If execlists's lite-restore is based on the common GEM context tag
rather than the per-intel_context LRCA, then a context switch between
two intel_contexts on the same engine derived from the same GEM context
will perform a lite-restore instead of a full context switch. We can
exploit this by poisoning the ringbuffer of the first context and trying
to trick a simple RING_TAIL update (i.e. lite-restore)
v2: Also check what happens if preempt ce[0] with ce[1] (both instances
on the same engine from the same parent context) [Tvrtko]
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
---
Don't let rq[0] out of your sight!
---
drivers/gpu/drm/i915/gt/intel_lrc.c | 18 ++-
drivers/gpu/drm/i915/gt/selftest_lrc.c | 180 +++++++++++++++++++++++++
2 files changed, 194 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 9877b6ba13df..431d3b8c3371 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -235,6 +235,16 @@ static void execlists_init_reg_state(u32 *reg_state,
const struct intel_ring *ring,
bool close);
+static void __context_pin_acquire(struct intel_context *ce)
+{
+ mutex_acquire(&ce->pin_mutex.dep_map, 2, 0, _RET_IP_);
+}
+
+static void __context_pin_release(struct intel_context *ce)
+{
+ mutex_release(&ce->pin_mutex.dep_map, 0, _RET_IP_);
+}
+
static void mark_eio(struct i915_request *rq)
{
if (!i915_request_signaled(rq))
@@ -2761,7 +2771,7 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled)
GEM_BUG_ON(!i915_vma_is_pinned(ce->state));
/* Proclaim we have exclusive access to the context image! */
- mutex_acquire(&ce->pin_mutex.dep_map, 2, 0, _THIS_IP_);
+ __context_pin_acquire(ce);
rq = active_request(rq);
if (!rq) {
@@ -2825,7 +2835,7 @@ static void __execlists_reset(struct intel_engine_cs *engine, bool stalled)
__execlists_reset_reg_state(ce, engine);
__execlists_update_reg_state(ce, engine);
ce->lrc_desc |= CTX_DESC_FORCE_RESTORE; /* paranoid: GPU was reset! */
- mutex_release(&ce->pin_mutex.dep_map, 0, _THIS_IP_);
+ __context_pin_release(ce);
unwind:
/* Push back any incomplete requests for replay after the reset. */
@@ -4439,7 +4449,7 @@ void intel_lr_context_reset(struct intel_engine_cs *engine,
bool scrub)
{
GEM_BUG_ON(!intel_context_is_pinned(ce));
- mutex_acquire(&ce->pin_mutex.dep_map, 2, 0, _THIS_IP_);
+ __context_pin_acquire(ce);
/*
* We want a simple context + ring to execute the breadcrumb update.
@@ -4465,7 +4475,7 @@ void intel_lr_context_reset(struct intel_engine_cs *engine,
intel_ring_update_space(ce->ring);
__execlists_update_reg_state(ce, engine);
- mutex_release(&ce->pin_mutex.dep_map, 0, _THIS_IP_);
+ __context_pin_release(ce);
}
#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c
index 93f2fcdc49bf..d33d7508b3f1 100644
--- a/drivers/gpu/drm/i915/gt/selftest_lrc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c
@@ -79,6 +79,184 @@ static int live_sanitycheck(void *arg)
return err;
}
+static int live_unlite_restore(struct drm_i915_private *i915, int prio)
+{
+ struct intel_engine_cs *engine;
+ struct i915_gem_context *ctx;
+ enum intel_engine_id id;
+ intel_wakeref_t wakeref;
+ struct igt_spinner spin;
+ int err = -ENOMEM;
+
+ /*
+ * Check that we can correctly context switch between 2 instances
+ * on the same engine from the same parent context.
+ */
+
+ mutex_lock(&i915->drm.struct_mutex);
+ wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+
+ if (igt_spinner_init(&spin, &i915->gt))
+ goto err_unlock;
+
+ ctx = kernel_context(i915);
+ if (!ctx)
+ goto err_spin;
+
+ err = 0;
+ for_each_engine(engine, i915, id) {
+ struct intel_context *ce[2] = {};
+ struct i915_request *rq[2];
+ struct igt_live_test t;
+ int n;
+
+ if (prio && !intel_engine_has_preemption(engine))
+ continue;
+
+ if (!intel_engine_can_store_dword(engine))
+ continue;
+
+ if (igt_live_test_begin(&t, i915, __func__, engine->name)) {
+ err = -EIO;
+ break;
+ }
+
+ for (n = 0; n < ARRAY_SIZE(ce); n++) {
+ struct intel_context *tmp;
+
+ tmp = intel_context_create(ctx, engine);
+ if (IS_ERR(tmp)) {
+ err = PTR_ERR(tmp);
+ goto err_ce;
+ }
+
+ err = intel_context_pin(tmp);
+ if (err) {
+ intel_context_put(tmp);
+ goto err_ce;
+ }
+
+ /*
+ * Setup the pair of contexts such that if we
+ * lite-restore using the RING_TAIL from ce[1] it
+ * will execute garbage from ce[0]->ring.
