1 files changed, 176 insertions, 86 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index df6965761046..f9b54236102d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -27,6 +27,30 @@
 #include "amdgpu.h"
 #include "amdgpu_sched.h"
 
+#define to_amdgpu_ctx_entity(e) \
+        container_of((e), struct amdgpu_ctx_entity, entity)
+
+const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = {
+        [AMDGPU_HW_IP_GFX]      =       1,
+        [AMDGPU_HW_IP_COMPUTE]  =       4,
+        [AMDGPU_HW_IP_DMA]      =       2,
+        [AMDGPU_HW_IP_UVD]      =       1,
+        [AMDGPU_HW_IP_VCE]      =       1,
+        [AMDGPU_HW_IP_UVD_ENC]  =       1,
+        [AMDGPU_HW_IP_VCN_DEC]  =       1,
+        [AMDGPU_HW_IP_VCN_ENC]  =       1,
+};
+
+static int amdgput_ctx_total_num_entities(void)
+{
+        unsigned i, num_entities = 0;
+
+        for (i = 0; i < AMDGPU_HW_IP_NUM; ++i)
+                num_entities += amdgpu_ctx_num_entities[i];
+
+        return num_entities;
+}
+
 static int amdgpu_ctx_priority_permit(struct drm_file *filp,
                                       enum drm_sched_priority priority)
 {
@@ -48,6 +72,7 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev,
                            struct drm_file *filp,
                            struct amdgpu_ctx *ctx)
 {
+        unsigned num_entities = amdgput_ctx_total_num_entities();
         unsigned i, j;
         int r;
 
@@ -60,19 +85,33 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev,
 
         memset(ctx, 0, sizeof(*ctx));
         ctx->adev = adev;
-        kref_init(&ctx->refcount);
-        spin_lock_init(&ctx->ring_lock);
-        ctx->fences = kcalloc(amdgpu_sched_jobs * AMDGPU_MAX_RINGS,
+
+        ctx->fences = kcalloc(amdgpu_sched_jobs * num_entities,
                               sizeof(struct dma_fence*), GFP_KERNEL);
         if (!ctx->fences)
                 return -ENOMEM;
 
-        mutex_init(&ctx->lock);
+        ctx->entities[0] = kcalloc(num_entities,
+                                   sizeof(struct amdgpu_ctx_entity),
+                                   GFP_KERNEL);
+        if (!ctx->entities[0]) {
+                r = -ENOMEM;
+                goto error_free_fences;
+        }
 
-        for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-                ctx->rings[i].sequence = 1;
-                ctx->rings[i].fences = &ctx->fences[amdgpu_sched_jobs * i];
+        for (i = 0; i < num_entities; ++i) {
+                struct amdgpu_ctx_entity *entity = &ctx->entities[0][i];
+
+                entity->sequence = 1;
+                entity->fences = &ctx->fences[amdgpu_sched_jobs * i];
         }
+        for (i = 1; i < AMDGPU_HW_IP_NUM; ++i)
+                ctx->entities[i] = ctx->entities[i - 1] +
+                        amdgpu_ctx_num_entities[i - 1];
+
+        kref_init(&ctx->refcount);
+        spin_lock_init(&ctx->ring_lock);
+        mutex_init(&ctx->lock);
 
         ctx->reset_counter = atomic_read(&adev->gpu_reset_counter);
         ctx->reset_counter_query = ctx->reset_counter;
@@ -80,31 +119,70 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev,
         ctx->init_priority = priority;
         ctx->override_priority = DRM_SCHED_PRIORITY_UNSET;
 
-        /* create context entity for each ring */
-        for (i = 0; i < adev->num_rings; i++) {
-                struct amdgpu_ring *ring = adev->rings[i];
-                struct drm_sched_rq *rq;
+        for (i = 0; i < AMDGPU_HW_IP_NUM; ++i) {
+                struct amdgpu_ring *rings[AMDGPU_MAX_RINGS];
+                struct drm_sched_rq *rqs[AMDGPU_MAX_RINGS];
+                unsigned num_rings;
+
+                switch (i) {
+                case AMDGPU_HW_IP_GFX:
+                        rings[0] = &adev->gfx.gfx_ring[0];
+                        num_rings = 1;
+                        break;
+                case AMDGPU_HW_IP_COMPUTE:
+                        for (j = 0; j < adev->gfx.num_compute_rings; ++j)
+                                rings[j] = &adev->gfx.compute_ring[j];
+                        num_rings = adev->gfx.num_compute_rings;
+                        break;
+                case AMDGPU_HW_IP_DMA:
+                        for (j = 0; j < adev->sdma.num_instances; ++j)
+                                rings[j] = &adev->sdma.instance[j].ring;
+                        num_rings = adev->sdma.num_instances;
+                        break;
+                case AMDGPU_HW_IP_UVD:
+                        rings[0] = &adev->uvd.inst[0].ring;
+                        num_rings = 1;
+                        break;
+                case AMDGPU_HW_IP_VCE:
+                        rings[0] = &adev->vce.ring[0];
+                        num_rings = 1;
+                        break;
+                case AMDGPU_HW_IP_UVD_ENC:
+                        rings[0] = &adev->uvd.inst[0].ring_enc[0];
+                        num_rings = 1;
+                        break;
+                case AMDGPU_HW_IP_VCN_DEC:
+                        rings[0] = &adev->vcn.ring_dec;
+                        num_rings = 1;
+                        break;
+                case AMDGPU_HW_IP_VCN_ENC:
+                        rings[0] = &adev->vcn.ring_enc[0];
+                        num_rings = 1;
+                        break;
+                case AMDGPU_HW_IP_VCN_JPEG:
+                        rings[0] = &adev->vcn.ring_jpeg;
+                        num_rings = 1;
+                        break;
+                }
 
-                rq = &ring->sched.sched_rq[priority];
+                for (j = 0; j < num_rings; ++j)
+                        rqs[j] = &rings[j]->sched.sched_rq[priority];
 
-                if (ring == &adev->gfx.kiq.ring)
-                        continue;
-
-                r = drm_sched_entity_init(&ctx->rings[i].entity,
-                                          &rq, 1, &ctx->guilty);
+                for (j = 0; j < amdgpu_ctx_num_entities[i]; ++j)
+                        r = drm_sched_entity_init(&ctx->entities[i][j].entity,
+                                                  rqs, num_rings, &ctx->guilty);
                 if (r)
-                        goto failed;
+                        goto error_cleanup_entities;
         }
 
-        r = amdgpu_queue_mgr_init(adev, &ctx->queue_mgr);
-        if (r)
-                goto failed;
-
         return 0;
 
-failed:
-        for (j = 0; j < i; j++)
-                drm_sched_entity_destroy(&ctx->rings[j].entity);
+error_cleanup_entities:
+        for (i = 0; i < num_entities; ++i)
+                drm_sched_entity_destroy(&ctx->entities[0][i].entity);
+        kfree(ctx->entities[0]);
+
+error_free_fences:
         kfree(ctx->fences);
         ctx->fences = NULL;
         return r;
@@ -113,25 +191,47 @@ failed:
 static void amdgpu_ctx_fini(struct kref *ref)
 {
         struct amdgpu_ctx *ctx = container_of(ref, struct amdgpu_ctx, refcount);
+        unsigned num_entities = amdgput_ctx_total_num_entities();
         struct amdgpu_device *adev = ctx->adev;
         unsigned i, j;
 
         if (!adev)
                 return;