summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/vce_v3_0.c')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vce_v3_0.c108
1 files changed, 66 insertions, 42 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index 6feed726e299..37ca685e5a9a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -43,9 +43,13 @@
#define GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT 0x04
#define GRBM_GFX_INDEX__VCE_INSTANCE_MASK 0x10
+#define GRBM_GFX_INDEX__VCE_ALL_PIPE 0x07
+
#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR0 0x8616
#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR1 0x8617
#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR2 0x8618
+#define mmGRBM_GFX_INDEX_DEFAULT 0xE0000000
+
#define VCE_STATUS_VCPU_REPORT_FW_LOADED_MASK 0x02
#define VCE_V3_0_FW_SIZE (384 * 1024)
@@ -54,6 +58,9 @@
#define FW_52_8_3 ((52 << 24) | (8 << 16) | (3 << 8))
+#define GET_VCE_INSTANCE(i) ((i) << GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT \
+ | GRBM_GFX_INDEX__VCE_ALL_PIPE)
+
static void vce_v3_0_mc_resume(struct amdgpu_device *adev, int idx);
static void vce_v3_0_set_ring_funcs(struct amdgpu_device *adev);
static void vce_v3_0_set_irq_funcs(struct amdgpu_device *adev);
@@ -134,7 +141,7 @@ static void vce_v3_0_set_vce_sw_clock_gating(struct amdgpu_device *adev,
accessible but the firmware will throttle the clocks on the
fly as necessary.
*/
- if (gated) {
+ if (!gated) {
data = RREG32(mmVCE_CLOCK_GATING_B);
data |= 0x1ff;
data &= ~0xef0000;
@@ -175,7 +182,7 @@ static void vce_v3_0_set_vce_sw_clock_gating(struct amdgpu_device *adev,
WREG32(mmVCE_UENC_CLOCK_GATING_2, data);
data = RREG32(mmVCE_UENC_REG_CLOCK_GATING);
- data &= ~0xffc00000;
+ data &= ~0x3ff;
WREG32(mmVCE_UENC_REG_CLOCK_GATING, data);
data = RREG32(mmVCE_UENC_DMA_DCLK_CTRL);
@@ -249,7 +256,7 @@ static int vce_v3_0_start(struct amdgpu_device *adev)
if (adev->vce.harvest_config & (1 << idx))
continue;
- WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, idx);
+ WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(idx));
vce_v3_0_mc_resume(adev, idx);
WREG32_FIELD(VCE_STATUS, JOB_BUSY, 1);
@@ -273,7 +280,7 @@ static int vce_v3_0_start(struct amdgpu_device *adev)
}
}
- WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, 0);
+ WREG32(mmGRBM_GFX_INDEX, mmGRBM_GFX_INDEX_DEFAULT);
mutex_unlock(&adev->grbm_idx_mutex);
return 0;
@@ -288,7 +295,7 @@ static int vce_v3_0_stop(struct amdgpu_device *adev)
if (adev->vce.harvest_config & (1 << idx))
continue;
- WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, idx);
+ WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(idx));
if (adev->asic_type >= CHIP_STONEY)
WREG32_P(mmVCE_VCPU_CNTL, 0, ~0x200001);
@@ -306,7 +313,7 @@ static int vce_v3_0_stop(struct amdgpu_device *adev)
vce_v3_0_set_vce_sw_clock_gating(adev, false);
}
- WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, 0);
+ WREG32(mmGRBM_GFX_INDEX, mmGRBM_GFX_INDEX_DEFAULT);
mutex_unlock(&adev->grbm_idx_mutex);
return 0;
