drm/i915: Pass the set of guilty engines to i915_reset()

Currently, we rely on inspecting the hangcheck state from within the
i915_reset() routines to determine which engines were guilty of the
hang. This is problematic for cases where we want to run
i915_handle_error() and call i915_reset() independently of hangcheck.
Instead of relying on the indirect parameter passing, turn it into an
explicit parameter providing the set of stalled engines which then are
treated as guilty until proven innocent.

While we are removing the implicit stalled parameter, also make the
reason into an explicit parameter to i915_reset(). We still need a
back-channel for i915_handle_error() to hand over the task to the locked
waiter, but let's keep that its own channel rather than incriminate
another.

This leaves stalled/seqno as being private to hangcheck, with no more
nefarious snooping by reset, be it whole-device or per-engine. \o/

The only real issue now is that this makes it crystal clear that we
don't actually do any testing of hangcheck per se in
drv_selftest/live_hangcheck, merely of resets!

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Jeff McGee <jeff.mcgee@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180406220354.18911-2-chris@chris-wilson.co.uk

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 7ce229c6f424f64c0d3c70daa9f7e93e06323392..f770be18b2d79d7b17897679c2a3067ed50359fc 100644 (file)
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1866,6 +1866,8 @@ static int i915_resume_switcheroo(struct drm_device *dev)
 /**
  * i915_reset - reset chip after a hang
  * @i915: #drm_i915_private to reset
+ * @stalled_mask: mask of the stalled engines with the guilty requests
+ * @reason: user error message for why we are resetting
  *
  * Reset the chip.  Useful if a hang is detected. Marks the device as wedged
  * on failure.
@@ -1880,7 +1882,9 @@ static int i915_resume_switcheroo(struct drm_device *dev)
  *   - re-init interrupt state
  *   - re-init display
  */
-void i915_reset(struct drm_i915_private *i915)
+void i915_reset(struct drm_i915_private *i915,
+               unsigned int stalled_mask,
+               const char *reason)
 {
        struct i915_gpu_error *error = &i915->gpu_error;
        int ret;
@@ -1899,9 +1903,8 @@ void i915_reset(struct drm_i915_private *i915)
        if (!i915_gem_unset_wedged(i915))
                goto wakeup;
 
-       if (error->reason)
-               dev_notice(i915->drm.dev,
-                          "Resetting chip for %s\n", error->reason);
+       if (reason)
+               dev_notice(i915->drm.dev, "Resetting chip for %s\n", reason);
        error->reset_count++;
 
        disable_irq(i915->drm.irq);
@@ -1944,7 +1947,7 @@ void i915_reset(struct drm_i915_private *i915)
                goto error;
        }
 
-       i915_gem_reset(i915);
+       i915_gem_reset(i915, stalled_mask);
        intel_overlay_reset(i915);
 
        /*

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 6b3f2f651def2b30c32c6358165e3bd838194e39..9bca104c409e6e946723da5baaf360d299342da4 100644 (file)
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2701,8 +2701,11 @@ extern void i915_driver_unload(struct drm_device *dev);
 extern int intel_gpu_reset(struct drm_i915_private *dev_priv, u32 engine_mask);
 extern bool intel_has_gpu_reset(struct drm_i915_private *dev_priv);
 
-extern void i915_reset(struct drm_i915_private *i915);
-extern int i915_reset_engine(struct intel_engine_cs *engine, const char *msg);
+extern void i915_reset(struct drm_i915_private *i915,
+                      unsigned int stalled_mask,
+                      const char *reason);
+extern int i915_reset_engine(struct intel_engine_cs *engine,
+                            const char *reason);
 
 extern bool intel_has_reset_engine(struct drm_i915_private *dev_priv);
 extern int intel_reset_guc(struct drm_i915_private *dev_priv);
@@ -3126,7 +3129,8 @@ static inline u32 i915_reset_engine_count(struct i915_gpu_error *error,
 struct i915_request *
 i915_gem_reset_prepare_engine(struct intel_engine_cs *engine);
 int i915_gem_reset_prepare(struct drm_i915_private *dev_priv);
-void i915_gem_reset(struct drm_i915_private *dev_priv);
+void i915_gem_reset(struct drm_i915_private *dev_priv,
+                   unsigned int stalled_mask);
 void i915_gem_reset_finish_engine(struct intel_engine_cs *engine);
 void i915_gem_reset_finish(struct drm_i915_private *dev_priv);
 void i915_gem_set_wedged(struct drm_i915_private *dev_priv);

