linux/arch/x86/kernel/irq_64.c

166 lines
3.7 KiB
C
Raw Normal View History

/*
* Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
*
* This file contains the lowest level x86_64-specific interrupt
* entry and irq statistics code. All the remaining irq logic is
* done by the generic kernel/irq/ code and in the
* x86_64-specific irq controller code. (e.g. i8259.c and
* io_apic.c.)
*/
#include <linux/kernel_stat.h>
#include <linux/interrupt.h>
#include <linux/seq_file.h>
#include <linux/module.h>
#include <linux/delay.h>
#include <linux/ftrace.h>
#include <linux/uaccess.h>
#include <linux/smp.h>
#include <asm/io_apic.h>
#include <asm/idle.h>
DEFINE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat);
EXPORT_PER_CPU_SYMBOL(irq_stat);
DEFINE_PER_CPU(struct pt_regs *, irq_regs);
EXPORT_PER_CPU_SYMBOL(irq_regs);
/*
* Probabilistic stack overflow check:
*
* Only check the stack in process context, because everything else
* runs on the big interrupt stacks. Checking reliably is too expensive,
* so we just check from interrupts.
*/
static inline void stack_overflow_check(struct pt_regs *regs)
{
#ifdef CONFIG_DEBUG_STACKOVERFLOW
u64 curbase = (u64)task_stack_page(current);
WARN_ONCE(regs->sp >= curbase &&
regs->sp <= curbase + THREAD_SIZE &&
regs->sp < curbase + sizeof(struct thread_info) +
sizeof(struct pt_regs) + 128,
"do_IRQ: %s near stack overflow (cur:%Lx,sp:%lx)\n",
current->comm, curbase, regs->sp);
#endif
}
/*
* do_IRQ handles all normal device IRQ's (the special
* SMP cross-CPU interrupts have their own specific
* handlers).
*/
asmlinkage unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 15:55:46 +02:00
{
struct pt_regs *old_regs = set_irq_regs(regs);
struct irq_desc *desc;
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 15:55:46 +02:00
/* high bit used in ret_from_ code */
unsigned vector = ~regs->orig_ax;
unsigned irq;
exit_idle();
irq_enter();
irq = __get_cpu_var(vector_irq)[vector];
stack_overflow_check(regs);
desc = irq_to_desc(irq);
if (likely(desc))
generic_handle_irq_desc(irq, desc);
else {
if (!disable_apic)
ack_APIC_irq();
if (printk_ratelimit())
printk(KERN_EMERG "%s: %d.%d No irq handler for vector\n",
__func__, smp_processor_id(), vector);
}
irq_exit();
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 15:55:46 +02:00
set_irq_regs(old_regs);
return 1;
}
#ifdef CONFIG_HOTPLUG_CPU
/* A cpu has been removed from cpu_online_mask. Reset irq affinities. */
void fixup_irqs(void)
{
unsigned int irq;
static int warned;
struct irq_desc *desc;
for_each_irq_desc(irq, desc) {
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
int break_affinity = 0;
int set_affinity = 1;
const struct cpumask *affinity;
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
if (!desc)
continue;
if (irq == 2)
continue;
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
/* interrupt's are disabled at this point */
spin_lock(&desc->lock);
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
affinity = desc->affinity;
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
if (!irq_has_action(irq) ||
cpumask_equal(affinity, cpu_online_mask)) {
spin_unlock(&desc->lock);
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
continue;
}
if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
break_affinity = 1;
affinity = cpu_all_mask;
}
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
if (desc->chip->mask)
desc->chip->mask(irq);
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
if (desc->chip->set_affinity)
desc->chip->set_affinity(irq, affinity);
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
else if (!(warned++))
set_affinity = 0;
if (desc->chip->unmask)
desc->chip->unmask(irq);
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
spin_unlock(&desc->lock);
x86_64 irq: use mask/unmask and proper locking in fixup_irqs() Force irq migration path during cpu offline, is not using proper locks and irq_chip mask/unmask routines. This will result in some races(especially the device generating the interrupt can see some inconsistent state, resulting in issues like stuck irq,..). Appended patch fixes the issue by taking proper lock and encapsulating irq_chip set_affinity() with a mask() before and an unmask() after. This fixes a MSI irq stuck issue reported by Darrick Wong. There are several more general bugs in this area(irq migration in the process context). For example, 1. Possibility of missing edge triggered irq. 2. Reliable method of migrating level triggered irq in the process context. We plan to look and close these in the near future. Eric says: In addition even with the fix from Suresh there is still at least one nasty hardware race in fixup_irqs(). However we exercise that code path rarely enough that we are unlikely to hit it in the real world, and that race seems to have existed since the code was merged. And a fix for that is not coming soon as it is an open investigation area if we can fix irq migration to work outside of irq context or if we have to rework the requirements imposed by the generic cpu hotplug and layer on fixup_irqs(). So this may come up again. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Reported-and-tested-by: Darrick Wong <djwong@us.ibm.com> Cc: Andi Kleen <ak@suse.de> Acked-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-26 00:52:35 +02:00
if (break_affinity && set_affinity)
printk("Broke affinity for irq %i\n", irq);
else if (!set_affinity)
printk("Cannot set affinity for irq %i\n", irq);
}
/* That doesn't seem sufficient. Give it 1ms. */
local_irq_enable();
mdelay(1);
local_irq_disable();
}
#endif
extern void call_softirq(void);
asmlinkage void do_softirq(void)
{
__u32 pending;
unsigned long flags;
if (in_interrupt())
return;
local_irq_save(flags);
pending = local_softirq_pending();
/* Switch to interrupt stack */
if (pending) {
call_softirq();
WARN_ON_ONCE(softirq_count());
}
local_irq_restore(flags);
}