mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
synced 2026-08-09 06:14:34 +02:00
Feng TangandAndrew Morton
c60ffec33d
When debugging RCU stall cases, usually all CPUs will respond to the NMI and print out the backtrace. But in some nasty or hardware related cases, some CPUs may fail to respond in 10 seconds, and very likely this is sign of severe issues. Paul McKenney has implemented the NMI backtrace stall check for x86, and for other architectures, it should be also helpful to at least print out those CPUs which failed to repond to the NMI, so that users can get an early heads-up for possible CPU hard stall. [feng.tang@linux.alibaba.com: avoid hard-coding "10" in two places and in a comment] Link: https://lore.kernel.org/ag-1ciG0FSomBf7q@U-2FWC9VHC-2323.local [akpm@linux-foundation.org: use __stringify()] Link: https://lore.kernel.org/20260521030336.92172-1-feng.tang@linux.alibaba.com Signed-off-by: Feng Tang <feng.tang@linux.alibaba.com> Reviewed-by: Petr Mladek <pmladek@suse.com> Cc: "Paul E . McKenney" <paulmck@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
133 lines
3.7 KiB
C
133 lines
3.7 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* NMI backtrace support
|
|
*
|
|
* Gratuitously copied from arch/x86/kernel/apic/hw_nmi.c by Russell King,
|
|
* with the following header:
|
|
*
|
|
* HW NMI watchdog support
|
|
*
|
|
* started by Don Zickus, Copyright (C) 2010 Red Hat, Inc.
|
|
*
|
|
* Arch specific calls to support NMI watchdog
|
|
*
|
|
* Bits copied from original nmi.c file
|
|
*/
|
|
#include <linux/cpumask.h>
|
|
#include <linux/delay.h>
|
|
#include <linux/kprobes.h>
|
|
#include <linux/stringify.h>
|
|
#include <linux/nmi.h>
|
|
#include <linux/cpu.h>
|
|
#include <linux/sched/debug.h>
|
|
|
|
#ifdef arch_trigger_cpumask_backtrace
|
|
/* For reliability, we're prepared to waste bits here. */
|
|
static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly;
|
|
|
|
/* "in progress" flag of arch_trigger_cpumask_backtrace */
|
|
static unsigned long backtrace_flag;
|
|
|
|
#define NMI_BT_TIMEOUT_SEC 10
|
|
|
|
/*
|
|
* When raise() is called it will be passed a pointer to the
|
|
* backtrace_mask. Architectures that call nmi_cpu_backtrace()
|
|
* directly from their raise() functions may rely on the mask
|
|
* they are passed being updated as a side effect of this call.
|
|
*/
|
|
void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
|
|
int exclude_cpu,
|
|
void (*raise)(cpumask_t *mask))
|
|
{
|
|
int i, this_cpu = get_cpu();
|
|
|
|
if (test_and_set_bit(0, &backtrace_flag)) {
|
|
/*
|
|
* If there is already a trigger_all_cpu_backtrace() in progress
|
|
* (backtrace_flag == 1), don't output double cpu dump infos.
|
|
*/
|
|
put_cpu();
|
|
return;
|
|
}
|
|
|
|
cpumask_copy(to_cpumask(backtrace_mask), mask);
|
|
if (exclude_cpu != -1)
|
|
cpumask_clear_cpu(exclude_cpu, to_cpumask(backtrace_mask));
|
|
|
|
/*
|
|
* Don't try to send an NMI to this cpu; it may work on some
|
|
* architectures, but on others it may not, and we'll get
|
|
* information at least as useful just by doing a dump_stack() here.
|
|
* Note that nmi_cpu_backtrace(NULL) will clear the cpu bit.
|
|
*/
|
|
if (cpumask_test_cpu(this_cpu, to_cpumask(backtrace_mask)))
|
|
nmi_cpu_backtrace(NULL);
|
|
|
|
if (!cpumask_empty(to_cpumask(backtrace_mask))) {
|
|
pr_info("Sending NMI from CPU %d to CPUs %*pbl:\n",
|
|
this_cpu, nr_cpumask_bits, to_cpumask(backtrace_mask));
|
|
nmi_backtrace_stall_snap(to_cpumask(backtrace_mask));
|
|
raise(to_cpumask(backtrace_mask));
|
|
}
|
|
|
|
/* Wait for up to NMI_BT_TIMEOUT_SEC seconds for all CPUs to do the backtrace */
|
|
for (i = 0; i < NMI_BT_TIMEOUT_SEC * 1000; i++) {
|
|
if (cpumask_empty(to_cpumask(backtrace_mask)))
|
|
break;
|
|
mdelay(1);
|
|
touch_softlockup_watchdog();
|
|
}
|
|
|
|
if (!cpumask_empty(to_cpumask(backtrace_mask))) {
|
|
pr_warn("After " __stringify(NMI_BT_TIMEOUT_SEC) " seconds, these CPUS still haven't responded to the NMI: %*pbl\n",
|
|
cpumask_pr_args(to_cpumask(backtrace_mask)));
|
|
|
|
nmi_backtrace_stall_check(to_cpumask(backtrace_mask));
|
|
}
|
|
|
|
/*
|
|
* Force flush any remote buffers that might be stuck in IRQ context
|
|
* and therefore could not run their irq_work.
|
|
*/
|
|
printk_trigger_flush();
|
|
|
|
clear_bit_unlock(0, &backtrace_flag);
|
|
put_cpu();
|
|
}
|
|
|
|
// Dump stacks even for idle CPUs.
|
|
static bool backtrace_idle;
|
|
module_param(backtrace_idle, bool, 0644);
|
|
|
|
bool nmi_cpu_backtrace(struct pt_regs *regs)
|
|
{
|
|
int cpu = smp_processor_id();
|
|
unsigned long flags;
|
|
|
|
if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) {
|
|
/*
|
|
* Allow nested NMI backtraces while serializing
|
|
* against other CPUs.
|
|
*/
|
|
printk_cpu_sync_get_irqsave(flags);
|
|
if (!READ_ONCE(backtrace_idle) && regs && cpu_in_idle(instruction_pointer(regs))) {
|
|
pr_warn("NMI backtrace for cpu %d skipped: idling at %pS\n",
|
|
cpu, (void *)instruction_pointer(regs));
|
|
} else {
|
|
pr_warn("NMI backtrace for cpu %d\n", cpu);
|
|
if (regs)
|
|
show_regs(regs);
|
|
else
|
|
dump_stack();
|
|
}
|
|
printk_cpu_sync_put_irqrestore(flags);
|
|
cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask));
|
|
return true;
|
|
}
|
|
|
|
return false;
|
|
}
|
|
NOKPROBE_SYMBOL(nmi_cpu_backtrace);
|
|
#endif
|