Skip to contentSkip to conversation

panic: Reduce CPU consumption after panic

PATCH Superseded [email protected] 4 participants 11 messages Apr 28, 2025 – May 7, 2025 x86

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

v2 cover panic: Reduce CPU consumption after panic

From: Carlos Bilbao <[email protected]>

Provide a priority-based mechanism to set the behavior of the kernel at the post-panic stage -- the current default is a waste of CPU except for cases with console that generate insightful output.

In v1 [1], I illustrated the potential to reduce unnecessary CPU usage with an experiment with VMs. The main delta is that, instead of a weak function that archs can overwrite, we provide a flexible priority-based mechanism (following suggestions by Sean Christopherson), panic_set_handling().

[1] https://lore.kernel.org/all/[email protected]/

Carlos:
panic: Allow for dynamic custom behavior after panic x86/panic: Add x86_panic_handler as default post-panic behavior

---
 arch/x86/kernel/setup.c | 12 ++++++++++++
 include/linux/panic.h   |  2 ++
 kernel/panic.c          | 27 +++++++++++++++++++++++++++
 3 files changed, 41 insertions(+)
#1 lore raw

PATCH v2 1/2 panic: Allow for dynamic custom behavior after panic

From: Carlos Bilbao <[email protected]>

Introduce panic_set_handling() to allow overriding the default post-panic behavior.

---
 include/linux/panic.h |  2 ++
 kernel/panic.c        | 27 +++++++++++++++++++++++++++
 2 files changed, 29 insertions(+)
include/linux/panic.h +2 -0
index 2494d51707ef..cf8d4a944407 100644
@@ -98,4 +98,6 @@ extern void add_taint(unsigned flag, enum lockdep_ok);
 extern int test_taint(unsigned flag);
 extern unsigned long get_taint(void);
 
+void panic_set_handling(void (*fn)(void), int priority);
+
 #endif	/* _LINUX_PANIC_H */
kernel/panic.c +27 -0
index a3889f38153d..2cdd83b4afb6 100644
@@ -276,6 +276,30 @@ static void panic_other_cpus_shutdown(bool crash_kexec)
 		crash_smp_send_stop();
 }
 
+/*
+ * This is the default function called after a kernel panic has been
+ * handled. Higher priority alternatives can be set with function
+ * panic_set_handling()
+ *
+ */
+static void after_panic_handling(void)
+{
+	mdelay(PANIC_TIMER_STEP);
+}
+
+static void (*panic_halt)(void) = after_panic_handling;
+static int panic_hlt_priority;
+
+void panic_set_handling(void (*fn)(void), int priority)
+{
+	if (priority <= panic_hlt_priority)
+		return;
+
+	panic_hlt_priority = priority;
+	panic_halt = fn;
+}
+EXPORT_SYMBOL_GPL(panic_set_handling);
+
 /**
  *	panic - halt the system
  *	@fmt: The text string to print
@@ -467,6 +491,9 @@ void panic(const char *fmt, ...)
 	console_flush_on_panic(CONSOLE_FLUSH_PENDING);
 	nbcon_atomic_flush_unsafe();
 
+	if (panic_halt)
+		panic_halt();
+
 	local_irq_enable();
 	for (i = 0; ; i += PANIC_TIMER_STEP) {
 		touch_softlockup_watchdog();
Signature · 1 line
2.47.1
#2 lore raw

Re: [PATCH v2 1/2] panic: Allow for dynamic custom behavior after panic

Quoted 10 lines · On Mon, Apr 28, 2025, [email protected] wrote:
kernel/panic.c +2 -0
index a3889f38153d..2cdd83b4afb6 100644
@@ -276,6 +276,30 @@ static void panic_other_cpus_shutdown(bool crash_kexec)
 		crash_smp_send_stop();
 }
 
+/*
+ * This is the default function called after a kernel panic has been

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 290

Quoted 10 lines
kernel/panic.c +10 -0
+ * handled. Higher priority alternatives can be set with function
+ * panic_set_handling()
+ *
+ */
+static void after_panic_handling(void)
+{
+	mdelay(PANIC_TIMER_STEP);
+}
+
+static void (*panic_halt)(void) = after_panic_handling;

The default implementation clearly doesn't halt, which makes this unnecessarily confusing. And if you're going to provide a default implementation, why bother checking for NULL in panic()? Just leave panic_halt NULL.

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 291

Quoted 1 line
kernel/panic.c +1 -0
+static int panic_hlt_priority;

Uber nit, pick one of halt or hlt.

