From 39c3fd58952d7599d367c84c1330b785d91d6088 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 2 Dec 2017 18:11:04 +0100 Subject: kernel/irq: Extend lockdep class for request mutex The IRQ code already has support for lockdep class for the lock mutex in an interrupt descriptor. Extend this to add a second class for the request mutex in the descriptor. Not having a class is resulting in false positive splats in some code paths. Signed-off-by: Andrew Lunn Signed-off-by: Thomas Gleixner Acked-by: linus.walleij@linaro.org Cc: grygorii.strashko@ti.com Cc: f.fainelli@gmail.com Link: https://lkml.kernel.org/r/1512234664-21555-1-git-send-email-andrew@lunn.ch --- kernel/irq/generic-chip.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c index c26c5bb..508c03d 100644 --- a/kernel/irq/generic-chip.c +++ b/kernel/irq/generic-chip.c @@ -364,10 +364,11 @@ irq_get_domain_generic_chip(struct irq_domain *d, unsigned int hw_irq) EXPORT_SYMBOL_GPL(irq_get_domain_generic_chip); /* - * Separate lockdep class for interrupt chip which can nest irq_desc - * lock. + * Separate lockdep classes for interrupt chip which can nest irq_desc + * lock and request mutex. */ static struct lock_class_key irq_nested_lock_class; +static struct lock_class_key irq_nested_request_class; /* * irq_map_generic_chip - Map a generic chip for an irq domain @@ -409,7 +410,8 @@ int irq_map_generic_chip(struct irq_domain *d, unsigned int virq, set_bit(idx, &gc->installed); if (dgc->gc_flags & IRQ_GC_INIT_NESTED_LOCK) - irq_set_lockdep_class(virq, &irq_nested_lock_class); + irq_set_lockdep_class(virq, &irq_nested_lock_class, + &irq_nested_request_class); if (chip->irq_calc_mask) chip->irq_calc_mask(data); @@ -479,7 +481,8 @@ void irq_setup_generic_chip(struct irq_chip_generic *gc, u32 msk, continue; if (flags & IRQ_GC_INIT_NESTED_LOCK) - irq_set_lockdep_class(i, &irq_nested_lock_class); + irq_set_lockdep_class(i, &irq_nested_lock_class, + &irq_nested_request_class); if (!(flags & IRQ_GC_NO_MASK)) { struct irq_data *d = irq_get_irq_data(i); -- cgit v1.1 From 11bca0a83f83f6093d816295668e74ef24595944 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 2 Dec 2017 09:13:04 -0800 Subject: genirq: Guard handle_bad_irq log messages An interrupt storm on a bad interrupt will cause the kernel log to be clogged. [ 60.089234] ->handle_irq(): ffffffffbe2f803f, [ 60.090455] 0xffffffffbf2af380 [ 60.090510] handle_bad_irq+0x0/0x2e5 [ 60.090522] ->irq_data.chip(): ffffffffbf2af380, [ 60.090553] IRQ_NOPROBE set [ 60.090584] ->handle_irq(): ffffffffbe2f803f, [ 60.090590] handle_bad_irq+0x0/0x2e5 [ 60.090596] ->irq_data.chip(): ffffffffbf2af380, [ 60.090602] 0xffffffffbf2af380 [ 60.090608] ->action(): (null) [ 60.090779] handle_bad_irq+0x0/0x2e5 This was seen when running an upstream kernel on Acer Chromebook R11. The system was unstable as result. Guard the log message with __printk_ratelimit to reduce the impact. This won't prevent the interrupt storm from happening, but at least the system remains stable. Signed-off-by: Guenter Roeck Signed-off-by: Thomas Gleixner Cc: Dmitry Torokhov Cc: Joe Perches Cc: Andy Shevchenko Cc: Mika Westerberg Link: https://bugzilla.kernel.org/show_bug.cgi?id=197953 Link: https://lkml.kernel.org/r/1512234784-21038-1-git-send-email-linux@roeck-us.net --- kernel/irq/debug.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/irq/debug.h b/kernel/irq/debug.h index 17f05ef..e4d3819 100644 --- a/kernel/irq/debug.h +++ b/kernel/irq/debug.h @@ -12,6 +12,11 @@ static inline void print_irq_desc(unsigned int irq, struct irq_desc *desc) { + static DEFINE_RATELIMIT_STATE(ratelimit, 5 * HZ, 5); + + if (!