+ */
+ memset(tmp->ring->vaddr,
+ POISON_INUSE, /* IPEHR: 0x5a5a5a5a [hung!] */
+ tmp->ring->vma->size);
+
+ ce[n] = tmp;
+ }
+ GEM_BUG_ON(!ce[1]->ring->size);
+ intel_ring_reset(ce[1]->ring, ce[1]->ring->size / 2);
+
+ local_bh_disable(); /* appease lockdep */
+ __context_pin_acquire(ce[1]);
+ __execlists_update_reg_state(ce[1], engine);
+ __context_pin_release(ce[1]);
+ local_bh_enable();
+
+ rq[0] = igt_spinner_create_request(&spin, ce[0], MI_ARB_CHECK);
+ if (IS_ERR(rq[0])) {
+ err = PTR_ERR(rq[0]);
+ goto err_ce;
+ }
+
+ i915_request_get(rq[0]);
+ i915_request_add(rq[0]);
+ GEM_BUG_ON(rq[0]->tail > ce[1]->ring->emit);
+
+ if (!igt_wait_for_spinner(&spin, rq[0])) {
+ i915_request_put(rq[0]);
+ goto err_ce;
+ }
+
+ rq[1] = i915_request_create(ce[1]);
+ if (IS_ERR(rq[1])) {
+ err = PTR_ERR(rq[1]);
+ i915_request_put(rq[0]);
+ goto err_ce;
+ }
+
+ if (!prio) {
+ /*
+ * Ensure we do the switch to ce[1] on completion.
+ *
+ * rq[0] is already submitted, so this should reduce
+ * to a no-op (a wait on a request on the same engine
+ * uses the submit fence, not the completion fence),
+ * but it will install a dependency on rq[1] for rq[0]
+ * that will prevent the pair being reordered by
+ * timeslicing.
+ */
+ i915_request_await_dma_fence(rq[1], &rq[0]->fence);
+ }
+
+ i915_request_get(rq[1]);
+ i915_request_add(rq[1]);
+ GEM_BUG_ON(rq[1]->tail <= rq[0]->tail);
+ i915_request_put(rq[0]);
+
+ if (prio) {
+ struct i915_sched_attr attr = {
+ .priority = prio,
+ };
+
+ /* Alternatively preempt the spinner with ce[1] */
+ engine->schedule(rq[1], &attr);
+ }
+
+ /* And switch back to ce[0] for good measure */
+ rq[0] = i915_request_create(ce[0]);
+ if (IS_ERR(rq[0])) {
+ err = PTR_ERR(rq[0]);
+ i915_request_put(rq[1]);
+ goto err_ce;
+ }
+
+ i915_request_await_dma_fence(rq[0], &rq[1]->fence);
+ i915_request_get(rq[0]);
+ i915_request_add(rq[0]);
+ GEM_BUG_ON(rq[0]->tail > rq[1]->tail);
+ i915_request_put(rq[1]);
+ i915_request_put(rq[0]);
+
+err_ce:
+ tasklet_kill(&engine->execlists.tasklet); /* flush submission */
+ igt_spinner_end(&spin);
+ for (n = 0; n < ARRAY_SIZE(ce); n++) {
+ if (IS_ERR_OR_NULL(ce[n]))
+ break;
+
+ intel_context_unpin(ce[n]);
+ intel_context_put(ce[n]);
+ }
+
+ if (igt_live_test_end(&t))
+ err = -EIO;
+ if (err)
+ break;
+ }
+
+ kernel_context_close(ctx);
+err_spin:
+ igt_spinner_fini(&spin);
+err_unlock:
+ intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+ mutex_unlock(&i915->drm.struct_mutex);
+ return err;
+}
+
+static int live_unlite_switch(void *arg)
+{
+ return live_unlite_restore(arg, 0);
+}
+
+static int live_unlite_preempt(void *arg)
+{
+ return live_unlite_restore(arg, I915_USER_PRIORITY(I915_PRIORITY_MAX));
+}
+
static int
emit_semaphore_chain(struct i915_request *rq, struct i915_vma *vma, int idx)
{
@@ -2178,6 +2356,8 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915)
{
static const struct i915_subtest tests[] = {
SUBTEST(live_sanitycheck),
+ SUBTEST(live_unlite_switch),
+ SUBTEST(live_unlite_preempt),
SUBTEST(live_timeslice_preempt),
SUBTEST(live_busywait_preempt),
SUBTEST(live_preempt),
--
2.23.0
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
