aboutsummaryrefslogtreecommitdiff
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_atpx_handler.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c9
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c27
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c12
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c40
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c20
-rw-r--r--drivers/gpu/drm/amd/amdgpu/nv.c1
-rw-r--r--drivers/gpu/drm/amd/amdgpu/soc15.c5
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c44
11 files changed, 114 insertions, 47 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
index 9b384a94d2f3..3e35a8f2c5e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
@@ -574,6 +574,7 @@ static const struct amdgpu_px_quirk amdgpu_px_quirk_list[] = {
{ 0x1002, 0x6900, 0x1002, 0x0124, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0x1002, 0x6900, 0x1028, 0x0812, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0x1002, 0x6900, 0x1028, 0x0813, AMDGPU_PX_QUIRK_FORCE_ATPX },
+ { 0x1002, 0x699f, 0x1028, 0x0814, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0x1002, 0x6900, 0x1025, 0x125A, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0x1002, 0x6900, 0x17AA, 0x3806, AMDGPU_PX_QUIRK_FORCE_ATPX },
{ 0, 0, 0, 0, 0 },
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 4e4094f842e7..8b26c970a3cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1143,6 +1143,9 @@ static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p,
num_deps = chunk->length_dw * 4 /
sizeof(struct drm_amdgpu_cs_chunk_sem);
+ if (p->post_deps)
+ return -EINVAL;
+
p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps),
GFP_KERNEL);
p->num_post_deps = 0;
@@ -1166,8 +1169,7 @@ static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p,
static int amdgpu_cs_process_syncobj_timeline_out_dep(struct amdgpu_cs_parser *p,
- struct amdgpu_cs_chunk
- *chunk)
+ struct amdgpu_cs_chunk *chunk)
{
struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps;
unsigned num_deps;
@@ -1177,6 +1179,9 @@ static int amdgpu_cs_process_syncobj_timeline_out_dep(struct amdgpu_cs_parser *p
num_deps = chunk->length_dw * 4 /
sizeof(struct drm_amdgpu_cs_chunk_syncobj);
+ if (p->post_deps)
+ return -EINVAL;
+
p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps),
GFP_KERNEL);
p->num_post_deps = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index f539a2a92774..7398b4850649 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -534,21 +534,24 @@ int amdgpu_ctx_wait_prev_fence(struct amdgpu_ctx *ctx,
struct drm_sched_entity *entity)
{
struct amdgpu_ctx_entity *centity = to_amdgpu_ctx_entity(entity);
- unsigned idx = centity->sequence & (amdgpu_sched_jobs - 1);
- struct dma_fence *other = centity->fences[idx];
+ struct dma_fence *other;
+ unsigned idx;
+ long r;
- if (other) {
- signed long r;
- r = dma_fence_wait(other, true);
- if (r < 0) {
- if (r != -ERESTARTSYS)
- DRM_ERROR("Error (%ld) waiting for fence!\n", r);
+ spin_lock(&ctx->ring_lock);
+ idx = centity->sequence & (amdgpu_sched_jobs - 1);
+ other = dma_fence_get(centity->fences[idx]);
+ spin_unlock(&ctx->ring_lock);
- return r;
- }
- }
+ if (!other)
+ return 0;
- return 0;
+ r = dma_fence_wait(other, true);
+ if (r < 0 && r != -ERESTARTSYS)
+ DRM_ERROR("Error (%ld) waiting for fence!\n", r);
+
+ dma_fence_put(other);
+ return r;
}
void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
index df8a23554831..f6ac1e9548f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h
@@ -32,7 +32,6 @@ struct amdgpu_gds {
uint32_t gws_size;
uint32_t oa_size;
uint32_t gds_compute_max_wave_id;
- uint32_t vgt_gs_max_wave_id;
};
struct amdgpu_gds_reg_offset {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 99f14fcc1460..19661c645703 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -30,6 +30,7 @@
#define AMDGPU_VCN_FIRMWARE_OFFSET 256
#define AMDGPU_VCN_MAX_ENC_RINGS 3
+#define VCN_DEC_KMD_CMD 0x80000000
#define VCN_DEC_CMD_FENCE 0x00000000
#define VCN_DEC_CMD_TRAP 0x00000001
#define VCN_DEC_CMD_WRITE_REG 0x00000004
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 32773b7523d2..f41287f9000d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4206,15 +4206,6 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 header, control = 0;
- /* Prevent a hw deadlock due to a wave ID mismatch between ME and GDS.
