[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87o8pjh1i0.fsf@gmx.net>
Date: Tue, 16 Jun 2020 10:13:27 +0200
From: Stephen Berman <stephen.berman@....net>
To: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Cc: Thomas Gleixner <tglx@...utronix.de>,
Peter Zijlstra <peterz@...radead.org>,
linux-kernel@...r.kernel.org, linux-acpi@...r.kernel.org
Subject: Re: power-off delay/hang due to commit 6d25be57 (mainline)
On Tue, 16 Jun 2020 09:38:27 +0200 Sebastian Andrzej Siewior <bigeasy@...utronix.de> wrote:
> On 2020-06-16 09:14:37 [+0200], Stephen Berman wrote:
>>
>> I set CONFIG_ACPI_DEBUG=y, applied the patch and rebuilt 5.6.4, but:
>>
>> $ cat /sys/kernel/debug/tracing/trace > trace.txt
>> cat: /sys/kernel/debug/tracing/trace: No such file or directory
>>
>> Here are all the 5.6.4 config options with "TRACING" or "TRACE"; do I
>> need to set any that aren't set?
>
> I see that "Kernel hacking ---> Tracers" is enabled. You should have
> one tracer enabled for that trace_printk() to work. You have context
> switch tracer so it should be enough.
>
> You might need CONFIG_DEBUG_FS (but it is set in your 5.1 config) or it
> is just not mounted. So I have here:
>
> | debugfs on /sys/kernel/debug type debugfs (rw,nosuid,nodev,noexec,relatime)
> | tracefs on /sys/kernel/debug/tracing type tracefs (rw,nosuid,nodev,noexec,relatime)
>
> Does it work for you if you mount it?
# mount -t debugfs debugfs /sys/kernel/debug
# mount -t tracefs tracefs /sys/kernel/debug/tracing
# mount
...
debugfs on /sys/kernel/debug type debugfs (rw,relatime)
tracefs on /sys/kernel/debug/tracing type tracefs (rw,relatime)
Yes, thanks, that did it. Trace attached.
Steve Berman
Download attachment "trace.txt.gz" of type "application/octet-stream" (10705 bytes)
Powered by blists - more mailing lists