Re: [BUG] NULL pointer dereference probably caused by kernel/trace/ring_buffer.c

[Steven Rostedt]

Yes it's been reported. At my next layover, I'll take a look at it.

-- Steve 

On October 31, 2022 9:02:12 AM EDT, Roland Ruckerbauer <roland.rucky@xxxxxxxxx> wrote:
>Somewhere between kernel 6.0.2 and 6.0.5 I started getting crashes
>when https://github.com/mchehab/rasdaemon is starting.
>
>After investigating a bit, I found the following commits:
>a6b4d8d6656f ring-buffer: Fix race between reset page and reading page
>fa76ee6fea9c ring-buffer: Add ring_buffer_wake_waiters()
>7bf3c4d84204 ring-buffer: Check pending waiters when doing wake ups as well
>692cc072c800 ring-buffer: Have the shortest_full queue be the shortest
>not longest
>57af2334ca70 ring-buffer: Allow splice to read previous partially read pages
>
>I guess one of them must have broken something rasdaemon is using,
>since the crash is in ring_buffer_wake_waiters(), and it was first
>introduced by this commit series.
>
>The crash report is attached in dmesg.log.
>
>For me the crash is easily reproducible. For testing I run 6.0.5
>upstream kernel, but I guess everything since 6.0.3 is affected as
>well.
>
>My hardware (hopefully irrelevant): x570 amd platform (ryzen 5000)
>
>If I can help in any way, please say so (testing patches etc ...).
>
>Thanks,
>Roland Ruckerbauer

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity and top posting.