linux/kernel/irq/spurious.c

348 lines
8.2 KiB
C
Raw Normal View History

/*
* linux/kernel/irq/spurious.c
*
* Copyright (C) 1992, 1998-2004 Linus Torvalds, Ingo Molnar
*
* This file contains spurious interrupt handling.
*/
#include <linux/jiffies.h>
#include <linux/irq.h>
#include <linux/module.h>
#include <linux/kallsyms.h>
#include <linux/interrupt.h>
#include <linux/moduleparam.h>
#include <linux/timer.h>
#include "internals.h"
static int irqfixup __read_mostly;
#define POLL_SPURIOUS_IRQ_INTERVAL (HZ/10)
static void poll_spurious_irqs(unsigned long dummy);
static DEFINE_TIMER(poll_spurious_irq_timer, poll_spurious_irqs, 0, 0);
static int irq_poll_cpu;
static atomic_t irq_poll_active;
/*
* We wait here for a poller to finish.
*
* If the poll runs on this CPU, then we yell loudly and return
* false. That will leave the interrupt line disabled in the worst
* case, but it should never happen.
*
* We wait until the poller is done and then recheck disabled and
* action (about to be disabled). Only if it's still active, we return
* true and let the handler run.
*/
bool irq_wait_for_poll(struct irq_desc *desc)
{
if (WARN_ONCE(irq_poll_cpu == smp_processor_id(),
"irq poll in progress on cpu %d for irq %d\n",
smp_processor_id(), desc->irq_data.irq))
return false;
#ifdef CONFIG_SMP
do {
raw_spin_unlock(&desc->lock);
while (desc->status & IRQ_INPROGRESS)
cpu_relax();
raw_spin_lock(&desc->lock);
} while (desc->status & IRQ_INPROGRESS);
/* Might have been disabled in meantime */
return !(desc->status & IRQ_DISABLED) && desc->action;
#else
return false;
#endif
}
/*
* Recovery handler for misrouted interrupts.
*/
static int try_one_irq(int irq, struct irq_desc *desc, bool force)
{
irqreturn_t ret = IRQ_NONE;
struct irqaction *action;
raw_spin_lock(&desc->lock);
/* PER_CPU and nested thread interrupts are never polled */
if (desc->status & (IRQ_PER_CPU | IRQ_NESTED_THREAD))
goto out;
/*
* Do not poll disabled interrupts unless the spurious
* disabled poller asks explicitely.
*/
if ((desc->status & IRQ_DISABLED) && !force)
goto out;
/*
* All handlers must agree on IRQF_SHARED, so we test just the
* first. Check for action->next as well.
*/
action = desc->action;
if (!action || !(action->flags & IRQF_SHARED) ||
(action->flags & __IRQF_TIMER) || !action->next)
goto out;
/* Already running on another processor */
if (desc->status & IRQ_INPROGRESS) {
/*
* Already running: If it is shared get the other
* CPU to go looking for our mystery interrupt too
*/
desc->status |= IRQ_PENDING;
goto out;
}
/* Mark it poll in progress */
desc->status |= IRQ_POLL_INPROGRESS;
do {
if (handle_irq_event(desc) == IRQ_HANDLED)
ret = IRQ_HANDLED;
action = desc->action;
} while ((desc->status & IRQ_PENDING) && action);
desc->status &= ~IRQ_POLL_INPROGRESS;
out:
raw_spin_unlock(&desc->lock);
return ret == IRQ_HANDLED;
}
static int misrouted_irq(int irq)
{
struct irq_desc *desc;
int i, ok = 0;
if (atomic_inc_return(&irq_poll_active) == 1)
goto out;
irq_poll_cpu = smp_processor_id();
for_each_irq_desc(i, desc) {
if (!i)
continue;
if (i == irq) /* Already tried */
continue;
if (try_one_irq(i, desc, false))
ok = 1;
}
out:
atomic_dec(&irq_poll_active);
/* So the caller can adjust the irq error counts */
return ok;
}
