drm/amdgpu: move VM fields into job

They are the same for all IBs.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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

@@ -283,7 +283,8 @@ struct amdgpu_ring_funcs {
 	int (*parse_cs)(struct amdgpu_cs_parser *p, uint32_t ib_idx);
 	/* command emit functions */
 	void (*emit_ib)(struct amdgpu_ring *ring,
-			struct amdgpu_ib *ib, bool ctx_switch);
+			struct amdgpu_ib *ib,
+			unsigned vm_id, bool ctx_switch);
 	void (*emit_fence)(struct amdgpu_ring *ring, uint64_t addr,
 			   uint64_t seq, unsigned flags);
 	void (*emit_pipeline_sync)(struct amdgpu_ring *ring);
@@ -741,11 +742,6 @@ struct amdgpu_ib {
 	uint64_t			gpu_addr;
 	uint32_t			*ptr;
 	struct amdgpu_user_fence        *user;
-	unsigned			vm_id;
-	uint64_t			vm_pd_addr;
-	uint32_t			gds_base, gds_size;
-	uint32_t			gws_base, gws_size;
-	uint32_t			oa_base, oa_size;
 	uint32_t			flags;
 	/* resulting sequence number */
 	uint64_t			sequence;
@@ -1262,6 +1258,11 @@ struct amdgpu_job {
 	uint32_t		num_ibs;
 	void			*owner;
 	uint64_t		ctx;
+	unsigned		vm_id;
+	uint64_t		vm_pd_addr;
+	uint32_t		gds_base, gds_size;
+	uint32_t		gws_base, gws_size;
+	uint32_t		oa_base, oa_size;
 	struct amdgpu_user_fence uf;
 };
 #define to_amdgpu_job(sched_job)		\
@@ -2221,7 +2222,7 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 #define amdgpu_ring_get_rptr(r) (r)->funcs->get_rptr((r))
 #define amdgpu_ring_get_wptr(r) (r)->funcs->get_wptr((r))
 #define amdgpu_ring_set_wptr(r) (r)->funcs->set_wptr((r))
-#define amdgpu_ring_emit_ib(r, ib, c) (r)->funcs->emit_ib((r), (ib), (c))
+#define amdgpu_ring_emit_ib(r, ib, vm_id, c) (r)->funcs->emit_ib((r), (ib), (vm_id), (c))
 #define amdgpu_ring_emit_pipeline_sync(r) (r)->funcs->emit_pipeline_sync((r))
 #define amdgpu_ring_emit_vm_flush(r, vmid, addr) (r)->funcs->emit_vm_flush((r), (vmid), (addr))
 #define amdgpu_ring_emit_fence(r, addr, seq, flags) (r)->funcs->emit_fence((r), (addr), (seq), (flags))

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

@@ -473,6 +473,9 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 		goto error_validate;
 
 	if (p->bo_list) {
+		struct amdgpu_bo *gds = p->bo_list->gds_obj;
+		struct amdgpu_bo *gws = p->bo_list->gws_obj;
+		struct amdgpu_bo *oa = p->bo_list->oa_obj;
 		struct amdgpu_vm *vm = &fpriv->vm;
 		unsigned i;
 
@@ -481,6 +484,19 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 
 			p->bo_list->array[i].bo_va = amdgpu_vm_bo_find(vm, bo);
 		}
+
+		if (gds) {
+			p->job->gds_base = amdgpu_bo_gpu_offset(gds);
+			p->job->gds_size = amdgpu_bo_size(gds);
+		}
+		if (gws) {
+			p->job->gws_base = amdgpu_bo_gpu_offset(gws);
+			p->job->gws_size = amdgpu_bo_size(gws);
+		}
+		if (oa) {
+			p->job->oa_base = amdgpu_bo_gpu_offset(oa);
+			p->job->oa_size = amdgpu_bo_size(oa);
+		}
 	}
 
 error_validate:
@@ -744,26 +760,6 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev,
 		j++;
 	}
 
-	/* add GDS resources to first IB */
-	if (parser->bo_list) {
-		struct amdgpu_bo *gds = parser->bo_list->gds_obj;
-		struct amdgpu_bo *gws = parser->bo_list->gws_obj;
-		struct amdgpu_bo *oa = parser->bo_list->oa_obj;
-		struct amdgpu_ib *ib = &parser->job->ibs[0];
-
-		if (gds) {
-			ib->gds_base = amdgpu_bo_gpu_offset(gds);
-			ib->gds_size = amdgpu_bo_size(gds);
-		}
-		if (gws) {
-			ib->gws_base = amdgpu_bo_gpu_offset(gws);
-			ib->gws_size = amdgpu_bo_size(gws);
-		}
-		if (oa) {
-			ib->oa_base = amdgpu_bo_gpu_offset(oa);
-			ib->oa_size = amdgpu_bo_size(oa);
-		}
-	}
 	/* wrap the last IB with user fence */
 	if (parser->job->uf.bo) {
 		struct amdgpu_ib *ib = &parser->job->ibs[parser->job->num_ibs - 1];

