lkml.org 
[lkml]   [2026]   [May]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v2] locking/rtmutex: Annotate API and implementation
On 2026-05-05 09:50:11 [+0200], Bart Van Assche wrote:
>
> On 5/5/26 9:32 AM, Sebastian Andrzej Siewior wrote:
> > I see. I don't know why CONTEXT_ANALYSIS_rtmutex was skipped. Was this
> > just a regular compile with newer clang or was there more to it? I do
> > remember clang was not complaining but then context-analysis flag was
> > missing…
>
> Hi Sebastian,

Hi Bart,

> This was noticed by enabling CONFIG_WARN_CONTEXT_ANALYSIS_ALL=n. I'm not
> sure why this was not noticed earlier. Maybe as the result of an
> oversight?

I have here clang-22 and make with CONFIG_WARN_CONTEXT_ANALYSIS_ALL=n
and CONFIG_WARN_CONTEXT_ANALYSIS=y. I don't see any warning without your
patch during "make kernel/locking/".

The change
CONTEXT_ANALYSIS_rtmutex.o := y

is not needed because rtmutex.o is not compiled on its own:
| $ git grep "rtmutex\.c" kernel/locking/
| kernel/locking/rtmutex_api.c:#include "rtmutex.c"
| kernel/locking/rwsem.c:#include "rtmutex.c"
| kernel/locking/spinlock_rt.c:#include "rtmutex.c"
| kernel/locking/ww_rt_mutex.c:#include "rtmutex.c"

or do I miss something here?
I tried with PREEMPT_RT enabled and disabled. Also with and without
LOCKDEP in case it makes a change.

> Thanks,
>
> Bart.

Sebastian

\
 
 \ /
  Last update: 2026-05-05 16:05    [W:0.131 / U:2.279 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and my Meterkast|Read the blog