@@ -320,11 +327,12 @@ static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev)
{
u32 tmp;
- /* Fiji, Stoney, Polaris10, Polaris11 are single pipe */
+ /* Fiji, Stoney, Polaris10, Polaris11, Polaris12 are single pipe */
if ((adev->asic_type == CHIP_FIJI) ||
(adev->asic_type == CHIP_STONEY) ||
(adev->asic_type == CHIP_POLARIS10) ||
- (adev->asic_type == CHIP_POLARIS11))
+ (adev->asic_type == CHIP_POLARIS11) ||
+ (adev->asic_type == CHIP_POLARIS12))
return AMDGPU_VCE_HARVEST_VCE1;
/* Tonga and CZ are dual or single pipe */
@@ -395,8 +403,7 @@ static int vce_v3_0_sw_init(void *handle)
for (i = 0; i < adev->vce.num_rings; i++) {
ring = &adev->vce.ring[i];
sprintf(ring->name, "vce%d", i);
- r = amdgpu_ring_init(adev, ring, 512, VCE_CMD_NO_OP, 0xf,
- &adev->vce.irq, 0, AMDGPU_RING_TYPE_VCE);
+ r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0);
if (r)
return r;
}
@@ -586,17 +593,17 @@ static bool vce_v3_0_check_soft_reset(void *handle)
* VCE team suggest use bit 3--bit 6 for busy status check
*/
mutex_lock(&adev->grbm_idx_mutex);
- WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0);
+ WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(0));
if (RREG32(mmVCE_STATUS) & AMDGPU_VCE_STATUS_BUSY_MASK) {
srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE0, 1);
srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE1, 1);
}
- WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0x10);
+ WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(1));
if (RREG32(mmVCE_STATUS) & AMDGPU_VCE_STATUS_BUSY_MASK) {
srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE0, 1);
srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE1, 1);
}
- WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0);
+ WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(0));
mutex_unlock(&adev->grbm_idx_mutex);
if (srbm_soft_reset) {
@@ -734,7 +741,7 @@ static int vce_v3_0_set_clockgating_state(void *handle,
if (adev->vce.harvest_config & (1 << i))
continue;
- WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, i);
+ WREG32(mmGRBM_GFX_INDEX, GET_VCE_INSTANCE(i));
if (enable) {
/* initialize VCE_CLOCK_GATING_A: Clock ON/OFF delay */
@@ -753,7 +760,7 @@ static int vce_v3_0_set_clockgating_state(void *handle,
vce_v3_0_set_vce_sw_clock_gating(adev, enable);
}
- WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, 0);
+ WREG32(mmGRBM_GFX_INDEX, mmGRBM_GFX_INDEX_DEFAULT);
mutex_unlock(&adev->grbm_idx_mutex);
return 0;
@@ -814,28 +821,7 @@ static void vce_v3_0_emit_pipeline_sync(struct amdgpu_ring *ring)
amdgpu_ring_write(ring, seq);
}
-static unsigned vce_v3_0_ring_get_emit_ib_size(struct amdgpu_ring *ring)
-{
- return
- 5; /* vce_v3_0_ring_emit_ib */
-}
-
-static unsigned vce_v3_0_ring_get_dma_frame_size(struct amdgpu_ring *ring)
-{
- return
- 4 + /* vce_v3_0_emit_pipeline_sync */
- 6; /* amdgpu_vce_ring_emit_fence x1 no user fence */
-}
-
-static unsigned vce_v3_0_ring_get_dma_frame_size_vm(struct amdgpu_ring *ring)
-{
- return
- 6 + /* vce_v3_0_emit_vm_flush */
