The patch titled
Subject: nmi_backtrace: add more trigger_*_cpu_backtrace() methods
has been added to the -mm tree. Its filename is
nmi_backtrace-add-more-trigger__cpu_backtrace-methods.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/nmi_backtrace-add-more-trigger__cpu_backtrace-methods.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/nmi_backtrace-add-more-trigger__cpu_backtrace-methods.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Chris Metcalf <cmetcalf@xxxxxxxxxxxx>
Subject: nmi_backtrace: add more trigger_*_cpu_backtrace() methods
Patch series "improvements to the nmi_backtrace code" v9.
This patch series modifies the trigger_xxx_backtrace() NMI-based remote
backtracing code to make it more flexible, and makes a few small
improvements along the way.
The motivation comes from the task isolation code, where there are
scenarios where we want to be able to diagnose a case where some cpu is
about to interrupt a task-isolated cpu. It can be helpful to see both
where the interrupting cpu is, and also an approximation of where the cpu
that is being interrupted is. The nmi_backtrace framework allows us to
discover the stack of the interrupted cpu.
I've tested that the change works as desired on tile, and build-tested
x86, arm, mips, and sparc64. For x86 I confirmed that the generic
cpuidle stuff as well as the architecture-specific routines are in the
new cpuidle section. For arm, mips, and sparc I just build-tested it
and made sure the generic cpuidle routines were in the new cpuidle
section, but I didn't attempt to figure out which the
platform-specific idle routines might be. That might be more usefully
done by someone with platform experience in follow-up patches.
This patch (of 4):
Currently you can only request a backtrace of either all cpus, or all cpus
but yourself. It can also be helpful to request a remote backtrace of a
single cpu, and since we want that, the logical extension is to support a
cpumask as the underlying primitive.
This change modifies the existing lib/nmi_backtrace.c code to take a
cpumask as its basic primitive, and modifies the linux/nmi.h code to use
the new "cpumask" method instead.
The existing clients of nmi_backtrace (arm and x86) are converted
to using the new cpumask approach in this change.
The other users of the backtracing API (sparc64 and mips) are converted to
use the cpumask approach rather than the all/allbutself approach. The
mips code ignored the "include_self" boolean but with this change it will
now also dump a local backtrace if requested.
Link: http://lkml.kernel.org/r/1472487169-14923-2-git-send-email-cmetcalf@xxxxxxxxxxxx
Signed-off-by: Chris Metcalf <cmetcalf@xxxxxxxxxxxx>
Tested-by: Daniel Thompson <daniel.thompson@xxxxxxxxxx> [arm]
Reviewed-by: Aaron Tomlin <atomlin@xxxxxxxxxx>
Reviewed-by: Petr Mladek <pmladek@xxxxxxxx>
Cc: Rafael J. Wysocki" <rjw@xxxxxxxxxxxxx>
Cc: Russell King <linux@xxxxxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxx>
Cc: Ralf Baechle <ralf@xxxxxxxxxxxxxx>
Cc: David Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
arch/arm/include/asm/irq.h | 5 ++--
arch/arm/kernel/smp.c | 4 +--
arch/mips/include/asm/irq.h | 5 ++--
arch/mips/kernel/process.c | 11 ++++++++--
arch/sparc/include/asm/irq_64.h | 5 ++--
arch/sparc/kernel/process_64.c | 10 ++++-----
arch/x86/include/asm/irq.h | 5 ++--
arch/x86/kernel/apic/hw_nmi.c | 18 ++++++++---------
include/linux/nmi.h | 31 +++++++++++++++++++++++++-----
lib/nmi_backtrace.c | 17 ++++++++--------
10 files changed, 72 insertions(+), 39 deletions(-)
diff -puN arch/arm/include/asm/irq.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/arm/include/asm/irq.h
--- a/arch/arm/include/asm/irq.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/arm/include/asm/irq.h
@@ -36,8 +36,9 @@ extern void set_handle_irq(void (*handle
#endif
#ifdef CONFIG_SMP
-extern void arch_trigger_all_cpu_backtrace(bool);
-#define arch_trigger_all_cpu_backtrace(x) arch_trigger_all_cpu_backtrace(x)
+extern void arch_trigger_cpumask_backtrace(const cpumask_t *mask,
+ bool exclude_self);
+#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif
static inline int nr_legacy_irqs(void)
diff -puN arch/arm/kernel/smp.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/arm/kernel/smp.c
--- a/arch/arm/kernel/smp.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/arm/kernel/smp.c
@@ -760,7 +760,7 @@ static void raise_nmi(cpumask_t *mask)