__ratelimit(&ratelimit)) + return; + printk("irq %d, desc: %p, depth: %d, count: %d, unhandled: %d\n", irq, desc, desc->depth, desc->irq_count, desc->irqs_unhandled); printk("->handle_irq(): %p, ", desc->handle_irq); -- cgit v1.1 From da5dd9e854d2edd6b02ebfe28583052f922104da Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 29 Dec 2017 10:42:10 +0100 Subject: genirq/msi: Handle reactivation only on success When analyzing the fallout of the x86 vector allocation rework it turned out that the error handling in msi_domain_alloc_irqs() is broken. If MSI_FLAG_MUST_REACTIVATE is set for a MSI domain then it clears the activation flag for a successfully initialized msi descriptor. If a subsequent initialization fails then the error handling code path does not deactivate the interrupt because the activation flag got cleared. Move the clearing of the activation flag outside of the initialization loop so that an eventual failure can be cleaned up correctly. Fixes: 22d0b12f3560 ("genirq/irqdomain: Add force reactivation flag to irq domains") Signed-off-by: Thomas Gleixner Tested-by: Alexandru Chirvasitu Tested-by: Andy Shevchenko Cc: Dou Liyang Cc: Pavel Machek Cc: Maciej W. Rozycki Cc: Mikael Pettersson Cc: Josh Poulson Cc: Mihai Costache Cc: Stephen Hemminger Cc: Marc Zyngier Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang Cc: Dexuan Cui Cc: Simon Xiao Cc: Saeed Mahameed Cc: Jork Loeser Cc: Bjorn Helgaas Cc: devel@linuxdriverproject.org Cc: KY Srinivasan Cc: Alan Cox Cc: Sakari Ailus , Cc: linux-media@vger.kernel.org --- kernel/irq/msi.c | 35 +++++++++++++++++++++++++++-------- 1 file changed, 27 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index edb987b..9ba9543 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -339,6 +339,13 @@ int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev, return ret; } +static bool msi_check_reservation_mode(struct msi_domain_info *info) +{ + if (!(info->flags & MSI_FLAG_MUST_REACTIVATE)) + return false; + return true; +} + /** * msi_domain_alloc_irqs - Allocate interrupts from a MSI interrupt domain * @domain: The domain to allocate from @@ -353,9 +360,11 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, { struct msi_domain_info *info = domain->host_data; struct msi_domain_ops *ops = info->ops; - msi_alloc_info_t arg; + struct irq_data *irq_data; struct msi_desc *desc; + msi_alloc_info_t arg; int i, ret, virq; + bool can_reserve; ret = msi_domain_prepare_irqs(domain, dev, nvec, &arg); if (ret) @@ -385,6 +394,8 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, if (ops->msi_finish) ops->msi_finish(&arg, 0); + can_reserve = msi_check_reservation_mode(info); + for_each_msi_entry(desc, dev) { virq = desc->irq; if (desc->nvec_used == 1) @@ -397,15 +408,23 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, * the MSI entries before the PCI layer enables MSI in the * card. Otherwise the card latches a random msi message. */ - if (info->flags & MSI_FLAG_ACTIVATE_EARLY) { - struct irq_data *irq_data; + if (!(info->flags & MSI_FLAG_ACTIVATE_EARLY)) + continue; + irq_data = irq_domain_get_irq_data(domain, desc->irq); + ret = irq_domain_activate_irq(irq_data, true); + if (ret) + goto cleanup; + } + + /* + * If these interrupts use reservation mode, clear the activated bit + * so request_irq() will assign the final vector. + */ + if (can_reserve) { + for_each_msi_entry(desc, dev) { irq_data = irq_domain_get_irq_data(domain, desc->irq); - ret = irq_domain_activate_irq(irq_data, true); - if (ret) - goto cleanup; - if (info->flags & MSI_FLAG_MUST_REACTIVATE) - irqd_clr_activated(irq_data); + irqd_clr_activated(irq_data); } } return 0; -- cgit v1.1 From 69790ba92b8d67eaee5e50b30a5b696d40664caf Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 29 Dec 2017 16:44:34 +0100 Subject: genirq: Introduce