- 4 + /* vce_v3_0_emit_pipeline_sync */
- 6 + 6; /* amdgpu_vce_ring_emit_fence x2 vm fence */
-}
-
-const struct amd_ip_funcs vce_v3_0_ip_funcs = {
+static const struct amd_ip_funcs vce_v3_0_ip_funcs = {
.name = "vce_v3_0",
.early_init = vce_v3_0_early_init,
.late_init = NULL,
@@ -856,10 +842,17 @@ const struct amd_ip_funcs vce_v3_0_ip_funcs = {
};
static const struct amdgpu_ring_funcs vce_v3_0_ring_phys_funcs = {
+ .type = AMDGPU_RING_TYPE_VCE,
+ .align_mask = 0xf,
+ .nop = VCE_CMD_NO_OP,
.get_rptr = vce_v3_0_ring_get_rptr,
.get_wptr = vce_v3_0_ring_get_wptr,
.set_wptr = vce_v3_0_ring_set_wptr,
.parse_cs = amdgpu_vce_ring_parse_cs,
+ .emit_frame_size =
+ 4 + /* vce_v3_0_emit_pipeline_sync */
+ 6, /* amdgpu_vce_ring_emit_fence x1 no user fence */
+ .emit_ib_size = 5, /* vce_v3_0_ring_emit_ib */
.emit_ib = amdgpu_vce_ring_emit_ib,
.emit_fence = amdgpu_vce_ring_emit_fence,
.test_ring = amdgpu_vce_ring_test_ring,
@@ -868,15 +861,21 @@ static const struct amdgpu_ring_funcs vce_v3_0_ring_phys_funcs = {
.pad_ib = amdgpu_ring_generic_pad_ib,
.begin_use = amdgpu_vce_ring_begin_use,
.end_use = amdgpu_vce_ring_end_use,
- .get_emit_ib_size = vce_v3_0_ring_get_emit_ib_size,
- .get_dma_frame_size = vce_v3_0_ring_get_dma_frame_size,
};
static const struct amdgpu_ring_funcs vce_v3_0_ring_vm_funcs = {
+ .type = AMDGPU_RING_TYPE_VCE,
+ .align_mask = 0xf,
+ .nop = VCE_CMD_NO_OP,
.get_rptr = vce_v3_0_ring_get_rptr,
.get_wptr = vce_v3_0_ring_get_wptr,
.set_wptr = vce_v3_0_ring_set_wptr,
- .parse_cs = NULL,
+ .parse_cs = amdgpu_vce_ring_parse_cs_vm,
+ .emit_frame_size =
+ 6 + /* vce_v3_0_emit_vm_flush */
+ 4 + /* vce_v3_0_emit_pipeline_sync */
+ 6 + 6, /* amdgpu_vce_ring_emit_fence x2 vm fence */
+ .emit_ib_size = 4, /* amdgpu_vce_ring_emit_ib */
.emit_ib = vce_v3_0_ring_emit_ib,
.emit_vm_flush = vce_v3_0_emit_vm_flush,
.emit_pipeline_sync = vce_v3_0_emit_pipeline_sync,
@@ -887,8 +886,6 @@ static const struct amdgpu_ring_funcs vce_v3_0_ring_vm_funcs = {
.pad_ib = amdgpu_ring_generic_pad_ib,
.begin_use = amdgpu_vce_ring_begin_use,
.end_use = amdgpu_vce_ring_end_use,
- .get_emit_ib_size = vce_v3_0_ring_get_emit_ib_size,
- .get_dma_frame_size = vce_v3_0_ring_get_dma_frame_size_vm,
};
static void vce_v3_0_set_ring_funcs(struct amdgpu_device *adev)
@@ -916,3 +913,30 @@ static void vce_v3_0_set_irq_funcs(struct amdgpu_device *adev)
adev->vce.irq.num_types = 1;
adev->vce.irq.funcs = &vce_v3_0_irq_funcs;
};
+
+const struct amdgpu_ip_block_version vce_v3_0_ip_block =
+{
+ .type = AMD_IP_BLOCK_TYPE_VCE,
+ .major = 3,
+ .minor = 0,
+ .rev = 0,
+ .funcs = &vce_v3_0_ip_funcs,
+};
+
+const struct amdgpu_ip_block_version vce_v3_1_ip_block =
+{
+ .type = AMD_IP_BLOCK_TYPE_VCE,
+ .major = 3,
+ .minor = 1,
+ .rev = 0,
+ .funcs = &vce_v3_0_ip_funcs,
+};
+
+const struct amdgpu_ip_block_version vce_v3_4_ip_block =
+{
+ .type = AMD_IP_BLOCK_TYPE_VCE,
+ .major = 3,
+ .minor = 4,
+ .rev = 0,
+ .funcs = &vce_v3_0_ip_funcs,
+};
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-28 03:31:54 +0000