summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c4
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c5
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c8
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c36
-rw-r--r--drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c5
6 files changed, 51 insertions, 9 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 817116e53d44..95a05b03f799 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -119,9 +119,11 @@
* - 3.57.0 - Compute tunneling on GFX10+
* - 3.58.0 - Add GFX12 DCC support
* - 3.59.0 - Cleared VRAM
+ * - 3.60.0 - Add AMDGPU_TILING_GFX12_DCC_WRITE_COMPRESS_DISABLE (Vulkan requirement)
+ * - 3.61.0 - Contains fix for RV/PCO compute queues
*/
#define KMS_DRIVER_MAJOR 3
-#define KMS_DRIVER_MINOR 59
+#define KMS_DRIVER_MINOR 61
#define KMS_DRIVER_PATCHLEVEL 0
/*
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index babe94ade247..e5fc80ed06ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -3815,9 +3815,10 @@ int psp_init_cap_microcode(struct psp_context *psp, const char *chip_name)
if (err == -ENODEV) {
dev_warn(adev->dev, "cap microcode does not exist, skip\n");
err = 0;
- goto out;
+ } else {
+ dev_err(adev->dev, "fail to initialize cap microcode\n");
}
- dev_err(adev->dev, "fail to initialize cap microcode\n");
+ goto out;
}
info = &adev->firmware.ucode[AMDGPU_UCODE_ID_CAP];
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index ff286940ab43..01ae2f88dec8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -309,7 +309,7 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev,
mutex_lock(&adev->mman.gtt_window_lock);
while (src_mm.remaining) {
uint64_t from, to, cur_size, tiling_flags;
- uint32_t num_type, data_format, max_com;
+ uint32_t num_type, data_format, max_com, write_compress_disable;
struct dma_fence *next;
/* Never copy more than 256MiB at once to avoid a timeout */
@@ -340,9 +340,13 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev,
max_com = AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_MAX_COMPRESSED_BLOCK);
num_type = AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_NUMBER_TYPE);
data_format = AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_DATA_FORMAT);
+ write_compress_disable =
+ AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_WRITE_COMPRESS_DISABLE);
copy_flags |= (AMDGPU_COPY_FLAGS_SET(MAX_COMPRESSED, max_com) |
AMDGPU_COPY_FLAGS_SET(NUMBER_TYPE, num_type) |
- AMDGPU_COPY_FLAGS_SET(DATA_FORMAT, data_format));
+ AMDGPU_COPY_FLAGS_SET(DATA_FORMAT, data_format) |
+ AMDGPU_COPY_FLAGS_SET(WRITE_COMPRESS_DISABLE,
+ write_compress_disable));
}
r = amdgpu_copy_buffer(ring, from, to, cur_size, resv,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 461fb8090ae0..208b7d1d8a27 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -119,6 +119,8 @@ struct amdgpu_copy_mem {
#define AMDGPU_COPY_FLAGS_NUMBER_TYPE_MASK 0x07
#define AMDGPU_COPY_FLAGS_DATA_FORMAT_SHIFT 8
#define AMDGPU_COPY_FLAGS_DATA_FORMAT_MASK 0x3f
+#define AMDGPU_COPY_FLAGS_WRITE_COMPRESS_DISABLE_SHIFT 14
+#define AMDGPU_COPY_FLAGS_WRITE_COMPRESS_DISABLE_MASK 0x1
#define AMDGPU_COPY_FLAGS_SET(field, value) \
(((__u32)(value) & AMDGPU_COPY_FLAGS_##field##_MASK) << AMDGPU_COPY_FLAGS_##field##_SHIFT)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index fa572b40989e..0dce4421418c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -7437,6 +7437,38 @@ static void gfx_v9_0_ring_emit_cleaner_shader(struct amdgpu_ring *ring)
amdgpu_ring_write(ring, 0); /* RESERVED field, programmed to zero */
}
+static void gfx_v9_0_ring_begin_use_compute(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+ struct amdgpu_ip_block *gfx_block =
+ amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GFX);
+
+ amdgpu_gfx_enforce_isolation_ring_begin_use(ring);
+
+ /* Raven and PCO APUs seem to have stability issues
+ * with compute and gfxoff and gfx pg. Disable gfx pg during
+ * submission and allow again afterwards.
