[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <58C520A0-D263-4F08-B1CB-D32C043865F2@vmware.com>
Date: Tue, 29 Nov 2022 18:02:01 +0000
From: Nadav Amit <namit@...are.com>
To: Steven Rostedt <rostedt@...dmis.org>
CC: Peter Zijlstra <peterz@...radead.org>,
Arnd Bergmann <arnd@...db.de>,
Thomas Gleixner <tglx@...utronix.de>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
kernel list <linux-kernel@...r.kernel.org>,
"linux-ia64@...r.kernel.org" <linux-ia64@...r.kernel.org>,
"linux-um@...ts.infradead.org" <linux-um@...ts.infradead.org>,
Linux-Arch <linux-arch@...r.kernel.org>,
Linux-MM <linux-mm@...ck.org>, Andy Lutomirski <luto@...nel.org>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
Dave Hansen <dave.hansen@...ux.intel.com>,
X86 ML <x86@...nel.org>, Richard Weinberger <richard@....at>,
Anton Ivanov <anton.ivanov@...bridgegreys.com>,
Johannes Berg <johannes@...solutions.net>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH 3/3] compiler: inline does not imply notrace
On Nov 29, 2022, at 7:06 AM, Steven Rostedt <rostedt@...dmis.org> wrote:
> On Tue, 29 Nov 2022 04:25:38 +0000
> Nadav Amit <namit@...are.com> wrote:
>
>
>> I will need to further debug it, but this issue does not occur every time.
>>
>> The kernel didn’t crash exactly - it’s more of a deadlock. I have lockdep
>> enabled, so it is not a deadlock that lockdep knows. Could it be that
>> somehow things just slowed down due to IPIs and mostly-disabled IRQs? I have
>> no idea. I would need to recreate the scenario.
>
> You have lockdep enabled and you are running function tracing with stack
> trace on? So you are doing a stack trace on *every* function that is traced?
>
> I don't think you hit a deadlock, I think you hit a live lock. You could
> possibly slow the system down so much that when an interrupt finishes it's
> time for it to be triggered again, and you never make forward progress.
It might be the issue. Perhaps I have a bug, because my code was supposed to
either enable stack-tracing with selected functions or create a trace all
function but *without* stack-tracing.
Thanks for the pointer and sorry for the noise.
Regards,
Nadav
Powered by blists - more mailing lists