Re: [PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

Hi, On 9/6/23 10:34, Christian König wrote: Am 05.09.23 um 16:29 schrieb Thomas Hellström: Hi, Christian On 9/5/23 15:14, Christian König wrote: Am 05.09.23 um 10:58 schrieb Thomas Hellström: If *any* object of a certain WW mutex class is locked, lockdep will consider *all* mutexes of that c

Re: [PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

Am 05.09.23 um 16:29 schrieb Thomas Hellström: Hi, Christian On 9/5/23 15:14, Christian König wrote: Am 05.09.23 um 10:58 schrieb Thomas Hellström: If *any* object of a certain WW mutex class is locked, lockdep will consider *all* mutexes of that class as locked. Also the lock allocation tra

Re: [PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

Hi, Christian On 9/5/23 15:14, Christian König wrote: Am 05.09.23 um 10:58 schrieb Thomas Hellström: If *any* object of a certain WW mutex class is locked, lockdep will consider *all* mutexes of that class as locked. Also the lock allocation tracking code will apparently register only the addre

Re: [PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

Am 05.09.23 um 10:58 schrieb Thomas Hellström: If *any* object of a certain WW mutex class is locked, lockdep will consider *all* mutexes of that class as locked. Also the lock allocation tracking code will apparently register only the address of the first mutex locked in a sequence. This has the

Re: [PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

On 9/5/23 10:58, Thomas Hellström wrote: If *any* object of a certain WW mutex class is locked, lockdep will consider *all* mutexes of that class as locked. Also the lock allocation tracking code will apparently register only the address of the first mutex locked in a sequence. This has the odd c

Re: [PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

On Tue, 5 Sep 2023 10:58:32 +0200 Thomas Hellström wrote: > If *any* object of a certain WW mutex class is locked, lockdep will > consider *all* mutexes of that class as locked. Also the lock allocation > tracking code will apparently register only the address of the first > mutex locked in a se

[PATCH 3/3] drm/drm_exec: Work around a WW mutex lockdep oddity

If *any* object of a certain WW mutex class is locked, lockdep will consider *all* mutexes of that class as locked. Also the lock allocation tracking code will apparently register only the address of the first mutex locked in a sequence. This has the odd consequence that if that first mutex is unlo