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 306d7a805eb70643ce72d576ecc1b5e3adcf352b..28ab0beff86c08a75f23856597c789eaa324bae5 100644 (file)
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3213,7 +3213,8 @@ void i915_gem_reset_engine(struct intel_engine_cs *engine,
        engine->reset_hw(engine, request);
 }
 
-void i915_gem_reset(struct drm_i915_private *dev_priv)
+void i915_gem_reset(struct drm_i915_private *dev_priv,
+                   unsigned int stalled_mask)
 {
        struct intel_engine_cs *engine;
        enum intel_engine_id id;
@@ -3227,7 +3228,7 @@ void i915_gem_reset(struct drm_i915_private *dev_priv)
 
                i915_gem_reset_engine(engine,
                                      engine->hangcheck.active_request,
-                                     engine->hangcheck.stalled);
+                                     stalled_mask & ENGINE_MASK(id));
                ctx = fetch_and_zero(&engine->last_retired_context);
                if (ctx)
                        engine->context_unpin(engine, ctx);

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h
index ac5760673cc9f689772318676b393227571c3f97..c05b6034d718555f343a7f575e29cf9fa36c0554 100644 (file)
--- a/drivers/gpu/drm/i915/i915_gpu_error.h
+++ b/drivers/gpu/drm/i915/i915_gpu_error.h
@@ -269,6 +269,9 @@ struct i915_gpu_error {
        /** Number of times an engine has been reset */
        u32 reset_engine_count[I915_NUM_ENGINES];
 
+       /** Set of stalled engines with guilty requests, in the current reset */
+       u32 stalled_mask;
+
        /** Reason for the current *global* reset */
        const char *reason;
 

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index c2f878ace0ea21feca0b451690b04935fc1effa8..b03d18561b558cc31c99a6f162e92833476a51b4 100644 (file)
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2961,7 +2961,8 @@ static irqreturn_t gen11_irq_handler(int irq, void *arg)
 }
 
 static void i915_reset_device(struct drm_i915_private *dev_priv,
-                             const char *msg)
+                             u32 engine_mask,
+                             const char *reason)
 {
        struct i915_gpu_error *error = &dev_priv->gpu_error;
        struct kobject *kobj = &dev_priv->drm.primary->kdev->kobj;
@@ -2979,9 +2980,11 @@ static void i915_reset_device(struct drm_i915_private *dev_priv,
        i915_wedge_on_timeout(&w, dev_priv, 5*HZ) {
                intel_prepare_reset(dev_priv);
 
-               error->reason = msg;
+               error->reason = reason;
+               error->stalled_mask = engine_mask;
 
                /* Signal that locked waiters should reset the GPU */
+               smp_mb__before_atomic();
                set_bit(I915_RESET_HANDOFF, &error->flags);
                wake_up_all(&error->wait_queue);
 
@@ -2990,7 +2993,7 @@ static void i915_reset_device(struct drm_i915_private *dev_priv,
                 */
                do {
                        if (mutex_trylock(&dev_priv->drm.struct_mutex)) {
-                               i915_reset(dev_priv);
+                               i915_reset(dev_priv, engine_mask, reason);
                                mutex_unlock(&dev_priv->drm.struct_mutex);
                        }
                } while (wait_on_bit_timeout(&error->flags,
@@ -2998,6 +3001,7 @@ static void i915_reset_device(struct drm_i915_private *dev_priv,
                                             TASK_UNINTERRUPTIBLE,
                                             1));
 
+               error->stalled_mask = 0;
                error->reason = NULL;
 
                intel_finish_reset(dev_priv);
@@ -3122,7 +3126,7 @@ void i915_handle_error(struct drm_i915_private *dev_priv,
                                    TASK_UNINTERRUPTIBLE);
        }
 
