On Fri, Sep 8, 2023 at 2:43 PM Tj <linux@xxxxxx> wrote: > > On 23/08/2023 14:56, Deucher, Alexander wrote: > > >> -----Original Message----- > >> From: amd-gfx <amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of Tj > >> Sent: Wednesday, August 23, 2023 4:54 AM > >> To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx > >> Subject: 6.5.0-rc7: RIP: 0010:radeon_gem_va_ioctl+0x3dc/0x4f0 [radeon] > >> > >> Recently, and I think especially since 6.5.0-rc5 (but possibly earlier), on resume > >> from S3 suspend the host sees: > >> > > > > If this is a regression can you bisect? There are very few changes for the radeon driver these days. > > > > Alex > Typical - as soon as it was reported it stopped occurring! > > It would have been difficult to bisect though because I have no idea which version worked - if any. I could only mention the versions I'd noticed it in and as I build and test most versions some don't get suspend/resume cycles and some I'm testing other things and don't check the logs for everything. I think it's fixed with this patch: https://cgit.freedesktop.org/drm/drm-misc/commit/?h=drm-misc-fixes&id=dcbad727513d277144aee482b2ffbcd2255c37aa If so the issue would only happen under memory pressure. Alex
