000000000000 0000000000000000 00000000ffff98f3 0000000000000000 $12 : 8000000409d73fe0 0000000000009c00 ffffffff846547c8 000000000000af3b $16 : 80000004096bab68 80000004096babd0 0000000000000000 80000004096ba800 $20 : 0000000000000000 0000000000000000 ffffffff81090000 0000000000000008 $24 : 0000000000000061 ffffffff808637b0 $28 : 8000000409d70000 8000000409d73cf0 80000000271bd300 ffffffff80c7804c Hi : 000000000000002a Lo : 000000000000003f epc : ffffffff80de37ec netif_carrier_off+0xc/0x58 ra : ffffffff80c7804c phy_state_machine+0x48c/0x4f8 Status: 14009ce3 KX SX UX KERNEL EXL IE Cause : 00800008 (ExcCode 02) BadVA : 0000000000000048 PrId : 000d9501 (Cavium Octeon III) Modules linked in: Process kworker/12:1 (pid: 1502, threadinfo=8000000409d70000, task=80000004021ed100, tls=0000000000000000) Stack : 8000000409a54000 80000004096bab68 80000000271bd300 80000000271c1e00 0000000000000000 ffffffff808a1708 8000000409a54000 80000000271bd300 80000000271bd320 8000000409a54030 ffffffff80ff0f00 0000000000000001 ffffffff81090000 ffffffff808a1ac0 8000000402182080 ffffffff84650000 8000000402182080 ffffffff84650000 ffffffff80ff0000 8000000409a54000 ffffffff808a1970 0000000000000000 80000004099e8000 8000000402099240 0000000000000000 ffffffff808a8598 0000000000000000 8000000408eeeb00 8000000409a54000 00000000810a1d00 0000000000000000 8000000409d73de8 8000000409d73de8 0000000000000088 000000000c009c00 8000000409d73e08 8000000409d73e08 8000000402182080 ffffffff808a84d0 8000000402182080 ... Call Trace: [] netif_carrier_off+0xc/0x58 [] phy_state_machine+0x48c/0x4f8 [] process_one_work+0x158/0x368 [] worker_thread+0x150/0x4c0 [] kthread+0xc8/0xe0 [] ret_from_kernel_thread+0x14/0x1c The original motivation for this change originated from Marc Gonzales indicating that his network driver did not have its adjust_link callback executing with phydev->link = 0 while he was expecting it. PHYLIB has never made any such guarantees ever because phy_stop() merely just tells the workqueue to move into PHY_HALTED state which will happen asynchronously. Reported-by: Geert Uytterhoeven Reported-by: David Daney Fixes: 7ad813f20853 ("net: phy: Correctly process PHY_HALTED in phy_stop_machine()") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/phy.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 5068c58..d0626bf 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -749,9 +749,6 @@ void phy_stop_machine(struct phy_device *phydev) if (phydev->state > PHY_UP && phydev->state != PHY_HALTED) phydev->state = PHY_UP; mutex_unlock(&phydev->lock); - - /* Now we can run the state machine synchronously */ - phy_state_machine(&phydev->state_queue.work); } /** -- cgit v1.1 From 0b89ede629637c03b9a728fb08bfe6ed51de9be7 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 31 Aug 2017 12:30:54 +0200 Subject: s390/mm: fork vs. 5 level page tabel The mm->context.asce field of a new process is not set up correctly in case of a fork with a 5 level page table. Add the missing case to init_new_context(). Fixes: 1aea9b3f9210 ("s390/mm: implement 5 level pages tables") Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/mmu_context.h | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index 4541ac4..24bc416 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -44,6 +44,11 @@ static inline int init_new_context(struct task_struct *tsk, mm->context.asce = __pa(mm->pgd) | _ASCE_TABLE_LENGTH | _ASCE_USER_BITS | _ASCE_TYPE_REGION3; break; + case -PAGE_SIZE: + /* forked 5-level task, set new asce with new_mm->pgd */ + mm->context.asce = __pa(mm->pgd) | _ASCE_TABLE_LENGTH | + _ASCE_USER_BITS | _ASCE_TYPE_REGION1; + break; case 1UL << 53: /* forked 4-level task, set new asce with new mm->pgd */ mm->context.asce = __pa(mm->pgd) | _ASCE_TABLE_LENGTH | -- cgit v1.1 From 8ab867cb0806a8e195de161fd8883a0578d1d050 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 31 Aug 2017 13:18:22 +0200 Subject: s390/mm: fix BUG_ON in crst_table_upgrade A 31-bit compat process can force a BUG_ON in crst_table_upgrade with specific, invalid mmap calls, e.g. mmap((void*) 0x7fff8000, 0x10000, 3, 32, -1, 0) The arch_get_unmapped_area[_topdown] functions miss an if condition in the decision to do a page table upgrade. Fixes: 9b11c7912d00 ("s390/mm: simplify arch_get_unmapped_area[_topdown]") Cc: # v4.12+ Signed-off-by: Martin Schwidefsky --- arch/s390/mm/mmap.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c index 2e10d2b..5bea139 100644 --- a/arch/s390/mm/mmap.c +++ b/arch/s390/mm/mmap.c @@ -119,7 +119,8 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, return addr; check_asce_limit: - if (addr + len > current->mm->context.asce_limit) { + if (addr + len > current->mm->context.asce_limit && + addr + len <= TASK_SIZE) { rc = crst_table_upgrade(mm, addr + len); if (rc) return (unsigned long) rc; @@ -183,7 +184,8 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, } check_asce_limit: - if (addr + len > current->mm->context.asce_limit) { + if (addr + len > current->mm->context.asce_limit && + addr + len <= TASK_SIZE) { rc = crst_table_upgrade(mm, addr + len); if (rc) return (unsigned long) rc; -- cgit v1.1 From f5308d1b83eba20e69df5e0926ba7257c8dd9074 Mon Sep 17 00:00:00 2001 From: Cameron Gutman Date: Thu, 31 Aug 2017 11:52:20 -0700 Subject: Input: xpad - fix PowerA init quirk for some gamepad models The PowerA gamepad initialization quirk worked with the PowerA wired gamepad I had around (0x24c6:0x543a), but a user reported [0] that it didn't work for him, even though our gamepads shared the same vendor and product IDs. When I initially implemented the PowerA quirk, I wanted to avoid actually triggering the rumble action during init. My tests showed that my gamepad would work correctly even if it received a rumble of 0 intensity, so that's what I went with. Unfortunately, this apparently isn't true for all models (perhaps a firmware difference?). This non-working gamepad seems to require the real magic rumble packet that the Microsoft driver sends, which actually vibrates the gamepad. To counteract this effect, I still send the old zero-rumble PowerA quirk packet which cancels the rumble effect before the motors can spin up enough to vibrate. [0]: https://github.com/paroj/xpad/issues/48#issuecomment-313904867 Reported-by: Kyle Beauchamp Tested-by: Kyle Beauchamp Fixes: 81093c9848a7 ("Input: xpad - support some quirky Xbox One pads") Cc: stable@vger.kernel.org # v4.12 Signed-off-by: Cameron Gutman Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 24 +++++++++++++++++++----- 1 file changed, 19 insertions(+), 5 deletions(-) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 298a6ba..ca0e19a 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -476,10 +476,21 @@ static const u8 xboxone_hori_init[] = { }; /* - * A rumble packet is required for some PowerA pads to start + * A specific rumble packet is required for some PowerA pads to start * sending input reports. One of those pads is (0x24c6:0x543a). */ -static const u8 xboxone_zerorumble_init[] = { +static const u8 xboxone_rumblebegin_init[] = { + 0x09, 0x00, 0x00, 0x09, 0x00, 0x0F, 0x00, 0x00, + 0x1D, 0x1D, 0xFF, 0x00, 0x00 +}; + +/* + * A rumble packet with zero FF intensity will immediately + * terminate the rumbling required to init PowerA pads. + * This should happen fast enough that the motors don't + * spin up to enough speed to actually vibrate the gamepad. + */ +static const u8 xboxone_rumbleend_init[] = { 0x09, 0x00, 0x00, 0x09, 0x00, 0x0F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }; @@ -494,9 +505,12 @@ static const struct xboxone_init_packet xboxone_init_packets[] = { XBOXONE_INIT_PKT(0x0e6f, 0x0165, xboxone_hori_init), XBOXONE_INIT_PKT(0x0f0d, 0x0067, xboxone_hori_init), XBOXONE_INIT_PKT(0x0000, 0x0000, xboxone_fw2015_init), - XBOXONE_INIT_PKT(0x24c6, 0x541a, xboxone_zerorumble_init), - XBOXONE_INIT_PKT(0x24c6, 0x542a, xboxone_zerorumble_init), - XBOXONE_INIT_PKT(0x24c6, 0x543a, xboxone_zerorumble_init), + XBOXONE_INIT_PKT(0x24c6, 0x541a, xboxone_rumblebegin_init), + XBOXONE_INIT_PKT(0x24c6, 0x542a, xboxone_rumblebegin_init), + XBOXONE_INIT_PKT(0x24c6, 0x543a, xboxone_rumblebegin_init), + XBOXONE_INIT_PKT(0x24c6, 0x541a, xboxone_rumbleend_init), + XBOXONE_INIT_PKT(0x24c6, 0x542a, xboxone_rumbleend_init), + XBOXONE_INIT_PKT(0x24c6, 0x543a, xboxone_rumbleend_init), }; struct xpad_output_packet { -- cgit v1.1 From f581a0dd744fe32b0a8805e279c59ec1ac676d60 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 31 Aug 2017 16:47:43 +0200 Subject: wl1251: add a missing spin_lock_init() wl1251: add a missing spin_lock_init() This fixes the following kernel warning: [ 5668.771453] BUG: spinlock bad magic on CPU#0, kworker/u2:3/9745 [ 5668.771850] lock: 0xce63ef20, .magic: 00000000, .owner: /-1, .owner_cpu: 0 [ 5668.772277] CPU: 0 PID: 9745 Comm: kworker/u2:3 Tainted: G W 4.12.0-03002-gec979a4-dirty #40 [ 5668.772796] Hardware name: Nokia RX-51 board [ 5668.773071] Workqueue: phy1 wl1251_irq_work [ 5668.773345] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [ 5668.773803] [] (show_stack) from [] (do_raw_spin_lock+0x6c/0xa0) [ 5668.774230] [] (do_raw_spin_lock) from [] (_raw_spin_lock_irqsave+0x10/0x18) [ 5668.774658] [] (_raw_spin_lock_irqsave) from [] (wl1251_op_tx+0x38/0x5c) [ 5668.775115] [] (wl1251_op_tx) from [] (ieee80211_tx_frags+0x188/0x1c0) [ 5668.775543] [] (ieee80211_tx_frags) from [] (__ieee80211_tx+0x6c/0x130) [ 5668.775970] [] (__ieee80211_tx) from [] (ieee80211_tx+0xdc/0x104) [ 5668.776367] [] (ieee80211_tx) from [] (__ieee80211_subif_start_xmit+0x454/0x8c8) [ 5668.776824] [] (__ieee80211_subif_start_xmit) from [] (ieee80211_subif_start_xmit+0x30/0x2fc) [ 5668.777343] [] (ieee80211_subif_start_xmit) from [] (dev_hard_start_xmit+0x80/0x118) ... by adding the missing spin_lock_init(). Reported-by: Pavel Machek Cc: Kalle Valo Signed-off-by: Cong Wang Acked-by: Pavel Machek Signed-off-by: Kalle Valo Signed-off-by: Pavel Machek Cc: stable@kernel.org Signed-off-by: David S. Miller --- drivers/net/wireless/ti/wl1251/main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/ti/wl1251/main.c b/drivers/net/wireless/ti/wl1251/main.c index 08f0477..9915d83 100644 --- a/drivers/net/wireless/ti/wl1251/main.c +++ b/drivers/net/wireless/ti/wl1251/main.c @@ -1571,6 +1571,7 @@ struct ieee80211_hw *wl1251_alloc_hw(void) wl->state = WL1251_STATE_OFF; mutex_init(&wl->mutex); + spin_lock_init(&wl->wl_lock); wl->tx_mgmt_frm_rate = DEFAULT_HW_GEN_TX_RATE; wl->tx_mgmt_frm_mod = DEFAULT_HW_GEN_MODULATION_TYPE; -- cgit v1.1 From dd2bc473482eedc60c29cf00ad12568ce40ce511 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 4 Aug 2017 11:22:31 +0800 Subject: ceph: fix readpage from fscache ceph_readpage() unlocks page prematurely prematurely in the case that page is reading from fscache. Caller of readpage expects that page is uptodate when it get unlocked. So page shoule get locked by completion callback of fscache_read_or_alloc_pages() Cc: stable@vger.kernel.org # 4.1+, needs backporting for < 4.7 Signed-off-by: "Yan, Zheng" Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov --- fs/ceph/addr.c | 24 +++++++++++++++--------- fs/ceph/cache.c | 12 +++--------- 2 files changed, 18 insertions(+), 18 deletions(-) diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 5083628..1bc709f 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -189,7 +189,7 @@ static int ceph_releasepage(struct page *page, gfp_t g) /* * read a single page, without unlocking it. */ -static int readpage_nounlock(struct file *filp, struct page *page) +static int ceph_do_readpage(struct file *filp, struct page *page) { struct inode *inode = file_inode(filp); struct ceph_inode_info *ci = ceph_inode(inode); @@ -219,7 +219,7 @@ static int readpage_nounlock(struct file *filp, struct page *page) err = ceph_readpage_from_fscache(inode, page); if (err == 0) - goto out; + return -EINPROGRESS; dout("readpage inode %p file %p page %p index %lu\n", inode, filp, page, page->index); @@ -249,8 +249,11 @@ out: static int ceph_readpage(struct file *filp, struct page *page) { - int r = readpage_nounlock(filp, page); - unlock_page(page); + int r = ceph_do_readpage(filp, page); + if (r != -EINPROGRESS) + unlock_page(page); + else + r = 0; return r; } @@ -1237,7 +1240,7 @@ retry_locked: goto retry_locked; r = writepage_nounlock(page, NULL); if (r < 0) - goto fail_nosnap; + goto fail_unlock; goto retry_locked; } @@ -1265,11 +1268,14 @@ retry_locked: } /* we need to read it. */ - r = readpage_nounlock(file, page); - if (r < 0) - goto fail_nosnap; + r = ceph_do_readpage(file, page); + if (r < 0) { + if (r == -EINPROGRESS) + return -EAGAIN; + goto fail_unlock; + } goto retry_locked; -fail_nosnap: +fail_unlock: unlock_page(page); return r; } diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c index fd11728..337f886 100644 --- a/fs/ceph/cache.c +++ b/fs/ceph/cache.c @@ -297,13 +297,7 @@ void ceph_fscache_file_set_cookie(struct inode *inode, struct file *filp) } } -static void ceph_vfs_readpage_complete(struct page *page, void *data, int error) -{ - if (!error) - SetPageUptodate(page); -} - -static void ceph_vfs_readpage_complete_unlock(struct page *page, void *data, int error) +static void ceph_readpage_from_fscache_complete(struct page *page, void *data, int error) { if (!error) SetPageUptodate(page); @@ -331,7 +325,7 @@ int ceph_readpage_from_fscache(struct inode *inode, struct page *page) return -ENOBUFS; ret = fscache_read_or_alloc_page(ci->fscache, page, - ceph_vfs_readpage_complete, NULL, + ceph_readpage_from_fscache_complete, NULL, GFP_KERNEL); switch (ret) { @@ -360,7 +354,7 @@ int ceph_readpages_from_fscache(struct inode *inode, return -ENOBUFS; ret = fscache_read_or_alloc_pages(ci->fscache, mapping, pages, nr_pages, - ceph_vfs_readpage_complete_unlock, + ceph_readpage_from_fscache_complete, NULL, mapping_gfp_mask(mapping)); switch (ret) { -- cgit v1.1 From a4d1a885251382250ec315482bdd8ca52dd61e6a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:26 -0400 Subject: dax: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Replace all mmu_notifier_invalidate_page() calls by *_invalidate_range() and make sure it is bracketed by calls to *_invalidate_range_start()/end(). Note that because we can not presume the pmd value or pte value we have to assume the worst and unconditionaly report an invalidation as happening. Signed-off-by: Jérôme Glisse Cc: Dan Williams Cc: Ross Zwisler Cc: Bernhard Held Cc: Adam Borowski Cc: Andrea Arcangeli Cc: Radim Krčmář Cc: Wanpeng Li Cc: Paolo Bonzini Cc: Takashi Iwai Cc: Nadav Amit Cc: Mike Galbraith Cc: Kirill A. Shutemov Cc: axie Cc: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dax.c | 19 +++++++++++-------- include/linux/mm.h | 1 + mm/memory.c | 26 +++++++++++++++++++++----- 3 files changed, 33 insertions(+), 13 deletions(-) diff --git a/fs/dax.c b/fs/dax.c index 865d42c..ab925dc 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -646,11 +646,10 @@ static void dax_mapping_entry_mkclean(struct address_space *mapping, pte_t pte, *ptep = NULL; pmd_t *pmdp = NULL; spinlock_t *ptl; - bool changed; i_mmap_lock_read(mapping); vma_interval_tree_foreach(vma, &mapping->i_mmap, index, index) { - unsigned long address; + unsigned long address, start, end; cond_resched(); @@ -658,8 +657,13 @@ static void dax_mapping_entry_mkclean(struct address_space *mapping, continue; address = pgoff_address(index, vma); - changed = false; - if (follow_pte_pmd(vma->vm_mm, address, &ptep, &pmdp, &ptl)) + + /* + * Note because we provide start/end to follow_pte_pmd it will + * call mmu_notifier_invalidate_range_start() on our behalf + * before taking any lock. + */ + if (follow_pte_pmd(vma->vm_mm, address, &start, &end, &ptep, &pmdp, &ptl)) continue; if (pmdp) { @@ -676,7 +680,7 @@ static void dax_mapping_entry_mkclean(struct address_space *mapping, pmd = pmd_wrprotect(pmd); pmd = pmd_mkclean(pmd); set_pmd_at(vma->vm_mm, address, pmdp, pmd); - changed = true; + mmu_notifier_invalidate_range(vma->vm_mm, start, end); unlock_pmd: spin_unlock(ptl); #endif @@ -691,13 +695,12 @@ unlock_pmd: pte = pte_wrprotect(pte); pte = pte_mkclean(pte); set_pte_at(vma->vm_mm, address, ptep, pte); - changed = true; + mmu_notifier_invalidate_range(vma->vm_mm, start, end); unlock_pte: pte_unmap_unlock(ptep, ptl); } - if (changed) - mmu_notifier_invalidate_page(vma->vm_mm, address); + mmu_notifier_invalidate_range_end(vma->vm_mm, start, end); } i_mmap_unlock_read(mapping); } diff --git a/include/linux/mm.h b/include/linux/mm.h index 46b9ac5..c1f6c95 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1260,6 +1260,7 @@ int copy_page_range(struct mm_struct *dst, struct mm_struct *src, void unmap_mapping_range(struct address_space *mapping, loff_t const holebegin, loff_t const holelen, int even_cows); int follow_pte_pmd(struct mm_struct *mm, unsigned long address, + unsigned long *start, unsigned long *end, pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp); int follow_pfn(struct vm_area_struct *vma, unsigned long address, unsigned long *pfn); diff --git a/mm/memory.c b/mm/memory.c index fe2fba2..56e48e4 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4008,7 +4008,8 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address) #endif /* __PAGETABLE_PMD_FOLDED */ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address, - pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp) + unsigned long *start, unsigned long *end, + pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp) { pgd_t *pgd; p4d_t *p4d; @@ -4035,17 +4036,29 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address, if (!pmdpp) goto out; + if (start && end) { + *start = address & PMD_MASK; + *end = *start + PMD_SIZE; + mmu_notifier_invalidate_range_start(mm, *start, *end); + } *ptlp = pmd_lock(mm, pmd); if (pmd_huge(*pmd)) { *pmdpp = pmd; return 0; } spin_unlock(*ptlp); + if (start && end) + mmu_notifier_invalidate_range_end(mm, *start, *end); } if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) goto out; + if (start && end) { + *start = address & PAGE_MASK; + *end = *start + PAGE_SIZE; + mmu_notifier_invalidate_range_start(mm, *start, *end); + } ptep = pte_offset_map_lock(mm, pmd, address, ptlp); if (!pte_present(*ptep)) goto unlock; @@ -4053,6 +4066,8 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address, return 0; unlock: pte_unmap_unlock(ptep, *ptlp); + if (start && end) + mmu_notifier_invalidate_range_end(mm, *start, *end); out: return -EINVAL; } @@ -4064,20 +4079,21 @@ static inline int follow_pte(struct mm_struct *mm, unsigned long address, /* (void) is needed to make gcc happy */ (void) __cond_lock(*ptlp, - !