drm/i915: Separate out reset flags from the reset counter

In preparation for introducing a per-engine reset, we can first separate
the mixing of the reset state from the global reset counter.

The loss of atomicity in updating the reset state poses a small problem
for handling the waiters. For requests, this is solved by advancing the
seqno so that a waiter waking up after the reset knows the request is
complete. For pending flips, we still rely on the increment of the
global reset epoch (as well as the reset-in-progress flag) to signify
when the hardware was reset.

The advantage, now that we do not inspect the reset state during reset
itself i.e. we no longer emit requests during reset, is that we can use
the atomic updates of the state flags to ensure that only one reset
worker is active.

v2: Mika spotted that I transformed the i915_gem_wait_for_error() wakeup
into a waiter wakeup.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470414607-32453-6-git-send-email-arun.siluvery@linux.intel.com
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-7-chris@chris-wilson.co.uk

1 files changed, 14 insertions, 32 deletions

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index f39bede7664c..dced7e72b625 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1405,9 +1405,10 @@ struct i915_gpu_error {
 	 * State variable controlling the reset flow and count
 	 *
 	 * This is a counter which gets incremented when reset is triggered,
-	 * and again when reset has been handled. So odd values (lowest bit set)
-	 * means that reset is in progress and even values that
-	 * (reset_counter >> 1):th reset was successfully completed.
+	 *
+	 * Before the reset commences, the I915_RESET_IN_PROGRESS bit is set
+	 * meaning that any waiters holding onto the struct_mutex should
+	 * relinquish the lock immediately in order for the reset to start.
 	 *
 	 * If reset is not completed succesfully, the I915_WEDGE bit is
 	 * set meaning that hardware is terminally sour and there is no
@@ -1422,10 +1423,11 @@ struct i915_gpu_error {
 	 * naturally enforces the correct ordering between the bail-out of the
 	 * waiter and the gpu reset work code.
 	 */
-	atomic_t reset_counter;
+	unsigned long reset_count;
 
-#define I915_RESET_IN_PROGRESS_FLAG	1
-#define I915_WEDGED			(1 << 31)
+	unsigned long flags;
+#define I915_RESET_IN_PROGRESS	0
+#define I915_WEDGED		(BITS_PER_LONG - 1)
 
 	/**
 	 * Waitqueue to signal when a hang is detected. Used to for waiters
@@ -3241,44 +3243,24 @@ i915_gem_find_active_request(struct intel_engine_cs *engine);
 
 void i915_gem_retire_requests(struct drm_i915_private *dev_priv);
 
-static inline u32 i915_reset_counter(struct i915_gpu_error *error)
-{
-	return atomic_read(&error->reset_counter);
-}
-
-static inline bool __i915_reset_in_progress(u32 reset)
-{
-	return unlikely(reset & I915_RESET_IN_PROGRESS_FLAG);
-}
-
-static inline bool __i915_reset_in_progress_or_wedged(u32 reset)
-{
-	return unlikely(reset & (I915_RESET_IN_PROGRESS_FLAG | I915_WEDGED));
-}
-
-static inline bool __i915_terminally_wedged(u32 reset)
-{
-	return unlikely(reset & I915_WEDGED);
-}
-
 static inline bool i915_reset_in_progress(struct i915_gpu_error *error)
 {
-	return __i915_reset_in_progress(i915_reset_counter(error));
+	return unlikely(test_bit(I915_RESET_IN_PROGRESS, &error->flags));
 }
 
-static inline bool i915_reset_in_progress_or_wedged(struct i915_gpu_error *error)
+static inline bool i915_terminally_wedged(struct i915_gpu_error *error)
 {
-	return __i915_reset_in_progress_or_wedged(i915_reset_counter(error));
+	return unlikely(test_bit(I915_WEDGED, &error->flags));
 }
 
-static inline bool i915_terminally_wedged(struct i915_gpu_error *error)
+static inline bool i915_reset_in_progress_or_wedged(struct i915_gpu_error *error)
 {
-	return __i915_terminally_wedged(i915_reset_counter(error));
+	return i915_reset_in_progress(error) | i915_terminally_wedged(error);
 }
 
 static inline u32 i915_reset_count(struct i915_gpu_error *error)
 {
-	return ((i915_reset_counter(error) & ~I915_WEDGED) + 1) / 2;
+	return READ_ONCE(error->reset_count);
 }
 
 void i915_gem_reset(struct drm_device *dev);