From b137b9942a07843c64a934cfdb7d43155e507e13 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Mon, 23 Aug 2010 20:25:32 +0000 Subject: ACPI: Don't report current_now if battery reports in mWh ACPI batteries can report in units of either current or energy. Right now we expose the current_now file even if the battery is reporting energy units, resulting in a file that should contain mA instead containing mW. Don't expose this value unless the battery is reporting current. Signed-off-by: Matthew Garrett Cc: Arjan van de Ven Acked-by: Rafael J. Wysocki Acked-by: Zhang Rui Signed-off-by: Len Brown --- drivers/acpi/battery.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers') diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index dc58402..9841720 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -273,7 +273,6 @@ static enum power_supply_property energy_battery_props[] = { POWER_SUPPLY_PROP_CYCLE_COUNT, POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN, POWER_SUPPLY_PROP_VOLTAGE_NOW, - POWER_SUPPLY_PROP_CURRENT_NOW, POWER_SUPPLY_PROP_POWER_NOW, POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN, POWER_SUPPLY_PROP_ENERGY_FULL, -- cgit v1.1 From 0e52247a2ed1f211f0c4f682dc999610a368903f Mon Sep 17 00:00:00 2001 From: Cam Macdonell Date: Tue, 7 Sep 2010 17:25:20 -0700 Subject: PCI: fix pci_resource_alignment prototype This fixes the prototype for both pci_resource_alignment() and pci_sriov_resource_alignment(). Patch started as debugging effort from Cam Macdonell. Cc: Cam Macdonell Cc: Avi Kivity [chrisw: add iov bits] Signed-off-by: Chris Wright Signed-off-by: Jesse Barnes --- drivers/pci/iov.c | 2 +- drivers/pci/pci.h | 5 +++-- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'drivers') diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c index ce6a366..553d8ee 100644 --- a/drivers/pci/iov.c +++ b/drivers/pci/iov.c @@ -608,7 +608,7 @@ int pci_iov_resource_bar(struct pci_dev *dev, int resno, * the VF BAR size multiplied by the number of VFs. The alignment * is just the VF BAR size. */ -int pci_sriov_resource_alignment(struct pci_dev *dev, int resno) +resource_size_t pci_sriov_resource_alignment(struct pci_dev *dev, int resno) { struct resource tmp; enum pci_bar_type type; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 7754a67..6beb11b 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -264,7 +264,8 @@ extern int pci_iov_init(struct pci_dev *dev); extern void pci_iov_release(struct pci_dev *dev); extern int pci_iov_resource_bar(struct pci_dev *dev, int resno, enum pci_bar_type *type); -extern int pci_sriov_resource_alignment(struct pci_dev *dev, int resno); +extern resource_size_t pci_sriov_resource_alignment(struct pci_dev *dev, + int resno); extern void pci_restore_iov_state(struct pci_dev *dev); extern int pci_iov_bus_range(struct pci_bus *bus); @@ -320,7 +321,7 @@ static inline int pci_ats_enabled(struct pci_dev *dev) } #endif /* CONFIG_PCI_IOV */ -static inline int pci_resource_alignment(struct pci_dev *dev, +static inline resource_size_t pci_resource_alignment(struct pci_dev *dev, struct resource *res) { #ifdef CONFIG_PCI_IOV -- cgit v1.1 From ee05d6939ed17b55e9c2466af32c208e0d547eb8 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Tue, 14 Sep 2010 15:15:52 +0200 Subject: vhost-net: fix range checking in mrg bufs case In mergeable buffer case, we use headcount, log_num and seg as indexes in same-size arrays, and we know that headcount <= seg and log_num equals either 0 or seg. Therefore, the right thing to do is range-check seg, not headcount as we do now: these will be different if guest chains s/g descriptors (this does not happen now, but we can not trust the guest). Long term, we should add BUG_ON checks to verify two other indexes are what we think they should be. Reported-by: Jason Wang Signed-off-by: Michael S. Tsirkin --- drivers/vhost/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index 29e850a..7c80082 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -243,7 +243,7 @@ static int get_rx_bufs(struct vhost_virtqueue *vq, int r, nlogs = 0; while (datalen > 0) { - if (unlikely(headcount >= VHOST_NET_MAX_SG)) { + if (unlikely(seg >= VHOST_NET_MAX_SG)) { r = -ENOBUFS; goto err; } -- cgit v1.1 From 37880c909c6a22674d3c0f83f2737264b4e60fe1 Mon Sep 17 00:00:00 2001 From: christophe leroy Date: Thu, 16 Sep 2010 09:05:25 +0200 Subject: spi/mpc8xxx: fix buffer overrun on large transfers It fixes an issue when sending-only or receiving-only more than PAGE_SIZE bytes. Signed-off-by: christophe leroy Acked-by: Joakim Tjernlund Signed-off-by: Grant Likely --- drivers/spi/spi_mpc8xxx.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/spi/spi_mpc8xxx.c b/drivers/spi/spi_mpc8xxx.c index d31b57f..1dd86b8 100644 --- a/drivers/spi/spi_mpc8xxx.c +++ b/drivers/spi/spi_mpc8xxx.c @@ -408,11 +408,17 @@ static void mpc8xxx_spi_cpm_bufs_start(struct mpc8xxx_spi *mspi) xfer_ofs = mspi->xfer_in_progress->len - mspi->count; - out_be32(&rx_bd->cbd_bufaddr, mspi->rx_dma + xfer_ofs); + if (mspi->rx_dma == mspi->dma_dummy_rx) + out_be32(&rx_bd->cbd_bufaddr, mspi->rx_dma); + else + out_be32(&rx_bd->cbd_bufaddr, mspi->rx_dma + xfer_ofs); out_be16(&rx_bd->cbd_datlen, 0); out_be16(&rx_bd->cbd_sc, BD_SC_EMPTY | BD_SC_INTRPT | BD_SC_WRAP); - out_be32(&tx_bd->cbd_bufaddr, mspi->tx_dma + xfer_ofs); + if (mspi->tx_dma == mspi->dma_dummy_tx) + out_be32(&tx_bd->cbd_bufaddr, mspi->tx_dma); + else + out_be32(&tx_bd->cbd_bufaddr, mspi->tx_dma + xfer_ofs); out_be16(&tx_bd->cbd_datlen, xfer_len); out_be16(&tx_bd->cbd_sc, BD_SC_READY | BD_SC_INTRPT | BD_SC_WRAP | BD_SC_LAST); -- cgit v1.1 From 7acc7c683a747689aaaaad4fce1683fc3f85e552 Mon Sep 17 00:00:00 2001 From: Wey-Yi Guy Date: Wed, 8 Sep 2010 08:30:20 -0700 Subject: iwlwifi: do not perferm force reset while doing scan When uCode error condition detected, driver try to perform either rf reset or firmware reload in order bring device back to working condition. If rf reset is required and scan is in process, there is no need to issue rf reset since scan already reset the rf. If firmware reload is required and scan is in process, skip the reload request. There is a possibility firmware reload during scan cause problem. [ 485.804046] WARNING: at net/mac80211/main.c:310 ieee80211_restart_hw+0x28/0x62() [ 485.804049] Hardware name: Latitude E6400 [ 485.804052] ieee80211_restart_hw called with hardware scan in progress [ 485.804054] Modules linked in: iwlagn iwlcore bnep sco rfcomm l2cap crc16 bluetooth [last unloaded: iwlcore] [ 485.804069] Pid: 812, comm: kworker/u:3 Tainted: G W 2.6.36-rc3-wl+ #74 [ 485.804072] Call Trace: [ 485.804079] [] warn_slowpath_common+0x60/0x75 [ 485.804084] [] warn_slowpath_fmt+0x26/0x2a [ 485.804089] [] ieee80211_restart_hw+0x28/0x62 [ 485.804102] [] iwl_bg_restart+0x113/0x150 [iwlagn] [ 485.804108] [] process_one_work+0x181/0x25c [ 485.804119] [] ? iwl_bg_restart+0x0/0x150 [iwlagn] [ 485.804124] [] worker_thread+0xf9/0x1f2 [ 485.804128] [] ? worker_thread+0x0/0x1f2 [ 485.804133] [] kthread+0x64/0x69 [ 485.804137] [] ? kthread+0x0/0x69 [ 485.804141] [] kernel_thread_helper+0x6/0x10 [ 485.804145] ---[ end trace 3d4ebdc02d524bbb ]--- [ 485.804148] WG> 1 [ 485.804153] Pid: 812, comm: kworker/u:3 Tainted: G W 2.6.36-rc3-wl+ #74 [ 485.804156] Call Trace: [ 485.804161] [] ? ieee80211_restart_hw+0x5c/0x62 [ 485.804172] [] iwl_bg_restart+0x118/0x150 [iwlagn] [ 485.804177] [] process_one_work+0x181/0x25c [ 485.804188] [] ? iwl_bg_restart+0x0/0x150 [iwlagn] [ 485.804192] [] worker_thread+0xf9/0x1f2 [ 485.804197] [] ? worker_thread+0x0/0x1f2 [ 485.804201] [] kthread+0x64/0x69 [ 485.804205] [] ? kthread+0x0/0x69 [ 485.804209] [] kernel_thread_helper+0x6/0x10 Signed-off-by: Wey-Yi Guy --- drivers/net/wireless/iwlwifi/iwl-core.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'drivers') diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c index 07dbc27..e23c406 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.c +++ b/drivers/net/wireless/iwlwifi/iwl-core.c @@ -2613,6 +2613,11 @@ int iwl_force_reset(struct iwl_priv *priv, int mode, bool external) if (test_bit(STATUS_EXIT_PENDING, &priv->status)) return -EINVAL; + if (test_bit(STATUS_SCANNING, &priv->status)) { + IWL_DEBUG_INFO(priv, "scan in progress.\n"); + return -EINVAL; + } + if (mode >= IWL_MAX_FORCE_RESET) { IWL_DEBUG_INFO(priv, "invalid reset request.