(res = __follow_pte_pmd(mm, address, ptepp, NULL, - ptlp))); + !(res = __follow_pte_pmd(mm, address, NULL, NULL, + ptepp, NULL, ptlp))); return res; } int follow_pte_pmd(struct mm_struct *mm, unsigned long address, + unsigned long *start, unsigned long *end, pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp) { int res; /* (void) is needed to make gcc happy */ (void) __cond_lock(*ptlp, - !(res = __follow_pte_pmd(mm, address, ptepp, pmdpp, - ptlp))); + !(res = __follow_pte_pmd(mm, address, start, end, + ptepp, pmdpp, ptlp))); return res; } EXPORT_SYMBOL(follow_pte_pmd); -- cgit v1.1 From 369ea8242c0fb5239b4ddf0dc568f694bd244de4 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:27 -0400 Subject: mm/rmap: update to new mmu_notifier semantic v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Replace all mmu_notifier_invalidate_page() calls by *_invalidate_range() and make sure it is bracketed by calls to *_invalidate_range_start()/end(). Note that because we can not presume the pmd value or pte value we have to assume the worst and unconditionaly report an invalidation as happening. Changed since v2: - try_to_unmap_one() only one call to mmu_notifier_invalidate_range() - compute end with PAGE_SIZE << compound_order(page) - fix PageHuge() case in try_to_unmap_one() Signed-off-by: Jérôme Glisse Reviewed-by: Andrea Arcangeli Cc: Dan Williams Cc: Ross Zwisler Cc: Bernhard Held Cc: Adam Borowski Cc: Radim Krčmář Cc: Wanpeng Li Cc: Paolo Bonzini Cc: Takashi Iwai Cc: Nadav Amit Cc: Mike Galbraith Cc: Kirill A. Shutemov Cc: axie Cc: Andrew Morton Signed-off-by: Linus Torvalds --- mm/rmap.c | 35 ++++++++++++++++++++++++++++++++--- 1 file changed, 32 insertions(+), 3 deletions(-) diff --git a/mm/rmap.c b/mm/rmap.c index c8993c6..c570f82 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -887,11 +887,21 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, .address = address, .flags = PVMW_SYNC, }; + unsigned long start = address, end; int *cleaned = arg; + /* + * We have to assume the worse case ie pmd for invalidation. Note that + * the page can not be free from this function. + */ + end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page))); + mmu_notifier_invalidate_range_start(vma->vm_mm, start, end); + while (page_vma_mapped_walk(&pvmw)) { + unsigned long cstart, cend; int ret = 0; - address = pvmw.address; + + cstart = address = pvmw.address; if (pvmw.pte) { pte_t entry; pte_t *pte = pvmw.pte; @@ -904,6 +914,7 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, entry = pte_wrprotect(entry); entry = pte_mkclean(entry); set_pte_at(vma->vm_mm, address, pte, entry); + cend = cstart + PAGE_SIZE; ret = 1; } else { #ifdef CONFIG_TRANSPARENT_HUGE_PAGECACHE @@ -918,6 +929,8 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, entry = pmd_wrprotect(entry); entry = pmd_mkclean(entry); set_pmd_at(vma->vm_mm, address, pmd, entry); + cstart &= PMD_MASK; + cend = cstart + PMD_SIZE; ret = 1; #else /* unexpected pmd-mapped page? */ @@ -926,11 +939,13 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma, } if (ret) { - mmu_notifier_invalidate_page(vma->vm_mm, address); + mmu_notifier_invalidate_range(vma->vm_mm, cstart, cend); (*cleaned)++; } } + mmu_notifier_invalidate_range_end(vma->vm_mm, start, end); + return true; } @@ -1324,6 +1339,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, pte_t pteval; struct page *subpage; bool ret = true; + unsigned long start = address, end; enum ttu_flags flags = (enum ttu_flags)arg; /* munlock has nothing to gain from examining un-locked vmas */ @@ -1335,6 +1351,14 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, flags & TTU_MIGRATION, page); } + /* + * We have to assume the worse case ie pmd for invalidation. Note that + * the page can not be free in this function as call of try_to_unmap() + * must hold a reference on the page. + */ + end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page))); + mmu_notifier_invalidate_range_start(vma->vm_mm, start, end); + while (page_vma_mapped_walk(&pvmw)) { /* * If the page is mlock()d, we cannot swap it out. @@ -1445,6 +1469,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, if (unlikely(PageSwapBacked(page) != PageSwapCache(page))) { WARN_ON_ONCE(1); ret = false; + /* We have to invalidate as we cleared the pte */ page_vma_mapped_walk_done(&pvmw); break; } @@ -1490,8 +1515,12 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma, discard: page_remove_rmap(subpage, PageHuge(page)); put_page(page); - mmu_notifier_invalidate_page(mm, address); + mmu_notifier_invalidate_range(mm, address, + address + PAGE_SIZE); } + + mmu_notifier_invalidate_range_end(vma->vm_mm, start, end); + return ret; } -- cgit v1.1 From d1d5762e4767324ce4ec32f9b2d2aaccfd87f664 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:28 -0400 Subject: powerpc/powernv: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and now are bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Cc: linuxppc-dev@lists.ozlabs.org Cc: Alistair Popple Cc: Michael Ellerman Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- arch/powerpc/platforms/powernv/npu-dma.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/arch/powerpc/platforms/powernv/npu-dma.c b/arch/powerpc/platforms/powernv/npu-dma.c index b5d960d..4c7b859 100644 --- a/arch/powerpc/platforms/powernv/npu-dma.c +++ b/arch/powerpc/platforms/powernv/npu-dma.c @@ -614,15 +614,6 @@ static void pnv_npu2_mn_change_pte(struct mmu_notifier *mn, mmio_invalidate(npu_context, 1, address, true); } -static void pnv_npu2_mn_invalidate_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - struct npu_context *npu_context = mn_to_npu_context(mn); - - mmio_invalidate(npu_context, 1, address, true); -} - static void pnv_npu2_mn_invalidate_range(struct mmu_notifier *mn, struct mm_struct *mm, unsigned long start, unsigned long end) @@ -640,7 +631,6 @@ static void pnv_npu2_mn_invalidate_range(struct mmu_notifier *mn, static const struct mmu_notifier_ops nv_nmmu_notifier_ops = { .release = pnv_npu2_mn_release, .change_pte = pnv_npu2_mn_change_pte, - .invalidate_page = pnv_npu2_mn_invalidate_page, .invalidate_range = pnv_npu2_mn_invalidate_range, }; -- cgit v1.1 From c90270a9f11b455102874c15c4c9c1dabb0d030b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:29 -0400 Subject: drm/amdgpu: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Reviewed-by: Christian König Cc: amd-gfx@lists.freedesktop.org Cc: Felix Kuehling Cc: Christian König Cc: Alex Deucher Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c | 31 ------------------------------- 1 file changed, 31 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c index 6558a3e..e1cde6b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c @@ -147,36 +147,6 @@ static void amdgpu_mn_invalidate_node(struct amdgpu_mn_node *node, } /** - * amdgpu_mn_invalidate_page - callback to notify about mm change - * - * @mn: our notifier - * @mn: the mm this callback is about - * @address: address of invalidate page - * - * Invalidation of a single page. Blocks for all BOs mapping it - * and unmap them by move them into system domain again. - */ -static void amdgpu_mn_invalidate_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - struct amdgpu_mn *rmn = container_of(mn, struct amdgpu_mn, mn); - struct interval_tree_node *it; - - mutex_lock(&rmn->lock); - - it = interval_tree_iter_first(&rmn->objects, address, address); - if (it) { - struct amdgpu_mn_node *node; - - node = container_of(it, struct amdgpu_mn_node, it); - amdgpu_mn_invalidate_node(node, address, address); - } - - mutex_unlock(&rmn->lock); -} - -/** * amdgpu_mn_invalidate_range_start - callback to notify about mm change * * @mn: our notifier @@ -215,7 +185,6 @@ static void amdgpu_mn_invalidate_range_start(struct mmu_notifier *mn, static const struct mmu_notifier_ops amdgpu_mn_ops = { .release = amdgpu_mn_release, - .invalidate_page = amdgpu_mn_invalidate_page, .invalidate_range_start = amdgpu_mn_invalidate_range_start, }; -- cgit v1.1 From b1a89257f28e28eccd65f777ff622c3aab6d8526 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:30 -0400 Subject: IB/umem: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Tested-by: Leon Romanovsky Cc: linux-rdma@vger.kernel.org Cc: Artemy Kovalyov Cc: Doug Ledford Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/infiniband/core/umem_odp.c | 19 ------------------- 1 file changed, 19 deletions(-) diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index 8c4ec56..55e8f5e 