IRQD_CAN_RESERVE flag Add a new flag to mark interrupts which can use reservation mode. This is going to be used in subsequent patches to disable reservation mode for a certain class of MSI devices. Signed-off-by: Thomas Gleixner Tested-by: Alexandru Chirvasitu Tested-by: Andy Shevchenko Cc: Dou Liyang Cc: Pavel Machek Cc: Maciej W. Rozycki Cc: Mikael Pettersson Cc: Josh Poulson Cc: Mihai Costache Cc: Stephen Hemminger Cc: Marc Zyngier Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang Cc: Dexuan Cui Cc: Simon Xiao Cc: Saeed Mahameed Cc: Jork Loeser Cc: Bjorn Helgaas Cc: devel@linuxdriverproject.org Cc: KY Srinivasan Cc: Alan Cox Cc: Sakari Ailus , Cc: linux-media@vger.kernel.org --- kernel/irq/debugfs.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/irq/debugfs.c b/kernel/irq/debugfs.c index 7f608ac..acfaaef 100644 --- a/kernel/irq/debugfs.c +++ b/kernel/irq/debugfs.c @@ -113,6 +113,7 @@ static const struct irq_bit_descr irqdata_states[] = { BIT_MASK_DESCR(IRQD_SETAFFINITY_PENDING), BIT_MASK_DESCR(IRQD_AFFINITY_MANAGED), BIT_MASK_DESCR(IRQD_MANAGED_SHUTDOWN), + BIT_MASK_DESCR(IRQD_CAN_RESERVE), BIT_MASK_DESCR(IRQD_FORWARDED_TO_VCPU), -- cgit v1.1 From 702cb0a02813299d6911b775c637906ae21b737d Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 29 Dec 2017 16:59:06 +0100 Subject: genirq/irqdomain: Rename early argument of irq_domain_activate_irq() The 'early' argument of irq_domain_activate_irq() is actually used to denote reservation mode. To avoid confusion, rename it before abuse happens. No functional change. Fixes: 72491643469a ("genirq/irqdomain: Update irq_domain_ops.activate() signature") Signed-off-by: Thomas Gleixner Cc: Alexandru Chirvasitu Cc: Andy Shevchenko Cc: Dou Liyang Cc: Pavel Machek Cc: Maciej W. Rozycki Cc: Mikael Pettersson Cc: Josh Poulson Cc: Mihai Costache Cc: Stephen Hemminger Cc: Marc Zyngier Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang Cc: Dexuan Cui Cc: Simon Xiao Cc: Saeed Mahameed Cc: Jork Loeser Cc: Bjorn Helgaas Cc: devel@linuxdriverproject.org Cc: KY Srinivasan Cc: Alan Cox Cc: Sakari Ailus , Cc: linux-media@vger.kernel.org --- kernel/irq/internals.h | 2 +- kernel/irq/irqdomain.c | 13 +++++++------ 2 files changed, 8 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index 07d08ca..ab19371 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -440,7 +440,7 @@ static inline bool irq_fixup_move_pending(struct irq_desc *desc, bool fclear) #endif /* !CONFIG_GENERIC_PENDING_IRQ */ #if !defined(CONFIG_IRQ_DOMAIN) || !defined(CONFIG_IRQ_DOMAIN_HIERARCHY) -static inline int irq_domain_activate_irq(struct irq_data *data, bool early) +static inline int irq_domain_activate_irq(struct irq_data *data, bool reserve) { irqd_set_activated(data); return 0; diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 4f4f600..62068ad 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -1693,7 +1693,7 @@ static void __irq_domain_deactivate_irq(struct irq_data *irq_data) } } -static int __irq_domain_activate_irq(struct irq_data *irqd, bool early) +static int __irq_domain_activate_irq(struct irq_data *irqd, bool reserve) { int ret = 0; @@ -1702,9 +1702,9 @@ static int __irq_domain_activate_irq(struct irq_data *irqd, bool early) if (irqd->parent_data) ret = __irq_domain_activate_irq(irqd->parent_data, - early); + reserve); if (!ret && domain->ops->activate) { - ret = domain->ops->activate(domain, irqd, early); + ret = domain->ops->activate(domain, irqd, reserve); /* Rollback in case of error */ if (ret && irqd->parent_data) __irq_domain_deactivate_irq(irqd->parent_data); @@ -1716,17 +1716,18 @@ static int __irq_domain_activate_irq(struct irq_data *irqd, bool early) /** * irq_domain_activate_irq - Call domain_ops->activate