* [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU
@ 2023-08-04 14:00 Douglas Anderson
2023-08-04 14:00 ` [PATCH v4 2/2] watchdog/hardlockup: Avoid large stack frames in watchdog_hardlockup_check() Douglas Anderson
2023-08-07 7:35 ` [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU Chen-Yu Tsai
0 siblings, 2 replies; 4+ messages in thread
From: Douglas Anderson @ 2023-08-04 14:00 UTC (permalink / raw)
To: Andrew Morton
Cc: Petr Mladek, Michal Hocko, Douglas Anderson, Arnd Bergmann,
Borislav Petkov, Christophe Leroy, Darrick J. Wong, Dave Hansen,
David S. Miller, Frederic Weisbecker, Gaosheng Cui,
Greg Kroah-Hartman, Guilherme G. Piccoli, H. Peter Anvin,
Huacai Chen, Ingo Molnar, Jason A. Donenfeld, Jianmin Lv,
Jinyang He, Josh Poimboeuf, Kees Cook, Lecopzer Chen,
Marc Zyngier, Mark Rutland, Michael Ellerman, Nicholas Piggin,
Paul E. McKenney, Peter Zijlstra, Philippe Mathieu-Daudé,
Pingfan Liu, Qing Zhang, Russell King (Oracle), Russell King,
Stephen Rothwell, Thomas Bogendoerfer, Thomas Gleixner, Tom Rix,
Ulf Hansson, Valentin Schneider, WANG Xuerui, linux-arm-kernel,
linux-kernel, linux-mips, linuxppc-dev, loongarch, sparclinux,
x86
The APIs that allow backtracing across CPUs have always had a way to
exclude the current CPU. This convenience means callers didn't need to
find a place to allocate a CPU mask just to handle the common case.
Let's extend the API to take a CPU ID to exclude instead of just a
boolean. This isn't any more complex for the API to handle and allows
the hardlockup detector to exclude a different CPU (the one it already
did a trace for) without needing to find space for a CPU mask.
Arguably, this new API also encourages safer behavior. Specifically if
the caller wants to avoid tracing the current CPU (maybe because they
already traced the current CPU) this makes it more obvious to the
caller that they need to make sure that the current CPU ID can't
change.
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
---
Changes in v4:
- Renamed trigger_allbutself_cpu_backtrace() for when trigger is unsupported.
Changes in v3:
- ("nmi_backtrace: Allow excluding an arbitrary CPU") new for v3.
arch/arm/include/asm/irq.h | 2 +-
arch/arm/kernel/smp.c | 4 ++--
arch/loongarch/include/asm/irq.h | 2 +-
arch/loongarch/kernel/process.c | 4 ++--
arch/mips/include/asm/irq.h | 2 +-
arch/mips/kernel/process.c | 4 ++--
arch/powerpc/include/asm/irq.h | 2 +-
arch/powerpc/kernel/stacktrace.c | 4 ++--
arch/powerpc/kernel/watchdog.c | 4 ++--
arch/sparc/include/asm/irq_64.h | 2 +-
arch/sparc/kernel/process_64.c | 6 +++---
arch/x86/include/asm/irq.h | 2 +-
arch/x86/kernel/apic/hw_nmi.c | 4 ++--
include/linux/nmi.h | 14 +++++++-------
kernel/watchdog.c | 2 +-
lib/nmi_backtrace.c | 6 +++---
16 files changed, 32 insertions(+), 32 deletions(-)
diff --git a/arch/arm/include/asm/irq.h b/arch/arm/include/asm/irq.h
index 18605f1b3580..26c1d2ced4ce 100644
--- a/arch/arm/include/asm/irq.h
+++ b/arch/arm/include/asm/irq.h
@@ -32,7 +32,7 @@ void handle_IRQ(unsigned int, struct pt_regs *);
#include <linux/cpumask.h>
extern void arch_trigger_cpumask_backtrace(const cpumask_t *mask,
- bool exclude_self);
+ int exclude_cpu);
#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 6756203e45f3..3431c0553f45 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -846,7 +846,7 @@ static void raise_nmi(cpumask_t *mask)
__ipi_send_mask(ipi_desc[IPI_CPU_BACKTRACE], mask);
}
-void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
{
- nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_nmi);
+ nmi_trigger_cpumask_backtrace(mask, exclude_cpu, raise_nmi);
}
diff --git a/arch/loongarch/include/asm/irq.h b/arch/loongarch/include/asm/irq.h
index a115e8999c69..218b4da0ea90 100644
--- a/arch/loongarch/include/asm/irq.h
+++ b/arch/loongarch/include/asm/irq.h
@@ -40,7 +40,7 @@ void spurious_interrupt(void);
#define NR_IRQS_LEGACY 16