+ */
+ if (gfx_block && amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 1, 0))
+ gfx_v9_0_set_powergating_state(gfx_block, AMD_PG_STATE_UNGATE);
+}
+
+static void gfx_v9_0_ring_end_use_compute(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+ struct amdgpu_ip_block *gfx_block =
+ amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GFX);
+
+ /* Raven and PCO APUs seem to have stability issues
+ * with compute and gfxoff and gfx pg. Disable gfx pg during
+ * submission and allow again afterwards.
+ */
+ if (gfx_block && amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 1, 0))
+ gfx_v9_0_set_powergating_state(gfx_block, AMD_PG_STATE_GATE);
+
+ amdgpu_gfx_enforce_isolation_ring_end_use(ring);
+}
+
static const struct amd_ip_funcs gfx_v9_0_ip_funcs = {
.name = "gfx_v9_0",
.early_init = gfx_v9_0_early_init,
@@ -7613,8 +7645,8 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = {
.emit_wave_limit = gfx_v9_0_emit_wave_limit,
.reset = gfx_v9_0_reset_kcq,
.emit_cleaner_shader = gfx_v9_0_ring_emit_cleaner_shader,
- .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use,
- .end_use = amdgpu_gfx_enforce_isolation_ring_end_use,
+ .begin_use = gfx_v9_0_ring_begin_use_compute,
+ .end_use = gfx_v9_0_ring_end_use_compute,
};
static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_kiq = {
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
index 9c17df2cf37b..7e10e94624e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
@@ -1741,11 +1741,12 @@ static void sdma_v7_0_emit_copy_buffer(struct amdgpu_ib *ib,
uint32_t byte_count,
uint32_t copy_flags)
{
- uint32_t num_type, data_format, max_com;
+ uint32_t num_type, data_format, max_com, write_cm;
max_com = AMDGPU_COPY_FLAGS_GET(copy_flags, MAX_COMPRESSED);
data_format = AMDGPU_COPY_FLAGS_GET(copy_flags, DATA_FORMAT);
num_type = AMDGPU_COPY_FLAGS_GET(copy_flags, NUMBER_TYPE);
+ write_cm = AMDGPU_COPY_FLAGS_GET(copy_flags, WRITE_COMPRESS_DISABLE) ? 2 : 1;
ib->ptr[ib->length_dw++] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_COPY) |
SDMA_PKT_COPY_LINEAR_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR) |
@@ -1762,7 +1763,7 @@ static void sdma_v7_0_emit_copy_buffer(struct amdgpu_ib *ib,
if ((copy_flags & (AMDGPU_COPY_FLAGS_READ_DECOMPRESSED | AMDGPU_COPY_FLAGS_WRITE_COMPRESSED)))
ib->ptr[ib->length_dw++] = SDMA_DCC_DATA_FORMAT(data_format) | SDMA_DCC_NUM_TYPE(num_type) |
((copy_flags & AMDGPU_COPY_FLAGS_READ_DECOMPRESSED) ? SDMA_DCC_READ_CM(2) : 0) |
- ((copy_flags & AMDGPU_COPY_FLAGS_WRITE_COMPRESSED) ? SDMA_DCC_WRITE_CM(1) : 0) |
+ ((copy_flags & AMDGPU_COPY_FLAGS_WRITE_COMPRESSED) ? SDMA_DCC_WRITE_CM(write_cm) : 0) |
SDMA_DCC_MAX_COM(max_com) | SDMA_DCC_MAX_UCOM(1);
else
ib->ptr[ib->length_dw++] = 0;
generated by cgit v1.2.3 (git 2.46.0) at 2025-09-28 06:11:14 +0000