On Mon, Apr 24, 2017 at 02:22:21PM -0700, Michel Thierry wrote: > > > On 20/04/17 17:17, Michel Thierry wrote: > >>Hmm. Interesting. This relies on i915_gem_retire_requests() (i.e. > >>struct_mutex) to skip replaying innocent requests, but here we should be > >>asserting that we do have the hung request. > >> > >>i.e. > >>request = i915_gem_find_active_request(engine); > >>if (!request) > >> goto skip. > >> > >>Bonus points for tying that into i915_gem_reset_prepare_engine() so that > >>we only seach for the active_request once. > >> > > Will this do it? > https://patchwork.freedesktop.org/patch/152494/ (ignore the > DRM_ERROR I still have to change) That's a little uglier than I was hoping for :( > I'm not sure about reusing the active request in full-reset (what if > we have more than one engine hung?). We can have more than one active request, just don't forget a if (i915_gem_request_completed(engine->hangcheck.hung_request)) /* skip the reset! */ -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