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -166,24 +166,6 @@ static int invalidate_page_trampoline(struct ib_umem *item, u64 start, return 0; } -static void ib_umem_notifier_invalidate_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - struct ib_ucontext *context = container_of(mn, struct ib_ucontext, mn); - - if (!context->invalidate_range) - return; - - ib_ucontext_notifier_start_account(context); - down_read(&context->umem_rwsem); - rbt_ib_umem_for_each_in_range(&context->umem_tree, address, - address + PAGE_SIZE, - invalidate_page_trampoline, NULL); - up_read(&context->umem_rwsem); - ib_ucontext_notifier_end_account(context); -} - static int invalidate_range_start_trampoline(struct ib_umem *item, u64 start, u64 end, void *cookie) { @@ -237,7 +219,6 @@ static void ib_umem_notifier_invalidate_range_end(struct mmu_notifier *mn, static const struct mmu_notifier_ops ib_umem_notifiers = { .release = ib_umem_notifier_release, - .invalidate_page = ib_umem_notifier_invalidate_page, .invalidate_range_start = ib_umem_notifier_invalidate_range_start, .invalidate_range_end = ib_umem_notifier_invalidate_range_end, }; -- cgit v1.1 From 7def96f0a97310c92e6145c776164af479fb1591 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:31 -0400 Subject: IB/hfi1: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Cc: linux-rdma@vger.kernel.org Cc: Dean Luick Cc: Ira Weiny Cc: Doug Ledford Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/infiniband/hw/hfi1/mmu_rb.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/infiniband/hw/hfi1/mmu_rb.c b/drivers/infiniband/hw/hfi1/mmu_rb.c index ccbf52c..e4b56a0 100644 --- a/drivers/infiniband/hw/hfi1/mmu_rb.c +++ b/drivers/infiniband/hw/hfi1/mmu_rb.c @@ -67,8 +67,6 @@ struct mmu_rb_handler { static unsigned long mmu_node_start(struct mmu_rb_node *); static unsigned long mmu_node_last(struct mmu_rb_node *); -static inline void mmu_notifier_page(struct mmu_notifier *, struct mm_struct *, - unsigned long); static inline void mmu_notifier_range_start(struct mmu_notifier *, struct mm_struct *, unsigned long, unsigned long); @@ -82,7 +80,6 @@ static void do_remove(struct mmu_rb_handler *handler, static void handle_remove(struct work_struct *work); static const struct mmu_notifier_ops mn_opts = { - .invalidate_page = mmu_notifier_page, .invalidate_range_start = mmu_notifier_range_start, }; @@ -285,12 +282,6 @@ void hfi1_mmu_rb_remove(struct mmu_rb_handler *handler, handler->ops->remove(handler->ops_arg, node); } -static inline void mmu_notifier_page(struct mmu_notifier *mn, - struct mm_struct *mm, unsigned long addr) -{ - mmu_notifier_mem_invalidate(mn, mm, addr, addr + PAGE_SIZE); -} - static inline void mmu_notifier_range_start(struct mmu_notifier *mn, struct mm_struct *mm, unsigned long start, -- cgit v1.1 From f0d1c713d66d7b16527568e79528964cf4b5513d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:32 -0400 Subject: iommu/amd: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Cc: Suravee Suthikulpanit Cc: iommu@lists.linux-foundation.org Cc: Joerg Roedel Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/iommu/amd_iommu_v2.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/iommu/amd_iommu_v2.c b/drivers/iommu/amd_iommu_v2.c index 6629c47..dccf5b7 100644 --- a/drivers/iommu/amd_iommu_v2.c +++ b/drivers/iommu/amd_iommu_v2.c @@ -391,13 +391,6 @@ static int mn_clear_flush_young(struct mmu_notifier *mn, return 0; } -static void mn_invalidate_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - __mn_flush_page(mn, address); -} - static void mn_invalidate_range(struct mmu_notifier *mn, struct mm_struct *mm, unsigned long start, unsigned long end) @@ -436,7 +429,6 @@ static void mn_release(struct mmu_notifier *mn, struct mm_struct *mm) static const struct mmu_notifier_ops iommu_mn = { .release = mn_release, .clear_flush_young = mn_clear_flush_young, - .invalidate_page = mn_invalidate_page, .invalidate_range = mn_invalidate_range, }; -- cgit v1.1 From 30ef7d2c056cb7ef22c16c6fe0756105d584985c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:33 -0400 Subject: iommu/intel: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Cc: David Woodhouse Cc: iommu@lists.linux-foundation.org Cc: Joerg Roedel Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/iommu/intel-svm.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/iommu/intel-svm.c b/drivers/iommu/intel-svm.c index f167c0d..f620dcc 100644 --- a/drivers/iommu/intel-svm.c +++ b/drivers/iommu/intel-svm.c @@ -223,14 +223,6 @@ static void intel_change_pte(struct mmu_notifier *mn, struct mm_struct *mm, intel_flush_svm_range(svm, address, 1, 1, 0); } -static void intel_invalidate_page(struct mmu_notifier *mn, struct mm_struct *mm, - unsigned long address) -{ - struct intel_svm *svm = container_of(mn, struct intel_svm, notifier); - - intel_flush_svm_range(svm, address, 1, 1, 0); -} - /* Pages have been freed at this point */ static void intel_invalidate_range(struct mmu_notifier *mn, struct mm_struct *mm, @@ -285,7 +277,6 @@ static void intel_mm_release(struct mmu_notifier *mn, struct mm_struct *mm) static const struct mmu_notifier_ops intel_mmuops = { .release = intel_mm_release, .change_pte = intel_change_pte, - .invalidate_page = intel_invalidate_page, .invalidate_range = intel_invalidate_range, }; -- cgit v1.1 From 192e856451d8d5e15aad470dbbb5daf5de3d2c0f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:34 -0400 Subject: misc/mic/scif: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Cc: Sudeep Dutt Cc: Ashutosh Dixit Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/misc/mic/scif/scif_dma.c | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/drivers/misc/mic/scif/scif_dma.c b/drivers/misc/mic/scif/scif_dma.c index 64d5760..63d6246 100644 --- a/drivers/misc/mic/scif/scif_dma.c +++ b/drivers/misc/mic/scif/scif_dma.c @@ -200,16 +200,6 @@ static void scif_mmu_notifier_release(struct mmu_notifier *mn, schedule_work(&scif_info.misc_work); } -static void scif_mmu_notifier_invalidate_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - struct scif_mmu_notif *mmn; - - mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier); - scif_rma_destroy_tcw(mmn, address, PAGE_SIZE); -} - static void scif_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, struct mm_struct *mm, unsigned long start, @@ -235,7 +225,6 @@ static void scif_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn, static const struct mmu_notifier_ops scif_mmu_notifier_ops = { .release = scif_mmu_notifier_release, .clear_flush_young = NULL, - .invalidate_page = scif_mmu_notifier_invalidate_page, .invalidate_range_start = scif_mmu_notifier_invalidate_range_start, .invalidate_range_end = scif_mmu_notifier_invalidate_range_end}; -- cgit v1.1 From a4870125ca9fcd221fbc5f067967e56cfb4d9fbd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:35 -0400 Subject: sgi-gru: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Cc: Dimitri Sivanich Cc: Jack Steiner Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/misc/sgi-gru/grutlbpurge.c | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/drivers/misc/sgi-gru/grutlbpurge.c b/drivers/misc/sgi-gru/grutlbpurge.c index e936d43..9918eda 100644 --- a/drivers/misc/sgi-gru/grutlbpurge.c +++ b/drivers/misc/sgi-gru/grutlbpurge.c @@ -247,17 +247,6 @@ static void gru_invalidate_range_end(struct mmu_notifier *mn, gru_dbg(grudev, "gms %p, start 0x%lx, end 0x%lx\n", gms, start, end); } -static void gru_invalidate_page(struct mmu_notifier *mn, struct mm_struct *mm, - unsigned long address) -{ - struct gru_mm_struct *gms = container_of(mn, struct gru_mm_struct, - ms_notifier); - - STAT(mmu_invalidate_page); - gru_flush_tlb_range(gms, address, PAGE_SIZE); - gru_dbg(grudev, "gms %p, address 0x%lx\n", gms, address); -} - static void gru_release(struct mmu_notifier *mn, struct mm_struct *mm) { struct gru_mm_struct *gms = container_of(mn, struct gru_mm_struct, @@ -269,7 +258,6 @@ static void gru_release(struct mmu_notifier *mn, struct mm_struct *mm) static const struct mmu_notifier_ops gru_mmuops = { - .invalidate_page = gru_invalidate_page, .invalidate_range_start = gru_invalidate_range_start, .invalidate_range_end = gru_invalidate_range_end, .release = gru_release, -- cgit v1.1 From a81461b0546cbf57128f34e3dd6013bcdbdf0371 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:36 -0400 Subject: xen/gntdev: update to new mmu_notifier semantic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Signed-off-by: Jérôme Glisse Reviewed-by: Boris Ostrovsky Cc: Konrad Rzeszutek Wilk Cc: Roger Pau Monné Cc: xen-devel@lists.xenproject.org (moderated for non-subscribers) Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- drivers/xen/gntdev.