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 295

Quoted 4 lines
kernel/panic.c +4 -0
+
+void panic_set_handling(void (*fn)(void), int priority)
+{
+	if (priority <= panic_hlt_priority)

If panic_halt is NULL by default, maybe do?

	if (panic_halt && priority <= panic_halt_priority)

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 301

Quoted 6 lines
kernel/panic.c +6 -0
+		return;
+
+	panic_hlt_priority = priority;
+	panic_halt = fn;
+}
+EXPORT_SYMBOL_GPL(panic_set_handling);

This doesn't seem like something that should be exported unless it's absolutely necessary, and it shouldn't be necessary as of this series.

Quoted 17 lines
kernel/panic.c +4 -0
+
 /**
  *	panic - halt the system
  *	@fmt: The text string to print
@@ -467,6 +491,9 @@ void panic(const char *fmt, ...)
 	console_flush_on_panic(CONSOLE_FLUSH_PENDING);
 	nbcon_atomic_flush_unsafe();
 
+	if (panic_halt)
+		panic_halt();
+
 	local_irq_enable();
 	for (i = 0; ; i += PANIC_TIMER_STEP) {
 		touch_softlockup_watchdog();
Signature · 1 line
2.47.1
#3 lore raw

Re: [PATCH v2 1/2] panic: Allow for dynamic custom behavior after panic

Hello,

Sean, thanks for getting back to me so fast!

Quoted 9 lines · Sean Christopherson
Quoted 8 lines · On Mon, Apr 28, 2025, [email protected] wrote:
kernel/panic.c +0 -0
index a3889f38153d..2cdd83b4afb6 100644
@@ -276,6 +276,30 @@ static void panic_other_cpus_shutdown(bool crash_kexec)
 		crash_smp_send_stop();
 }
 
Quoted 15 lines

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 290

Quoted 12 lines
kernel/panic.c +12 -0
+/*
+ * This is the default function called after a kernel panic has been
+ * handled. Higher priority alternatives can be set with function
+ * panic_set_handling()
+ *
+ */
+static void after_panic_handling(void)
+{
+	mdelay(PANIC_TIMER_STEP);
+}
+
+static void (*panic_halt)(void) = after_panic_handling;

The default implementation clearly doesn't halt, which makes this unnecessarily confusing. And if you're going to provide a default implementation, why bother checking for NULL in panic()? Just leave panic_halt NULL.

Agreed.

Quoted 3 lines

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 291

Quoted 2 lines
kernel/panic.c +1 -0
+static int panic_hlt_priority;

Uber nit, pick one of halt or hlt.

True, will use halt.

Quoted 25 lines

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 295

Quoted 4 lines
kernel/panic.c +4 -0
+
+void panic_set_handling(void (*fn)(void), int priority)
+{
+	if (priority <= panic_hlt_priority)

If panic_halt is NULL by default, maybe do?

	if (panic_halt && priority <= panic_halt_priority)

kernel/panic.c static void panic_other_cpus_shutdown(bool crash_kexec) · line 301

Quoted 6 lines
kernel/panic.c +6 -0
+		return;
+
+	panic_hlt_priority = priority;
+	panic_halt = fn;
+}
+EXPORT_SYMBOL_GPL(panic_set_handling);

This doesn't seem like something that should be exported unless it's absolutely necessary, and it shouldn't be necessary as of this series.

Quoted 8 lines
kernel/panic.c +1 -0
+
 /**
  *	panic - halt the system
  *	@fmt: The text string to print
@@ -467,6 +491,9 @@ void panic(const char *fmt, ...)
 	console_flush_on_panic(CONSOLE_FLUSH_PENDING);
 	nbcon_atomic_flush_unsafe();
 
Quoted 9 lines
Quoted 9 lines
kernel/panic.c +3 -0
+	if (panic_halt)
+		panic_halt();
+
 	local_irq_enable();
 	for (i = 0; ; i += PANIC_TIMER_STEP) {
 		touch_softlockup_watchdog();
Signature · 1 line
2.47.1

Agree with all. Will fix in v3.

Thanks,

Carlos

#4 lore raw

PATCH v2 2/2 x86/panic: Add x86_panic_handler as default post-panic behavior

In reply to [email protected]

From: Carlos Bilbao <[email protected]>

Add function x86_panic_handler() as the default behavior for x86 for post-panic stage via panic_set_handling(). Instead of busy-wait loop, it will halt if there's no console to save CPU cycles.