- * This resets the wave ID counters. (needed by transform feedback)
- * TODO: This might only be needed on a VMID switch when we change
- * the GDS OA mapping, not sure.
- */
- amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
- amdgpu_ring_write(ring, mmVGT_GS_MAX_WAVE_ID);
- amdgpu_ring_write(ring, ring->adev->gds.vgt_gs_max_wave_id);
-
if (ib->flags & AMDGPU_IB_FLAG_CE)
header = PACKET3(PACKET3_INDIRECT_BUFFER_CNST, 2);
else
@@ -4961,7 +4952,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = {
5 + /* HDP_INVL */
8 + 8 + /* FENCE x2 */
2, /* SWITCH_BUFFER */
- .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_gfx */
+ .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_gfx */
.emit_ib = gfx_v10_0_ring_emit_ib_gfx,
.emit_fence = gfx_v10_0_ring_emit_fence,
.emit_pipeline_sync = gfx_v10_0_ring_emit_pipeline_sync,
@@ -5112,7 +5103,6 @@ static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev)
default:
adev->gds.gds_size = 0x10000;
adev->gds.gds_compute_max_wave_id = 0x4ff;
- adev->gds.vgt_gs_max_wave_id = 0x3ff;
break;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 751567f78567..ee1ccdcf2d30 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1321,6 +1321,39 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
return 0;
}
+static int gfx_v8_0_csb_vram_pin(struct amdgpu_device *adev)
+{
+ int r;
+
+ r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, false);
+ if (unlikely(r != 0))
+ return r;
+
+ r = amdgpu_bo_pin(adev->gfx.rlc.clear_state_obj,
+ AMDGPU_GEM_DOMAIN_VRAM);
+ if (!r)
+ adev->gfx.rlc.clear_state_gpu_addr =
+ amdgpu_bo_gpu_offset(adev->gfx.rlc.clear_state_obj);
+
+ amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
+
+ return r;
+}
+
+static void gfx_v8_0_csb_vram_unpin(struct amdgpu_device *adev)
+{
+ int r;
+
+ if (!adev->gfx.rlc.clear_state_obj)
+ return;
+
+ r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, true);
+ if (likely(r == 0)) {
+ amdgpu_bo_unpin(adev->gfx.rlc.clear_state_obj);
+ amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
+ }
+}
+
static void gfx_v8_0_mec_fini(struct amdgpu_device *adev)
{
amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL);
@@ -4785,6 +4818,10 @@ static int gfx_v8_0_hw_init(void *handle)
gfx_v8_0_init_golden_registers(adev);
gfx_v8_0_constants_init(adev);
+ r = gfx_v8_0_csb_vram_pin(adev);
+ if (r)
+ return r;
+
r = adev->gfx.rlc.funcs->resume(adev);
if (r)
return r;
@@ -4901,6 +4938,9 @@ static int gfx_v8_0_hw_fini(void *handle)
else
pr_err("rlc is busy, skip halt rlc\n");
amdgpu_gfx_rlc_exit_safe_mode(adev);
+
+ gfx_v8_0_csb_vram_unpin(adev);
+
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 1cf639a51178..c066e1d3f981 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -596,14 +596,18 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev)
case CHIP_VEGA20:
break;
case CHIP_RAVEN:
- if (adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8)
- break;
- if ((adev->gfx.rlc_fw_version != 106 &&
- adev->gfx.rlc_fw_version < 531) ||
- (adev->gfx.rlc_fw_version == 53815) ||
- (adev->gfx.rlc_feature_version < 1) ||
- !adev->gfx.rlc.is_rlc_v2_1)
+ if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8)
+ &&((adev->gfx.rlc_fw_version != 106 &&
+ adev->gfx.rlc_fw_version < 531) ||
+ (adev->gfx.rlc_fw_version == 53815) ||
+ (adev->gfx.rlc_feature_version < 1) ||
+ !adev->gfx.rlc.is_rlc_v2_1))
adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
+
+ if (adev->pm.pp_feature & PP_GFXOFF_MASK)
+ adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG |
+ AMD_PG_SUPPORT_CP |
+ AMD_PG_SUPPORT_RLC_SMU_HS;
break;
default:
break;
@@ -4869,7 +4873,7 @@ static void gfx_v9_0_ring_soft_recovery(struct amdgpu_ring *ring, unsigned vmid)
value = REG_SET_FIELD(value, SQ_CMD, MODE, 0x01);
value = REG_SET_FIELD(value, SQ_CMD, CHECK_VMID, 1);
value = REG_SET_FIELD(value, SQ_CMD, VM_ID, vmid);
- WREG32(mmSQ_CMD, value);
+ WREG32_SOC15(GC, 0, mmSQ_CMD, value);
}