#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
-void arch_trigger_cpumask_backtrace(const struct cpumask *mask, bool exclude_self);
+void arch_trigger_cpumask_backtrace(const struct cpumask *mask, int exclude_cpu);
#define MAX_IO_PICS 2
#define NR_IRQS (64 + (256 * MAX_IO_PICS))
diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c
index 2e04eb07abb6..778e8d09953e 100644
--- a/arch/loongarch/kernel/process.c
+++ b/arch/loongarch/kernel/process.c
@@ -345,9 +345,9 @@ static void raise_backtrace(cpumask_t *mask)
}
}
-void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
{
- nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_backtrace);
+ nmi_trigger_cpumask_backtrace(mask, exclude_cpu, raise_backtrace);
}
#ifdef CONFIG_64BIT
diff --git a/arch/mips/include/asm/irq.h b/arch/mips/include/asm/irq.h
index 75abfa834ab7..3a848e7e69f7 100644
--- a/arch/mips/include/asm/irq.h
+++ b/arch/mips/include/asm/irq.h
@@ -77,7 +77,7 @@ extern int cp0_fdc_irq;
extern int get_c0_fdc_int(void);
void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
- bool exclude_self);
+ int exclude_cpu);
#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif /* _ASM_IRQ_H */
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index a3225912c862..5387ed0a5186 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -750,9 +750,9 @@ static void raise_backtrace(cpumask_t *mask)
}
}
-void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
{
- nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_backtrace);
+ nmi_trigger_cpumask_backtrace(mask, exclude_cpu, raise_backtrace);
}
int mips_get_process_fp_mode(struct task_struct *task)
diff --git a/arch/powerpc/include/asm/irq.h b/arch/powerpc/include/asm/irq.h
index f257cacb49a9..ba1a5974e714 100644
--- a/arch/powerpc/include/asm/irq.h
+++ b/arch/powerpc/include/asm/irq.h
@@ -55,7 +55,7 @@ int irq_choose_cpu(const struct cpumask *mask);
#if defined(CONFIG_PPC_BOOK3S_64) && defined(CONFIG_NMI_IPI)
extern void arch_trigger_cpumask_backtrace(const cpumask_t *mask,
- bool exclude_self);
+ int exclude_cpu);
#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif
diff --git a/arch/powerpc/kernel/stacktrace.c b/arch/powerpc/kernel/stacktrace.c
index 5de8597eaab8..b15f15dcacb5 100644
--- a/arch/powerpc/kernel/stacktrace.c
+++ b/arch/powerpc/kernel/stacktrace.c
@@ -221,8 +221,8 @@ static void raise_backtrace_ipi(cpumask_t *mask)
}
}
-void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
{
- nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_backtrace_ipi);
+ nmi_trigger_cpumask_backtrace(mask, exclude_cpu, raise_backtrace_ipi);
}
#endif /* defined(CONFIG_PPC_BOOK3S_64) && defined(CONFIG_NMI_IPI) */
diff --git a/arch/powerpc/kernel/watchdog.c b/arch/powerpc/kernel/watchdog.c
index edb2dd1f53eb..8c464a5d8246 100644
--- a/arch/powerpc/kernel/watchdog.c
+++ b/arch/powerpc/kernel/watchdog.c
@@ -245,7 +245,7 @@ static void watchdog_smp_panic(int cpu)
__cpumask_clear_cpu(c, &wd_smp_cpus_ipi);
}
} else {
- trigger_allbutself_cpu_backtrace();
+ trigger_allbutcpu_cpu_backtrace(cpu);
cpumask_clear(&wd_smp_cpus_ipi);
}
@@ -416,7 +416,7 @@ DEFINE_INTERRUPT_HANDLER_NMI(soft_nmi_interrupt)
xchg(&__wd_nmi_output, 1); // see wd_lockup_ipi
if (sysctl_hardlockup_all_cpu_backtrace)
- trigger_allbutself_cpu_backtrace();
+ trigger_allbutcpu_cpu_backtrace(cpu);
if (hardlockup_panic)
nmi_panic(regs, "Hard LOCKUP");
diff --git a/arch/sparc/include/asm/irq_64.h b/arch/sparc/include/asm/irq_64.h
index b436029f1ced..8c4c0c87f998 100644
--- a/arch/sparc/include/asm/irq_64.h
+++ b/arch/sparc/include/asm/irq_64.h
@@ -87,7 +87,7 @@ static inline unsigned long get_softint(void)
}
void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
- bool exclude_self);
+ int exclude_cpu);
#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
extern void *hardirq_stack[NR_CPUS];
