From 5692251c254a3d561316c4e8e10c77e470b60658 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sat, 28 Apr 2018 12:15:32 +0100 Subject: [PATCH] drm/i915/lrc: Scrub the GPU state of the guilty hanging request MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Previously, we just reset the ring register in the context image such that we could skip over the broken batch and emit the closing breadcrumb. However, on resume the context image and GPU state would be reloaded, which may have been left in an inconsistent state by the reset. The presumption was that at worst it would just cause another reset and skip again until it recovered, however it seems just as likely to cause an unrecoverable hang. Instead of risking loading an incomplete context image, restore it back to the default state. v2: Fix up off-by-one from including the ppHSWP in with the register state. v3: Use a ring local to compact a few lines. v4: Beware setting the ring local before checking for a NULL request. References: https://bugs.freedesktop.org/show_bug.cgi?id=105304 Signed-off-by: Chris Wilson Cc: Mika Kuoppala Cc: Michał Winiarski Cc: Michel Thierry Cc: Tvrtko Ursulin Reviewed-by: Michel Thierry #v2 Link: https://patchwork.freedesktop.org/patch/msgid/20180428111532.15819-1-chris@chris-wilson.co.uk --- drivers/gpu/drm/i915/intel_lrc.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 87eb3a688424..58cad2448184 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -1803,8 +1803,8 @@ static void reset_common_ring(struct intel_engine_cs *engine, struct i915_request *request) { struct intel_engine_execlists * const execlists = &engine->execlists; - struct intel_context *ce; unsigned long flags; + u32 *regs; GEM_TRACE("%s request global=%x, current=%d\n", engine->name, request ? request->global_seqno : 0, @@ -1854,14 +1854,24 @@ static void reset_common_ring(struct intel_engine_cs *engine, * future request will be after userspace has had the opportunity * to recreate its own state. */ - ce = &request->ctx->engine[engine->id]; - execlists_init_reg_state(ce->lrc_reg_state, - request->ctx, engine, ce->ring); + regs = request->ctx->engine[engine->id].lrc_reg_state; + if (engine->default_state) { + void *defaults; + + defaults = i915_gem_object_pin_map(engine->default_state, + I915_MAP_WB); + if (!IS_ERR(defaults)) { + memcpy(regs, /* skip restoring the vanilla PPHWSP */ + defaults + LRC_STATE_PN * PAGE_SIZE, + engine->context_size - PAGE_SIZE); + i915_gem_object_unpin_map(engine->default_state); + } + } + execlists_init_reg_state(regs, request->ctx, engine, request->ring); /* Move the RING_HEAD onto the breadcrumb, past the hanging batch */ - ce->lrc_reg_state[CTX_RING_BUFFER_START+1] = - i915_ggtt_offset(ce->ring->vma); - ce->lrc_reg_state[CTX_RING_HEAD+1] = request->postfix; + regs[CTX_RING_BUFFER_START + 1] = i915_ggtt_offset(request->ring->vma); + regs[CTX_RING_HEAD + 1] = request->postfix; request->ring->head = request->postfix; intel_ring_update_space(request->ring); -- 2.30.2