\n"); return -EINVAL; -- cgit v1.1 From 04746ff1289f75af26af279eb4b0b3e231677ee4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 17 Sep 2010 22:58:08 -0700 Subject: qlcnic: dont assume NET_IP_ALIGN is 2 qlcnic driver allocates rx skbs and gives to hardware too bytes of extra storage, allowing for corruption of kernel data. NET_IP_ALIGN being 0 on some platforms (including x86), drivers should not assume it's 2. rds_ring->skb_size = rds_ring->dma_size + NET_IP_ALIGN; ... skb = dev_alloc_skb(rds_ring->skb_size); skb_reserve(skb, 2); pci_map_single(pdev, skb->data, rds_ring->dma_size, PCI_DMA_FROMDEVICE); (and rds_ring->skb_size == rds_ring->dma_size) -> bug Because of extra alignment (1500 + 32) -> four extra bytes are available before the struct skb_shared_info, so corruption is not noticed. Note: this driver could use netdev_alloc_skb_ip_align() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/qlcnic/qlcnic_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/net/qlcnic/qlcnic_init.c b/drivers/net/qlcnic/qlcnic_init.c index 75ba744..60ab753 100644 --- a/drivers/net/qlcnic/qlcnic_init.c +++ b/drivers/net/qlcnic/qlcnic_init.c @@ -1316,7 +1316,7 @@ qlcnic_alloc_rx_skb(struct qlcnic_adapter *adapter, return -ENOMEM; } - skb_reserve(skb, 2); + skb_reserve(skb, NET_IP_ALIGN); dma = pci_map_single(pdev, skb->data, rds_ring->dma_size, PCI_DMA_FROMDEVICE); -- cgit v1.1 From 7cdffc86528ec9c55c83c649b6d64cadeb558136 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Sat, 18 Sep 2010 10:19:13 +0200 Subject: pcmcia: preserve configuration information if request_io fails partly If pcmcia_request_io() only fails partly -- for the second of two requested resources -- preserve the configuration settings for the first one. Signed-off-by: Dominik Brodowski --- drivers/pcmcia/pcmcia_resource.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'drivers') diff --git a/drivers/pcmcia/pcmcia_resource.c b/drivers/pcmcia/pcmcia_resource.c index a5c1765..9ba4dad 100644 --- a/drivers/pcmcia/pcmcia_resource.c +++ b/drivers/pcmcia/pcmcia_resource.c @@ -595,7 +595,13 @@ int pcmcia_request_io(struct pcmcia_device *p_dev) if (c->io[1].end) { ret = alloc_io_space(s, &c->io[1], p_dev->io_lines); if (ret) { + struct resource tmp = c->io[0]; + /* release the previously allocated resource */ release_io_space(s, &c->io[0]); + /* but preserve the settings, for they worked... */ + c->io[0].end = resource_size(&tmp); + c->io[0].start = tmp.start; + c->io[0].flags = tmp.flags; goto out; } } else -- cgit v1.1 From 863636828f1fcd9fdc15e24d620aa53cf18b432f Mon Sep 17 00:00:00 2001 From: Simon Guinot Date: Fri, 17 Sep 2010 23:33:51 +0200 Subject: dmaengine: fix interrupt clearing for mv_xor When using simultaneously the two DMA channels on a same engine, some transfers are never completed. For example, an endless lock can occur while writing heavily on a RAID5 array (with async-tx offload support enabled). Note that this issue can also be reproduced by using the DMA test client. On a same engine, the interrupt cause register is shared between two DMA channels. This patch make sure that the cause bit is only cleared for the requested channel. Signed-off-by: Simon Guinot Tested-by: Luc Saillard Acked-by: Saeed Bishara Signed-off-by: Nicolas Pitre --- drivers/dma/mv_xor.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index 86c5ae9..411d5bf 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -162,7 +162,7 @@ static int mv_is_err_intr(u32 intr_cause) static void mv_xor_device_clear_eoc_cause(struct mv_xor_chan *chan) { - u32 val = (1 << (1 + (chan->idx * 16))); + u32 val = ~(1 << (chan->idx * 16)); dev_dbg(chan->device->common.dev, "%s, val 0x%08x\n", __func__, val); __raw_writel(val, XOR_INTR_CAUSE(chan)); } -- cgit v1.1 From f539dfedbd169e5ed47912bb517c75976ab556f3 Mon Sep 17 00:00:00 2001 From: Simon Guinot Date: Sun, 19 Sep 2010 15:30:59 +0200 Subject: leds: leds-ns2: fix locking This patch replace all the lock functions with the irq safe variant. The ns2_led_{set,get}_mode() functions must be safe in all context. For example, the trigger timer call led_set_brightness() in a softirq context. Signed-off-by: Simon Guinot Signed-off-by: Nicolas Pitre --- drivers/leds/leds-ns2.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'drivers') diff --git a/drivers/leds/leds-ns2.c b/drivers/leds/leds-ns2.c index 74dce4b..350eb34 100644 --- a/drivers/leds/leds-ns2.c +++ b/drivers/leds/leds-ns2.c @@ -81,7 +81,7 @@ static int ns2_led_get_mode(struct ns2_led_data *led_dat, int cmd_level; int slow_level; - read_lock(&led_dat->rw_lock); + read_lock_irq(&led_dat->rw_lock); cmd_level = gpio_get_value(led_dat->cmd); slow_level = gpio_get_value(led_dat->slow); @@ -95,7 +95,7 @@ static int ns2_led_get_mode(struct ns2_led_data *led_dat, } } - read_unlock(&led_dat->rw_lock); + read_unlock_irq(&led_dat->rw_lock); return ret; } @@ -104,8 +104,9 @@ static void ns2_led_set_mode(struct ns2_led_data *led_dat, enum ns2_led_modes mode) { int i; + unsigned long flags; - write_lock(&led_dat->rw_lock); + write_lock_irqsave(&led_dat->rw_lock, flags); for (i = 0; i < ARRAY_SIZE(ns2_led_modval); i++) { if (mode == ns2_led_modval[i].mode) { @@ -116,7 +117,7 @@ static void ns2_led_set_mode(struct ns2_led_data *led_dat, } } - write_unlock(&led_dat->rw_lock); + write_unlock_irqrestore(&led_dat->rw_lock, flags); } static void ns2_led_set(struct led_classdev *led_cdev, -- cgit v1.1 From 8df8fd27123054b02007361bd5483775db84b4a8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 20 Sep 2010 02:28:59 +0000 Subject: qlcnic: dont set skb->truesize skb->truesize is set in core network. Dont change it unless dealing with fragments. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/qlcnic/qlcnic_init.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'drivers') diff --git a/drivers/net/qlcnic/qlcnic_init.c b/drivers/net/qlcnic/qlcnic_init.c index 60ab753..2c7cf0b6 100644 --- a/drivers/net/qlcnic/qlcnic_init.c +++ b/drivers/net/qlcnic/qlcnic_init.c @@ -1404,7 +1404,6 @@ qlcnic_process_rcv(struct qlcnic_adapter *adapter, if (pkt_offset) skb_pull(skb, pkt_offset); - skb->truesize = skb->len + sizeof(struct sk_buff); skb->protocol = eth_type_trans(skb, netdev); napi_gro_receive(&sds_ring->napi, skb); @@ -1466,8 +1465,6 @@ qlcnic_process_lro(struct qlcnic_adapter *adapter, skb_put(skb, lro_length + data_offset); - skb->truesize = skb->len + sizeof(struct sk_buff) + skb_headroom(skb); - skb_pull(skb, l2_hdr_offset); skb->protocol = eth_type_trans(skb, netdev); @@ -1700,8 +1697,6 @@ qlcnic_process_rcv_diag(struct qlcnic_adapter *adapter, if (pkt_offset) skb_pull(skb, pkt_offset); - skb->truesize = skb->len + sizeof(struct sk_buff); - if (!qlcnic_check_loopback_buff(skb->data)) adapter->diag_cnt++; -- cgit v1.1 From 7e96dc7045bff8758804b047c0dfb6868f182500 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Sep 2010 13:04:04 -0700 Subject: netxen: dont set skb->truesize skb->truesize is set in core network. Dont change it unless dealing with fragments. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/netxen/netxen_nic_init.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'drivers') diff --git a/drivers/net/netxen/netxen_nic_init.c b/drivers/net/netxen/netxen_nic_init.c index cabae7b..b075a35 100644 --- a/drivers/net/netxen/netxen_nic_init.c +++ b/drivers/net/netxen/netxen_nic_init.c @@ -1540,7 +1540,6 @@ netxen_process_rcv(struct netxen_adapter *adapter, if (pkt_offset) skb_pull(skb, pkt_offset); - skb->truesize = skb->len + sizeof(struct sk_buff); skb->protocol = eth_type_trans(skb, netdev); napi_gro_receive(&sds_ring->napi, skb); @@ -1602,8 +1601,6 @@ netxen_process_lro(struct netxen_adapter *adapter, skb_put(skb, lro_length + data_offset); - skb->truesize = skb->len + sizeof(struct sk_buff) + skb_headroom(skb); - skb_pull(skb, l2_hdr_offset); skb->protocol = eth_type_trans(skb, netdev); -- cgit v1.1 From 9eecabcb9a924f1e11ba670365fd4babe423045c Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Tue, 21 Sep 2010 22:28:23 +0100 Subject: intel-iommu: Abort IOMMU setup for igfx if BIOS gave no shadow GTT space Yet another BIOS bug; Lenovo this time (X201). Red Hat bug #593516. Signed-off-by: David Woodhouse --- drivers/pci/intel-iommu.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'drivers') diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index c3ceebb..dee88c6c 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -3761,6 +3761,23 @@ static void __devinit quirk_iommu_rwbf(struct pci_dev *dev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2a40, quirk_iommu_rwbf); +static void __devinit quirk_calpella_no_shadow_gtt(struct pci_dev *dev) +{ + unsigned short ggc; + + if (pci_read_config_word(dev, 0x52, &ggc)) + return; + + if (!(ggc & 0x800)) { + printk(KERN_INFO "DMAR: BIOS has allocated no shadow GTT; disabling IOMMU for graphics\n"); + dmar_map_gfx = 0; + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0040, quirk_calpella_no_shadow_gtt); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_calpella_no_shadow_gtt); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0062, quirk_calpella_no_shadow_gtt); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x006a, quirk_calpella_no_shadow_gtt); + /* On Tylersburg chipsets, some BIOSes have been known to enable the ISOCH DMAR unit for the Azalia sound device, but not give it any TLB entries, which causes it to deadlock. Check for that. We do -- cgit v1.1 From eecfd57f6429d9d8e10be186566ef99fced55163 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Wed, 25 Aug 2010 21:17:34 +0100 Subject: intel-iommu: Use symbolic values instead of magic numbers in Lenovo w/a Commit 9eecabcb9a924f1e11ba670365fd4babe423045c ("intel-iommu: Abort IOMMU setup for igfx if BIOS gave no shadow GTT space") uses a bunch of magic numbers. Provide #defines for those to make it look slightly saner. Signed-off-by: Adam Jackson Signed-off-by: David Woodhouse --- drivers/pci/intel-iommu.