-       i915_reset_device(dev_priv, msg);
+       i915_reset_device(dev_priv, engine_mask, msg);
 
        for_each_engine(engine, dev_priv, tmp) {
                clear_bit(I915_RESET_ENGINE + engine->id,

diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index a9d0bde16443f7d7884ef538e20fb864d63c0f7a..629f3e860592a9ddbdadae248a656854af1de76f 100644 (file)
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -1185,11 +1185,13 @@ static bool __i915_spin_request(const struct i915_request *rq,
 
 static bool __i915_wait_request_check_and_reset(struct i915_request *request)
 {
-       if (likely(!i915_reset_handoff(&request->i915->gpu_error)))
+       struct i915_gpu_error *error = &request->i915->gpu_error;
+
+       if (likely(!i915_reset_handoff(error)))
                return false;
 
        __set_current_state(TASK_RUNNING);
-       i915_reset(request->i915);
+       i915_reset(request->i915, error->stalled_mask, error->reason);
        return true;
 }
 

diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
index acfb4dcc9fb53a705b6c21372c31cafc4ee5b1a3..24f913f26a7b888e5ee185738a08191db0f9458f 100644 (file)
--- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
+++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
@@ -437,7 +437,7 @@ static int igt_global_reset(void *arg)
        mutex_lock(&i915->drm.struct_mutex);
        reset_count = i915_reset_count(&i915->gpu_error);
 
-       i915_reset(i915);
+       i915_reset(i915, ALL_ENGINES, NULL);
 
        if (i915_reset_count(&i915->gpu_error) == reset_count) {
                pr_err("No GPU reset recorded!\n");
@@ -881,17 +881,18 @@ static int igt_reset_engines(void *arg)
        return 0;
 }
 
-static u32 fake_hangcheck(struct i915_request *rq)
+static u32 fake_hangcheck(struct i915_request *rq, u32 mask)
 {
-       u32 reset_count;
+       struct i915_gpu_error *error = &rq->i915->gpu_error;
+       u32 reset_count = i915_reset_count(error);
 
-       rq->engine->hangcheck.stalled = true;
-       rq->engine->hangcheck.seqno = intel_engine_get_seqno(rq->engine);
+       error->stalled_mask = mask;
 
-       reset_count = i915_reset_count(&rq->i915->gpu_error);
+       /* set_bit() must be after we have setup the backchannel (mask) */
+       smp_mb__before_atomic();
+       set_bit(I915_RESET_HANDOFF, &error->flags);
 
-       set_bit(I915_RESET_HANDOFF, &rq->i915->gpu_error.flags);
-       wake_up_all(&rq->i915->gpu_error.wait_queue);
+       wake_up_all(&error->wait_queue);
 
        return reset_count;
 }
@@ -939,7 +940,7 @@ static int igt_wait_reset(void *arg)
                goto out_rq;
        }
 
-       reset_count = fake_hangcheck(rq);
+       reset_count = fake_hangcheck(rq, ALL_ENGINES);
 
        timeout = i915_request_wait(rq, I915_WAIT_LOCKED, 10);
        if (timeout < 0) {
@@ -1075,9 +1076,9 @@ static int igt_reset_queue(void *arg)
                                goto fini;
                        }
 
-                       reset_count = fake_hangcheck(prev);
+                       reset_count = fake_hangcheck(prev, ENGINE_MASK(id));
 
-                       i915_reset(i915);
+                       i915_reset(i915, ENGINE_MASK(id), NULL);
 
                        GEM_BUG_ON(test_bit(I915_RESET_HANDOFF,
                                            &i915->gpu_error.flags));
@@ -1150,7 +1151,7 @@ static int igt_handle_error(void *arg)
        if (!intel_has_reset_engine(i915))
                return 0;
 
-       if (!intel_engine_can_store_dword(i915->engine[RCS]))
+       if (!engine || !intel_engine_can_store_dword(engine))
                return 0;
 
        mutex_lock(&i915->drm.struct_mutex);
@@ -1186,10 +1187,7 @@ static int igt_handle_error(void *arg)
        /* Temporarily disable error capture */
        error = xchg(&i915->gpu_error.first_error, (void *)-1);
 
-       engine->hangcheck.stalled = true;
-       engine->hangcheck.seqno = intel_engine_get_seqno(engine);
-
-       i915_handle_error(i915, intel_engine_flag(engine), 0, NULL);
+       i915_handle_error(i915, ENGINE_MASK(engine->id), 0, NULL);
 
        xchg(&i915->gpu_error.first_error, error);