1 files changed, 65 insertions, 35 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 46b590af2fd2..688bef1649b5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 
 #include <drm/drm.h>
+#include <drm/drm_drv.h>
 
 #include "amdgpu.h"
 #include "amdgpu_pm.h"
@@ -81,15 +82,15 @@ MODULE_FIRMWARE(FIRMWARE_VEGA20);
 
 static void amdgpu_vce_idle_work_handler(struct work_struct *work);
 static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
-				     struct amdgpu_bo *bo,
 				     struct dma_fence **fence);
 static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 				      bool direct, struct dma_fence **fence);
 
 /**
- * amdgpu_vce_init - allocate memory, load vce firmware
+ * amdgpu_vce_sw_init - allocate memory, load vce firmware
  *
  * @adev: amdgpu_device pointer
+ * @size: size for the new BO
  *
  * First step to get VCE online, allocate memory and load the firmware
  */
@@ -178,7 +179,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
 	version_major = (ucode_version >> 20) & 0xfff;
 	version_minor = (ucode_version >> 8) & 0xfff;
 	binary_id = ucode_version & 0xff;
-	DRM_INFO("Found VCE firmware Version: %hhd.%hhd Binary ID: %hhd\n",
+	DRM_INFO("Found VCE firmware Version: %d.%d Binary ID: %d\n",
 		version_major, version_minor, binary_id);
 	adev->vce.fw_version = ((version_major << 24) | (version_minor << 16) |
 				(binary_id << 8));
@@ -203,7 +204,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
 }
 
 /**
- * amdgpu_vce_fini - free memory
+ * amdgpu_vce_sw_fini - free memory
  *
  * @adev: amdgpu_device pointer
  *
@@ -216,7 +217,6 @@ int amdgpu_vce_sw_fini(struct amdgpu_device *adev)
 	if (adev->vce.vcpu_bo == NULL)
 		return 0;
 
-	cancel_delayed_work_sync(&adev->vce.idle_work);
 	drm_sched_entity_destroy(&adev->vce.entity);
 
 	amdgpu_bo_free_kernel(&adev->vce.vcpu_bo, &adev->vce.gpu_addr,
@@ -240,12 +240,13 @@ int amdgpu_vce_sw_fini(struct amdgpu_device *adev)
 int amdgpu_vce_entity_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
-	struct drm_sched_rq *rq;
+	struct drm_gpu_scheduler *sched;
 	int r;
 
 	ring = &adev->vce.ring[0];
-	rq = &ring->sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL];
-	r = drm_sched_entity_init(&adev->vce.entity, &rq, 1, NULL);
+	sched = &ring->sched;
+	r = drm_sched_entity_init(&adev->vce.entity, DRM_SCHED_PRIORITY_NORMAL,
+				  &sched, 1, NULL);
 	if (r != 0) {
 		DRM_ERROR("Failed setting up VCE run queue.\n");
 		return r;
@@ -291,7 +292,7 @@ int amdgpu_vce_resume(struct amdgpu_device *adev)
 	void *cpu_addr;
 	const struct common_firmware_header *hdr;
 	unsigned offset;
-	int r;
+	int r, idx;
 
 	if (adev->vce.vcpu_bo == NULL)
 		return -EINVAL;
@@ -311,8 +312,12 @@ int amdgpu_vce_resume(struct amdgpu_device *adev)
 
 	hdr = (const struct common_firmware_header *)adev->vce.fw->data;
 	offset = le32_to_cpu(hdr->ucode_array_offset_bytes);
-	memcpy_toio(cpu_addr, adev->vce.fw->data + offset,
-		    adev->vce.fw->size - offset);
+
+	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
+		memcpy_toio(cpu_addr, adev->vce.fw->data + offset,
+			    adev->vce.fw->size - offset);
+		drm_dev_exit(idx);
+	}
 
 	amdgpu_bo_kunmap(adev->vce.vcpu_bo);
 
@@ -427,31 +432,40 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp)
 /**
  * amdgpu_vce_get_create_msg - generate a VCE create msg
  *
- * @adev: amdgpu_device pointer
  * @ring: ring we should submit the msg to
  * @handle: VCE session handle to use
+ * @bo: amdgpu object for which we query the offset
  * @fence: optional fence to return
  *
  * Open up a stream for HW test
  */
 static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
-				     struct amdgpu_bo *bo,
 				     struct dma_fence **fence)
 {
 	const unsigned ib_size_dw = 1024;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
+	struct amdgpu_ib ib_msg;
 	struct dma_fence *f = NULL;
 	uint64_t addr;
 	int i, r;
 
-	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job);
+	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4,
+				     AMDGPU_IB_POOL_DIRECT, &job);
 	if (r)
 		return r;
 
-	ib = &job->ibs[0];
+	memset(&ib_msg, 0, sizeof(ib_msg));
+	/* only one gpu page is needed, alloc +1 page to make addr aligned. */
+	r = amdgpu_ib_get(ring->adev, NULL, AMDGPU_GPU_PAGE_SIZE * 2,
+			  AMDGPU_IB_POOL_DIRECT,
+			  &ib_msg);
+	if (r)
+		goto err;
 
-	addr = amdgpu_bo_gpu_offset(bo);
+	ib = &job->ibs[0];
+	/* let addr point to page boundary */
+	addr = AMDGPU_GPU_PAGE_ALIGN(ib_msg.gpu_addr);
 
 	/* stitch together an VCE create msg */
 	ib->length_dw = 0;
@@ -491,6 +505,7 @@ static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 		ib->ptr[i] = 0x0;
 
 	r = amdgpu_job_submit_direct(job, ring, &f);
+	amdgpu_ib_free(ring->adev, &ib_msg, f);
 	if (r)
 		goto err;
 
@@ -507,9 +522,9 @@ err:
 /**
  * amdgpu_vce_get_destroy_msg - generate a VCE destroy msg
  *
- * @adev: amdgpu_device pointer
  * @ring: ring we should submit the msg to
  * @handle: VCE session handle to use
+ * @direct: direct or delayed pool
  * @fence: optional fence to return
  *
  * Close up a stream for HW test or if userspace failed to do so
@@ -523,7 +538,9 @@ static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 	struct dma_fence *f = NULL;
 	int i, r;
 
-	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job);
+	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4,
+				     direct ? AMDGPU_IB_POOL_DIRECT :
+				     AMDGPU_IB_POOL_DELAYED, &job);
 	if (r)
 		return r;
 
@@ -569,9 +586,10 @@ err:
 }
 
 /**
- * amdgpu_vce_cs_validate_bo - make sure not to cross 4GB boundary
+ * amdgpu_vce_validate_bo - make sure not to cross 4GB boundary
  *
  * @p: parser context
+ * @ib_idx: indirect buffer to use
  * @lo: address of lower dword
  * @hi: address of higher dword
  * @size: minimum size
@@ -621,9 +639,11 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, uint32_t ib_idx,
  * amdgpu_vce_cs_reloc - command submission relocation
  *
  * @p: parser context
+ * @ib_idx: indirect buffer to use
  * @lo: address of lower dword
  * @hi: address of higher dword
  * @size: minimum size
+ * @index: bs/fb index
  *
  * Patch relocation inside command stream with real buffer address
  */
@@ -651,7 +671,7 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx,
 
 	if ((addr + (uint64_t)size) >
 	    (mapping->last + 1) * AMDGPU_GPU_PAGE_SIZE) {
-		DRM_ERROR("BO to small for addr 0x%010Lx %d %d\n",
+		DRM_ERROR("BO too small for addr 0x%010Lx %d %d\n",
 			  addr, lo, hi);
 		return -EINVAL;
 	}
@@ -707,10 +727,10 @@ static int amdgpu_vce_validate_handle(struct amdgpu_cs_parser *p,
 }
 
 /**
- * amdgpu_vce_cs_parse - parse and validate the command stream
+ * amdgpu_vce_ring_parse_cs - parse and validate the command stream
  *
  * @p: parser context
- *
+ * @ib_idx: indirect buffer to use
  */
 int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx)
 {
@@ -943,10 +963,10 @@ out:
 }
 
 /**
- * amdgpu_vce_cs_parse_vm - parse the command stream in VM mode
+ * amdgpu_vce_ring_parse_cs_vm - parse the command stream in VM mode
  *
  * @p: parser context
- *
+ * @ib_idx: indirect buffer to use
  */
 int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx)
 {
@@ -1036,7 +1056,9 @@ out:
  * amdgpu_vce_ring_emit_ib - execute indirect buffer
  *
  * @ring: engine to use
+ * @job: job to retrieve vmid from
  * @ib: the IB to execute
+ * @flags: unused
  *
  */
 void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring,
@@ -1054,7 +1076,9 @@ void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring,
  * amdgpu_vce_ring_emit_fence - add a fence command to the ring
  *
  * @ring: engine to use
- * @fence: the fence
+ * @addr: address
+ * @seq: sequence number
+ * @flags: fence related flags
  *
  */
 void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
@@ -1112,25 +1136,19 @@ int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring)
  * amdgpu_vce_ring_test_ib - test if VCE IBs are working
  *
  * @ring: the engine to test on
+ * @timeout: timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT
  *
  */
 int amdgpu_vce_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 {
 	struct dma_fence *fence = NULL;
-	struct amdgpu_bo *bo = NULL;
 	long r;
 
 	/* skip vce ring1/2 ib test for now, since it's not reliable */
 	if (ring != &ring->adev->vce.ring[0])
 		return 0;
 
-	r = amdgpu_bo_create_reserved(ring->adev, 512, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
-				      &bo, NULL, NULL);
-	if (r)
-		return r;
-
-	r = amdgpu_vce_get_create_msg(ring, 1, bo, NULL);
+	r = amdgpu_vce_get_create_msg(ring, 1, NULL);
 	if (r)
 		goto error;
 
@@ -1146,7 +1164,19 @@ int amdgpu_vce_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 
 error:
 	dma_fence_put(fence);
-	amdgpu_bo_unreserve(bo);
-	amdgpu_bo_unref(&bo);
 	return r;
 }
+
+enum amdgpu_ring_priority_level amdgpu_vce_get_ring_prio(int ring)
+{
+	switch(ring) {
+	case 0:
+		return AMDGPU_RING_PRIO_0;
+	case 1:
+		return AMDGPU_RING_PRIO_1;
+	case 2:
+		return AMDGPU_RING_PRIO_2;
+	default:
+		return AMDGPU_RING_PRIO_0;
+	}
+}