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index b51d8fb0ecdc..1ea3f37fa985 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -236,7 +236,7 @@ static void __global_reg_poll(struct global_reg_snapshot *gp)
}
}
-void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
{
struct thread_info *tp = current_thread_info();
struct pt_regs *regs = get_irq_regs();
@@ -252,7 +252,7 @@ void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot));
- if (cpumask_test_cpu(this_cpu, mask) && !exclude_self)
+ if (cpumask_test_cpu(this_cpu, mask) && this_cpu != exclude_cpu)
__global_reg_self(tp, regs, this_cpu);
smp_fetch_global_regs();
@@ -260,7 +260,7 @@ void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
for_each_cpu(cpu, mask) {
struct global_reg_snapshot *gp;
- if (exclude_self && cpu == this_cpu)
+ if (cpu == exclude_cpu)
continue;
gp = &global_cpu_snapshot[cpu].reg;
diff --git a/arch/x86/include/asm/irq.h b/arch/x86/include/asm/irq.h
index 29e083b92813..836c170d3087 100644
--- a/arch/x86/include/asm/irq.h
+++ b/arch/x86/include/asm/irq.h
@@ -42,7 +42,7 @@ extern void init_ISA_irqs(void);
#ifdef CONFIG_X86_LOCAL_APIC
void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
- bool exclude_self);
+ int exclude_cpu);
#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif
diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c
index 34a992e275ef..d6e01f924299 100644
--- a/arch/x86/kernel/apic/hw_nmi.c
+++ b/arch/x86/kernel/apic/hw_nmi.c
@@ -34,9 +34,9 @@ static void nmi_raise_cpu_backtrace(cpumask_t *mask)
apic->send_IPI_mask(mask, NMI_VECTOR);
}
-void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
{
- nmi_trigger_cpumask_backtrace(mask, exclude_self,
+ nmi_trigger_cpumask_backtrace(mask, exclude_cpu,
nmi_raise_cpu_backtrace);
}
diff --git a/include/linux/nmi.h b/include/linux/nmi.h
index e3e6a64b98e0..7cf7801856a1 100644
--- a/include/linux/nmi.h
+++ b/include/linux/nmi.h
@@ -157,31 +157,31 @@ static inline void touch_nmi_watchdog(void)
#ifdef arch_trigger_cpumask_backtrace
static inline bool trigger_all_cpu_backtrace(void)
{
- arch_trigger_cpumask_backtrace(cpu_online_mask, false);
+ arch_trigger_cpumask_backtrace(cpu_online_mask, -1);
return true;
}
-static inline bool trigger_allbutself_cpu_backtrace(void)
+static inline bool trigger_allbutcpu_cpu_backtrace(int exclude_cpu)
{
- arch_trigger_cpumask_backtrace(cpu_online_mask, true);
+ arch_trigger_cpumask_backtrace(cpu_online_mask, exclude_cpu);
return true;
}
static inline bool trigger_cpumask_backtrace(struct cpumask *mask)
{
- arch_trigger_cpumask_backtrace(mask, false);
+ arch_trigger_cpumask_backtrace(mask, -1);
return true;
}
static inline bool trigger_single_cpu_backtrace(int cpu)
{
- arch_trigger_cpumask_backtrace(cpumask_of(cpu), false);
+ arch_trigger_cpumask_backtrace(cpumask_of(cpu), -1);
return true;
}
/* generic implementation */
void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
- bool exclude_self,
+ int exclude_cpu,
void (*raise)(cpumask_t *mask));
bool nmi_cpu_backtrace(struct pt_regs *regs);
@@ -190,7 +190,7 @@ static inline bool trigger_all_cpu_backtrace(void)
{
return false;
}
-static inline bool trigger_allbutself_cpu_backtrace(void)
+static inline bool trigger_allbutcpu_cpu_backtrace(void)
{
return false;
}
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index be38276a365f..085d7a78f62f 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -523,7 +523,7 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer)
dump_stack();
if (softlockup_all_cpu_backtrace) {
- trigger_allbutself_cpu_backtrace();
+ trigger_allbutcpu_cpu_backtrace(smp_processor_id());
clear_bit_unlock(0, &soft_lockup_nmi_warn);
}
diff --git a/lib/nmi_backtrace.c b/lib/nmi_backtrace.c
index 5274bbb026d7..33c154264bfe 100644
--- a/lib/nmi_backtrace.c
+++ b/lib/nmi_backtrace.c
@@ -34,7 +34,7 @@ static unsigned long backtrace_flag;
* they are passed being updated as a side effect of this call.