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index dee88c6c..4dfce3d 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -3761,14 +3761,24 @@ static void __devinit quirk_iommu_rwbf(struct pci_dev *dev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2a40, quirk_iommu_rwbf); +#define GGC 0x52 +#define GGC_MEMORY_SIZE_MASK (0xf << 8) +#define GGC_MEMORY_SIZE_NONE (0x0 << 8) +#define GGC_MEMORY_SIZE_1M (0x1 << 8) +#define GGC_MEMORY_SIZE_2M (0x3 << 8) +#define GGC_MEMORY_VT_ENABLED (0x8 << 8) +#define GGC_MEMORY_SIZE_2M_VT (0x9 << 8) +#define GGC_MEMORY_SIZE_3M_VT (0xa << 8) +#define GGC_MEMORY_SIZE_4M_VT (0xb << 8) + static void __devinit quirk_calpella_no_shadow_gtt(struct pci_dev *dev) { unsigned short ggc; - if (pci_read_config_word(dev, 0x52, &ggc)) + if (pci_read_config_word(dev, GGC, &ggc)) return; - if (!(ggc & 0x800)) { + if (!(ggc & GGC_MEMORY_VT_ENABLED)) { printk(KERN_INFO "DMAR: BIOS has allocated no shadow GTT; disabling IOMMU for graphics\n"); dmar_map_gfx = 0; } -- cgit v1.1 From 5222454cf941b09205d502242dd1e44fd62aa551 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Wed, 22 Sep 2010 12:31:46 +1000 Subject: drm/radeon: don't allow device to be opened if powered down If the switcheroo has switched the device off, don't let X open it. Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_kms.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers') diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 5eee3c4..8fbbe1c 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -203,6 +203,10 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) */ int radeon_driver_firstopen_kms(struct drm_device *dev) { + struct radeon_device *rdev = dev->dev_private; + + if (rdev->powered_down) + return -EINVAL; return 0; } -- cgit v1.1 From 0fbecd400dd0a82d465b3086f209681e8c54cb0f Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Tue, 21 Sep 2010 02:15:15 +0200 Subject: drm/ttm: Clear the ghost cpu_writers flag on ttm_buffer_object_transfer. It makes sense for a BO to move after a process has requested exclusive RW access on it (e.g. because the BO used to be located in unmappable VRAM and we intercepted the CPU access from the fault handler). If we let the ghost object inherit cpu_writers from the original object, ttm_bo_release_list() will raise a kernel BUG when the ghost object is destroyed. This can be reproduced with the nouveau driver on nv5x. Reported-by: Marcin Slusarz Reviewed-by: Jerome Glisse Tested-by: Marcin Slusarz Signed-off-by: Francisco Jerez Signed-off-by: Dave Airlie --- drivers/gpu/drm/ttm/ttm_bo_util.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers') diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 7cffb3e..3451a82 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -351,6 +351,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, INIT_LIST_HEAD(&fbo->lru); INIT_LIST_HEAD(&fbo->swap); fbo->vm_node = NULL; + atomic_set(&fbo->cpu_writers, 0); fbo->sync_obj = driver->sync_obj_ref(bo->sync_obj); kref_init(&fbo->list_kref); -- cgit v1.1 From 5786aee8bf6d747ea59595601a19e78ad33d6929 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 22 Sep 2010 12:31:53 +0200 Subject: vhost: fix log ctx signalling The log eventfd signalling got put in dead code. We didn't notice because qemu currently does polling instead of eventfd select. Signed-off-by: Michael S. Tsirkin --- drivers/vhost/vhost.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'drivers') diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index c579dcc..dd3d6f7 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -858,11 +858,12 @@ int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log, if (r < 0) return r; len -= l; - if (!len) + if (!len) { + if (vq->log_ctx) + eventfd_signal(vq->log_ctx, 1); return 0; + } } - if (vq->log_ctx) - eventfd_signal(vq->log_ctx, 1); /* Length written exceeds what we have stored. This is a bug. */ BUG(); return 0; -- cgit v1.1 From ec5a32f67c603b11d68eb283d94eb89a4f6cfce1 Mon Sep 17 00:00:00 2001 From: Luca Tettamanti Date: Wed, 22 Sep 2010 10:41:58 +0000 Subject: atl1: fix resume adapter->cmb.cmb is initialized when the device is opened and freed when it's closed. Accessing it unconditionally during resume results either in a crash (NULL pointer dereference, when the interface has not been opened yet) or data corruption (when the interface has been used and brought down adapter->cmb.cmb points to a deallocated memory area). Cc: stable@kernel.org Signed-off-by: Luca Tettamanti Acked-by: Chris Snook Signed-off-by: David S. Miller --- drivers/net/atlx/atl1.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/net/atlx/atl1.c b/drivers/net/atlx/atl1.c index 63b9ba0..bbd6e30 100644 --- a/drivers/net/atlx/atl1.c +++ b/drivers/net/atlx/atl1.c @@ -2847,10 +2847,11 @@ static int atl1_resume(struct pci_dev *pdev) pci_enable_wake(pdev, PCI_D3cold, 0); atl1_reset_hw(&adapter->hw); - adapter->cmb.cmb->int_stats = 0; - if (netif_running(netdev)) + if (netif_running(netdev)) { + adapter->cmb.cmb->int_stats = 0; atl1_up(adapter); + } netif_device_attach(netdev); return 0; -- cgit v1.1 From 3f5a2a713aad28480d86b0add00c68484b54febc Mon Sep 17 00:00:00 2001 From: Luca Tettamanti Date: Wed, 22 Sep 2010 10:42:31 +0000 Subject: atl1: zero out CMB and SBM in atl1_free_ring_resources They are allocated in atl1_setup_ring_resources, zero out the pointers in atl1_free_ring_resources (like the other resources). Signed-off-by: Luca Tettamanti Acked-by: Chris Snook Signed-off-by: David S. Miller --- drivers/net/atlx/atl1.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'drivers') diff --git a/drivers/net/atlx/atl1.c b/drivers/net/atlx/atl1.c index bbd6e30..c73be28 100644 --- a/drivers/net/atlx/atl1.c +++ b/drivers/net/atlx/atl1.c @@ -1251,6 +1251,12 @@ static void atl1_free_ring_resources(struct atl1_adapter *adapter) rrd_ring->desc = NULL; rrd_ring->dma = 0; + + adapter->cmb.dma = 0; + adapter->cmb.cmb = NULL; + + adapter->smb.dma = 0; + adapter->smb.smb = NULL; } static void atl1_setup_mac_ctrl(struct atl1_adapter *adapter) -- cgit v1.1 From 9f9ff20d46c6728b092f34b6a642e1e81ab5e254 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 14 Aug 2010 11:01:45 +0200 Subject: dma/shdma: move dereference below the NULL check "param" can be NULL here, so only dereference it after the check. Signed-off-by: Dan Carpenter Signed-off-by: Dan Williams --- drivers/dma/shdma.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/dma/shdma.c b/drivers/dma/shdma.c index fb64cf3..eb6b54d 100644 --- a/drivers/dma/shdma.c +++ b/drivers/dma/shdma.c @@ -580,7 +580,6 @@ static struct dma_async_tx_descriptor *sh_dmae_prep_slave_sg( sh_chan = to_sh_chan(chan); param = chan->private; - slave_addr = param->config->addr; /* Someone calling slave DMA on a public channel? */ if (!param || !sg_len) { @@ -589,6 +588,8 @@ static struct dma_async_tx_descriptor *sh_dmae_prep_slave_sg( return NULL; } + slave_addr = param->config->addr; + /* * if (param != NULL), this is a successfully requested slave channel, * therefore param->config != NULL too. -- cgit v1.1 From a9e31765e7d528858e1b0c202b823cf4df7577ca Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 22 Sep 2010 13:04:53 -0700 Subject: ipmi: fix acpi probe print After d9e1b6c45059ccf ("ipmi: fix ACPI detection with regspacing") we get [ 11.026326] ipmi_si: probing via ACPI [ 11.030019] ipmi_si 00:09: (null) regsize 1 spacing 1 irq 0 [ 11.035594] ipmi_si: Adding ACPI-specified kcs state machine on an old system with only one range for ipmi kcs range. Try to fix it by adding another res pointer. Signed-off-by: Yinghai Lu Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_si_intf.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'drivers') diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 3822b4f..2be457a 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2126,7 +2126,7 @@ static int __devinit ipmi_pnp_probe(struct pnp_dev *dev, { struct acpi_device *acpi_dev; struct smi_info *info; - struct resource *res; + struct resource *res, *res_second; acpi_handle handle; acpi_status status; unsigned long long tmp; @@ -2182,13 +2182,13 @@ static int __devinit ipmi_pnp_probe(struct pnp_dev *dev, info->io.addr_data = res->start; info->io.regspacing = DEFAULT_REGSPACING; - res = pnp_get_resource(dev, + res_second = pnp_get_resource(dev, (info->io.addr_type == IPMI_IO_ADDR_SPACE) ? IORESOURCE_IO : IORESOURCE_MEM, 1); - if (res) { - if (res->start > info->io.addr_data) - info->io.regspacing = res->start - info->io.addr_data; + if (res_second) { + if (res_second->start > info->io.addr_data) + info->io.regspacing = res_second->start - info->io.addr_data; } info->io.regsize = DEFAULT_REGSPACING; info->io.regshift = 0; -- cgit v1.1 From d544b7a40ad3423676b8876aad64fc5f87296b2d Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 22 Sep 2010 13:04:57 -0700 Subject: vmware balloon: rename module In an effort to minimize customer confusion we want to unify naming convention for VMware-provided kernel modules. This change renames the balloon driver from vmware_ballon to vmw_balloon. We expect to follow this naming convention (vmw_) for all modules that are part of mainline kernel and/or being