static void poll_spurious_irqs(unsigned long dummy)
{
struct irq_desc *desc;
int i;
if (atomic_inc_return(&irq_poll_active) != 1)
goto out;
irq_poll_cpu = smp_processor_id();
for_each_irq_desc(i, desc) {
unsigned int status;
if (!i)
continue;
/* Racy but it doesn't matter */
status = desc->status;
barrier();
if (!(status & IRQ_SPURIOUS_DISABLED))
continue;
genirq: try_one_irq() must be called with irq disabled Prarit reported: ================================= [ INFO: inconsistent lock state ] 2.6.32-rc5 #1 --------------------------------- inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage. swapper/0 [HC0[0]:SC1[1]:HE1:SE0] takes: (&irq_desc_lock_class){?.-...}, at: [<ffffffff810c264e>] try_one_irq+0x32/0x138 {IN-HARDIRQ-W} state was registered at: [<ffffffff81095160>] __lock_acquire+0x2fc/0xd5d [<ffffffff81095cb4>] lock_acquire+0xf3/0x12d [<ffffffff814cdadd>] _spin_lock+0x40/0x89 [<ffffffff810c3389>] handle_level_irq+0x30/0x105 [<ffffffff81014e0e>] handle_irq+0x95/0xb7 [<ffffffff810141bd>] do_IRQ+0x6a/0xe0 [<ffffffff81012813>] ret_from_intr+0x0/0x16 irq event stamp: 195096 hardirqs last enabled at (195096): [<ffffffff814cd7f7>] _spin_unlock_irq+0x3a/0x5c hardirqs last disabled at (195095): [<ffffffff814cdbdd>] _spin_lock_irq+0x29/0x95 softirqs last enabled at (195088): [<ffffffff81068c92>] __do_softirq+0x1c1/0x1ef softirqs last disabled at (195093): [<ffffffff8101304c>] call_softirq+0x1c/0x30 other info that might help us debug this: 1 lock held by swapper/0: #0: (kernel/irq/spurious.c:21){+.-...}, at: [<ffffffff81070cf2>] run_timer_softirq+0x1a9/0x315 stack backtrace: Pid: 0, comm: swapper Not tainted 2.6.32-rc5 #1 Call Trace: <IRQ> [<ffffffff81093e94>] valid_state+0x187/0x1ae [<ffffffff81093fe4>] mark_lock+0x129/0x253 [<ffffffff810951d4>] __lock_acquire+0x370/0xd5d [<ffffffff81095cb4>] lock_acquire+0xf3/0x12d [<ffffffff814cdadd>] _spin_lock+0x40/0x89 [<ffffffff810c264e>] try_one_irq+0x32/0x138 [<ffffffff810c2795>] poll_all_shared_irqs+0x41/0x6d [<ffffffff810c27dd>] poll_spurious_irqs+0x1c/0x49 [<ffffffff81070d82>] run_timer_softirq+0x239/0x315 [<ffffffff81068bd3>] __do_softirq+0x102/0x1ef [<ffffffff8101304c>] call_softirq+0x1c/0x30 [<ffffffff81014b65>] do_softirq+0x59/0xca [<ffffffff810686ad>] irq_exit+0x58/0xae [<ffffffff81029b84>] smp_apic_timer_interrupt+0x94/0xba [<ffffffff81012a33>] apic_timer_interrupt+0x13/0x20 The reason is that try_one_irq() is called from hardirq context with interrupts disabled and from softirq context (poll_all_shared_irqs()) with interrupts enabled. Disable interrupts before calling it from poll_all_shared_irqs(). Reported-and-tested-by: Prarit Bhargava <prarit@redhat.com> Signed-off-by: Yong Zhang <yong.zhang0@gmail.com> LKML-Reference: <1257563773-4620-1-git-send-email-yong.zhang0@gmail.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2009-11-07 11:16:13 +08:00
local_irq_disable();
try_one_irq(i, desc, true);