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c index f3bf8f4..8236059 100644 --- a/drivers/xen/gntdev.c +++ b/drivers/xen/gntdev.c @@ -484,13 +484,6 @@ static void mn_invl_range_start(struct mmu_notifier *mn, mutex_unlock(&priv->lock); } -static void mn_invl_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - mn_invl_range_start(mn, mm, address, address + PAGE_SIZE); -} - static void mn_release(struct mmu_notifier *mn, struct mm_struct *mm) { @@ -522,7 +515,6 @@ static void mn_release(struct mmu_notifier *mn, static const struct mmu_notifier_ops gntdev_mmu_ops = { .release = mn_release, - .invalidate_page = mn_invl_page, .invalidate_range_start = mn_invl_range_start, }; -- cgit v1.1 From fb1522e099f0c69f36655af233a64e3f55941f5b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:37 -0400 Subject: KVM: update to new mmu_notifier semantic v2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Calls to mmu_notifier_invalidate_page() were replaced by calls to mmu_notifier_invalidate_range() and are now bracketed by calls to mmu_notifier_invalidate_range_start()/end() Remove now useless invalidate_page callback. Changed since v1 (Linus Torvalds) - remove now useless kvm_arch_mmu_notifier_invalidate_page() Signed-off-by: Jérôme Glisse Tested-by: Mike Galbraith Tested-by: Adam Borowski Cc: Paolo Bonzini Cc: Radim Krčmář Cc: kvm@vger.kernel.org Cc: Kirill A. Shutemov Cc: Andrew Morton Cc: Andrea Arcangeli Signed-off-by: Linus Torvalds --- arch/arm/include/asm/kvm_host.h | 6 ------ arch/arm64/include/asm/kvm_host.h | 6 ------ arch/mips/include/asm/kvm_host.h | 5 ----- arch/powerpc/include/asm/kvm_host.h | 5 ----- arch/x86/include/asm/kvm_host.h | 2 -- arch/x86/kvm/x86.c | 11 ---------- virt/kvm/kvm_main.c | 42 ------------------------------------- 7 files changed, 77 deletions(-) diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h index 127e2dd..4a879f6 100644 --- a/arch/arm/include/asm/kvm_host.h +++ b/arch/arm/include/asm/kvm_host.h @@ -225,12 +225,6 @@ int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices); int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end); int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); -/* We do not have shadow page tables, hence the empty hooks */ -static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, - unsigned long address) -{ -} - struct kvm_vcpu *kvm_arm_get_running_vcpu(void); struct kvm_vcpu __percpu **kvm_get_running_vcpus(void); void kvm_arm_halt_guest(struct kvm *kvm); diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index d686300..e923b58 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -326,12 +326,6 @@ void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end); int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); -/* We do not have shadow page tables, hence the empty hooks */ -static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, - unsigned long address) -{ -} - struct kvm_vcpu *kvm_arm_get_running_vcpu(void); struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void); void kvm_arm_halt_guest(struct kvm *kvm); diff --git a/arch/mips/include/asm/kvm_host.h b/arch/mips/include/asm/kvm_host.h index 2998479..a9af1d2 100644 --- a/arch/mips/include/asm/kvm_host.h +++ b/arch/mips/include/asm/kvm_host.h @@ -938,11 +938,6 @@ void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end); int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); -static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, - unsigned long address) -{ -} - /* Emulation */ int kvm_get_inst(u32 *opc, struct kvm_vcpu *vcpu, u32 *out); enum emulation_result update_pc(struct kvm_vcpu *vcpu, u32 cause); diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index 8b3f123..e372ed8 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -67,11 +67,6 @@ extern int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end); extern int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); extern void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); -static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, - unsigned long address) -{ -} - #define HPTEG_CACHE_NUM (1 << 15) #define HPTEG_HASH_BITS_PTE 13 #define HPTEG_HASH_BITS_PTE_LONG 12 diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index f4d120a..92c9032 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1375,8 +1375,6 @@ int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu); int kvm_cpu_get_interrupt(struct kvm_vcpu *v); void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event); void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu); -void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, - unsigned long address); void kvm_define_shared_msr(unsigned index, u32 msr); int kvm_set_shared_msr(unsigned index, u64 val, u64 mask); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 05a5e57..272320e 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6734,17 +6734,6 @@ void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_vcpu_reload_apic_access_page); -void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, - unsigned long address) -{ - /* - * The physical address of apic access page is stored in the VMCS. - * Update it when it becomes invalid. - */ - if (address == gfn_to_hva(kvm, APIC_DEFAULT_PHYS_BASE >> PAGE_SHIFT)) - kvm_make_all_cpus_request(kvm, KVM_REQ_APIC_PAGE_RELOAD); -} - /* * Returns 1 to let vcpu_run() continue the guest execution loop without * exiting to the userspace. Otherwise, the value will be returned to the diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 15252d7..4d81f6d 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -322,47 +322,6 @@ static inline struct kvm *mmu_notifier_to_kvm(struct mmu_notifier *mn) return container_of(mn, struct kvm, mmu_notifier); } -static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address) -{ - struct kvm *kvm = mmu_notifier_to_kvm(mn); - int need_tlb_flush, idx; - - /* - * When ->invalidate_page runs, the linux pte has been zapped - * already but the page is still allocated until - * ->invalidate_page returns. So if we increase the sequence - * here the kvm page fault will notice if the spte can't be - * established because the page is going to be freed. If - * instead the kvm page fault establishes the spte before - * ->invalidate_page runs, kvm_unmap_hva will release it - * before returning. - * - * The sequence increase only need to be seen at spin_unlock - * time, and not at spin_lock time. - * - * Increasing the sequence after the spin_unlock would be - * unsafe because the kvm page fault could then establish the - * pte after kvm_unmap_hva returned, without noticing the page - * is going to be freed. - */ - idx = srcu_read_lock(&kvm->srcu); - spin_lock(&kvm->mmu_lock); - - kvm->mmu_notifier_seq++; - need_tlb_flush = kvm_unmap_hva(kvm, address) | kvm->tlbs_dirty; - /* we've to flush the tlb before the pages can be freed */ - if (need_tlb_flush) - kvm_flush_remote_tlbs(kvm); - - spin_unlock(&kvm->mmu_lock); - - kvm_arch_mmu_notifier_invalidate_page(kvm, address); - - srcu_read_unlock(&kvm->srcu, idx); -} - static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn, struct mm_struct *mm, unsigned long address, @@ -510,7 +469,6 @@ static void kvm_mmu_notifier_release(struct mmu_notifier *mn, } static const struct mmu_notifier_ops kvm_mmu_notifier_ops = { - .invalidate_page = kvm_mmu_notifier_invalidate_page, .invalidate_range_start = kvm_mmu_notifier_invalidate_range_start, .invalidate_range_end = kvm_mmu_notifier_invalidate_range_end, .clear_flush_young = kvm_mmu_notifier_clear_flush_young, -- cgit v1.1 From 5f32b265400de723ab0db23101a75ac073bdd980 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Date: Thu, 31 Aug 2017 17:17:38 -0400 Subject: mm/mmu_notifier: kill invalidate_page MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The invalidate_page callback suffered from two pitfalls. First it used to happen after the page table lock was release and thus a new page might have setup before the call to invalidate_page() happened. This is in a weird way fixed by commit c7ab0d2fdc84 ("mm: convert try_to_unmap_one() to use page_vma_mapped_walk()") that moved the callback under the page table lock but this also broke several existing users of the mmu_notifier API that assumed they could sleep inside this callback. The second pitfall was invalidate_page() being the only callback not taking a range of address in respect to invalidation but was giving an address and a page. Lots of the callback implementers assumed this could never be THP and thus failed to invalidate the appropriate range for THP. By killing this callback we unify the mmu_notifier callback API to always take a virtual address range as input. Finally this also simplifies the end user life as there is now two clear choices: - invalidate_range_start()/end() callback (which allow you to sleep) - invalidate_range() where you can not sleep but happen right after page table update under page table lock Signed-off-by: Jérôme Glisse Cc: Bernhard Held Cc: Adam Borowski Cc: Andrea Arcangeli Cc: Radim Krčmář Cc: Wanpeng Li Cc: Paolo Bonzini Cc: Takashi Iwai Cc: Nadav Amit Cc: Mike Galbraith Cc: Kirill A. Shutemov Cc: axie Cc: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmu_notifier.h | 25 ------------------------- mm/mmu_notifier.c | 14 -------------- 2 files changed, 39 