smp_cross_call(mask, IPI_CPU_BACKTRACE);
}
-void arch_trigger_all_cpu_backtrace(bool include_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
{
- nmi_trigger_all_cpu_backtrace(include_self, raise_nmi);
+ nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_nmi);
}
diff -puN arch/mips/include/asm/irq.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/mips/include/asm/irq.h
--- a/arch/mips/include/asm/irq.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/mips/include/asm/irq.h
@@ -51,7 +51,8 @@ extern int cp0_fdc_irq;
extern int get_c0_fdc_int(void);
-void arch_trigger_all_cpu_backtrace(bool);
-#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace
+void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
+ bool exclude_self);
+#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif /* _ASM_IRQ_H */
diff -puN arch/mips/kernel/process.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/mips/kernel/process.c
--- a/arch/mips/kernel/process.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/mips/kernel/process.c
@@ -569,9 +569,16 @@ static void arch_dump_stack(void *info)
dump_stack();
}
-void arch_trigger_all_cpu_backtrace(bool include_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
{
- smp_call_function(arch_dump_stack, NULL, 1);
+ long this_cpu = get_cpu();
+
+ if (cpumask_test_cpu(this_cpu, mask) && !exclude_self)
+ dump_stack();
+
+ smp_call_function_many(mask, arch_dump_stack, NULL, 1);
+
+ put_cpu();
}
int mips_get_process_fp_mode(struct task_struct *task)
diff -puN arch/sparc/include/asm/irq_64.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/sparc/include/asm/irq_64.h
--- a/arch/sparc/include/asm/irq_64.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/sparc/include/asm/irq_64.h
@@ -86,8 +86,9 @@ static inline unsigned long get_softint(
return retval;
}
-void arch_trigger_all_cpu_backtrace(bool);
-#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace
+void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
+ bool exclude_self);
+#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
extern void *hardirq_stack[NR_CPUS];
extern void *softirq_stack[NR_CPUS];
diff -puN arch/sparc/kernel/process_64.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/sparc/kernel/process_64.c
--- a/arch/sparc/kernel/process_64.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/sparc/kernel/process_64.c
@@ -239,7 +239,7 @@ static void __global_reg_poll(struct glo
}
}
-void arch_trigger_all_cpu_backtrace(bool include_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
{
struct thread_info *tp = current_thread_info();
struct pt_regs *regs = get_irq_regs();
@@ -255,15 +255,15 @@ void arch_trigger_all_cpu_backtrace(bool
memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot));
- if (include_self)
+ if (cpumask_test_cpu(this_cpu, mask) && !exclude_self)
__global_reg_self(tp, regs, this_cpu);
smp_fetch_global_regs();
- for_each_online_cpu(cpu) {
+ for_each_cpu(cpu, mask) {
struct global_reg_snapshot *gp;
- if (!include_self && cpu == this_cpu)
+ if (exclude_self && cpu == this_cpu)
continue;
gp = &global_cpu_snapshot[cpu].reg;
@@ -300,7 +300,7 @@ void arch_trigger_all_cpu_backtrace(bool
static void sysrq_handle_globreg(int key)
{
- arch_trigger_all_cpu_backtrace(true);
+ trigger_all_cpu_backtrace();
}
static struct sysrq_key_op sparc_globalreg_op = {
diff -puN arch/x86/include/asm/irq.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/x86/include/asm/irq.h
--- a/arch/x86/include/asm/irq.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/x86/include/asm/irq.h
@@ -50,8 +50,9 @@ extern int vector_used_by_percpu_irq(uns
extern void init_ISA_irqs(void);
#ifdef CONFIG_X86_LOCAL_APIC
-void arch_trigger_all_cpu_backtrace(bool);
-#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace
+void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
+ bool exclude_self);
+#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace
#endif
#endif /* _ASM_X86_IRQ_H */
diff -puN arch/x86/kernel/apic/hw_nmi.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods arch/x86/kernel/apic/hw_nmi.c
--- a/arch/x86/kernel/apic/hw_nmi.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/arch/x86/kernel/apic/hw_nmi.c
@@ -26,32 +26,32 @@ u64 hw_nmi_get_sample_period(int watchdo
}
#endif
-#ifdef arch_trigger_all_cpu_backtrace
+#ifdef arch_trigger_cpumask_backtrace
static void nmi_raise_cpu_backtrace(cpumask_t *mask)
