Am 06.05.19 um 14:20 schrieb Mikhail Gavrilov: > [CAUTION: External Email] > > On Sun, 5 May 2019 at 15:18, Christian König > <ckoenig.leichtzumerken@xxxxxxxxx> wrote: >> Yeah, but for most end users we need to get the GPU working as fast as >> possible on a lockup. >> >> Saving all the state (which actually can be a couple of gigabytes if you >> include all textures etc..) is not really an option then. >> >> What we could probably do rather easily is to add a function to run a >> script instead of a GPU reset on lockup detection. >> > This would be useful if this script would run not instead GPU reset > and before GPU reset. That won't work. The kernel can't wait for spawned processes to finish because it is holding locks. The script could as last operation trigger a manual reset, but that would not be the same as a timeout reset because you don't know the cause of it and would always need to do a full engine reset. Sorry, but what you are suggesting here (collect data and then reset) is not easily doable. Christian. > > -- > Best Regards, > Mike Gavrilov. _______________________________________________ amd-gfx mailing list amd-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/amd-gfx