deletions(-) diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h index c91b3bc..7b2e31b 100644 --- a/include/linux/mmu_notifier.h +++ b/include/linux/mmu_notifier.h @@ -95,17 +95,6 @@ struct mmu_notifier_ops { pte_t pte); /* - * Before this is invoked any secondary MMU is still ok to - * read/write to the page previously pointed to by the Linux - * pte because the page hasn't been freed yet and it won't be - * freed until this returns. If required set_page_dirty has to - * be called internally to this method. - */ - void (*invalidate_page)(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long address); - - /* * invalidate_range_start() and invalidate_range_end() must be * paired and are called only when the mmap_sem and/or the * locks protecting the reverse maps are held. If the subsystem @@ -220,8 +209,6 @@ extern int __mmu_notifier_test_young(struct mm_struct *mm, unsigned long address); extern void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address, pte_t pte); -extern void __mmu_notifier_invalidate_page(struct mm_struct *mm, - unsigned long address); extern void __mmu_notifier_invalidate_range_start(struct mm_struct *mm, unsigned long start, unsigned long end); extern void __mmu_notifier_invalidate_range_end(struct mm_struct *mm, @@ -268,13 +255,6 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm, __mmu_notifier_change_pte(mm, address, pte); } -static inline void mmu_notifier_invalidate_page(struct mm_struct *mm, - unsigned long address) -{ - if (mm_has_notifiers(mm)) - __mmu_notifier_invalidate_page(mm, address); -} - static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm, unsigned long start, unsigned long end) { @@ -442,11 +422,6 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm, { } -static inline void mmu_notifier_invalidate_page(struct mm_struct *mm, - unsigned long address) -{ -} - static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm, unsigned long start, unsigned long end) { diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c index 54ca545..3142852 100644 --- a/mm/mmu_notifier.c +++ b/mm/mmu_notifier.c @@ -174,20 +174,6 @@ void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address, srcu_read_unlock(&srcu, id); } -void __mmu_notifier_invalidate_page(struct mm_struct *mm, - unsigned long address) -{ - struct mmu_notifier *mn; - int id; - - id = srcu_read_lock(&srcu); - hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) { - if (mn->ops->invalidate_page) - mn->ops->invalidate_page(mn, mm, address); - } - srcu_read_unlock(&srcu, id); -} - void __mmu_notifier_invalidate_range_start(struct mm_struct *mm, unsigned long start, unsigned long end) { -- cgit v1.1 From e746bf730a76fe53b82c9e6b6da72d58e9ae3565 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Thu, 31 Aug 2017 16:15:20 -0700 Subject: mm,page_alloc: don't call __node_reclaim() with oom_lock held. We are doing a last second memory allocation attempt before calling out_of_memory(). But since slab shrinker functions might indirectly wait for other thread's __GFP_DIRECT_RECLAIM && !__GFP_NORETRY memory allocations via sleeping locks, calling slab shrinker functions from node_reclaim() from get_page_from_freelist() with oom_lock held has possibility of deadlock. Therefore, make sure that last second memory allocation attempt does not call slab shrinker functions. Link: http://lkml.kernel.org/r/1503577106-9196-1-git-send-email-penguin-kernel@I-love.SAKURA.ne.jp Signed-off-by: Tetsuo Handa Acked-by: Michal Hocko Cc: Mel Gorman Cc: Vlastimil Babka Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 7a58eb5..1423da8 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3291,10 +3291,13 @@ __alloc_pages_may_oom(gfp_t gfp_mask, unsigned int order, /* * Go through the zonelist yet one more time, keep very high watermark * here, this is only to catch a parallel oom killing, we must fail if - * we're still under heavy pressure. + * we're still under heavy pressure. But make sure that this reclaim + * attempt shall not depend on __GFP_DIRECT_RECLAIM && !__GFP_NORETRY + * allocation which will never fail due to oom_lock already held. */ - page = get_page_from_freelist(gfp_mask | __GFP_HARDWALL, order, - ALLOC_WMARK_HIGH|ALLOC_CPUSET, ac); + page = get_page_from_freelist((gfp_mask | __GFP_HARDWALL) & + ~__GFP_DIRECT_RECLAIM, order, + ALLOC_WMARK_HIGH|ALLOC_CPUSET, ac); if (page) goto out; -- cgit v1.1 From 22cf8bc6cb0d48c6d31da8eccbd4201ab9b0c4c6 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 31 Aug 2017 16:15:23 -0700 Subject: kernel/kthread.c: kthread_worker: don't hog the cpu If the worker thread continues getting work, it will hog the cpu and rcu stall complains. Make it a good citizen. This is triggered in a loop block device test. Link: http://lkml.kernel.org/r/5de0a179b3184e1a2183fc503448b0269f24d75b.1503697127.git.shli@fb.com Signed-off-by: Shaohua Li Cc: Petr Mladek Cc: Thomas Gleixner Cc: Tejun Heo Cc: Oleg Nesterov Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kthread.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/kthread.c b/kernel/kthread.c index 26db528..1c19edf 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -637,6 +637,7 @@ repeat: schedule(); try_to_freeze(); + cond_resched(); goto repeat; } EXPORT_SYMBOL_GPL(kthread_worker_fn); -- cgit v1.1 From 355627f518978b5167256d27492fe0b343aaf2f2 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Thu, 31 Aug 2017 16:15:26 -0700 Subject: mm, uprobes: fix multiple free of ->uprobes_state.xol_area Commit 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable") made it possible to kill a forking task while it is waiting to acquire its ->mmap_sem for write, in dup_mmap(). However, it was overlooked that this introduced an new error path before the new mm_struct's ->uprobes_state.xol_area has been set to NULL after being copied from the old mm_struct by the memcpy in dup_mm(). For a task that has previously hit a uprobe tracepoint, this resulted in the 'struct xol_area' being freed multiple times if the task was killed at just the right time while forking. Fix it by setting ->uprobes_state.xol_area to NULL in mm_init() rather than in uprobe_dup_mmap(). With CONFIG_UPROBE_EVENTS=y, the bug can be reproduced by the same C program given by commit 2b7e8665b4ff ("fork: fix incorrect fput of ->exe_file causing use-after-free"), provided that a uprobe tracepoint has been set on the fork_thread() function. For example: $ gcc reproducer.c -o reproducer -lpthread $ nm reproducer | grep fork_thread 0000000000400719 t fork_thread $ echo "p $PWD/reproducer:0x719" > /sys/kernel/debug/tracing/uprobe_events $ echo 1 > /sys/kernel/debug/tracing/events/uprobes/enable $ ./reproducer Here is the use-after-free reported by KASAN: BUG: KASAN: use-after-free in uprobe_clear_state+0x1c4/0x200 Read of size 8 at addr ffff8800320a8b88 by task reproducer/198 CPU: 1 PID: 198 Comm: reproducer Not tainted 4.13.0-rc7-00015-g36fde05f3fb5 #255 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-20170228_101828-anatol 04/01/2014 Call Trace: dump_stack+0xdb/0x185 print_address_description+0x7e/0x290 kasan_report+0x23b/0x350 __asan_report_load8_noabort+0x19/0x20 uprobe_clear_state+0x1c4/0x200 mmput+0xd6/0x360 do_exit+0x740/0x1670 do_group_exit+0x13f/0x380 get_signal+0x597/0x17d0 do_signal+0x99/0x1df0 exit_to_usermode_loop+0x166/0x1e0 syscall_return_slowpath+0x258/0x2c0 entry_SYSCALL_64_fastpath+0xbc/0xbe ... Allocated by task 199: save_stack_trace+0x1b/0x20 kasan_kmalloc+0xfc/0x180 kmem_cache_alloc_trace+0xf3/0x330 __create_xol_area+0x10f/0x780 uprobe_notify_resume+0x1674/0x2210 exit_to_usermode_loop+0x150/0x1e0 prepare_exit_to_usermode+0x14b/0x180 retint_user+0x8/0x20 Freed by task 199: save_stack_trace+0x1b/0x20 kasan_slab_free+0xa8/0x1a0 kfree+0xba/0x210 uprobe_clear_state+0x151/0x200 mmput+0xd6/0x360 copy_process.part.8+0x605f/0x65d0 _do_fork+0x1a5/0xbd0 SyS_clone+0x19/0x20 do_syscall_64+0x22f/0x660 return_from_SYSCALL_64+0x0/0x7a Note: without KASAN, you may instead see a "Bad page state" message, or simply a general protection fault. Link: http://lkml.kernel.org/r/20170830033303.17927-1-ebiggers3@gmail.com Fixes: 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for write killable") Signed-off-by: Eric Biggers Reported-by: Oleg Nesterov Acked-by: Oleg Nesterov Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Dmitry Vyukov Cc: Ingo Molnar Cc: Konstantin Khlebnikov Cc: Mark Rutland Cc: Michal Hocko Cc: Peter Zijlstra Cc: Vlastimil Babka Cc: [4.7+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/events/uprobes.c | 2 -- kernel/fork.c | 8 ++++++++ 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 0e137f9..267f6ef 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1262,8 +1262,6 @@ void uprobe_end_dup_mmap(void) void uprobe_dup_mmap(struct mm_struct *oldmm, struct mm_struct *newmm) { - newmm->uprobes_state.xol_area = NULL; - if (test_bit(MMF_HAS_UPROBES, &oldmm->flags)) { set_bit(MMF_HAS_UPROBES, &newmm->flags); /* unconditionally, dup_mmap() skips VM_DONTCOPY vmas */ diff --git