*/
void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
- bool exclude_self,
+ int exclude_cpu,
void (*raise)(cpumask_t *mask))
{
int i, this_cpu = get_cpu();
@@ -49,8 +49,8 @@ void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
}
cpumask_copy(to_cpumask(backtrace_mask), mask);
- if (exclude_self)
- cpumask_clear_cpu(this_cpu, to_cpumask(backtrace_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
--
2.41.0.585.gd2178a4bd4-goog
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH v4 2/2] watchdog/hardlockup: Avoid large stack frames in watchdog_hardlockup_check()
2023-08-04 14:00 [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU Douglas Anderson
@ 2023-08-04 14:00 ` Douglas Anderson
2023-08-07 13:39 ` Petr Mladek
2023-08-07 7:35 ` [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU Chen-Yu Tsai
1 sibling, 1 reply; 4+ messages in thread
From: Douglas Anderson @ 2023-08-04 14:00 UTC (permalink / raw)
To: Andrew Morton
Cc: Petr Mladek, Michal Hocko, Douglas Anderson, kernel test robot,
Lecopzer Chen, Pingfan Liu, linux-kernel
After commit 77c12fc95980 ("watchdog/hardlockup: add a "cpu" param to
watchdog_hardlockup_check()") we started storing a `struct cpumask` on
the stack in watchdog_hardlockup_check(). On systems with
CONFIG_NR_CPUS set to 8192 this takes up 1K on the stack. That
triggers warnings with `CONFIG_FRAME_WARN` set to 1024.
We'll use the new trigger_allbutcpu_cpu_backtrace() to avoid needing
to use a CPU mask at all.
Fixes: 77c12fc95980 ("watchdog/hardlockup: add a "cpu" param to watchdog_hardlockup_check()")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202307310955.pLZDhpnl-lkp@intel.com
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
---
(no changes since v3)
Changes in v3:
- Rebase on change introducing trigger_allbutcpu_cpu_backtrace().
Changes in v2:
- Allocate space when userspace requests all cpus be backtraced.
kernel/watchdog.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index 085d7a78f62f..d145305d95fe 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -151,9 +151,6 @@ void watchdog_hardlockup_check(unsigned int cpu, struct pt_regs *regs)
*/
if (is_hardlockup(cpu)) {
unsigned int this_cpu = smp_processor_id();
- struct cpumask backtrace_mask;
-
- cpumask_copy(&backtrace_mask, cpu_online_mask);
/* Only print hardlockups once. */
if (per_cpu(watchdog_hardlockup_warned, cpu))
@@ -167,10 +164,8 @@ void watchdog_hardlockup_check(unsigned int cpu, struct pt_regs *regs)
show_regs(regs);
else
dump_stack();
- cpumask_clear_cpu(cpu, &backtrace_mask);
} else {
- if (trigger_single_cpu_backtrace(cpu))
- cpumask_clear_cpu(cpu, &backtrace_mask);
+ trigger_single_cpu_backtrace(cpu);
}
/*
@@ -179,7 +174,7 @@ void watchdog_hardlockup_check(unsigned int cpu, struct pt_regs *regs)
*/
if (sysctl_hardlockup_all_cpu_backtrace &&
!test_and_set_bit(0, &watchdog_hardlockup_all_cpu_dumped))
- trigger_cpumask_backtrace(&backtrace_mask);
+ trigger_allbutcpu_cpu_backtrace(cpu);
if (hardlockup_panic)
nmi_panic(regs, "Hard LOCKUP");
