1 files changed, 114 insertions, 157 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index 0dce59f4f6e2..9505ae96fbec 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -112,6 +112,8 @@ static void sdma_v5_0_set_ring_funcs(struct amdgpu_device *adev);
 static void sdma_v5_0_set_buffer_funcs(struct amdgpu_device *adev);
 static void sdma_v5_0_set_vm_pte_funcs(struct amdgpu_device *adev);
 static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev);
+static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring);
+static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring);
 
 static const struct soc15_reg_golden golden_settings_sdma_5[] = {
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_CHICKEN_BITS, 0xffbf1f0f, 0x03ab0107),
@@ -369,67 +371,36 @@ static uint64_t sdma_v5_0_ring_get_wptr(struct amdgpu_ring *ring)
 static void sdma_v5_0_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	uint32_t *wptr_saved;
-	uint32_t *is_queue_unmap;
-	uint64_t aggregated_db_index;
-	uint32_t mqd_size = adev->mqds[AMDGPU_HW_IP_DMA].mqd_size;
 
 	DRM_DEBUG("Setting write pointer\n");
-	if (ring->is_mes_queue) {
-		wptr_saved = (uint32_t *)(ring->mqd_ptr + mqd_size);
-		is_queue_unmap = (uint32_t *)(ring->mqd_ptr + mqd_size +
-					      sizeof(uint32_t));
-		aggregated_db_index =
-			amdgpu_mes_get_aggregated_doorbell_index(adev,
-			AMDGPU_MES_PRIORITY_LEVEL_NORMAL);
-
+	if (ring->use_doorbell) {
+		DRM_DEBUG("Using doorbell -- "
+			  "wptr_offs == 0x%08x "
+			  "lower_32_bits(ring->wptr) << 2 == 0x%08x "
+			  "upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
+			  ring->wptr_offs,
+			  lower_32_bits(ring->wptr << 2),
+			  upper_32_bits(ring->wptr << 2));
+		/* XXX check if swapping is necessary on BE */
 		atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
 			     ring->wptr << 2);
-		*wptr_saved = ring->wptr << 2;
-		if (*is_queue_unmap) {
-			WDOORBELL64(aggregated_db_index, ring->wptr << 2);
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-					ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		} else {
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-					ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-
-			if (*is_queue_unmap)
-				WDOORBELL64(aggregated_db_index,
-					    ring->wptr << 2);
-		}
+		DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
+			  ring->doorbell_index, ring->wptr << 2);
+		WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
 	} else {
-		if (ring->use_doorbell) {
-			DRM_DEBUG("Using doorbell -- "
-				  "wptr_offs == 0x%08x "
-				  "lower_32_bits(ring->wptr) << 2 == 0x%08x "
-				  "upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
-				  ring->wptr_offs,
-				  lower_32_bits(ring->wptr << 2),
-				  upper_32_bits(ring->wptr << 2));
-			/* XXX check if swapping is necessary on BE */
-			atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
-				     ring->wptr << 2);
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-				  ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		} else {
-			DRM_DEBUG("Not using doorbell -- "
-				  "mmSDMA%i_GFX_RB_WPTR == 0x%08x "
-				  "mmSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
-				  ring->me,
-				  lower_32_bits(ring->wptr << 2),
-				  ring->me,
-				  upper_32_bits(ring->wptr << 2));
-			WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
-					     ring->me, mmSDMA0_GFX_RB_WPTR),
-					lower_32_bits(ring->wptr << 2));
-			WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
-					     ring->me, mmSDMA0_GFX_RB_WPTR_HI),
-					upper_32_bits(ring->wptr << 2));
-		}
+		DRM_DEBUG("Not using doorbell -- "
+			  "mmSDMA%i_GFX_RB_WPTR == 0x%08x "
+			  "mmSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
+			  ring->me,
+			  lower_32_bits(ring->wptr << 2),
+			  ring->me,
+			  upper_32_bits(ring->wptr << 2));
+		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
+							     ring->me, mmSDMA0_GFX_RB_WPTR),
+				lower_32_bits(ring->wptr << 2));
+		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
+							     ring->me, mmSDMA0_GFX_RB_WPTR_HI),
+				upper_32_bits(ring->wptr << 2));
 	}
 }
 
@@ -575,11 +546,9 @@ static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
 	}
 
 	if (flags & AMDGPU_FENCE_FLAG_INT) {
-		uint32_t ctx = ring->is_mes_queue ?
-			(ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0;
 		/* generate an interrupt */
 		amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_TRAP));
-		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(ctx));
+		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0));
 	}
 }
 
@@ -588,15 +557,15 @@ static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
  * sdma_v5_0_gfx_stop - stop the gfx async dma engines
  *
  * @adev: amdgpu_device pointer
- *
+ * @inst_mask: mask of dma engine instances to be disabled
  * Stop the gfx async dma ring buffers (NAVI10).
  */
-static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev)
+static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev, uint32_t inst_mask)
 {
 	u32 rb_cntl, ib_cntl;
 	int i;
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
 		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
@@ -688,9 +657,11 @@ static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable)
 {
 	u32 f32_cntl;
 	int i;
+	uint32_t inst_mask;
 
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
 	if (!enable) {
-		sdma_v5_0_gfx_stop(adev);
+		sdma_v5_0_gfx_stop(adev, 1 << inst_mask);
 		sdma_v5_0_rlc_stop(adev);
 	}
 
@@ -1046,33 +1017,22 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 	int r;
 	u32 tmp;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
-			return r;
-		}
-
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+		return r;
 	}
 
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
+
 	r = amdgpu_ring_alloc(ring, 20);
 	if (r) {
 		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
-		if (!ring->is_mes_queue)
-			amdgpu_device_wb_free(adev, index);
+		amdgpu_device_wb_free(adev, index);
 		return r;
 	}
 
@@ -1085,10 +1045,7 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (ring->is_mes_queue)
-			tmp = le32_to_cpu(*cpu_ptr);
-		else
-			tmp = le32_to_cpu(adev->wb.wb[index]);
+		tmp = le32_to_cpu(adev->wb.wb[index]);
 		if (tmp == 0xDEADBEEF)
 			break;
 		if (amdgpu_emu_mode == 1)
@@ -1100,8 +1057,7 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 	if (i >= adev->usec_timeout)
 		r = -ETIMEDOUT;
 
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 
 	return r;
 }
@@ -1124,38 +1080,24 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	long r;
 	u32 tmp = 0;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = (void *)amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+		return r;
+	}
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
 
-		r = amdgpu_ib_get(adev, NULL, 256,
-					AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
-			goto err0;
-		}
+	r = amdgpu_ib_get(adev, NULL, 256,
+			  AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+		goto err0;
 	}
 
 	ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
@@ -1183,10 +1125,7 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		goto err1;
 	}
 
-	if (ring->is_mes_queue)
-		tmp = le32_to_cpu(*cpu_ptr);
-	else
-		tmp = le32_to_cpu(adev->wb.wb[index]);
+	tmp = le32_to_cpu(adev->wb.wb[index]);
 
 	if (tmp == 0xDEADBEEF)
 		r = 0;
@@ -1197,8 +1136,7 @@ err1:
 	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err0:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -1389,6 +1327,36 @@ static void sdma_v5_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
 	amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask);
 }
 
+static int sdma_v5_0_soft_reset_engine(struct amdgpu_device *adev, u32 instance_id)
+{
+	u32 grbm_soft_reset;
+	u32 tmp;
+
+	grbm_soft_reset = REG_SET_FIELD(0,
+					GRBM_SOFT_RESET, SOFT_RESET_SDMA0,
+					1);
+	grbm_soft_reset <<= instance_id;
+
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	tmp |= grbm_soft_reset;
+	DRM_DEBUG("GRBM_SOFT_RESET=0x%08X\n", tmp);
+	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+
+	udelay(50);
+
+	tmp &= ~grbm_soft_reset;
+	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	return 0;
+}
+
+static const struct amdgpu_sdma_funcs sdma_v5_0_sdma_funcs = {
+	.stop_kernel_queue = &sdma_v5_0_stop_queue,
+	.start_kernel_queue = &sdma_v5_0_restore_queue,
+	.soft_reset_kernel_queue = &sdma_v5_0_soft_reset_engine,
+};
+
 static int sdma_v5_0_early_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -1431,6 +1399,7 @@ static int sdma_v5_0_sw_init(struct amdgpu_ip_block *ip_block)
 		return r;
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
+		adev->sdma.instance[i].funcs = &sdma_v5_0_sdma_funcs;
 		ring = &adev->sdma.instance[i].ring;
 		ring->ring_obj = NULL;
 		ring->use_doorbell = true;
@@ -1572,32 +1541,25 @@ static int sdma_v5_0_soft_reset(struct amdgpu_ip_block *ip_block)
 static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
-	int i, j, r;
-	u32 rb_cntl, ib_cntl, f32_cntl, freeze, cntl, preempt, soft_reset, stat1_reg;
+	u32 inst_id = ring->me;
 
-	if (amdgpu_sriov_vf(adev))
-		return -EINVAL;
+	return amdgpu_sdma_reset_engine(adev, inst_id);
+}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		if (ring == &adev->sdma.instance[i].ring)
-			break;
-	}
+static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring)
+{
+	u32 f32_cntl, freeze, cntl, stat1_reg;
+	struct amdgpu_device *adev = ring->adev;
+	int i, j, r = 0;
 
-	if (i == adev->sdma.num_instances) {
-		DRM_ERROR("sdma instance not found\n");
+	if (amdgpu_sriov_vf(adev))
 		return -EINVAL;
-	}
 
+	i = ring->me;
 	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	/* stop queue */
-	ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
-	ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
-
-	rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
-	rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
+	sdma_v5_0_gfx_stop(adev, 1 << i);
 
 	/* engine stop SDMA1_F32_CNTL.HALT to 1 and SDMAx_FREEZE freeze bit to 1 */
 	freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE));
@@ -1628,30 +1590,25 @@ static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 	cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL));
 	cntl = REG_SET_FIELD(cntl, SDMA0_CNTL, UTC_L1_ENABLE, 0);
 	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), cntl);
+err0:
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
+	return r;
+}
 
-	/* soft reset SDMA_GFX_PREEMPT.IB_PREEMPT = 0 mmGRBM_SOFT_RESET.SOFT_RESET_SDMA0/1 = 1 */
-	preempt = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT));
-	preempt = REG_SET_FIELD(preempt, SDMA0_GFX_PREEMPT, IB_PREEMPT, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT), preempt);
-
-	soft_reset = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-	soft_reset |= 1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i;
-
-	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset);
-
-	udelay(50);
-
-	soft_reset &= ~(1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i);
-	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset);
+static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	u32 inst_id = ring->me;
+	u32 freeze;
+	int r;
 
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	/* unfreeze*/
-	freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE));
+	freeze = RREG32(sdma_v5_0_get_reg_offset(adev, inst_id, mmSDMA0_FREEZE));
 	freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze);
+	WREG32(sdma_v5_0_get_reg_offset(adev, inst_id, mmSDMA0_FREEZE), freeze);
 
-	r = sdma_v5_0_gfx_resume_instance(adev, i, true);
-
-err0:
+	r = sdma_v5_0_gfx_resume_instance(adev, inst_id, true);
 	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	return r;
 }