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

@@ -74,8 +74,6 @@ int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			ib->gpu_addr = amdgpu_sa_bo_gpu_addr(ib->sa_bo);
 	}
 
-	ib->vm_id = 0;
-
 	return 0;
 }
 
@@ -147,7 +145,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		return -EINVAL;
 	}
 
-	if (vm && !ibs->vm_id) {
+	if (vm && !job->vm_id) {
 		dev_err(adev->dev, "VM IB without ID\n");
 		return -EINVAL;
 	}
@@ -162,10 +160,10 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		patch_offset = amdgpu_ring_init_cond_exec(ring);
 
 	if (vm) {
-		r = amdgpu_vm_flush(ring, ib->vm_id, ib->vm_pd_addr,
-				    ib->gds_base, ib->gds_size,
-				    ib->gws_base, ib->gws_size,
-				    ib->oa_base, ib->oa_size);
+		r = amdgpu_vm_flush(ring, job->vm_id, job->vm_pd_addr,
+				    job->gds_base, job->gds_size,
+				    job->gws_base, job->gws_size,
+				    job->oa_base, job->oa_size);
 		if (r) {
 			amdgpu_ring_undo(ring);
 			return r;
@@ -187,7 +185,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && skip_preamble)
 			continue;
 
-		amdgpu_ring_emit_ib(ring, ib, need_ctx_switch);
+		amdgpu_ring_emit_ib(ring, ib, job ? job->vm_id : 0,
+				    need_ctx_switch);
 		need_ctx_switch = false;
 	}
 
@@ -197,8 +196,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	r = amdgpu_fence_emit(ring, &hwf);
 	if (r) {
 		dev_err(adev->dev, "failed to emit fence (%d)\n", r);
-		if (ib->vm_id)
-			amdgpu_vm_reset_id(adev, ib->vm_id);
+		if (job && job->vm_id)
+			amdgpu_vm_reset_id(adev, job->vm_id);
 		amdgpu_ring_undo(ring);
 		return r;
 	}

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

@@ -142,23 +142,15 @@ static struct fence *amdgpu_job_dependency(struct amd_sched_job *sched_job)
 
 	struct fence *fence = amdgpu_sync_get_fence(&job->sync);
 
-	if (fence == NULL && vm && !job->ibs->vm_id) {
+	if (fence == NULL && vm && !job->vm_id) {
 		struct amdgpu_ring *ring = job->ring;
-		unsigned i, vm_id;
-		uint64_t vm_pd_addr;
 		int r;
 
 		r = amdgpu_vm_grab_id(vm, ring, &job->sync,
 				      &job->base.s_fence->base,
-				      &vm_id, &vm_pd_addr);
+				      &job->vm_id, &job->vm_pd_addr);
 		if (r)
 			DRM_ERROR("Error getting VM ID (%d)\n", r);
-		else {
-			for (i = 0; i < job->num_ibs; ++i) {
-				job->ibs[i].vm_id = vm_id;
-				job->ibs[i].vm_pd_addr = vm_pd_addr;
-			}
-		}
 
 		fence = amdgpu_sync_get_fence(&job->sync);
 	}

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

@@ -762,7 +762,8 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx)
  * @ib: the IB to execute
  *
  */
-void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib, bool ctx_switch)
+void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib,
+			     unsigned vm_id, bool ctx_switch)
 {
 	amdgpu_ring_write(ring, VCE_CMD_IB);
 	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));

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

@@ -34,7 +34,8 @@ int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 			       bool direct, struct fence **fence);
 void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp);
 int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx);
-void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib, bool ctx_switch);
+void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib,
+			     unsigned vm_id, bool ctx_switch);
 void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
 				unsigned flags);
 int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring);