--
2.41.0.585.gd2178a4bd4-goog
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU
2023-08-04 14:00 [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU Douglas Anderson
2023-08-04 14:00 ` [PATCH v4 2/2] watchdog/hardlockup: Avoid large stack frames in watchdog_hardlockup_check() Douglas Anderson
@ 2023-08-07 7:35 ` Chen-Yu Tsai
1 sibling, 0 replies; 4+ messages in thread
From: Chen-Yu Tsai @ 2023-08-07 7:35 UTC (permalink / raw)
To: Douglas Anderson
Cc: Andrew Morton, Petr Mladek, Michal Hocko, Arnd Bergmann,
Borislav Petkov, Christophe Leroy, Darrick J. Wong, Dave Hansen,
David S. Miller, Frederic Weisbecker, Gaosheng Cui,
Greg Kroah-Hartman, Guilherme G. Piccoli, H. Peter Anvin,
Huacai Chen, Ingo Molnar, Jason A. Donenfeld, Jianmin Lv,
Jinyang He, Josh Poimboeuf, Kees Cook, Lecopzer Chen,
Marc Zyngier, Mark Rutland, Michael Ellerman, Nicholas Piggin,
Paul E. McKenney, Peter Zijlstra, Philippe Mathieu-Daudé,
Pingfan Liu, Qing Zhang, Russell King (Oracle), Russell King,
Stephen Rothwell, Thomas Bogendoerfer, Thomas Gleixner, Tom Rix,
Ulf Hansson, Valentin Schneider, WANG Xuerui, linux-arm-kernel,
linux-kernel, linux-mips, linuxppc-dev, loongarch, sparclinux,
x86
On Fri, Aug 4, 2023 at 10:01 PM Douglas Anderson <dianders@chromium.org> wrote:
>
> The APIs that allow backtracing across CPUs have always had a way to
> exclude the current CPU. This convenience means callers didn't need to
> find a place to allocate a CPU mask just to handle the common case.
>
> Let's extend the API to take a CPU ID to exclude instead of just a
> boolean. This isn't any more complex for the API to handle and allows
> the hardlockup detector to exclude a different CPU (the one it already
> did a trace for) without needing to find space for a CPU mask.
>
> Arguably, this new API also encourages safer behavior. Specifically if
> the caller wants to avoid tracing the current CPU (maybe because they
> already traced the current CPU) this makes it more obvious to the
> caller that they need to make sure that the current CPU ID can't
> change.
>
> Acked-by: Michal Hocko <mhocko@suse.com>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---
>
> Changes in v4:
> - Renamed trigger_allbutself_cpu_backtrace() for when trigger is unsupported.
>
> Changes in v3:
> - ("nmi_backtrace: Allow excluding an arbitrary CPU") new for v3.
>
> arch/arm/include/asm/irq.h | 2 +-
> arch/arm/kernel/smp.c | 4 ++--
> arch/loongarch/include/asm/irq.h | 2 +-
> arch/loongarch/kernel/process.c | 4 ++--
> arch/mips/include/asm/irq.h | 2 +-
> arch/mips/kernel/process.c | 4 ++--
> arch/powerpc/include/asm/irq.h | 2 +-
> arch/powerpc/kernel/stacktrace.c | 4 ++--
> arch/powerpc/kernel/watchdog.c | 4 ++--
> arch/sparc/include/asm/irq_64.h | 2 +-
> arch/sparc/kernel/process_64.c | 6 +++---
> arch/x86/include/asm/irq.h | 2 +-
> arch/x86/kernel/apic/hw_nmi.c | 4 ++--
> include/linux/nmi.h | 14 +++++++-------
> kernel/watchdog.c | 2 +-
> lib/nmi_backtrace.c | 6 +++---
> 16 files changed, 32 insertions(+), 32 deletions(-)
>
[...]