a/kernel/fork.c b/kernel/fork.c index cbbea27..b7e9e57 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -785,6 +785,13 @@ static void mm_init_owner(struct mm_struct *mm, struct task_struct *p) #endif } +static void mm_init_uprobes_state(struct mm_struct *mm) +{ +#ifdef CONFIG_UPROBES + mm->uprobes_state.xol_area = NULL; +#endif +} + static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p, struct user_namespace *user_ns) { @@ -812,6 +819,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p, #if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS mm->pmd_huge_pte = NULL; #endif + mm_init_uprobes_state(mm); if (current->mm) { mm->flags = current->mm->flags & MMF_INIT_MASK; -- cgit v1.1 From c461ad6a63b37ba74632e90c063d14823c884247 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 31 Aug 2017 16:15:30 -0700 Subject: mm, madvise: ensure poisoned pages are removed from per-cpu lists Wendy Wang reported off-list that a RAS HWPOISON-SOFT test case failed and bisected it to the commit 479f854a207c ("mm, page_alloc: defer debugging checks of pages allocated from the PCP"). The problem is that a page that was poisoned with madvise() is reused. The commit removed a check that would trigger if DEBUG_VM was enabled but re-enabling the check only fixes the problem as a side-effect by printing a bad_page warning and recovering. The root of the problem is that an madvise() can leave a poisoned page on the per-cpu list. This patch drains all per-cpu lists after pages are poisoned so that they will not be reused. Wendy reports that the test case in question passes with this patch applied. While this could be done in a targeted fashion, it is over-complicated for such a rare operation. Link: http://lkml.kernel.org/r/20170828133414.7qro57jbepdcyz5x@techsingularity.net Fixes: 479f854a207c ("mm, page_alloc: defer debugging checks of pages allocated from the PCP") Signed-off-by: Mel Gorman Reported-by: Wang, Wendy Tested-by: Wang, Wendy Acked-by: David Rientjes Acked-by: Vlastimil Babka Cc: "Hansen, Dave" Cc: "Luck, Tony" Cc: Naoya Horiguchi Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/madvise.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/mm/madvise.c b/mm/madvise.c index 23ed525..4d7d1e5 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -613,6 +613,7 @@ static int madvise_inject_error(int behavior, unsigned long start, unsigned long end) { struct page *page; + struct zone *zone; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -646,6 +647,11 @@ static int madvise_inject_error(int behavior, if (ret) return ret; } + + /* Ensure that all poisoned pages are removed from per-cpu lists */ + for_each_populated_zone(zone) + drain_all_pages(zone); + return 0; } #endif -- cgit v1.1 From c03567a8e8d5cf2aaca40e605c48f319dc2ead57 Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Thu, 31 Aug 2017 16:15:33 -0700 Subject: include/linux/compiler.h: don't perform compiletime_assert with -O0 Commit c7acec713d14 ("kernel.h: handle pointers to arrays better in container_of()") made use of __compiletime_assert() from container_of() thus increasing the usage of this macro, allowing developers to notice type conflicts in usage of container_of() at compile time. However, the implementation of __compiletime_assert relies on compiler optimizations to report an error. This means that if a developer uses "-O0" with any code that performs container_of(), the compiler will always report an error regardless of whether there is an actual problem in the code. This patch disables compile_time_assert when optimizations are disabled to allow such code to compile with CFLAGS="-O0". Example compilation failure: ./include/linux/compiler.h:547:38: error: call to `__compiletime_assert_94' declared with attribute error: pointer type mismatch in container_of() _compiletime_assert(condition, msg, __compiletime_assert_, __LINE__) ^ ./include/linux/compiler.h:530:4: note: in definition of macro `__compiletime_assert' prefix ## suffix(); \ ^~~~~~ ./include/linux/compiler.h:547:2: note: in expansion of macro `_compiletime_assert' _compiletime_assert(condition, msg, __compiletime_assert_, __LINE__) ^~~~~~~~~~~~~~~~~~~ ./include/linux/build_bug.h:46:37: note: in expansion of macro `compiletime_assert' #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg) ^~~~~~~~~~~~~~~~~~ ./include/linux/kernel.h:860:2: note: in expansion of macro `BUILD_BUG_ON_MSG' BUILD_BUG_ON_MSG(!__same_type(*(ptr), ((type *)0)->member) && \ ^~~~~~~~~~~~~~~~ [akpm@linux-foundation.org: use do{}while(0), per Michal] Link: http://lkml.kernel.org/r/20170829230114.11662-1-joe@ovn.org Fixes: c7acec713d14c6c ("kernel.h: handle pointers to arrays better in container_of()") Signed-off-by: Joe Stringer Cc: Ian Abbott Cc: Arnd Bergmann Cc: Michal Nazarewicz Cc: Kees Cook Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compiler.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/include/linux/compiler.h b/include/linux/compiler.h index eca8ad7..043b60d 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -517,7 +517,8 @@ static __always_inline void __write_once_size(volatile void *p, void *res, int s # define __compiletime_error_fallback(condition) do { } while (0) #endif -#define __compiletime_assert(condition, msg, prefix, suffix) \ +#ifdef __OPTIMIZE__ +# define __compiletime_assert(condition, msg, prefix, suffix) \ do { \ bool __cond = !(condition); \ extern void prefix ## suffix(void) __compiletime_error(msg); \ @@ -525,6 +526,9 @@ static __always_inline void __write_once_size(volatile void *p, void *res, int s prefix ## suffix(); \ __compiletime_error_fallback(__cond); \ } while (0) +#else +# define __compiletime_assert(condition, msg, prefix, suffix) do { } while (0) +#endif #define _compiletime_assert(condition, msg, prefix, suffix) \ __compiletime_assert(condition, msg, prefix, suffix) -- cgit v1.1 From e66186920bff278b18ebe460c710c7b0e0cfdf6e Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 31 Aug 2017 16:15:36 -0700 Subject: scripts/dtc: fix '%zx' warning dtc uses an incorrect format specifier for printing a uint64_t value. uint64_t may be either 'unsigned long' or 'unsigned long long' depending on the host architecture. Fix this by using %llx and casting to unsigned long long, which ensures that we always have a wide enough variable to print 64 bits of hex. HOSTCC scripts/dtc/checks.o scripts/dtc/checks.c: In function 'check_simple_bus_reg': scripts/dtc/checks.c:876:2: warning: format '%zx' expects argument of type 'size_t', but argument 4 has type 'uint64_t' [-Wformat=] snprintf(unit_addr, sizeof(unit_addr), "%zx", reg); ^ scripts/dtc/checks.c:876:2: warning: format '%zx' expects argument of type 'size_t', but argument 4 has type 'uint64_t' [-Wformat=] Link: http://lkml.kernel.org/r/20170829222034.GJ20805@n2100.armlinux.org.uk Fixes: 828d4cdd012c ("dtc: check.c fix compile error") Signed-off-by: Russell King Cc: Rob Herring Cc: Frank Rowand Cc: Shuah Khan Cc: David Gibson Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/dtc/checks.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/dtc/checks.c b/scripts/dtc/checks.c index 4b72b53..62ea8f8 100644 --- a/scripts/dtc/checks.c +++ b/scripts/dtc/checks.c @@ -873,7 +873,7 @@ static void check_simple_bus_reg(struct check *c, struct dt_info *dti, struct no while (size--) reg = (reg << 32) | fdt32_to_cpu(*(cells++)); - snprintf(unit_addr, sizeof(unit_addr), "%zx", reg); + snprintf(unit_addr, sizeof(unit_addr), "%llx", (unsigned long long)reg); if (!streq(unitname, unit_addr)) FAIL(c, dti, "Node %s simple-bus unit address format error, expected \"%s\"", node->fullpath, unit_addr); -- cgit v1.1 From c227390c91a355300f47f9bef0aefbdfaaca1500 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Thu, 31 Aug 2017 16:46:59 -0500 Subject: jfs should use MAX_LFS_FILESIZE when calculating s_maxbytes jfs had previously avoided the use of MAX_LFS_FILESIZE because it hadn't accounted for the whole 32-bit index range on 32-bit systems. That has been fixed by commit 0cc3b0ec23ce ("Clarify (and fix) MAX_LFS_FILESIZE macros"), so we can simplify the code now. Suggested by Andreas Dilger. Signed-off-by: Dave Kleikamp Reviewed-by: Andreas Dilger Cc: jfs-discussion@lists.sourceforge.net Signed-off-by: Linus Torvalds --- fs/jfs/super.