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

@@ -210,9 +210,10 @@ static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
  * Schedule an IB in the DMA ring (CIK).
  */
 static void cik_sdma_ring_emit_ib(struct amdgpu_ring *ring,
-			   struct amdgpu_ib *ib, bool ctx_switch)
+				  struct amdgpu_ib *ib,
+				  unsigned vm_id, bool ctx_switch)
 {
-	u32 extra_bits = ib->vm_id & 0xf;
+	u32 extra_bits = vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 4)

drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c

@@ -2030,7 +2030,8 @@ static void gfx_v7_0_ring_emit_fence_compute(struct amdgpu_ring *ring,
  * on the gfx ring for execution by the GPU.
  */
 static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
-				      struct amdgpu_ib *ib, bool ctx_switch)
+				      struct amdgpu_ib *ib,
+				      unsigned vm_id, bool ctx_switch)
 {
 	u32 header, control = 0;
 	u32 next_rptr = ring->wptr + 5;
@@ -2056,7 +2057,7 @@ static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 	else
 		header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw | (ib->vm_id << 24);
+	control |= ib->length_dw | (vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,
@@ -2069,7 +2070,8 @@ static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 }
 
 static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
-					  struct amdgpu_ib *ib, bool ctx_switch)
+					  struct amdgpu_ib *ib,
+					  unsigned vm_id, bool ctx_switch)
 {
 	u32 header, control = 0;
 	u32 next_rptr = ring->wptr + 5;
@@ -2084,7 +2086,7 @@ static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 
 	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw | (ib->vm_id << 24);
+	control |= ib->length_dw | (vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,

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

@@ -5646,7 +5646,8 @@ static void gfx_v8_0_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
 }
 
 static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
-				      struct amdgpu_ib *ib, bool ctx_switch)
+				      struct amdgpu_ib *ib,
+				      unsigned vm_id, bool ctx_switch)
 {
 	u32 header, control = 0;
 	u32 next_rptr = ring->wptr + 5;
@@ -5672,7 +5673,7 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 	else
 		header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw | (ib->vm_id << 24);
+	control |= ib->length_dw | (vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,
@@ -5685,7 +5686,8 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 }
 
 static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
-					  struct amdgpu_ib *ib, bool ctx_switch)
+					  struct amdgpu_ib *ib,
+					  unsigned vm_id, bool ctx_switch)
 {
 	u32 header, control = 0;
 	u32 next_rptr = ring->wptr + 5;
@@ -5701,7 +5703,7 @@ static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 
 	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw | (ib->vm_id << 24);
+	control |= ib->length_dw | (vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,

drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c

@@ -242,9 +242,10 @@ static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
  * Schedule an IB in the DMA ring (VI).
  */
 static void sdma_v2_4_ring_emit_ib(struct amdgpu_ring *ring,
-				   struct amdgpu_ib *ib, bool ctx_switch)
+				   struct amdgpu_ib *ib,
+				   unsigned vm_id, bool ctx_switch)
 {
-	u32 vmid = ib->vm_id & 0xf;
+	u32 vmid = vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 2)

drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c

@@ -400,9 +400,10 @@ static void sdma_v3_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
  * Schedule an IB in the DMA ring (VI).
  */
 static void sdma_v3_0_ring_emit_ib(struct amdgpu_ring *ring,
-				   struct amdgpu_ib *ib, bool ctx_switch)
+				   struct amdgpu_ib *ib,
+				   unsigned vm_id, bool ctx_switch)
 {
-	u32 vmid = ib->vm_id & 0xf;
+	u32 vmid = vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 2)

drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c

@@ -489,7 +489,8 @@ static int uvd_v4_2_ring_test_ring(struct amdgpu_ring *ring)
  * Write ring commands to execute the indirect buffer
  */
 static void uvd_v4_2_ring_emit_ib(struct amdgpu_ring *ring,
-				  struct amdgpu_ib *ib, bool ctx_switch)
+				  struct amdgpu_ib *ib,
+				  unsigned vm_id, bool ctx_switch)
 {
 	amdgpu_ring_write(ring, PACKET0(mmUVD_RBC_IB_BASE, 0));
 	amdgpu_ring_write(ring, ib->gpu_addr);

drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c

@@ -539,7 +539,8 @@ static int uvd_v5_0_ring_test_ring(struct amdgpu_ring *ring)
  * Write ring commands to execute the indirect buffer
  */
 static void uvd_v5_0_ring_emit_ib(struct amdgpu_ring *ring,
-				  struct amdgpu_ib *ib, bool ctx_switch)
+				  struct amdgpu_ib *ib,
+				  unsigned vm_id, bool ctx_switch)
 {
 	amdgpu_ring_write(ring, PACKET0(mmUVD_LMI_RBC_IB_64BIT_BAR_LOW, 0));
 	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));

drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c

@@ -631,7 +631,8 @@ static int uvd_v6_0_ring_test_ring(struct amdgpu_ring *ring)
  * Write ring commands to execute the indirect buffer
  */
 static void uvd_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
-				  struct amdgpu_ib *ib, bool ctx_switch)
+				  struct amdgpu_ib *ib,
+				  unsigned vm_id, bool ctx_switch)
 {
 	amdgpu_ring_write(ring, PACKET0(mmUVD_LMI_RBC_IB_64BIT_BAR_LOW, 0));
 	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));