genirq: try_one_irq() must be called with irq disabled Prarit reported: ================================= [ INFO: inconsistent lock state ] 2.6.32-rc5 #1 --------------------------------- inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage. swapper/0 [HC0[0]:SC1[1]:HE1:SE0] takes: (&irq_desc_lock_class){?.-...}, at: [<ffffffff810c264e>] try_one_irq+0x32/0x138 {IN-HARDIRQ-W} state was registered at: [<ffffffff81095160>] __lock_acquire+0x2fc/0xd5d [<ffffffff81095cb4>] lock_acquire+0xf3/0x12d [<ffffffff814cdadd>] _spin_lock+0x40/0x89 [<ffffffff810c3389>] handle_level_irq+0x30/0x105 [<ffffffff81014e0e>] handle_irq+0x95/0xb7 [<ffffffff810141bd>] do_IRQ+0x6a/0xe0 [<ffffffff81012813>] ret_from_intr+0x0/0x16 irq event stamp: 195096 hardirqs last enabled at (195096): [<ffffffff814cd7f7>] _spin_unlock_irq+0x3a/0x5c hardirqs last disabled at (195095): [<ffffffff814cdbdd>] _spin_lock_irq+0x29/0x95 softirqs last enabled at (195088): [<ffffffff81068c92>] __do_softirq+0x1c1/0x1ef softirqs last disabled at (195093): [<ffffffff8101304c>] call_softirq+0x1c/0x30 other info that might help us debug this: 1 lock held by swapper/0: #0: (kernel/irq/spurious.c:21){+.-...}, at: [<ffffffff81070cf2>] run_timer_softirq+0x1a9/0x315 stack backtrace: Pid: 0, comm: swapper Not tainted 2.6.32-rc5 #1 Call Trace: <IRQ> [<ffffffff81093e94>] valid_state+0x187/0x1ae [<ffffffff81093fe4>] mark_lock+0x129/0x253 [<ffffffff810951d4>] __lock_acquire+0x370/0xd5d [<ffffffff81095cb4>] lock_acquire+0xf3/0x12d [<ffffffff814cdadd>] _spin_lock+0x40/0x89 [<ffffffff810c264e>] try_one_irq+0x32/0x138 [<ffffffff810c2795>] poll_all_shared_irqs+0x41/0x6d [<ffffffff810c27dd>] poll_spurious_irqs+0x1c/0x49 [<ffffffff81070d82>] run_timer_softirq+0x239/0x315 [<ffffffff81068bd3>] __do_softirq+0x102/0x1ef [<ffffffff8101304c>] call_softirq+0x1c/0x30 [<ffffffff81014b65>] do_softirq+0x59/0xca [<ffffffff810686ad>] irq_exit+0x58/0xae [<ffffffff81029b84>] smp_apic_timer_interrupt+0x94/0xba [<ffffffff81012a33>] apic_timer_interrupt+0x13/0x20 The reason is that try_one_irq() is called from hardirq context with interrupts disabled and from softirq context (poll_all_shared_irqs()) with interrupts enabled. Disable interrupts before calling it from poll_all_shared_irqs(). Reported-and-tested-by: Prarit Bhargava <prarit@redhat.com> Signed-off-by: Yong Zhang <yong.zhang0@gmail.com> LKML-Reference: <1257563773-4620-1-git-send-email-yong.zhang0@gmail.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2009-11-07 11:16:13 +08:00
local_irq_enable();
}
out:
atomic_dec(&irq_poll_active);
mod_timer(&poll_spurious_irq_timer,
jiffies + POLL_SPURIOUS_IRQ_INTERVAL);
}
/*
* If 99,900 of the previous 100,000 interrupts have not been handled
* then assume that the IRQ is stuck in some manner. Drop a diagnostic
* and try to turn the IRQ off.
*
* (The other 100-of-100,000 interrupts may have been a correctly
* functioning device sharing an IRQ with the failing one)
*/
static void
__report_bad_irq(unsigned int irq, struct irq_desc *desc,
irqreturn_t action_ret)
{
struct irqaction *action;
unsigned long flags;
if (action_ret != IRQ_HANDLED && action_ret != IRQ_NONE) {
printk(KERN_ERR "irq event %d: bogus return value %x\n",
irq, action_ret);
} else {
printk(KERN_ERR "irq %d: nobody cared (try booting with "
"the \"irqpoll\" option)\n", irq);
}
dump_stack();
printk(KERN_ERR "handlers:\n");
/*
* We need to take desc->lock here. note_interrupt() is called
* w/o desc->lock held, but IRQ_PROGRESS set. We might race
* with something else removing an action. It's ok to take
* desc->lock here. See synchronize_irq().
*/
raw_spin_lock_irqsave(&desc->lock, flags);
action = desc->action;
while (action) {
printk(KERN_ERR "[<%p>]", action->handler);
print_symbol(" (%s)",
(unsigned long)action->handler);
printk("\n");
action = action->next;
}
raw_spin_unlock_irqrestore(&desc->lock, flags);
}
static void
report_bad_irq(unsigned int irq, struct irq_desc *desc, irqreturn_t action_ret)
{
static int count = 100;
if (count > 0) {
count--;
__report_bad_irq(irq, desc, action_ret);
}
}
static inline int
try_misrouted_irq(unsigned int irq, struct irq_desc *desc,
irqreturn_t action_ret)
{
struct irqaction *action;
if (!irqfixup)
return 0;
/* We didn't actually handle the IRQ - see if it was misrouted? */
if (action_ret == IRQ_NONE)
return 1;
/*
* But for 'irqfixup == 2' we also do it for handled interrupts if
* they are marked as IRQF_IRQPOLL (or for irq zero, which is the
* traditional PC timer interrupt.. Legacy)
*/
if (irqfixup < 2)
return 0;
if (!irq)
return 1;
/*
* Since we don't get the descriptor lock, "action" can
* change under us. We don't really care, but we don't
* want to follow a NULL pointer. So tell the compiler to
* just load it once by using a barrier.
*/
action = desc->action;
barrier();
return action && (action->flags & IRQF_IRQPOLL);
}
void note_interrupt(unsigned int irq, 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 21:55:46 +08:00
irqreturn_t action_ret)
{
if (desc->status & IRQ_POLL_INPROGRESS)
return;
if (unlikely(action_ret != IRQ_HANDLED)) {
/*
* If we are seeing only the odd spurious IRQ caused by
* bus asynchronicity then don't eventually trigger an error,
* otherwise the counter becomes a doomsday timer for otherwise
* working systems
*/
if (time_after(jiffies, desc->last_unhandled + HZ/10))
desc->irqs_unhandled = 1;
else
desc->irqs_unhandled++;
desc->last_unhandled = jiffies;
if (unlikely(action_ret != IRQ_NONE))
report_bad_irq(irq, desc, action_ret);
}
if (unlikely(try_misrouted_irq(irq, desc, action_ret))) {
int ok = misrouted_irq(irq);
if (action_ret == IRQ_NONE)
desc->irqs_unhandled -= ok;
}
desc->irq_count++;
if (likely(desc->irq_count < 100000))
return;
desc->irq_count = 0;
if (unlikely(desc->irqs_unhandled > 99900)) {
/*
* The interrupt is stuck
*/
__report_bad_irq(irq, desc, action_ret);
/*
* Now kill the IRQ
*/
printk(KERN_EMERG "Disabling IRQ #%d\n", irq);
desc->status |= IRQ_SPURIOUS_DISABLED;
desc->depth++;
irq_disable(desc);
mod_timer(&poll_spurious_irq_timer,
jiffies + POLL_SPURIOUS_IRQ_INTERVAL);
}
desc->irqs_unhandled = 0;
}
int noirqdebug __read_mostly;
int noirqdebug_setup(char *str)
{
noirqdebug = 1;
printk(KERN_INFO "IRQ lockup detection disabled\n");
return 1;
}
__setup("noirqdebug", noirqdebug_setup);
module_param(noirqdebug, bool, 0644);
MODULE_PARM_DESC(noirqdebug, "Disable irq lockup detection when true");
static int __init irqfixup_setup(char *str)
{
irqfixup = 1;
printk(KERN_WARNING "Misrouted IRQ fixup support enabled.\n");
printk(KERN_WARNING "This may impact system performance.\n");
return 1;
}
__setup("irqfixup", irqfixup_setup);
module_param(irqfixup, int, 0644);
static int __init irqpoll_setup(char *str)
{
irqfixup = 2;
printk(KERN_WARNING "Misrouted IRQ fixup and polling support "
"enabled\n");
printk(KERN_WARNING "This may significantly impact system "
"performance\n");
return 1;
}
__setup("irqpoll", irqpoll_setup);