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 78b41e1..60726ae 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -619,16 +619,10 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) if (!sb->s_root) goto out_no_root; - /* logical blocks are represented by 40 bits in pxd_t, etc. */ - sb->s_maxbytes = ((u64) sb->s_blocksize) << 40; -#if BITS_PER_LONG == 32 - /* - * Page cache is indexed by long. - * I would use MAX_LFS_FILESIZE, but it's only half as big + /* logical blocks are represented by 40 bits in pxd_t, etc. + * and page cache is indexed by long */ - sb->s_maxbytes = min(((u64) PAGE_SIZE << 32) - 1, - (u64)sb->s_maxbytes); -#endif + sb->s_maxbytes = min(((loff_t)sb->s_blocksize) << 40, MAX_LFS_FILESIZE); sb->s_time_gran = 1; return 0; -- cgit v1.1 From 25cc72a33835ed8a6f53180a822cadab855852ac Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Fri, 1 Sep 2017 10:52:31 +0200 Subject: mlxsw: spectrum: Forbid linking to devices that have uppers The mlxsw driver relies on NETDEV_CHANGEUPPER events to configure the device in case a port is enslaved to a master netdev such as bridge or bond. Since the driver ignores events unrelated to its ports and their uppers, it's possible to engineer situations in which the device's data path differs from the kernel's. One example to such a situation is when a port is enslaved to a bond that is already enslaved to a bridge. When the bond was enslaved the driver ignored the event - as the bond wasn't one of its uppers - and therefore a bridge port instance isn't created in the device. Until such configurations are supported forbid them by checking that the upper device doesn't have uppers of its own. Fixes: 0d65fc13042f ("mlxsw: spectrum: Implement LAG port join/leave") Signed-off-by: Ido Schimmel Reported-by: Nogah Frankel Tested-by: Nogah Frankel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 6 ++++++ include/linux/netdevice.h | 2 ++ net/core/dev.c | 3 ++- 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 60bf8f2..c6a3e61b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -4139,6 +4139,8 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *lower_dev, return -EINVAL; if (!info->linking) break; + if (netdev_has_any_upper_dev(upper_dev)) + return -EINVAL; if (netif_is_lag_master(upper_dev) && !mlxsw_sp_master_lag_check(mlxsw_sp, upper_dev, info->upper_info)) @@ -4258,6 +4260,10 @@ static int mlxsw_sp_netdevice_port_vlan_event(struct net_device *vlan_dev, upper_dev = info->upper_dev; if (!netif_is_bridge_master(upper_dev)) return -EINVAL; + if (!info->linking) + break; + if (netdev_has_any_upper_dev(upper_dev)) + return -EINVAL; break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 779b235..c99ba79 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3866,6 +3866,8 @@ int netdev_walk_all_upper_dev_rcu(struct net_device *dev, bool netdev_has_upper_dev_all_rcu(struct net_device *dev, struct net_device *upper_dev); +bool netdev_has_any_upper_dev(struct net_device *dev); + void *netdev_lower_get_next_private(struct net_device *dev, struct list_head **iter); void *netdev_lower_get_next_private_rcu(struct net_device *dev, diff --git a/net/core/dev.c b/net/core/dev.c index 818dfa6..86b4b0a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5668,12 +5668,13 @@ EXPORT_SYMBOL(netdev_has_upper_dev_all_rcu); * Find out if a device is linked to an upper device and return true in case * it is. The caller must hold the RTNL lock. */ -static bool netdev_has_any_upper_dev(struct net_device *dev) +bool netdev_has_any_upper_dev(struct net_device *dev) { ASSERT_RTNL(); return !list_empty(&dev->adj_list.upper); } +EXPORT_SYMBOL(netdev_has_any_upper_dev); /** * netdev_master_upper_dev_get - Get master upper device -- cgit v1.1 From 79e99bdd60b484af9afe0147e85a13e66d5c1cdb Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Fri, 1 Sep 2017 12:22:25 +0300 Subject: bridge: switchdev: Clear forward mark when transmitting packet Commit 6bc506b4fb06 ("bridge: switchdev: Add forward mark support for stacked devices") added the 'offload_fwd_mark' bit to the skb in order to allow drivers to indicate to the bridge driver that they already forwarded the packet in L2. In case the bit is set, before transmitting the packet from each port, the port's mark is compared with the mark stored in the skb's control block. If both marks are equal, we know the packet arrived from a switch device that already forwarded the packet and it's not re-transmitted. However, if the packet is transmitted from the bridge device itself (e.g., br0), we should clear the 'offload_fwd_mark' bit as the mark stored in the skb's control block isn't valid. This scenario can happen in rare cases where a packet was trapped during L3 forwarding and forwarded by the kernel to a bridge device. Fixes: 6bc506b4fb06 ("bridge: switchdev: Add forward mark support for stacked devices") Signed-off-by: Ido Schimmel Reported-by: Yotam Gigi Tested-by: Yotam Gigi Reviewed-by: Jiri Pirko Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_device.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 861ae2a..5a7be3b 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -53,6 +53,9 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) brstats->tx_bytes += skb->len; u64_stats_update_end(&brstats->syncp); +#ifdef CONFIG_NET_SWITCHDEV + skb->offload_fwd_mark = 0; +#endif BR_INPUT_SKB_CB(skb)->brdev = dev; skb_reset_mac_header(skb); -- cgit v1.1 From e8a732d1bc3ac313e22249c13a153c3fe54aa577 Mon Sep 17 00:00:00 2001 From: Yossi Kuperman Date: Fri, 1 Sep 2017 14:42:30 +0200 Subject: udp: fix secpath leak After commit dce4551cb2ad ("udp: preserve head state for IP_CMSG_PASSSEC") we preserve the secpath for the whole skb lifecycle, but we also end up leaking a reference to it. We must clear the head state on skb reception, if secpath is present. Fixes: dce4551cb2ad ("udp: preserve head state for IP_CMSG_PASSSEC") Signed-off-by: Yossi Kuperman Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- net/ipv4/udp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index a6dc48d..6234480 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1176,7 +1176,7 @@ static void udp_set_dev_scratch(struct sk_buff *skb) scratch->csum_unnecessary = !!skb_csum_unnecessary(skb); scratch->is_linear = !skb_is_nonlinear(skb); #endif - if (likely(!skb->_skb_refdst)) + if (likely(!skb->_skb_refdst && !skb_sec_path(skb))) scratch->_tsize_state |= UDP_SKB_IS_STATELESS; } -- cgit v1.1 From 138e4ad67afd5c6c318b056b4d17c17f2c0ca5c0 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 1 Sep 2017 18:55:33 +0200 Subject: epoll: fix race between ep_poll_callback(POLLFREE) and ep_free()/ep_remove() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The race was introduced by me in commit 971316f0503a ("epoll: ep_unregister_pollwait() can use the freed pwq->whead"). I did not realize that nothing can protect eventpoll after ep_poll_callback() sets ->whead = NULL, only whead->lock can save us from the race with ep_free() or ep_remove(). Move ->whead = NULL to the end of ep_poll_callback() and add the necessary barriers. TODO: cleanup the ewake/EPOLLEXCLUSIVE logic, it was confusing even before this patch. Hopefully this explains use-after-free reported by syzcaller: BUG: KASAN: use-after-free in debug_spin_lock_before ... _raw_spin_lock_irqsave+0x4a/0x60 kernel/locking/spinlock.c:159 ep_poll_callback+0x29f/0xff0 fs/eventpoll.c:1148 this is spin_lock(eventpoll->lock), ... Freed by task 17774: ... kfree+0xe8/0x2c0 mm/slub.c:3883 ep_free+0x22c/0x2a0 fs/eventpoll.c:865 Fixes: 971316f0503a ("epoll: ep_unregister_pollwait() can use the freed pwq->whead") Reported-by: 范龙飞 Cc: stable@vger.kernel.org Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- fs/eventpoll.c | 42 ++++++++++++++++++++++++++---------------- 1 file changed, 26 insertions(+), 16 deletions(-) diff --git a/fs/eventpoll.c b/fs/eventpoll.c index e767e43..adbe328 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -600,8 +600,13 @@ static void ep_remove_wait_queue(struct eppoll_entry *pwq) wait_queue_head_t *whead; rcu_read_lock(); - /* If it is cleared by POLLFREE, it should be rcu-safe */ - whead = rcu_dereference(pwq->whead); + /* + * If it is cleared by POLLFREE, it should be rcu-safe. + * If we read NULL we need a barrier paired with + * smp_store_release() in ep_poll_callback(), otherwise + * we rely on whead->lock. + */ + whead = smp_load_acquire(&pwq->whead); if (whead) remove_wait_queue(whead, &pwq->wait); rcu_read_unlock(); @@ -1134,17 +1139,6 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, v struct eventpoll *ep = epi->ep; int ewake = 0; - if ((unsigned long)key & POLLFREE) { - ep_pwq_from_wait(wait)->whead = NULL; - /* - * whead = NULL above can race with ep_remove_wait_queue() - * which can do another remove_wait_queue() after us, so we - * can't use __remove_wait_queue(). whead->lock is held by - * the caller. - */ - list_del_init(&wait->entry); - } - spin_lock_irqsave(&ep->lock, flags); ep_set_busy_poll_napi_id(epi); @@ -1228,10 +1222,26 @@ out_unlock: if (pwake) ep_poll_safewake(&ep->poll_wait); - if (epi->event.events & EPOLLEXCLUSIVE) - return ewake; + if (!(epi->event.events & EPOLLEXCLUSIVE)) + ewake = 1; + + if ((unsigned long)key & POLLFREE) { + /* + * If we race with ep_remove_wait_queue() it can miss + * ->whead = NULL and do another remove_wait_queue() after + * us, so we can't use __remove_wait_queue(). + */ + list_del_init(&wait->entry); + /* + * ->whead != NULL protects us from the race with ep_free() + * or ep_remove(), ep_remove_wait_queue() takes whead->lock + * held by the caller. Once we nullify it, nothing protects + * ep/epi or even wait. + */ + smp_store_release(&ep_pwq_from_wait(wait)->whead, NULL); + } - return 1; + return ewake; } /* -- cgit v1.1