distributed by VMware, with the sole exception of vmxnet3 driver (since the name of mainline driver happens to match with the name used in VMware Tools). Signed-off-by: Dmitry Torokhov Acked-by: Bhavesh Davda Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/Kconfig | 2 +- drivers/misc/Makefile | 2 +- drivers/misc/vmw_balloon.c | 844 ++++++++++++++++++++++++++++++++++++++++++ drivers/misc/vmware_balloon.c | 844 ------------------------------------------ 4 files changed, 846 insertions(+), 846 deletions(-) create mode 100644 drivers/misc/vmw_balloon.c delete mode 100644 drivers/misc/vmware_balloon.c (limited to 'drivers') diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index 0b591b6..b743312 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -368,7 +368,7 @@ config VMWARE_BALLOON If unsure, say N. To compile this driver as a module, choose M here: the - module will be called vmware_balloon. + module will be called vmw_balloon. config ARM_CHARLCD bool "ARM Ltd. Character LCD Driver" diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile index 255a80d..42eab95 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -33,5 +33,5 @@ obj-$(CONFIG_IWMC3200TOP) += iwmc3200top/ obj-$(CONFIG_HMC6352) += hmc6352.o obj-y += eeprom/ obj-y += cb710/ -obj-$(CONFIG_VMWARE_BALLOON) += vmware_balloon.o +obj-$(CONFIG_VMWARE_BALLOON) += vmw_balloon.o obj-$(CONFIG_ARM_CHARLCD) += arm-charlcd.o diff --git a/drivers/misc/vmw_balloon.c b/drivers/misc/vmw_balloon.c new file mode 100644 index 0000000..2a1e804 --- /dev/null +++ b/drivers/misc/vmw_balloon.c @@ -0,0 +1,844 @@ +/* + * VMware Balloon driver. + * + * Copyright (C) 2000-2010, VMware, Inc. All Rights Reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; version 2 of the License and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. + * + * Maintained by: Dmitry Torokhov + */ + +/* + * This is VMware physical memory management driver for Linux. The driver + * acts like a "balloon" that can be inflated to reclaim physical pages by + * reserving them in the guest and invalidating them in the monitor, + * freeing up the underlying machine pages so they can be allocated to + * other guests. The balloon can also be deflated to allow the guest to + * use more physical memory. Higher level policies can control the sizes + * of balloons in VMs in order to manage physical memory resources. + */ + +//#define DEBUG +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_AUTHOR("VMware, Inc."); +MODULE_DESCRIPTION("VMware Memory Control (Balloon) Driver"); +MODULE_VERSION("1.2.1.1-k"); +MODULE_ALIAS("dmi:*:svnVMware*:*"); +MODULE_ALIAS("vmware_vmmemctl"); +MODULE_LICENSE("GPL"); + +/* + * Various constants controlling rate of inflaint/deflating balloon, + * measured in pages. + */ + +/* + * Rate of allocating memory when there is no memory pressure + * (driver performs non-sleeping allocations). + */ +#define VMW_BALLOON_NOSLEEP_ALLOC_MAX 16384U + +/* + * Rates of memory allocaton when guest experiences memory pressure + * (driver performs sleeping allocations). + */ +#define VMW_BALLOON_RATE_ALLOC_MIN 512U +#define VMW_BALLOON_RATE_ALLOC_MAX 2048U +#define VMW_BALLOON_RATE_ALLOC_INC 16U + +/* + * Rates for releasing pages while deflating balloon. + */ +#define VMW_BALLOON_RATE_FREE_MIN 512U +#define VMW_BALLOON_RATE_FREE_MAX 16384U +#define VMW_BALLOON_RATE_FREE_INC 16U + +/* + * When guest is under memory pressure, use a reduced page allocation + * rate for next several cycles. + */ +#define VMW_BALLOON_SLOW_CYCLES 4 + +/* + * Use __GFP_HIGHMEM to allow pages from HIGHMEM zone. We don't + * allow wait (__GFP_WAIT) for NOSLEEP page allocations. Use + * __GFP_NOWARN, to suppress page allocation failure warnings. + */ +#define VMW_PAGE_ALLOC_NOSLEEP (__GFP_HIGHMEM|__GFP_NOWARN) + +/* + * Use GFP_HIGHUSER when executing in a separate kernel thread + * context and allocation can sleep. This is less stressful to + * the guest memory system, since it allows the thread to block + * while memory is reclaimed, and won't take pages from emergency + * low-memory pools. + */ +#define VMW_PAGE_ALLOC_CANSLEEP (GFP_HIGHUSER) + +/* Maximum number of page allocations without yielding processor */ +#define VMW_BALLOON_YIELD_THRESHOLD 1024 + +/* Maximum number of refused pages we accumulate during inflation cycle */ +#define VMW_BALLOON_MAX_REFUSED 16 + +/* + * Hypervisor communication port definitions. + */ +#define VMW_BALLOON_HV_PORT 0x5670 +#define VMW_BALLOON_HV_MAGIC 0x456c6d6f +#define VMW_BALLOON_PROTOCOL_VERSION 2 +#define VMW_BALLOON_GUEST_ID 1 /* Linux */ + +#define VMW_BALLOON_CMD_START 0 +#define VMW_BALLOON_CMD_GET_TARGET 1 +#define VMW_BALLOON_CMD_LOCK 2 +#define VMW_BALLOON_CMD_UNLOCK 3 +#define VMW_BALLOON_CMD_GUEST_ID 4 + +/* error codes */ +#define VMW_BALLOON_SUCCESS 0 +#define VMW_BALLOON_FAILURE -1 +#define VMW_BALLOON_ERROR_CMD_INVALID 1 +#define VMW_BALLOON_ERROR_PPN_INVALID 2 +#define VMW_BALLOON_ERROR_PPN_LOCKED 3 +#define VMW_BALLOON_ERROR_PPN_UNLOCKED 4 +#define VMW_BALLOON_ERROR_PPN_PINNED 5 +#define VMW_BALLOON_ERROR_PPN_NOTNEEDED 6 +#define VMW_BALLOON_ERROR_RESET 7 +#define VMW_BALLOON_ERROR_BUSY 8 + +#define VMWARE_BALLOON_CMD(cmd, data, result) \ +({ \ + unsigned long __stat, __dummy1, __dummy2; \ + __asm__ __volatile__ ("inl (%%dx)" : \ + "=a"(__stat), \ + "=c"(__dummy1), \ + "=d"(__dummy2), \ + "=b"(result) : \ + "0"(VMW_BALLOON_HV_MAGIC), \ + "1"(VMW_BALLOON_CMD_##cmd), \ + "2"(VMW_BALLOON_HV_PORT), \ + "3"(data) : \ + "memory"); \ + result &= -1UL; \ + __stat & -1UL; \ +}) + +#ifdef CONFIG_DEBUG_FS +struct vmballoon_stats { + unsigned int timer; + + /* allocation statustics */ + unsigned int alloc; + unsigned int alloc_fail; + unsigned int sleep_alloc; + unsigned int sleep_alloc_fail; + unsigned int refused_alloc; + unsigned int refused_free; + unsigned int free; + + /* monitor operations */ + unsigned int lock; + unsigned int lock_fail; + unsigned int unlock; + unsigned int unlock_fail; + unsigned int target; + unsigned int target_fail; + unsigned int start; + unsigned int start_fail; + unsigned int guest_type; + unsigned int guest_type_fail; +}; + +#define STATS_INC(stat) (stat)++ +#else +#define STATS_INC(stat) +#endif + +struct vmballoon { + + /* list of reserved physical pages */ + struct list_head pages; + + /* transient list of non-balloonable pages */ + struct list_head refused_pages; + unsigned int n_refused_pages; + + /* balloon size in pages */ + unsigned int size; + unsigned int target; + + /* reset flag */ + bool reset_required; + + /* adjustment rates (pages per second) */ + unsigned int rate_alloc; + unsigned int rate_free; + + /* slowdown page allocations for next few cycles */ + unsigned int slow_allocation_cycles; + +#ifdef CONFIG_DEBUG_FS + /* statistics */ + struct vmballoon_stats stats; + + /* debugfs file exporting statistics */ + struct dentry *dbg_entry; +#endif + + struct sysinfo sysinfo; + + struct delayed_work dwork; +}; + +static struct vmballoon balloon; +static struct workqueue_struct *vmballoon_wq; + +/* + * Send "start" command to the host, communicating supported version + * of the protocol. + */ +static bool vmballoon_send_start(struct vmballoon *b) +{ + unsigned long status, dummy; + + STATS_INC(b->stats.start); + + status = VMWARE_BALLOON_CMD(START, VMW_BALLOON_PROTOCOL_VERSION, dummy); + if (status == VMW_BALLOON_SUCCESS) + return true; + + pr_debug("%s - failed, hv returns %ld\n", __func__, status); + STATS_INC(b->stats.start_fail); + return false; +} + +static bool vmballoon_check_status(struct vmballoon *b, unsigned long status) +{ + switch (status) { + case VMW_BALLOON_SUCCESS: + return true; + + case VMW_BALLOON_ERROR_RESET: + b->reset_required = true; + /* fall through */ + + default: + return false; + } +} + +/* + * Communicate guest type to the host so that it can adjust ballooning + * algorithm to the one most appropriate for the guest. This command + * is normally issued after sending "start" command and is part of + * standard reset sequence. + */ +static bool vmballoon_send_guest_id(struct vmballoon *b) +{ + unsigned long status, dummy; + + status = VMWARE_BALLOON_CMD(GUEST_ID, VMW_BALLOON_GUEST_ID, dummy); + + STATS_INC(b->stats.guest_type); + + if (vmballoon_check_status(b, status)) + return true; + + pr_debug("%s - failed, hv returns %ld\n", __func__, status); + STATS_INC(b->stats.guest_type_fail); + return false; +} + +/* + * Retrieve desired balloon size from the host. + */ +static bool vmballoon_send_get_target(struct vmballoon *b, u32 *new_target) +{ + unsigned long status; + unsigned long target; + unsigned long limit; + u32 limit32; + + /* + * si_meminfo() is cheap. Moreover, we want to provide dynamic + * max balloon size later. So let us call si_meminfo() every + * iteration. + */ + si_meminfo(&b->sysinfo); + limit = b->sysinfo.totalram; + + /* Ensure limit fits in 32-bits */ + limit32 = (u32)limit; + if (limit != limit32) + return false; + + /* update stats */ + STATS_INC(b->stats.target); + + status = VMWARE_BALLOON_CMD(GET_TARGET, limit, target); + if (vmballoon_check_status(b, status)) { + *new_target = target; + return true; + } + + pr_debug("%s - failed, hv returns %ld\n", __func__, status); + STATS_INC(b->stats.target_fail); + return false; +} + +/* + * Notify the host about allocated page so that host can use it without + * fear that guest will need it. Host may reject some pages, we need to + * check the return value and maybe submit a different page. + */ +static bool vmballoon_send_lock_page(struct vmballoon *b, unsigned long pfn) +{ + unsigned long status, dummy; + u32 pfn32; + + pfn32 = (u32)pfn; + if (pfn32 != pfn) + return false; + + STATS_INC(b->stats.lock); + + status = VMWARE_BALLOON_CMD(LOCK, pfn, dummy); + if (vmballoon_check_status(b, status)) + return true; + + pr_debug("%s - ppn %lx, hv returns %ld\n", __func__, pfn, status); + STATS_INC(b->stats.lock_fail); + return false; +} + +/* + * Notify the host that guest intends to release given page back into + * the pool of available (to the guest) pages. + */ +static bool vmballoon_send_unlock_page(struct vmballoon *b, unsigned long pfn) +{ + unsigned long status, dummy; + u32 pfn32; + + pfn32 = (u32)pfn; + if (pfn32 != pfn) + return false; + + STATS_INC(b->stats.unlock); + + status = VMWARE_BALLOON_CMD(UNLOCK, pfn, dummy); + if (vmballoon_check_status(b, status)) + return true; + + pr_debug("%s - ppn %lx, hv returns %ld\n", __func__, pfn, status); + STATS_INC(b->stats.unlock_fail); + return false; +} + +/* + * Quickly release all pages allocated for the balloon. This function is + * called when host decides to "reset" balloon for one reason or another. + * Unlike normal "deflate" we do not (shall not) notify host of the pages + * being released. + */ +static void vmballoon_pop(struct vmballoon *b) +{ + struct page *page, *next; + unsigned int count = 0; + + list_for_each_entry_safe(page, next, &b->pages, lru) { + list_del(&page->lru); + __free_page(page); + STATS_INC(b->stats.free); + b->size--; + + if (++count >= b->rate_free) { + count = 0; + cond_resched(); + } + } +} + +/* + * Perform standard reset sequence by popping the balloon (in case it + * is not empty) and then restarting protocol. This operation normally + * happens when host responds with VMW_BALLOON_ERROR_RESET to a command. + */ +static void vmballoon_reset(struct vmballoon *b) +{ + /* free all pages, skipping monitor unlock */ + vmballoon_pop(b); + + if (vmballoon_send_start(b)) { + b->reset_required = false; + if (!vmballoon_send_guest_id(b)) + pr_err("failed to send guest ID to the host\n"); + } +} + +/* + * Allocate (or reserve) a page for the balloon and notify the host. If host + * refuses the page put it on "refuse" list and allocate another one until host + * is satisfied. "Refused" pages are released at the end of inflation cycle + * (when we allocate b->rate_alloc pages). + */ +static int vmballoon_reserve_page(struct vmballoon *b, bool can_sleep) +{ + struct page *page; + gfp_t flags; + bool locked = false; + + do { + if (!can_sleep) + STATS_INC(b->stats.alloc); + else + STATS_INC(b->stats.sleep_alloc); + + flags = can_sleep ? VMW_PAGE_ALLOC_CANSLEEP : VMW_PAGE_ALLOC_NOSLEEP; + page = alloc_page(flags); + if (!page) { + if (!can_sleep) + STATS_INC(b->stats.alloc_fail); + else + STATS_INC(b->stats.sleep_alloc_fail); + return -ENOMEM; + } + + /* inform monitor */ + locked = vmballoon_send_lock_page(b, page_to_pfn(page)); + if (!locked) { + STATS_INC(b->stats.refused_alloc); + + if (b->reset_required) { + __free_page(page); + return -EIO; + } + + /* + * Place page on the list of non-balloonable pages + * and retry allocation, unless we already accumulated + * too many of them, in which case take a breather. + */ + list_add(&page->lru, &b->refused_pages); + if (++b->n_refused_pages >= VMW_BALLOON_MAX_REFUSED) + return -EIO; + } + } while (!locked); + + /* track allocated page */ + list_add(&page->lru, &b->pages); + + /* update balloon size */ + b->size++; + + return 0; +} + +/* + * Release the page allocated for the balloon. Note that we first notify + * the host so it can make sure the page will be available for the guest + * to use, if needed. + */ +static int vmballoon_release_page(struct vmballoon *b, struct page *page) +{ + if (!vmballoon_send_unlock_page(b, page_to_pfn(page))) + return -EIO; + + list_del(&page->lru); + + /* deallocate page */ + __free_page(page); + STATS_INC(b->stats.free); + + /* update balloon size */ + b->size--; + + return 0; +} + +/* + * Release pages that were allocated while attempting to inflate the + * balloon but were refused by the host for one reason or another. + */ +static void vmballoon_release_refused_pages(struct vmballoon *b) +{ + struct page *page, *next; + + list_for_each_entry_safe(page, next, &b->refused_pages, lru) { + list_del(&page->lru); + __free_page(page); + STATS_INC(b->stats.refused_free); + } + + b->n_refused_pages = 0; +} + +/* + * Inflate the balloon towards its target size. Note that we try to limit + * the rate of allocation to make sure we are not choking the rest of the + * system. + */ +static void vmballoon_inflate(struct vmballoon *b) +{ + unsigned int goal; + unsigned int rate; + unsigned int i; + unsigned int allocations = 0; + int error = 0; + bool alloc_can_sleep = false; + + pr_debug("%s - size: %d, target %d\n", __func__, b->size, b->target); + + /* + * First try NOSLEEP page allocations to inflate balloon. + * + * If we do not throttle nosleep allocations, we can drain all + * free pages in the guest quickly (if the balloon target is high). + * As a side-effect, draining free pages helps to inform (force) + * the guest to start swapping if balloon target is not met yet, + * which is a desired behavior. However, balloon driver can consume + * all available CPU cycles if too many pages are allocated in a + * second. Therefore, we throttle nosleep allocations even when + * the guest is not under memory pressure. OTOH, if we have already + * predicted that the guest is under memory pressure, then we + * slowdown page allocations considerably. + */ + + goal = b->target - b->size; + /* + * Start with no sleep allocation rate which may be higher + * than sleeping allocation rate. + */ + rate = b->slow_allocation_cycles ? + b->rate_alloc : VMW_BALLOON_NOSLEEP_ALLOC_MAX; + + pr_debug("%s - goal: %d, no-sleep rate: %d, sleep rate: %d\n", + __func__, goal, rate, b->rate_alloc); + + for (i = 0; i < goal; i++) { + + error = vmballoon_reserve_page(b, alloc_can_sleep); + if (error) { + if (error != -ENOMEM) { + /* + * Not a page allocation failure, stop this + * cycle. Maybe we'll get new target from + * the host soon. + */ + break; + } + + if (alloc_can_sleep) { + /* + * CANSLEEP page allocation failed, so guest + * is under severe memory pressure. Quickly + * decrease allocation rate. + */ + b->rate_alloc = max(b->rate_alloc / 2, + VMW_BALLOON_RATE_ALLOC_MIN); + break; + } + + /* + * NOSLEEP page allocation failed, so the guest is + * under memory pressure. Let us slow down page + * allocations for next few cycles so that the guest + * gets out of memory pressure. Also, if we already + * allocated b->rate_alloc pages, let's pause, + * otherwise switch to sleeping allocations. + */ + b->slow_allocation_cycles = VMW_BALLOON_SLOW_CYCLES; + + if (i >= b->rate_alloc) + break; + + alloc_can_sleep = true; + /* Lower rate for sleeping allocations. */ + rate = b->rate_alloc; + } + + if (++allocations > VMW_BALLOON_YIELD_THRESHOLD) { + cond_resched(); + allocations = 0; + } + + if (i >= rate) { + /* We allocated enough pages, let's take a break. */ + break; + } + } + + /* + * We reached our goal without failures so try increasing + * allocation rate. + */ + if (error == 0 && i >= b->rate_alloc) { + unsigned int mult = i / b->rate_alloc; + + b->rate_alloc = + min(b->rate_alloc + mult * VMW_BALLOON_RATE_ALLOC_INC, + VMW_BALLOON_RATE_ALLOC_MAX); + } + + vmballoon_release_refused_pages(b); +} + +/* + * Decrease the size of the balloon allowing guest to use more memory. + */ +static void vmballoon_deflate(struct vmballoon *b) +{ + struct page *page, *next; + unsigned int i = 0; + unsigned int goal; + int error; + + pr_debug("%s - size: %d, target %d\n", __func__, b->size, b->target); + + /* limit deallocation rate */ + goal = min(b->size - b->target, b->rate_free); + + pr_debug("%s - goal: %d, rate: %d\n", __func__, goal, b->rate_free); + + /* free pages to reach target */ + list_for_each_entry_safe(page, next, &b->pages, lru) { + error = vmballoon_release_page(b, page); + if (error) { + /* quickly decrease rate in case of error */ + b->rate_free = max(b->rate_free / 2, + VMW_BALLOON_RATE_FREE_MIN); + return; + } + + if (++i >= goal) + break; + } + + /* slowly increase rate if there were no errors */ + b->rate_free = min(b->rate_free + VMW_BALLOON_RATE_FREE_INC, + VMW_BALLOON_RATE_FREE_MAX); +} + +/* + * Balloon work function: reset protocol, if needed, get the new size and + * adjust balloon as needed. Repeat in 1 sec. + */ +static void vmballoon_work(struct work_struct *work) +{ + struct delayed_work *dwork = to_delayed_work(work); + struct vmballoon *b = container_of(dwork, struct vmballoon, dwork); + unsigned int target; + + STATS_INC(b->stats.timer); + + if (b->reset_required) + vmballoon_reset(b); + + if (b->slow_allocation_cycles > 0) + b->slow_allocation_cycles--; + + if (vmballoon_send_get_target(b, &target)) { + /* update target, adjust size */ + b->target = target; + + if (b->size < target) + vmballoon_inflate(b); + else if (b->size > target) + vmballoon_deflate(b); + } + + queue_delayed_work(vmballoon_wq, dwork, round_jiffies_relative(HZ)); +} + +/* + * DEBUGFS Interface + */ +#ifdef CONFIG_DEBUG_FS + +static int vmballoon_debug_show(struct seq_file *f, void *offset) +{ + struct vmballoon *b = f->private; + struct vmballoon_stats *stats = &b->stats; + + /* format size info */ + seq_printf(f, + "target: %8d pages\n" + "current: %8d pages\n", + b->target, b->size); + + /* format rate info */ + seq_printf(f, + "rateNoSleepAlloc: %8d pages/sec\n" + "rateSleepAlloc: %8d pages/sec\n" + "rateFree: %8d pages/sec\n", + VMW_BALLOON_NOSLEEP_ALLOC_MAX, + b->rate_alloc, b->rate_free); + + seq_printf(f, + "\n" + "timer: %8u\n" + "start: %8u (%4u failed)\n" + "guestType: %8u (%4u failed)\n" + "lock: %8u (%4u failed)\n" + "unlock: %8u (%4u failed)\n" + "target: %8u (%4u failed)\n" + "primNoSleepAlloc: %8u (%4u failed)\n" + "primCanSleepAlloc: %8u (%4u failed)\n" + "primFree: %8u\n" + "errAlloc: %8u\n" + "errFree: %8u\n", + stats->timer, + stats->start, stats->start_fail, + stats->guest_type, stats->guest_type_fail, + stats->lock, stats->lock_fail, + stats->unlock, stats->unlock_fail, + stats->target, stats->target_fail, + stats->alloc, stats->alloc_fail, + stats->sleep_alloc, stats->sleep_alloc_fail, + stats->free, + stats->refused_alloc, stats->refused_free); + + return 0; +} + +static int vmballoon_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, vmballoon_debug_show, inode->i_private); +} + +static const struct file_operations vmballoon_debug_fops = { + .owner = THIS_MODULE, + .open = vmballoon_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __init vmballoon_debugfs_init(struct vmballoon *b) +{ + int error; + + b->dbg_entry = debugfs_create_file("vmmemctl", S_IRUGO, NULL, b, + &vmballoon_debug_fops); + if (IS_ERR(b->dbg_entry)) { + error = PTR_ERR(b->dbg_entry); + pr_err("failed to create debugfs entry, error: %d\n", error); + return error; + } + + return 0; +} + +static void __exit vmballoon_debugfs_exit(struct vmballoon *b) +{ + debugfs_remove(b->dbg_entry); +} + +#else + +static inline int vmballoon_debugfs_init(struct vmballoon *b) +{ + return 0; +} + +static inline void vmballoon_debugfs_exit(struct vmballoon *b) +{ +} + +#endif /* CONFIG_DEBUG_FS */ + +static int __init vmballoon_init(void) +{ + int error; + + /* + * Check if we are running on VMware's hypervisor and bail out + * if we are not. + */ + if (x86_hyper != &x86_hyper_vmware) + return -ENODEV; + + vmballoon_wq = create_freezeable_workqueue("vmmemctl"); + if (!vmballoon_wq) { + pr_err("failed to create workqueue\n"); + return -ENOMEM; + } + + INIT_LIST_HEAD(&balloon.pages); + INIT_LIST_HEAD(&balloon.refused_pages); + + /* initialize rates */ + balloon.rate_alloc = VMW_BALLOON_RATE_ALLOC_MAX; + balloon.rate_free = VMW_BALLOON_RATE_FREE_MAX; + + INIT_DELAYED_WORK(&balloon.dwork, vmballoon_work); + + /* + * Start balloon. + */ + if (!vmballoon_send_start(&balloon)) { + pr_err("failed to send start command to the host\n"); + error = -EIO; + goto fail; + } + + if (!vmballoon_send_guest_id(&balloon)) { + pr_err("failed to send guest ID to the host\n"); + error = -EIO; + goto fail; + } + + error = vmballoon_debugfs_init(&balloon); + if (error) + goto fail; + + queue_delayed_work(vmballoon_wq, &balloon.dwork, 0); + + return 0; + +fail: + destroy_workqueue(vmballoon_wq); + return error; +} +module_init(vmballoon_init); + +static void __exit vmballoon_exit(void) +{ + cancel_delayed_work_sync(&balloon.dwork); + destroy_workqueue(vmballoon_wq); + + vmballoon_debugfs_exit(&balloon); + + /* + * Deallocate all reserved memory, and reset connection with monitor. + * Reset connection before deallocating memory to avoid potential for + * additional spurious resets from guest touching deallocated pages. + */ + vmballoon_send_start(&balloon); + vmballoon_pop(&balloon); +} +module_exit(vmballoon_exit); diff --git a/drivers/misc/vmware_balloon.c b/drivers/misc/vmware_balloon.c deleted file mode 100644 index 2a1e804..0000000 --- a/drivers/misc/vmware_balloon.c +++ /dev/null @@ -1,844 +0,0 @@ -/* - * VMware Balloon driver. - * - * Copyright (C) 2000-2010, VMware, Inc. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the - * Free Software Foundation; version 2 of the License and no later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. - * - * Maintained by: Dmitry Torokhov - */ - -/* - * This is VMware physical memory management driver for Linux. The driver - * acts like a "balloon" that can be inflated to reclaim physical pages by - * reserving them in the guest and invalidating them in the monitor, - * freeing up the underlying machine pages so they can be allocated to - * other guests. The balloon can also be deflated to allow the guest to - * use more physical memory. Higher level policies can control the sizes - * of balloons in VMs in order to manage physical memory resources. - */ - -//#define DEBUG -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -MODULE_AUTHOR("VMware, Inc."); -MODULE_DESCRIPTION("VMware Memory Control (Balloon) Driver"); -MODULE_VERSION("1.2.1.1-k"); -MODULE_ALIAS("dmi:*:svnVMware*:*"); -MODULE_ALIAS("vmware_vmmemctl"); -MODULE_LICENSE("GPL"); - -/* - * Various constants controlling rate of inflaint/deflating balloon, - * measured in pages. - */ - -/* - * Rate of allocating memory when there is no memory pressure - * (driver performs non-sleeping allocations). - */ -#define VMW_BALLOON_NOSLEEP_ALLOC_MAX 16384U - -/* - * Rates of memory allocaton when guest experiences memory pressure - * (driver performs sleeping allocations). - */ -#define VMW_BALLOON_RATE_ALLOC_MIN 512U -#define VMW_BALLOON_RATE_ALLOC_MAX 2048U -#define VMW_BALLOON_RATE_ALLOC_INC 16U - -/* - * Rates for releasing pages while deflating balloon. - */ -#define VMW_BALLOON_RATE_FREE_MIN 512U -#define VMW_BALLOON_RATE_FREE_MAX 16384U -#define VMW_BALLOON_RATE_FREE_INC 16U - -/* - * When guest is under memory pressure, use a reduced page allocation - * rate for next several cycles. - */ -#define VMW_BALLOON_SLOW_CYCLES 4 - -/* - * Use __GFP_HIGHMEM to allow pages from HIGHMEM zone. We don't - * allow wait (__GFP_WAIT) for NOSLEEP page allocations. Use - * __GFP_NOWARN, to suppress page allocation failure warnings. - */ -#define VMW_PAGE_ALLOC_NOSLEEP (__GFP_HIGHMEM|__GFP_NOWARN) - -/* - * Use GFP_HIGHUSER when executing in a separate kernel thread - * context and allocation can sleep. This is less stressful to - * the guest memory system, since it allows the thread to block - * while memory is reclaimed, and won't take pages from emergency - * low-memory pools. - */ -#define VMW_PAGE_ALLOC_CANSLEEP (GFP_HIGHUSER) - -/* Maximum number of page allocations without yielding processor */ -#define VMW_BALLOON_YIELD_THRESHOLD 1024 - -/* Maximum number of refused pages we accumulate during inflation cycle */ -#define VMW_BALLOON_MAX_REFUSED 16 - -/* - * Hypervisor communication port definitions. - */ -#define VMW_BALLOON_HV_PORT 0x5670 -#define VMW_BALLOON_HV_MAGIC 0x456c6d6f -#define VMW_BALLOON_PROTOCOL_VERSION 2 -#define VMW_BALLOON_GUEST_ID 1 /* Linux */ - -#define VMW_BALLOON_CMD_START 0 -#define VMW_BALLOON_CMD_GET_TARGET 1 -#define VMW_BALLOON_CMD_LOCK 2 -#define VMW_BALLOON_CMD_UNLOCK 3 -#define VMW_BALLOON_CMD_GUEST_ID 4 - -/* error codes */ -#define VMW_BALLOON_SUCCESS 0 -#define VMW_BALLOON_FAILURE -1 -#define VMW_BALLOON_ERROR_CMD_INVALID 1 -#define VMW_BALLOON_ERROR_PPN_INVALID 2 -#define VMW_BALLOON_ERROR_PPN_LOCKED 3 -#define VMW_BALLOON_ERROR_PPN_UNLOCKED 4 -#define VMW_BALLOON_ERROR_PPN_PINNED 5 -#define VMW_BALLOON_ERROR_PPN_NOTNEEDED 6 -#define VMW_BALLOON_ERROR_RESET 7 -#define VMW_BALLOON_ERROR_BUSY 8 - -#define VMWARE_BALLOON_CMD(cmd, data, result) \ -({ \ - unsigned long __stat, __dummy1, __dummy2; \ - __asm__ __volatile__ ("inl (%%dx)" : \ - "=a"(__stat), \ - "=c"(__dummy1), \ - "=d"(__dummy2), \ - "=b"(result) : \ - "0"(VMW_BALLOON_HV_MAGIC), \ - "1"(VMW_BALLOON_CMD_##cmd), \ - "2"(VMW_BALLOON_HV_PORT), \ - "3"(data) : \ - "memory"); \ - result &= -1UL; \ - __stat & -1UL; \ -}) - -#ifdef CONFIG_DEBUG_FS -struct vmballoon_stats { - unsigned int timer; - - /* allocation statustics */ - unsigned int alloc; - unsigned int alloc_fail; - unsigned int sleep_alloc; - unsigned int sleep_alloc_fail; - unsigned int refused_alloc; - unsigned int refused_free; - unsigned int free; - - /* monitor operations */ - unsigned int lock; - unsigned int lock_fail; - unsigned int unlock; - unsigned int unlock_fail; - unsigned int target; - unsigned int target_fail; - unsigned int start; - unsigned int start_fail; - unsigned int guest_type; - unsigned int guest_type_fail; -}; - -#define STATS_INC(stat) (stat)++ -#else -#define STATS_INC(stat) -#endif - -struct vmballoon { - - /* list of reserved physical pages */ - struct list_head pages; - - /* transient list of non-balloonable pages */ - struct list_head refused_pages; - unsigned int n_refused_pages; - - /* balloon size in pages */ - unsigned int size; - unsigned