> diff --git a/include/linux/nmi.h b/include/linux/nmi.h
> index e3e6a64b98e0..7cf7801856a1 100644
> --- a/include/linux/nmi.h
> +++ b/include/linux/nmi.h
> @@ -157,31 +157,31 @@ static inline void touch_nmi_watchdog(void)
> #ifdef arch_trigger_cpumask_backtrace
> static inline bool trigger_all_cpu_backtrace(void)
> {
> - arch_trigger_cpumask_backtrace(cpu_online_mask, false);
> + arch_trigger_cpumask_backtrace(cpu_online_mask, -1);
> return true;
> }
>
> -static inline bool trigger_allbutself_cpu_backtrace(void)
> +static inline bool trigger_allbutcpu_cpu_backtrace(int exclude_cpu)
> {
> - arch_trigger_cpumask_backtrace(cpu_online_mask, true);
> + arch_trigger_cpumask_backtrace(cpu_online_mask, exclude_cpu);
> return true;
> }
>
> static inline bool trigger_cpumask_backtrace(struct cpumask *mask)
> {
> - arch_trigger_cpumask_backtrace(mask, false);
> + arch_trigger_cpumask_backtrace(mask, -1);
> return true;
> }
>
> static inline bool trigger_single_cpu_backtrace(int cpu)
> {
> - arch_trigger_cpumask_backtrace(cpumask_of(cpu), false);
> + arch_trigger_cpumask_backtrace(cpumask_of(cpu), -1);
> return true;
> }
>
> /* generic implementation */
> void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
> - bool exclude_self,
> + int exclude_cpu,
> void (*raise)(cpumask_t *mask));
> bool nmi_cpu_backtrace(struct pt_regs *regs);
>
> @@ -190,7 +190,7 @@ static inline bool trigger_all_cpu_backtrace(void)
> {
> return false;
> }
> -static inline bool trigger_allbutself_cpu_backtrace(void)
> +static inline bool trigger_allbutcpu_cpu_backtrace(void)
^
The parameter here is still wrong. It should be "int exclude_cpu".
This patch in Andrew's queue is causing build errors on next-20230807 on arm64:
kernel/watchdog.c: In function ‘watchdog_timer_fn’:
kernel/watchdog.c:521:25: error: too many arguments to function
‘trigger_allbutcpu_cpu_backtrace’
521 |
trigger_allbutcpu_cpu_backtrace(smp_processor_id());
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from kernel/watchdog.c:17:
./include/linux/nmi.h:193:20: note: declared here
193 | static inline bool trigger_allbutcpu_cpu_backtrace(void)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
make[3]: *** [scripts/Makefile.build:243: kernel/watchdog.o] Error 1
ChenYu
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v4 2/2] watchdog/hardlockup: Avoid large stack frames in watchdog_hardlockup_check()
2023-08-04 14:00 ` [PATCH v4 2/2] watchdog/hardlockup: Avoid large stack frames in watchdog_hardlockup_check() Douglas Anderson
@ 2023-08-07 13:39 ` Petr Mladek
0 siblings, 0 replies; 4+ messages in thread
From: Petr Mladek @ 2023-08-07 13:39 UTC (permalink / raw)
To: Douglas Anderson
Cc: Andrew Morton, Michal Hocko, kernel test robot, Lecopzer Chen,
Pingfan Liu, linux-kernel
On Fri 2023-08-04 07:00:43, Douglas Anderson wrote:
> After commit 77c12fc95980 ("watchdog/hardlockup: add a "cpu" param to
> watchdog_hardlockup_check()") we started storing a `struct cpumask` on
> the stack in watchdog_hardlockup_check(). On systems with
> CONFIG_NR_CPUS set to 8192 this takes up 1K on the stack. That
> triggers warnings with `CONFIG_FRAME_WARN` set to 1024.
>
> We'll use the new trigger_allbutcpu_cpu_backtrace() to avoid needing
> to use a CPU mask at all.
>
> Fixes: 77c12fc95980 ("watchdog/hardlockup: add a "cpu" param to watchdog_hardlockup_check()")
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/r/202307310955.pLZDhpnl-lkp@intel.com
> Acked-by: Michal Hocko <mhocko@suse.com>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Best Regards,
Petr
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2023-08-07 13:40 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-08-04 14:00 [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU Douglas Anderson
2023-08-04 14:00 ` [PATCH v4 2/2] watchdog/hardlockup: Avoid large stack frames in watchdog_hardlockup_check() Douglas Anderson
2023-08-07 13:39 ` Petr Mladek
2023-08-07 7:35 ` [PATCH v4 1/2] nmi_backtrace: Allow excluding an arbitrary CPU Chen-Yu Tsai
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox