Merge drm-intel-next-queued into gvt-next

Backmerge to pull in
https://patchwork.freedesktop.org/patch/353621/?series=73544&rev=1

Signed-off-by: Zhenyu Wang <[email protected]>

1 files changed, 57 insertions, 166 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 7a6beb2e7c4e..f96464e2c157 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -28,19 +28,10 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 
-#include <drm/drm.h>
-
 #include "amdgpu.h"
 #include "amdgpu_pm.h"
 #include "amdgpu_vcn.h"
 #include "soc15d.h"
-#include "soc15_common.h"
-
-#include "vcn/vcn_1_0_offset.h"
-#include "vcn/vcn_1_0_sh_mask.h"
-
-/* 1 second timeout */
-#define VCN_IDLE_TIMEOUT	msecs_to_jiffies(1000)
 
 /* Firmware Names */
 #define FIRMWARE_RAVEN		"amdgpu/raven_vcn.bin"
@@ -84,6 +75,9 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 		break;
 	case CHIP_ARCTURUS:
 		fw_name = FIRMWARE_ARCTURUS;
+		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
+		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
+			adev->vcn.indirect_sram = true;
 		break;
 	case CHIP_RENOIR:
 		fw_name = FIRMWARE_RENOIR;
@@ -174,15 +168,15 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 			dev_err(adev->dev, "(%d) failed to allocate vcn bo\n", r);
 			return r;
 		}
-	}
 
-	if (adev->vcn.indirect_sram) {
-		r = amdgpu_bo_create_kernel(adev, 64 * 2 * 4, PAGE_SIZE,
-			    AMDGPU_GEM_DOMAIN_VRAM, &adev->vcn.dpg_sram_bo,
-			    &adev->vcn.dpg_sram_gpu_addr, &adev->vcn.dpg_sram_cpu_addr);
-		if (r) {
-			dev_err(adev->dev, "(%d) failed to allocate DPG bo\n", r);
-			return r;
+		if (adev->vcn.indirect_sram) {
+			r = amdgpu_bo_create_kernel(adev, 64 * 2 * 4, PAGE_SIZE,
+					AMDGPU_GEM_DOMAIN_VRAM, &adev->vcn.inst[i].dpg_sram_bo,
+					&adev->vcn.inst[i].dpg_sram_gpu_addr, &adev->vcn.inst[i].dpg_sram_cpu_addr);
+			if (r) {
+				dev_err(adev->dev, "VCN %d (%d) failed to allocate DPG bo\n", i, r);
+				return r;
+			}
 		}
 	}
 
@@ -193,15 +187,16 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev)
 {
 	int i, j;
 
-	if (adev->vcn.indirect_sram) {
-		amdgpu_bo_free_kernel(&adev->vcn.dpg_sram_bo,
-				      &adev->vcn.dpg_sram_gpu_addr,
-				      (void **)&adev->vcn.dpg_sram_cpu_addr);
-	}
+	cancel_delayed_work_sync(&adev->vcn.idle_work);
 
 	for (j = 0; j < adev->vcn.num_vcn_inst; ++j) {
 		if (adev->vcn.harvest_config & (1 << j))
 			continue;
+		if (adev->vcn.indirect_sram) {
+			amdgpu_bo_free_kernel(&adev->vcn.inst[j].dpg_sram_bo,
+						  &adev->vcn.inst[j].dpg_sram_gpu_addr,
+						  (void **)&adev->vcn.inst[j].dpg_sram_cpu_addr);
+		}
 		kvfree(adev->vcn.inst[j].saved_bo);
 
 		amdgpu_bo_free_kernel(&adev->vcn.inst[j].vcpu_bo,
@@ -212,8 +207,6 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev)
 
 		for (i = 0; i < adev->vcn.num_enc_rings; ++i)
 			amdgpu_ring_fini(&adev->vcn.inst[j].ring_enc[i]);
-
-		amdgpu_ring_fini(&adev->vcn.inst[j].ring_jpeg);
 	}
 
 	release_firmware(adev->vcn.fw);
@@ -294,6 +287,7 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 	for (j = 0; j < adev->vcn.num_vcn_inst; ++j) {
 		if (adev->vcn.harvest_config & (1 << j))
 			continue;
+
 		for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
 			fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_enc[i]);
 		}
@@ -306,26 +300,17 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 			else
 				new_state.fw_based = VCN_DPG_STATE__UNPAUSE;
 
-			if (amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_jpeg))
-				new_state.jpeg = VCN_DPG_STATE__PAUSE;
-			else
-				new_state.jpeg = VCN_DPG_STATE__UNPAUSE;
-
-			adev->vcn.pause_dpg_mode(adev, &new_state);
+			adev->vcn.pause_dpg_mode(adev, j, &new_state);
 		}
 
-		fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_jpeg);
 		fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_dec);
 		fences += fence[j];
 	}
 
 	if (fences == 0) {
 		amdgpu_gfx_off_ctrl(adev, true);
-		if (adev->asic_type < CHIP_ARCTURUS && adev->pm.dpm_enabled)
-			amdgpu_dpm_enable_uvd(adev, false);
-		else
-			amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
-							       AMD_PG_STATE_GATE);
+		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
+		       AMD_PG_STATE_GATE);
 	} else {
 		schedule_delayed_work(&adev->vcn.idle_work, VCN_IDLE_TIMEOUT);
 	}
@@ -338,11 +323,8 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring)
 
 	if (set_clocks) {
 		amdgpu_gfx_off_ctrl(adev, false);
-		if (adev->asic_type < CHIP_ARCTURUS && adev->pm.dpm_enabled)
-			amdgpu_dpm_enable_uvd(adev, true);
-		else
-			amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
-							       AMD_PG_STATE_UNGATE);
+		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
+		       AMD_PG_STATE_UNGATE);
 	}
 
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)	{
@@ -358,17 +340,10 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring)
 		else
 			new_state.fw_based = VCN_DPG_STATE__UNPAUSE;
 
-		if (amdgpu_fence_count_emitted(&adev->vcn.inst[ring->me].ring_jpeg))
-			new_state.jpeg = VCN_DPG_STATE__PAUSE;
-		else
-			new_state.jpeg = VCN_DPG_STATE__UNPAUSE;
-
 		if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC)
 			new_state.fw_based = VCN_DPG_STATE__PAUSE;
-		else if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG)
-			new_state.jpeg = VCN_DPG_STATE__PAUSE;
 
-		adev->vcn.pause_dpg_mode(adev, &new_state);
+		adev->vcn.pause_dpg_mode(adev, ring->me, &new_state);
 	}
 }
 
@@ -518,9 +493,14 @@ static int amdgpu_vcn_dec_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 
 int amdgpu_vcn_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 {
+	struct amdgpu_device *adev = ring->adev;
 	struct dma_fence *fence;
 	long r;
 
+	/* temporarily disable ib test for sriov */
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	r = amdgpu_vcn_dec_get_create_msg(ring, 1, NULL);
 	if (r)
 		goto error;
@@ -569,13 +549,14 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring)
 }
 
 static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
-			      struct dma_fence **fence)
+					 struct amdgpu_bo *bo,
+					 struct dma_fence **fence)
 {
 	const unsigned ib_size_dw = 16;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	struct dma_fence *f = NULL;
-	uint64_t dummy;
+	uint64_t addr;
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job);
@@ -583,14 +564,14 @@ static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 		return r;
 
 	ib = &job->ibs[0];
-	dummy = ib->gpu_addr + 1024;
+	addr = amdgpu_bo_gpu_offset(bo);
 
 	ib->length_dw = 0;
 	ib->ptr[ib->length_dw++] = 0x00000018;
 	ib->ptr[ib->length_dw++] = 0x00000001; /* session info */
 	ib->ptr[ib->length_dw++] = handle;
-	ib->ptr[ib->length_dw++] = upper_32_bits(dummy);
-	ib->ptr[ib->length_dw++] = dummy;
+	ib->ptr[ib->length_dw++] = upper_32_bits(addr);
+	ib->ptr[ib->length_dw++] = addr;
 	ib->ptr[ib->length_dw++] = 0x0000000b;
 
 	ib->ptr[ib->length_dw++] = 0x00000014;
@@ -621,13 +602,14 @@ err:
 }
 
 static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
-				struct dma_fence **fence)
+					  struct amdgpu_bo *bo,
+					  struct dma_fence **fence)
 {
 	const unsigned ib_size_dw = 16;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	struct dma_fence *f = NULL;
-	uint64_t dummy;
+	uint64_t addr;
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job);
@@ -635,14 +617,14 @@ static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 		return r;
 
 	ib = &job->ibs[0];
-	dummy = ib->gpu_addr + 1024;
+	addr = amdgpu_bo_gpu_offset(bo);
 
 	ib->length_dw = 0;
 	ib->ptr[ib->length_dw++] = 0x00000018;
 	ib->ptr[ib->length_dw++] = 0x00000001;
 	ib->ptr[ib->length_dw++] = handle;
-	ib->ptr[ib->length_dw++] = upper_32_bits(dummy);
-	ib->ptr[ib->length_dw++] = dummy;
+	ib->ptr[ib->length_dw++] = upper_32_bits(addr);
+	ib->ptr[ib->length_dw++] = addr;
 	ib->ptr[ib->length_dw++] = 0x0000000b;
 
 	ib->ptr[ib->length_dw++] = 0x00000014;
@@ -674,129 +656,38 @@ err:
 
 int amdgpu_vcn_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 {
+	struct amdgpu_device *adev = ring->adev;
 	struct dma_fence *fence = NULL;
+	struct amdgpu_bo *bo = NULL;
 	long r;
 
-	r = amdgpu_vcn_enc_get_create_msg(ring, 1, NULL);
-	if (r)
-		goto error;
-
-	r = amdgpu_vcn_enc_get_destroy_msg(ring, 1, &fence);
-	if (r)
-		goto error;
-
-	r = dma_fence_wait_timeout(fence, false, timeout);
-	if (r == 0)
-		r = -ETIMEDOUT;
-	else if (r > 0)
-		r = 0;
-
-error:
-	dma_fence_put(fence);
-	return r;
-}
-
-int amdgpu_vcn_jpeg_ring_test_ring(struct amdgpu_ring *ring)
-{
-	struct amdgpu_device *adev = ring->adev;
-	uint32_t tmp = 0;
-	unsigned i;
-	int r;
-
-	WREG32(adev->vcn.inst[ring->me].external.jpeg_pitch, 0xCAFEDEAD);
-	r = amdgpu_ring_alloc(ring, 3);
-	if (r)
-		return r;
-
-	amdgpu_ring_write(ring, PACKET0(adev->vcn.internal.jpeg_pitch, 0));
-	amdgpu_ring_write(ring, 0xDEADBEEF);
-	amdgpu_ring_commit(ring);
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(adev->vcn.inst[ring->me].external.jpeg_pitch);
-		if (tmp == 0xDEADBEEF)
-			break;
-		udelay(1);
-	}
-
-	if (i >= adev->usec_timeout)
-		r = -ETIMEDOUT;
-
-	return r;
-}
-
-static int amdgpu_vcn_jpeg_set_reg(struct amdgpu_ring *ring, uint32_t handle,
-		struct dma_fence **fence)
-{
-	struct amdgpu_device *adev = ring->adev;
-	struct amdgpu_job *job;
-	struct amdgpu_ib *ib;
-	struct dma_fence *f = NULL;
-	const unsigned ib_size_dw = 16;
-	int i, r;
+	/* temporarily disable ib test for sriov */
+	if (amdgpu_sriov_vf(adev))
+		return 0;
 
-	r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job);
+	r = amdgpu_bo_create_reserved(ring->adev, 128 * 1024, PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_VRAM,
+				      &bo, NULL, NULL);
 	if (r)
 		return r;
 
-	ib = &job->ibs[0];
-
-	ib->ptr[0] = PACKETJ(adev->vcn.internal.jpeg_pitch, 0, 0, PACKETJ_TYPE0);
-	ib->ptr[1] = 0xDEADBEEF;
-	for (i = 2; i < 16; i += 2) {
-		ib->ptr[i] = PACKETJ(0, 0, 0, PACKETJ_TYPE6);
-		ib->ptr[i+1] = 0;
-	}
-	ib->length_dw = 16;
-
-	r = amdgpu_job_submit_direct(job, ring, &f);
+	r = amdgpu_vcn_enc_get_create_msg(ring, 1, bo, NULL);
 	if (r)
-		goto err;
-
-	if (fence)
-		*fence = dma_fence_get(f);
-	dma_fence_put(f);
-
-	return 0;
-
-err:
-	amdgpu_job_free(job);
-	return r;
-}
-
-int amdgpu_vcn_jpeg_ring_test_ib(struct amdgpu_ring *ring, long timeout)
-{
-	struct amdgpu_device *adev = ring->adev;
-	uint32_t tmp = 0;
-	unsigned i;
-	struct dma_fence *fence = NULL;
-	long r = 0;
+		goto error;
 
-	r = amdgpu_vcn_jpeg_set_reg(ring, 1, &fence);
+	r = amdgpu_vcn_enc_get_destroy_msg(ring, 1, bo, &fence);
 	if (r)
 		goto error;
 
 	r = dma_fence_wait_timeout(fence, false, timeout);
-	if (r == 0) {
+	if (r == 0)
 		r = -ETIMEDOUT;
-		goto error;
-	} else if (r < 0) {
-		goto error;
-	} else {
+	else if (r > 0)
 		r = 0;
-	}
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(adev->vcn.inst[ring->me].external.jpeg_pitch);
-		if (tmp == 0xDEADBEEF)
-			break;
-		udelay(1);
-	}
 
-	if (i >= adev->usec_timeout)
-		r = -ETIMEDOUT;
-
-	dma_fence_put(fence);
 error:
+	dma_fence_put(fence);
+	amdgpu_bo_unreserve(bo);
+	amdgpu_bo_unref(&bo);
 	return r;
 }