recursively to activate * interrupt - * @irq_data: outermost irq_data associated with interrupt + * @irq_data: Outermost irq_data associated with interrupt + * @reserve: If set only reserve an interrupt vector instead of assigning one * * This is the second step to call domain_ops->activate to program interrupt * controllers, so the interrupt could actually get delivered. */ -int irq_domain_activate_irq(struct irq_data *irq_data, bool early) +int irq_domain_activate_irq(struct irq_data *irq_data, bool reserve) { int ret = 0; if (!irqd_is_activated(irq_data)) - ret = __irq_domain_activate_irq(irq_data, early); + ret = __irq_domain_activate_irq(irq_data, reserve); if (!ret) irqd_set_activated(irq_data); return ret; -- cgit v1.1 From bc976233a872c0f20f018fb1e89264a541584e25 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 29 Dec 2017 10:47:22 +0100 Subject: genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu Reported-by: Andy Shevchenko Signed-off-by: Thomas Gleixner Tested-by: Alexandru Chirvasitu Tested-by: Andy Shevchenko Cc: Dou Liyang Cc: Pavel Machek Cc: Maciej W. Rozycki Cc: Mikael Pettersson Cc: Josh Poulson Cc: Mihai Costache Cc: Stephen Hemminger Cc: Marc Zyngier Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang Cc: Dexuan Cui Cc: Simon Xiao Cc: Saeed Mahameed Cc: Jork Loeser Cc: Bjorn Helgaas Cc: devel@linuxdriverproject.org Cc: KY Srinivasan Cc: Alan Cox Cc: Sakari Ailus , Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos --- kernel/irq/msi.c | 37 +++++++++++++++++++++++++++++++++---- 1 file changed, 33 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index 9ba9543..2f3c4f5 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -339,11 +339,38 @@ int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev, return ret; } -static bool msi_check_reservation_mode(struct msi_domain_info *info) +/* + * Carefully check whether the device can use reservation mode. If + * reservation mode is enabled then the early activation will assign a + * dummy vector to the device. If the PCI/MSI device does not support + * masking of the entry then this can result in spurious interrupts when + * the device driver is not absolutely careful. But even then a malfunction + * of the hardware could result in a spurious interrupt on the dummy vector + * and render the device unusable. If the entry can be masked then the core + * logic will prevent the spurious interrupt and reservation mode can be + * used. For now reservation mode is restricted to PCI/MSI. + */ +static bool msi_check_reservation_mode(struct irq_domain *domain, + struct msi_domain_info *info, + struct device *dev) { + struct msi_desc *desc; + + if (domain->bus_token != DOMAIN_BUS_PCI_MSI) + return false; + if (!(info->flags & MSI_FLAG_MUST_REACTIVATE)) return false; - return true; + + if (IS_ENABLED(CONFIG_PCI_MSI) && pci_msi_ignore_mask) + return false; + + /* + * Checking the first MSI descriptor is sufficient. MSIX supports + * masking and MSI does so when the maskbit is set. + */ + desc = first_msi_entry(dev); + return desc->msi_attrib.is_msix || desc->msi_attrib.maskbit; } /** @@ -394,7 +421,7 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, if (ops->msi_finish) ops->msi_finish(&arg, 0); - can_reserve = msi_check_reservation_mode(info); + can_reserve = msi_check_reservation_mode(domain, info, dev); for_each_msi_entry(desc, dev) { virq = desc->irq; @@ -412,7 +439,9 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, continue; irq_data = irq_domain_get_irq_data(domain, desc->irq); - ret = irq_domain_activate_irq(irq_data, true); + if (!can_reserve) + irqd_clr_can_reserve(irq_data); + ret = irq_domain_activate_irq(irq_data, can_reserve); if (ret) goto cleanup; } -- cgit v1.1