{
apic->send_IPI_mask(mask, NMI_VECTOR);
}
-void arch_trigger_all_cpu_backtrace(bool include_self)
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
{
- nmi_trigger_all_cpu_backtrace(include_self, nmi_raise_cpu_backtrace);
+ nmi_trigger_cpumask_backtrace(mask, exclude_self,
+ nmi_raise_cpu_backtrace);
}
-static int
-arch_trigger_all_cpu_backtrace_handler(unsigned int cmd, struct pt_regs *regs)
+static int nmi_cpu_backtrace_handler(unsigned int cmd, struct pt_regs *regs)
{
if (nmi_cpu_backtrace(regs))
return NMI_HANDLED;
return NMI_DONE;
}
-NOKPROBE_SYMBOL(arch_trigger_all_cpu_backtrace_handler);
+NOKPROBE_SYMBOL(nmi_cpu_backtrace_handler);
-static int __init register_trigger_all_cpu_backtrace(void)
+static int __init register_nmi_cpu_backtrace_handler(void)
{
- register_nmi_handler(NMI_LOCAL, arch_trigger_all_cpu_backtrace_handler,
+ register_nmi_handler(NMI_LOCAL, nmi_cpu_backtrace_handler,
0, "arch_bt");
return 0;
}
-early_initcall(register_trigger_all_cpu_backtrace);
+early_initcall(register_nmi_cpu_backtrace_handler);
#endif
diff -puN include/linux/nmi.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods include/linux/nmi.h
--- a/include/linux/nmi.h~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/include/linux/nmi.h
@@ -35,21 +35,34 @@ static inline void hardlockup_detector_d
* base function. Return whether such support was available,
* to allow calling code to fall back to some other mechanism:
*/
-#ifdef arch_trigger_all_cpu_backtrace
+#ifdef arch_trigger_cpumask_backtrace
static inline bool trigger_all_cpu_backtrace(void)
{
- arch_trigger_all_cpu_backtrace(true);
-
+ arch_trigger_cpumask_backtrace(cpu_online_mask, false);
return true;
}
+
static inline bool trigger_allbutself_cpu_backtrace(void)
{
- arch_trigger_all_cpu_backtrace(false);
+ arch_trigger_cpumask_backtrace(cpu_online_mask, true);
+ return true;
+}
+
+static inline bool trigger_cpumask_backtrace(struct cpumask *mask)
+{
+ arch_trigger_cpumask_backtrace(mask, false);
+ return true;
+}
+
+static inline bool trigger_single_cpu_backtrace(int cpu)
+{
+ arch_trigger_cpumask_backtrace(cpumask_of(cpu), false);
return true;
}
/* generic implementation */
-void nmi_trigger_all_cpu_backtrace(bool include_self,
+void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
+ bool exclude_self,
void (*raise)(cpumask_t *mask));
bool nmi_cpu_backtrace(struct pt_regs *regs);
@@ -62,6 +75,14 @@ static inline bool trigger_allbutself_cp
{
return false;
}
+static inline bool trigger_cpumask_backtrace(struct cpumask *mask)
+{
+ return false;
+}
+static inline bool trigger_single_cpu_backtrace(int cpu)
+{
+ return false;
+}
#endif
#ifdef CONFIG_LOCKUP_DETECTOR
diff -puN lib/nmi_backtrace.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods lib/nmi_backtrace.c
--- a/lib/nmi_backtrace.c~nmi_backtrace-add-more-trigger__cpu_backtrace-methods
+++ a/lib/nmi_backtrace.c
@@ -17,20 +17,21 @@
#include <linux/kprobes.h>
#include <linux/nmi.h>
-#ifdef arch_trigger_all_cpu_backtrace
+#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_all_cpu_backtrace */
+/* "in progress" flag of arch_trigger_cpumask_backtrace */
static unsigned long backtrace_flag;
/*
- * When raise() is called it will be is passed a pointer to the
+ * 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_all_cpu_backtrace(bool include_self,
+void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
+ bool exclude_self,
void (*raise)(cpumask_t *mask))
{
int i, this_cpu = get_cpu();
@@ -44,13 +45,13 @@ void nmi_trigger_all_cpu_backtrace(bool
return;
}
- cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask);
- if (!include_self)
+ cpumask_copy(to_cpumask(backtrace_mask), mask);
+ if (exclude_self)
cpumask_clear_cpu(this_cpu, to_cpumask(backtrace_mask));
if (!cpumask_empty(to_cpumask(backtrace_mask))) {
- pr_info("Sending NMI to %s CPUs:\n",
- (include_self ? "all" : "other"));
+ pr_info("Sending NMI from CPU %d to CPUs %*pbl:\n",
+ this_cpu, nr_cpumask_bits, to_cpumask(backtrace_mask));
raise(to_cpumask(backtrace_mask));
}
_
Patches currently in -mm which might be from cmetcalf@xxxxxxxxxxxx are
nmi_backtrace-add-more-trigger__cpu_backtrace-methods.patch
nmi_backtrace-do-a-local-dump_stack-instead-of-a-self-nmi.patch
arch-tile-adopt-the-new-nmi_backtrace-framework.patch
nmi_backtrace-generate-one-line-reports-for-idle-cpus.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
