drm/amdgpu: avoid KIQ clashing with compute or KFD queues v2

Instead of picking an arbitrary queue for KIQ, search for one according
to policy. The queue must be unused.

Also report the KIQ as an unavailable resource to KFD.

In testing I ran into KCQ initialization issues when using pipes 2/3 of
MEC2 for the KIQ. Therefore the policy disallows grabbing one of these.

v2: fix (ring.me + 1) to (ring.me -1) in amdgpu_amdkfd_device_init
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -1803,8 +1803,8 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 		return NULL;
 }
 
-static inline bool amdgpu_is_mec_queue_enabled(struct amdgpu_device *adev,
-						int mec, int pipe, int queue)
+static inline int amdgpu_queue_to_bit(struct amdgpu_device *adev,
+				      int mec, int pipe, int queue)
 {
 	int bit = 0;
 
@@ -1813,7 +1813,24 @@ static inline bool amdgpu_is_mec_queue_enabled(struct amdgpu_device *adev,
 	bit += pipe * adev->gfx.mec.num_queue_per_pipe;
 	bit += queue;
 
-	return test_bit(bit, adev->gfx.mec.queue_bitmap);
+	return bit;
+}
+
+static inline void amdgpu_bit_to_queue(struct amdgpu_device *adev, int bit,
+				       int *mec, int *pipe, int *queue)
+{
+	*queue = bit % adev->gfx.mec.num_queue_per_pipe;
+	*pipe = (bit / adev->gfx.mec.num_queue_per_pipe)
+		% adev->gfx.mec.num_pipe_per_mec;
+	*mec = (bit / adev->gfx.mec.num_queue_per_pipe)
+	       / adev->gfx.mec.num_pipe_per_mec;
+
+}
+static inline bool amdgpu_is_mec_queue_enabled(struct amdgpu_device *adev,
+					       int mec, int pipe, int queue)
+{
+	return test_bit(amdgpu_queue_to_bit(adev, mec, pipe, queue),
+			adev->gfx.mec.queue_bitmap);
 }
 
 /*

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c

@@ -111,6 +111,14 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 				  adev->gfx.mec.queue_bitmap,
 				  KGD_MAX_QUEUES);
 
+		/* remove the KIQ bit as well */
+		if (adev->gfx.kiq.ring.ready)
+			clear_bit(amdgpu_queue_to_bit(adev,
+						      adev->gfx.kiq.ring.me - 1,
+						      adev->gfx.kiq.ring.pipe,
+						      adev->gfx.kiq.ring.queue),
+				  gpu_resources.queue_bitmap);
+
 		/* According to linux/bitmap.h we shouldn't use bitmap_clear if
 		 * nbits is not compile time constant */
 		last_valid_bit = adev->gfx.mec.num_mec

drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c

@@ -1378,6 +1378,37 @@ static void gfx_v8_0_mec_fini(struct amdgpu_device *adev)
 	}
 }
 
+static int gfx_v8_0_kiq_acquire(struct amdgpu_device *adev,
+				 struct amdgpu_ring *ring)
+{
+	int queue_bit;
+	int mec, pipe, queue;
+
+	queue_bit = adev->gfx.mec.num_mec
+		    * adev->gfx.mec.num_pipe_per_mec
+		    * adev->gfx.mec.num_queue_per_pipe;
+
+	while (queue_bit-- >= 0) {
+		if (test_bit(queue_bit, adev->gfx.mec.queue_bitmap))
+			continue;
+
+		amdgpu_bit_to_queue(adev, queue_bit, &mec, &pipe, &queue);
+
+		/* Using pipes 2/3 from MEC 2 seems cause problems */
+		if (mec == 1 && pipe > 1)
+			continue;
+
+		ring->me = mec + 1;
+		ring->pipe = pipe;
+		ring->queue = queue;
+
+		return 0;
+	}
+
+	dev_err(adev->dev, "Failed to find a queue for KIQ\n");
+	return -EINVAL;
+}
+
 static int gfx_v8_0_kiq_init_ring(struct amdgpu_device *adev,
 				  struct amdgpu_ring *ring,
 				  struct amdgpu_irq_src *irq)
@@ -1395,15 +1426,11 @@ static int gfx_v8_0_kiq_init_ring(struct amdgpu_device *adev,
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
 	ring->doorbell_index = AMDGPU_DOORBELL_KIQ;
-	if (adev->gfx.mec2_fw) {
-		ring->me = 2;
-		ring->pipe = 0;
-	} else {
-		ring->me = 1;
-		ring->pipe = 1;
-	}
 
-	ring->queue = 0;
+	r = gfx_v8_0_kiq_acquire(adev, ring);
+	if (r)
+		return r;
+
 	ring->eop_gpu_addr = kiq->eop_gpu_addr;
 	sprintf(ring->name, "kiq %d.%d.%d", ring->me, ring->pipe, ring->queue);
 	r = amdgpu_ring_init(adev, ring, 1024,

drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c

@@ -1031,6 +1031,37 @@ static int gfx_v9_0_kiq_init(struct amdgpu_device *adev)
 	return 0;
 }
 
+static int gfx_v9_0_kiq_acquire(struct amdgpu_device *adev,
+				 struct amdgpu_ring *ring)
+{
+	int queue_bit;
+	int mec, pipe, queue;
+
+	queue_bit = adev->gfx.mec.num_mec
+		    * adev->gfx.mec.num_pipe_per_mec
+		    * adev->gfx.mec.num_queue_per_pipe;
+
+	while (queue_bit-- >= 0) {
+		if (test_bit(queue_bit, adev->gfx.mec.queue_bitmap))
+			continue;
+
+		amdgpu_bit_to_queue(adev, queue_bit, &mec, &pipe, &queue);
+
+		/* Using pipes 2/3 from MEC 2 seems cause problems */
+		if (mec == 1 && pipe > 1)
+			continue;
+
+		ring->me = mec + 1;
+		ring->pipe = pipe;
+		ring->queue = queue;
+
+		return 0;
+	}
+
+	dev_err(adev->dev, "Failed to find a queue for KIQ\n");
+	return -EINVAL;
+}
+
 static int gfx_v9_0_kiq_init_ring(struct amdgpu_device *adev,
 				  struct amdgpu_ring *ring,
 				  struct amdgpu_irq_src *irq)
@@ -1048,13 +1079,10 @@ static int gfx_v9_0_kiq_init_ring(struct amdgpu_device *adev,
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
 	ring->doorbell_index = AMDGPU_DOORBELL_KIQ;
-	if (adev->gfx.mec2_fw) {
-		ring->me = 2;
-		ring->pipe = 0;
-	} else {
-		ring->me = 1;
-		ring->pipe = 1;
-	}
+
+	r = gfx_v9_0_kiq_acquire(adev, ring);
+	if (r)
+		return r;
 
 	ring->queue = 0;
 	ring->eop_gpu_addr = kiq->eop_gpu_addr;