---
 arch/x86/kernel/setup.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
arch/x86/kernel/setup.c +12 -0
index 9d2a13b37833..3bfef55e9adb 100644
@@ -16,6 +16,7 @@
 #include <linux/initrd.h>
 #include <linux/iscsi_ibft.h>
 #include <linux/memblock.h>
+#include <linux/panic.h>
 #include <linux/panic_notifier.h>
 #include <linux/pci.h>
 #include <linux/root_dev.h>
@@ -837,6 +838,15 @@ static void __init x86_report_nx(void)
 	}
 }
 
+
+static void x86_panic_handler(void)
+{
+	if (console_trylock()) {
+		console_unlock();
+		safe_halt();
+	}
+}
+
 /*
  * Determine if we were loaded by an EFI loader.  If so, then we have also been
  * passed the efi memmap, systab, etc., so we should use these data structures
@@ -1252,6 +1262,8 @@ void __init setup_arch(char **cmdline_p)
 #endif
 
 	unwind_init();
+
+	panic_set_handling(x86_panic_handler, 1);
 }
 
 #ifdef CONFIG_X86_32
Signature · 1 line
2.47.1
#5 lore raw

Re: [PATCH v2 2/2] x86/panic: Add x86_panic_handler as default post-panic behavior

Quoted 24 lines · On Mon, Apr 28, 2025, [email protected] wrote:

From: Carlos Bilbao <[email protected]>

Add function x86_panic_handler() as the default behavior for x86 for post-panic stage via panic_set_handling(). Instead of busy-wait loop, it will halt if there's no console to save CPU cycles.

---
 arch/x86/kernel/setup.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
arch/x86/kernel/setup.c +1 -0
index 9d2a13b37833..3bfef55e9adb 100644
@@ -16,6 +16,7 @@
 #include <linux/initrd.h>
 #include <linux/iscsi_ibft.h>
 #include <linux/memblock.h>
+#include <linux/panic.h>
 #include <linux/panic_notifier.h>
 #include <linux/pci.h>
 #include <linux/root_dev.h>

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 841

Quoted 4 lines
arch/x86/kernel/setup.c +1 -0
@@ -837,6 +838,15 @@ static void __init x86_report_nx(void)
 	}
 }
 
+

Spurious newline.

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 844

Quoted 3 lines
arch/x86/kernel/setup.c +3 -0
+static void x86_panic_handler(void)
+{
+	if (console_trylock()) {

A comment here would be very helpful. Even with the changelog saying "if there's no console", as an unfamiliar reader of console code, I have zero idea why being able to lock the console is an effective test for no console.

Quoted 4 lines
arch/x86/kernel/setup.c +4 -0
+		console_unlock();
+		safe_halt();
+	}
+}
#6 lore raw

Re: [PATCH v2 2/2] x86/panic: Add x86_panic_handler as default post-panic behavior

Hello,

Quoted 31 lines · Sean Christopherson
Quoted 24 lines · On Mon, Apr 28, 2025, [email protected] wrote:

From: Carlos Bilbao <[email protected]>

Add function x86_panic_handler() as the default behavior for x86 for post-panic stage via panic_set_handling(). Instead of busy-wait loop, it will halt if there's no console to save CPU cycles.

---
 arch/x86/kernel/setup.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
arch/x86/kernel/setup.c +1 -0
index 9d2a13b37833..3bfef55e9adb 100644
@@ -16,6 +16,7 @@
 #include <linux/initrd.h>
 #include <linux/iscsi_ibft.h>
 #include <linux/memblock.h>
+#include <linux/panic.h>
 #include <linux/panic_notifier.h>
 #include <linux/pci.h>
 #include <linux/root_dev.h>

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 841

Quoted 4 lines
arch/x86/kernel/setup.c +1 -0
@@ -837,6 +838,15 @@ static void __init x86_report_nx(void)
 	}
 }
 
+

Spurious newline.

Quoted 6 lines

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 844

Quoted 3 lines
arch/x86/kernel/setup.c +3 -0
+static void x86_panic_handler(void)
+{
+	if (console_trylock()) {

A comment here would be very helpful. Even with the changelog saying "if there's no console", as an unfamiliar reader of console code, I have zero idea why being able to lock the console is an effective test for no console.

Agree, will fix in v3.

Quoted 5 lines

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 848

Quoted 5 lines
arch/x86/kernel/setup.c +4 -0
+		console_unlock();
+		safe_halt();
+	}
+}

Thanks,

Carlos

#7 lore raw

Re: [PATCH v2 2/2] x86/panic: Add x86_panic_handler as default post-panic behavior

In reply to [email protected]
Quoted 24 lines · On 2025-04-28, [email protected] wrote:

From: Carlos Bilbao <[email protected]>

Add function x86_panic_handler() as the default behavior for x86 for post-panic stage via panic_set_handling(). Instead of busy-wait loop, it will halt if there's no console to save CPU cycles.

