Quoting Chris Wilson (2018-06-10 20:43:14) > After triggering the mm switch with a load of PD_DIR, which may be > deferred unto the MI_SET_CONTEXT on rcs, serialise the next commands > with that load by posting a read of PD_DIR (or else those subsequent > commands may access the stale page tables). My stress test failed over night. Much improved MTBF, but not eliminated yet. Next step is to follow the STORE_REG_MEM with an engine->emit_flush(EMIT_INVALIDATE). I think we are indeed on the right track, and that these are much more than random delays. -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