int target; - - /* reset flag */ - bool reset_required; - - /* adjustment rates (pages per second) */ - unsigned int rate_alloc; - unsigned int rate_free; - - /* slowdown page allocations for next few cycles */ - unsigned int slow_allocation_cycles; - -#ifdef CONFIG_DEBUG_FS - /* statistics */ - struct vmballoon_stats stats; - - /* debugfs file exporting statistics */ - struct dentry *dbg_entry; -#endif - - struct sysinfo sysinfo; - - struct delayed_work dwork; -}; - -static struct vmballoon balloon; -static struct workqueue_struct *vmballoon_wq; - -/* - * Send "start" command to the host, communicating supported version - * of the protocol. - */ -static bool vmballoon_send_start(struct vmballoon *b) -{ - unsigned long status, dummy; - - STATS_INC(b->stats.start); - - status = VMWARE_BALLOON_CMD(START, VMW_BALLOON_PROTOCOL_VERSION, dummy); - if (status == VMW_BALLOON_SUCCESS) - return true; - - pr_debug("%s - failed, hv returns %ld\n", __func__, status); - STATS_INC(b->stats.start_fail); - return false; -} - -static bool vmballoon_check_status(struct vmballoon *b, unsigned long status) -{ - switch (status) { - case VMW_BALLOON_SUCCESS: - return true; - - case VMW_BALLOON_ERROR_RESET: - b->reset_required = true; - /* fall through */ - - default: - return false; - } -} - -/* - * Communicate guest type to the host so that it can adjust ballooning - * algorithm to the one most appropriate for the guest. This command - * is normally issued after sending "start" command and is part of - * standard reset sequence. - */ -static bool vmballoon_send_guest_id(struct vmballoon *b) -{ - unsigned long status, dummy; - - status = VMWARE_BALLOON_CMD(GUEST_ID, VMW_BALLOON_GUEST_ID, dummy); - - STATS_INC(b->stats.guest_type); - - if (vmballoon_check_status(b, status)) - return true; - - pr_debug("%s - failed, hv returns %ld\n", __func__, status); - STATS_INC(b->stats.guest_type_fail); - return false; -} - -/* - * Retrieve desired balloon size from the host. - */ -static bool vmballoon_send_get_target(struct vmballoon *b, u32 *new_target) -{ - unsigned long status; - unsigned long target; - unsigned long limit; - u32 limit32; - - /* - * si_meminfo() is cheap. Moreover, we want to provide dynamic - * max balloon size later. So let us call si_meminfo() every - * iteration. - */ - si_meminfo(&b->sysinfo); - limit = b->sysinfo.totalram; - - /* Ensure limit fits in 32-bits */ - limit32 = (u32)limit; - if (limit != limit32) - return false; - - /* update stats */ - STATS_INC(b->stats.target); - - status = VMWARE_BALLOON_CMD(GET_TARGET, limit, target); - if (vmballoon_check_status(b, status)) { - *new_target = target; - return true; - } - - pr_debug("%s - failed, hv returns %ld\n", __func__, status); - STATS_INC(b->stats.target_fail); - return false; -} - -/* - * Notify the host about allocated page so that host can use it without - * fear that guest will need it. Host may reject some pages, we need to - * check the return value and maybe submit a different page. - */ -static bool vmballoon_send_lock_page(struct vmballoon *b, unsigned long pfn) -{ - unsigned long status, dummy; - u32 pfn32; - - pfn32 = (u32)pfn; - if (pfn32 != pfn) - return false; - - STATS_INC(b->stats.lock); - - status = VMWARE_BALLOON_CMD(LOCK, pfn, dummy); - if (vmballoon_check_status(b, status)) - return true; - - pr_debug("%s - ppn %lx, hv returns %ld\n", __func__, pfn, status); - STATS_INC(b->stats.lock_fail); - return false; -} - -/* - * Notify the host that guest intends to release given page back into - * the pool of available (to the guest) pages. - */ -static bool vmballoon_send_unlock_page(struct vmballoon *b, unsigned long pfn) -{ - unsigned long status, dummy; - u32 pfn32; - - pfn32 = (u32)pfn; - if (pfn32 != pfn) - return false; - - STATS_INC(b->stats.unlock); - - status = VMWARE_BALLOON_CMD(UNLOCK, pfn, dummy); - if (vmballoon_check_status(b, status)) - return true; - - pr_debug("%s - ppn %lx, hv returns %ld\n", __func__, pfn, status); - STATS_INC(b->stats.unlock_fail); - return false; -} - -/* - * Quickly release all pages allocated for the balloon. This function is - * called when host decides to "reset" balloon for one reason or another. - * Unlike normal "deflate" we do not (shall not) notify host of the pages - * being released. - */ -static void vmballoon_pop(struct vmballoon *b) -{ - struct page *page, *next; - unsigned int count = 0; - - list_for_each_entry_safe(page, next, &b->pages, lru) { - list_del(&page->lru); - __free_page(page); - STATS_INC(b->stats.free); - b->size--; - - if (++count >= b->rate_free) { - count = 0; - cond_resched(); - } - } -} - -/* - * Perform standard reset sequence by popping the balloon (in case it - * is not empty) and then restarting protocol. This operation normally - * happens when host responds with VMW_BALLOON_ERROR_RESET to a command. - */ -static void vmballoon_reset(struct vmballoon *b) -{ - /* free all pages, skipping monitor unlock */ - vmballoon_pop(b); - - if (vmballoon_send_start(b)) { - b->reset_required = false; - if (!vmballoon_send_guest_id(b)) - pr_err("failed to send guest ID to the host\n"); - } -} - -/* - * Allocate (or reserve) a page for the balloon and notify the host. If host - * refuses the page put it on "refuse" list and allocate another one until host - * is satisfied. "Refused" pages are released at the end of inflation cycle - * (when we allocate b->rate_alloc pages). - */ -static int vmballoon_reserve_page(struct vmballoon *b, bool can_sleep) -{ - struct page *page; - gfp_t flags; - bool locked = false; - - do { - if (!can_sleep) - STATS_INC(b->stats.alloc); - else - STATS_INC(b->stats.sleep_alloc); - - flags = can_sleep ? VMW_PAGE_ALLOC_CANSLEEP : VMW_PAGE_ALLOC_NOSLEEP; - page = alloc_page(flags); - if (!page) { - if (!can_sleep) - STATS_INC(b->stats.alloc_fail); - else - STATS_INC(b->stats.sleep_alloc_fail); - return -ENOMEM; - } - - /* inform monitor */ - locked = vmballoon_send_lock_page(b, page_to_pfn(page)); - if (!locked) { - STATS_INC(b->stats.refused_alloc); - - if (b->reset_required) { - __free_page(page); - return -EIO; - } - - /* - * Place page on the list of non-balloonable pages - * and retry allocation, unless we already accumulated - * too many of them, in which case take a breather. - */ - list_add(&page->lru, &b->refused_pages); - if (++b->n_refused_pages >= VMW_BALLOON_MAX_REFUSED) - return -EIO; - } - } while (!locked); - - /* track allocated page */ - list_add(&page->lru, &b->pages); - - /* update balloon size */ - b->size++; - - return 0; -} - -/* - * Release the page allocated for the balloon. Note that we first notify - * the host so it can make sure the page will be available for the guest - * to use, if needed. - */ -static int vmballoon_release_page(struct vmballoon *b, struct page *page) -{ - if (!vmballoon_send_unlock_page(b, page_to_pfn(page))) - return -EIO; - - list_del(&page->lru); - - /* deallocate page */ - __free_page(page); - STATS_INC(b->stats.free); - - /* update balloon size */ - b->size--; - - return 0; -} - -/* - * Release pages that were allocated while attempting to inflate the - * balloon but were refused by the host for one reason or another. - */ -static void vmballoon_release_refused_pages(struct vmballoon *b) -{ - struct page *page, *next; - - list_for_each_entry_safe(page, next, &b->refused_pages, lru) { - list_del(&page->lru); - __free_page(page); - STATS_INC(b->stats.refused_free); - } - - b->n_refused_pages = 0; -} - -/* - * Inflate the balloon towards its target size. Note that we try to limit - * the rate of allocation to make sure we are not choking the rest of the - * system. - */ -static void vmballoon_inflate(struct vmballoon *b) -{ - unsigned int goal; - unsigned int rate; - unsigned int i; - unsigned int allocations = 0; - int error = 0; - bool alloc_can_sleep = false; - - pr_debug("%s - size: %d, target %d\n", __func__, b->size, b->target); - - /* - * First try NOSLEEP page allocations to inflate balloon. - * - * If we do not throttle nosleep allocations, we can drain all - * free pages in the guest quickly (if the balloon target is high). - * As a side-effect, draining free pages helps to inform (force) - * the guest to start swapping if balloon target is not met yet, - * which is a desired behavior. However, balloon driver can consume - * all available CPU cycles if too many pages are allocated in a - * second. Therefore, we throttle nosleep allocations even when - * the guest is not under memory pressure. OTOH, if we have already - * predicted that the guest is under memory pressure, then we - * slowdown page allocations considerably. - */ - - goal = b->target - b->size; - /* - * Start with no sleep allocation rate which may be higher - * than sleeping allocation rate. - */ - rate = b->slow_allocation_cycles ? - b->rate_alloc : VMW_BALLOON_NOSLEEP_ALLOC_MAX; - - pr_debug("%s - goal: %d, no-sleep rate: %d, sleep rate: %d\n", - __func__, goal, rate, b->rate_alloc); - - for (i = 0; i < goal; i++) { - - error = vmballoon_reserve_page(b, alloc_can_sleep); - if (error) { - if (error != -ENOMEM) { - /* - * Not a page allocation failure, stop this - * cycle. Maybe we'll get new target from - * the host