---
 arch/x86/kernel/setup.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
arch/x86/kernel/setup.c +1 -0
index 9d2a13b37833..3bfef55e9adb 100644
@@ -16,6 +16,7 @@
 #include <linux/initrd.h>
 #include <linux/iscsi_ibft.h>
 #include <linux/memblock.h>
+#include <linux/panic.h>
 #include <linux/panic_notifier.h>
 #include <linux/pci.h>
 #include <linux/root_dev.h>

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 847

Quoted 10 lines
arch/x86/kernel/setup.c +7 -0
@@ -837,6 +838,15 @@ static void __init x86_report_nx(void)
 	}
 }
 
+
+static void x86_panic_handler(void)
+{
+	if (console_trylock()) {
+		console_unlock();
+		safe_halt();
+	}

I do not understand what you are trying to accomplish with the console_trylock(). At this point in the panic, all the messages are already output. The console lock is totally irrelevant.

Also, the console lock is only valid for legacy consoles.

I see no reason why you don't just use safe_halt() as your panic handler.

John Ogness

#8 lore raw

Re: [PATCH v2 2/2] x86/panic: Add x86_panic_handler as default post-panic behavior

Hello John,

Quoted 39 lines · John Ogness
Quoted 24 lines · On 2025-04-28, [email protected] wrote:

From: Carlos Bilbao <[email protected]>

Add function x86_panic_handler() as the default behavior for x86 for post-panic stage via panic_set_handling(). Instead of busy-wait loop, it will halt if there's no console to save CPU cycles.

---
 arch/x86/kernel/setup.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
arch/x86/kernel/setup.c +1 -0
index 9d2a13b37833..3bfef55e9adb 100644
@@ -16,6 +16,7 @@
 #include <linux/initrd.h>
 #include <linux/iscsi_ibft.h>
 #include <linux/memblock.h>
+#include <linux/panic.h>
 #include <linux/panic_notifier.h>
 #include <linux/pci.h>
 #include <linux/root_dev.h>

arch/x86/kernel/setup.c static void __init x86_report_nx(void) · line 847

Quoted 10 lines
arch/x86/kernel/setup.c +7 -0
@@ -837,6 +838,15 @@ static void __init x86_report_nx(void)
 	}
 }
 
+
+static void x86_panic_handler(void)
+{
+	if (console_trylock()) {
+		console_unlock();
+		safe_halt();
+	}

I do not understand what you are trying to accomplish with the console_trylock(). At this point in the panic, all the messages are already output. The console lock is totally irrelevant.

Quoted 4 lines

Also, the console lock is only valid for legacy consoles.

I see no reason why you don't just use safe_halt() as your panic handler.

Yes, in my original implementation I simply used halt, but I was trying to be cautious in case any remaining messages hadn't flushed. I wonder, can we be certain that, as you said, all output (e.g., a backtrace) has already been displayed on screen at this point? I'm not sure.

Quoted 2 lines

John Ogness

Thanks,

Carlos

#9 lore raw

Re: [PATCH v2 2/2] x86/panic: Add x86_panic_handler as default post-panic behavior

Quoted 8 lines · Carlos Bilbao
Quoted 2 lines

I see no reason why you don't just use safe_halt() as your panic handler.

Yes, in my original implementation I simply used halt, but I was trying to be cautious in case any remaining messages hadn't flushed. I wonder, can we be certain that, as you said, all output (e.g., a backtrace) has already been displayed on screen at this point? I'm not sure.

Well, without this series, the kernel goes into an infinite loop. So if the messages are not all out, they won't magically appear during the infinite loop either.

John

#10 lore raw

Re: [PATCH v2 2/2] x86/panic: Add x86_panic_handler as default post-panic behavior

Hello,

Quoted 11 lines · John Ogness
Quoted 7 lines · Carlos Bilbao
Quoted 2 lines

I see no reason why you don't just use safe_halt() as your panic handler.

Yes, in my original implementation I simply used halt, but I was trying to be cautious in case any remaining messages hadn't flushed. I wonder, can we be certain that, as you said, all output (e.g., a backtrace) has already been displayed on screen at this point? I'm not sure.

Well, without this series, the kernel goes into an infinite loop. So if the messages are not all out, they won't magically appear during the infinite loop either.

I thought there may be some async stuff going on :)

Quoted 2 lines

John

Thanks,

Carlos