[v1] x86/nmi: Rate limit unknown NMI messages

[PATCH] x86/nmi: Rate limit unknown NMI messages

Posted by Namhyung Kim 2 years ago

On some AMD machines, unknown NMI messages were printed on the console
continuously when using perf command with IBS.  It was reported that it
can slow down the kernel.  Let's ratelimit the unknown NMI messages.

Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Guilherme Amadio <amadio@gentoo.org>
Cc: Stephane Eranian <eranian@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 arch/x86/kernel/nmi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 17e955ab69fe..d23867901186 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -303,13 +303,13 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 
 	__this_cpu_add(nmi_stats.unknown, 1);
 
-	pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
-		 reason, smp_processor_id());
+	pr_emerg_ratelimited("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
+			     reason, smp_processor_id());
 
 	if (unknown_nmi_panic || panic_on_unrecovered_nmi)
 		nmi_panic(regs, "NMI: Not continuing");
 
-	pr_emerg("Dazed and confused, but trying to continue\n");
+	pr_emerg_ratelimited("Dazed and confused, but trying to continue\n");
 }
 NOKPROBE_SYMBOL(unknown_nmi_error);
 
-- 
2.43.0.472.g3155946c3a-goog

Re: [PATCH] x86/nmi: Rate limit unknown NMI messages

Posted by Ravi Bangoria 2 years ago

> On some AMD machines, unknown NMI messages were printed on the console
> continuously when using perf command with IBS.  It was reported that it
> can slow down the kernel.  Let's ratelimit the unknown NMI messages.

Continuously repeating of the same message is useless. So,

Acked-by: Ravi Bangoria <ravi.bangoria@amd.com>

Re: [PATCH] x86/nmi: Rate limit unknown NMI messages

Posted by Guilherme Amadio 2 years ago

On Fri, Dec 08, 2023 at 05:52:11PM -0800, Namhyung Kim wrote:
> On some AMD machines, unknown NMI messages were printed on the console
> continuously when using perf command with IBS.  It was reported that it
> can slow down the kernel.  Let's ratelimit the unknown NMI messages.
> 
> Cc: Ravi Bangoria <ravi.bangoria@amd.com>
> Cc: Guilherme Amadio <amadio@gentoo.org>

The rate of messages on the console were proportional to the sampling frequency
when I noticed this happening, so this is a welcome change. Thank you!

Acked-by: Guilherme Amadio <amadio@gentoo.org>

Best regards,
-Guilherme

> Cc: Stephane Eranian <eranian@google.com>
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> ---
>  arch/x86/kernel/nmi.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
> index 17e955ab69fe..d23867901186 100644
> --- a/arch/x86/kernel/nmi.c
> +++ b/arch/x86/kernel/nmi.c
> @@ -303,13 +303,13 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
>  
>  	__this_cpu_add(nmi_stats.unknown, 1);
>  
> -	pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
> -		 reason, smp_processor_id());
> +	pr_emerg_ratelimited("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
> +			     reason, smp_processor_id());
>  
>  	if (unknown_nmi_panic || panic_on_unrecovered_nmi)
>  		nmi_panic(regs, "NMI: Not continuing");
>  
> -	pr_emerg("Dazed and confused, but trying to continue\n");
> +	pr_emerg_ratelimited("Dazed and confused, but trying to continue\n");
>  }
>  NOKPROBE_SYMBOL(unknown_nmi_error);
>  
> -- 
> 2.43.0.472.g3155946c3a-goog
>

[tip: x86/misc] x86/nmi: Rate limit unknown NMI messages

Posted by tip-bot2 for Namhyung Kim 1 year, 11 months ago

The following commit has been merged into the x86/misc branch of tip:

Commit-ID:     e2fbc857d3c677ce96d9260577491e0d8f21b6f7
Gitweb:        https://git.kernel.org/tip/e2fbc857d3c677ce96d9260577491e0d8f21b6f7
Author:        Namhyung Kim <namhyung@kernel.org>
AuthorDate:    Fri, 08 Dec 2023 17:52:11 -08:00
Committer:     Borislav Petkov (AMD) <bp@alien8.de>
CommitterDate: Thu, 25 Jan 2024 17:29:06 +01:00

x86/nmi: Rate limit unknown NMI messages

On some AMD machines, unknown NMI messages were printed on the console
continuously when using perf command with IBS.  It was reported that it
can slow down the kernel. Ratelimit the unknown NMI messages.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Ravi Bangoria <ravi.bangoria@amd.com>
Acked-by: Guilherme Amadio <amadio@gentoo.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20231209015211.357983-1-namhyung@kernel.org
---
 arch/x86/kernel/nmi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 17e955a..d238679 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -303,13 +303,13 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 
 	__this_cpu_add(nmi_stats.unknown, 1);
 
-	pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
-		 reason, smp_processor_id());
+	pr_emerg_ratelimited("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
+			     reason, smp_processor_id());
 
 	if (unknown_nmi_panic || panic_on_unrecovered_nmi)
 		nmi_panic(regs, "NMI: Not continuing");
 
-	pr_emerg("Dazed and confused, but trying to continue\n");
+	pr_emerg_ratelimited("Dazed and confused, but trying to continue\n");
 }
 NOKPROBE_SYMBOL(unknown_nmi_error);