soon. - */ - break; - } - - if (alloc_can_sleep) { - /* - * CANSLEEP page allocation failed, so guest - * is under severe memory pressure. Quickly - * decrease allocation rate. - */ - b->rate_alloc = max(b->rate_alloc / 2, - VMW_BALLOON_RATE_ALLOC_MIN); - break; - } - - /* - * NOSLEEP page allocation failed, so the guest is - * under memory pressure. Let us slow down page - * allocations for next few cycles so that the guest - * gets out of memory pressure. Also, if we already - * allocated b->rate_alloc pages, let's pause, - * otherwise switch to sleeping allocations. - */ - b->slow_allocation_cycles = VMW_BALLOON_SLOW_CYCLES; - - if (i >= b->rate_alloc) - break; - - alloc_can_sleep = true; - /* Lower rate for sleeping allocations. */ - rate = b->rate_alloc; - } - - if (++allocations > VMW_BALLOON_YIELD_THRESHOLD) { - cond_resched(); - allocations = 0; - } - - if (i >= rate) { - /* We allocated enough pages, let's take a break. */ - break; - } - } - - /* - * We reached our goal without failures so try increasing - * allocation rate. - */ - if (error == 0 && i >= b->rate_alloc) { - unsigned int mult = i / b->rate_alloc; - - b->rate_alloc = - min(b->rate_alloc + mult * VMW_BALLOON_RATE_ALLOC_INC, - VMW_BALLOON_RATE_ALLOC_MAX); - } - - vmballoon_release_refused_pages(b); -} - -/* - * Decrease the size of the balloon allowing guest to use more memory. - */ -static void vmballoon_deflate(struct vmballoon *b) -{ - struct page *page, *next; - unsigned int i = 0; - unsigned int goal; - int error; - - pr_debug("%s - size: %d, target %d\n", __func__, b->size, b->target); - - /* limit deallocation rate */ - goal = min(b->size - b->target, b->rate_free); - - pr_debug("%s - goal: %d, rate: %d\n", __func__, goal, b->rate_free); - - /* free pages to reach target */ - list_for_each_entry_safe(page, next, &b->pages, lru) { - error = vmballoon_release_page(b, page); - if (error) { - /* quickly decrease rate in case of error */ - b->rate_free = max(b->rate_free / 2, - VMW_BALLOON_RATE_FREE_MIN); - return; - } - - if (++i >= goal) - break; - } - - /* slowly increase rate if there were no errors */ - b->rate_free = min(b->rate_free + VMW_BALLOON_RATE_FREE_INC, - VMW_BALLOON_RATE_FREE_MAX); -} - -/* - * Balloon work function: reset protocol, if needed, get the new size and - * adjust balloon as needed. Repeat in 1 sec. - */ -static void vmballoon_work(struct work_struct *work) -{ - struct delayed_work *dwork = to_delayed_work(work); - struct vmballoon *b = container_of(dwork, struct vmballoon, dwork); - unsigned int target; - - STATS_INC(b->stats.timer); - - if (b->reset_required) - vmballoon_reset(b); - - if (b->slow_allocation_cycles > 0) - b->slow_allocation_cycles--; - - if (vmballoon_send_get_target(b, &target)) { - /* update target, adjust size */ - b->target = target; - - if (b->size < target) - vmballoon_inflate(b); - else if (b->size > target) - vmballoon_deflate(b); - } - - queue_delayed_work(vmballoon_wq, dwork, round_jiffies_relative(HZ)); -} - -/* - * DEBUGFS Interface - */ -#ifdef CONFIG_DEBUG_FS - -static int vmballoon_debug_show(struct seq_file *f, void *offset) -{ - struct vmballoon *b = f->private; - struct vmballoon_stats *stats = &b->stats; - - /* format size info */ - seq_printf(f, - "target: %8d pages\n" - "current: %8d pages\n", - b->target, b->size); - - /* format rate info */ - seq_printf(f, - "rateNoSleepAlloc: %8d pages/sec\n" - "rateSleepAlloc: %8d pages/sec\n" - "rateFree: %8d pages/sec\n", - VMW_BALLOON_NOSLEEP_ALLOC_MAX, - b->rate_alloc, b->rate_free); - - seq_printf(f, - "\n" - "timer: %8u\n" - "start: %8u (%4u failed)\n" - "guestType: %8u (%4u failed)\n" - "lock: %8u (%4u failed)\n" - "unlock: %8u (%4u failed)\n" - "target: %8u (%4u failed)\n" - "primNoSleepAlloc: %8u (%4u failed)\n" - "primCanSleepAlloc: %8u (%4u failed)\n" - "primFree: %8u\n" - "errAlloc: %8u\n" - "errFree: %8u\n", - stats->timer, - stats->start, stats->start_fail, - stats->guest_type, stats->guest_type_fail, - stats->lock, stats->lock_fail, - stats->unlock, stats->unlock_fail, - stats->target, stats->target_fail, - stats->alloc, stats->alloc_fail, - stats->sleep_alloc, stats->sleep_alloc_fail, - stats->free, - stats->refused_alloc, stats->refused_free); - - return 0; -} - -static int vmballoon_debug_open(struct inode *inode, struct file *file) -{ - return single_open(file, vmballoon_debug_show, inode->i_private); -} - -static const struct file_operations vmballoon_debug_fops = { - .owner = THIS_MODULE, - .open = vmballoon_debug_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int __init vmballoon_debugfs_init(struct vmballoon *b) -{ - int error; - - b->dbg_entry = debugfs_create_file("vmmemctl", S_IRUGO, NULL, b, - &vmballoon_debug_fops); - if (IS_ERR(b->dbg_entry)) { - error = PTR_ERR(b->dbg_entry); - pr_err("failed to create debugfs entry, error: %d\n", error); - return error; - } - - return 0; -} - -static void __exit vmballoon_debugfs_exit(struct vmballoon *b) -{ - debugfs_remove(b->dbg_entry); -} - -#else - -static inline int vmballoon_debugfs_init(struct vmballoon *b) -{ - return 0; -} - -static inline void vmballoon_debugfs_exit(struct vmballoon *b) -{ -} - -#endif /* CONFIG_DEBUG_FS */ - -static int __init vmballoon_init(void) -{ - int error; - - /* - * Check if we are running on VMware's hypervisor and bail out - * if we are not. - */ - if (x86_hyper != &x86_hyper_vmware) - return -ENODEV; - - vmballoon_wq = create_freezeable_workqueue("vmmemctl"); - if (!vmballoon_wq) { - pr_err("failed to create workqueue\n"); - return -ENOMEM; - } - - INIT_LIST_HEAD(&balloon.pages); - INIT_LIST_HEAD(&balloon.refused_pages); - - /* initialize rates */ - balloon.rate_alloc = VMW_BALLOON_RATE_ALLOC_MAX; - balloon.rate_free = VMW_BALLOON_RATE_FREE_MAX; - - INIT_DELAYED_WORK(&balloon.dwork, vmballoon_work); - - /* - * Start balloon. - */ - if (!vmballoon_send_start(&balloon)) { - pr_err("failed to send start command to the host\n"); - error = -EIO; - goto fail; - } - - if (!vmballoon_send_guest_id(&balloon)) { - pr_err("failed to send guest ID to the host\n"); - error = -EIO; - goto fail; - } - - error = vmballoon_debugfs_init(&balloon); - if (error) - goto fail; - - queue_delayed_work(vmballoon_wq, &balloon.dwork, 0); - - return 0; - -fail: - destroy_workqueue(vmballoon_wq); - return error; -} -module_init(vmballoon_init); - -static void __exit vmballoon_exit(void) -{ - cancel_delayed_work_sync(&balloon.dwork); - destroy_workqueue(vmballoon_wq); - - vmballoon_debugfs_exit(&balloon); - - /* - * Deallocate all reserved memory, and reset connection with monitor. - * Reset connection before deallocating memory to avoid potential for - * additional spurious resets from guest touching deallocated pages. - */ - vmballoon_send_start(&balloon); - vmballoon_pop(&balloon); -} -module_exit(vmballoon_exit); -- cgit v1.1 From eba93fcc34d6c4387ce8fbb53bb7b685f91f3343 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 22 Sep 2010 13:04:59 -0700 Subject: drivers/rtc/rtc-ab3100.c: add missing platform_set_drvdata() in ab3100_rtc_probe() Otherwise, calling platform_get_drvdata() in ab3100_rtc_remove() returns NULL. Signed-off-by: Axel Lin Acked-by:Wan ZongShun Acked-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-ab3100.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers') diff --git a/drivers/rtc/rtc-ab3100.c b/drivers/rtc/rtc-ab3100.c index d26780e..261a07e 100644 --- a/drivers/rtc/rtc-ab3100.c +++ b/drivers/rtc/rtc-ab3100.c @@ -235,6 +235,7 @@ static int __init ab3100_rtc_probe(struct platform_device *pdev) err = PTR_ERR(rtc); return err; } + platform_set_drvdata(pdev, rtc); return 0; } @@ -244,6 +245,7 @@ static int __exit ab3100_rtc_remove(struct platform_device *pdev) struct rtc_device *rtc = platform_get_drvdata(pdev); rtc_device_unregister(rtc); + platform_set_drvdata(pdev, NULL); return 0; } -- cgit v1.1 From 85a00d9bbfb4704fbf368944b1cb9fed8f1598c5 Mon Sep 17 00:00:00 2001 From: Peter Jones Date: Wed, 22 Sep 2010 13:05:04 -0700 Subject: efifb: check that the base address is plausible on pci systems Some Apple machines have identical DMI data but different memory configurations for the video. Given that, check that the address in our table is actually within the range of a PCI BAR on a VGA device in the machine. This also fixes up the return value from set_system(), which has always been wrong, but never resulted in bad behavior since there's only ever been one matching entry in the dmi table. The patch 1) stops people's machines from crashing when we get their display wrong, which seems to be unfortunately inevitable, 2) allows us to support identical dmi data with differing video memory configurations This also adds me as the efifb maintainer, since I've effectively been acting as such for quite some time. Signed-off-by: Peter Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/efifb.c | 61 +++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 49 insertions(+), 12 deletions(-) (limited to 'drivers') diff --git a/drivers/video/efifb.c b/drivers/video/efifb.c index 815f84b..c082b61 100644 --- a/drivers/video/efifb.c +++ b/drivers/video/efifb.c @@ -13,7 +13,7 @@ #include #include #include - +#include #include