static void gfx_v9_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index 662612f89c70..9922bce3fd89 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -552,7 +552,6 @@ static int nv_common_early_init(void *handle)
AMD_CG_SUPPORT_BIF_LS;
adev->pg_flags = AMD_PG_SUPPORT_VCN |
AMD_PG_SUPPORT_VCN_DPG |
- AMD_PG_SUPPORT_MMHUB |
AMD_PG_SUPPORT_ATHUB;
adev->external_rev_id = adev->rev_id + 0x1;
break;
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 23265414d448..04fbf05d7176 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -992,11 +992,6 @@ static int soc15_common_early_init(void *handle)
adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN;
}
-
- if (adev->pm.pp_feature & PP_GFXOFF_MASK)
- adev->pg_flags |= AMD_PG_SUPPORT_GFX_PG |
- AMD_PG_SUPPORT_CP |
- AMD_PG_SUPPORT_RLC_SMU_HS;
break;
default:
/* FIXME: not supported yet */
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 1cfc2620b2dd..dfde886cc6bd 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -1485,7 +1485,7 @@ static void vcn_v2_0_dec_ring_insert_start(struct amdgpu_ring *ring)
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_DATA0_INTERNAL_OFFSET, 0));
amdgpu_ring_write(ring, 0);
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
- amdgpu_ring_write(ring, VCN_DEC_CMD_PACKET_START << 1);
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_PACKET_START << 1));
}
/**
@@ -1498,7 +1498,7 @@ static void vcn_v2_0_dec_ring_insert_start(struct amdgpu_ring *ring)
static void vcn_v2_0_dec_ring_insert_end(struct amdgpu_ring *ring)
{
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
- amdgpu_ring_write(ring, VCN_DEC_CMD_PACKET_END << 1);
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_PACKET_END << 1));
}
/**
@@ -1543,7 +1543,7 @@ static void vcn_v2_0_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
amdgpu_ring_write(ring, upper_32_bits(addr) & 0xff);
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
- amdgpu_ring_write(ring, VCN_DEC_CMD_FENCE << 1);
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_FENCE << 1));
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_DATA0_INTERNAL_OFFSET, 0));
amdgpu_ring_write(ring, 0);
@@ -1553,7 +1553,7 @@ static void vcn_v2_0_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
- amdgpu_ring_write(ring, VCN_DEC_CMD_TRAP << 1);
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_TRAP << 1));
}
/**
@@ -1597,7 +1597,7 @@ static void vcn_v2_0_dec_ring_emit_reg_wait(struct amdgpu_ring *ring,
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
- amdgpu_ring_write(ring, VCN_DEC_CMD_REG_READ_COND_WAIT << 1);
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_REG_READ_COND_WAIT << 1));
}
static void vcn_v2_0_dec_ring_emit_vm_flush(struct amdgpu_ring *ring,
@@ -1626,7 +1626,7 @@ static void vcn_v2_0_dec_ring_emit_wreg(struct amdgpu_ring *ring,
amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
- amdgpu_ring_write(ring, VCN_DEC_CMD_WRITE_REG << 1);
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_WRITE_REG << 1));
}
/**
@@ -2079,6 +2079,36 @@ static int vcn_v2_0_process_interrupt(struct amdgpu_device *adev,
return 0;
}
+static int vcn_v2_0_dec_ring_test_ring(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+ uint32_t tmp = 0;
+ unsigned i;
+ int r;
+
+ WREG32(adev->vcn.external.scratch9, 0xCAFEDEAD);
+ r = amdgpu_ring_alloc(ring, 4);
+ if (r)
+ return r;
+ amdgpu_ring_write(ring, PACKET0(mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET, 0));
+ amdgpu_ring_write(ring, VCN_DEC_KMD_CMD | (VCN_DEC_CMD_PACKET_START << 1));
+ amdgpu_ring_write(ring, PACKET0(adev->vcn.internal.scratch9, 0));
+ amdgpu_ring_write(ring, 0xDEADBEEF);
+ amdgpu_ring_commit(ring);
+ for (i = 0; i < adev->usec_timeout; i++) {
+ tmp = RREG32(adev->vcn.external.scratch9);
+ if (tmp == 0xDEADBEEF)
+ break;
+ DRM_UDELAY(1);
+ }
+
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
+ return r;
+}
+
+
static int vcn_v2_0_set_powergating_state(void *handle,
enum amd_powergating_state state)
{
@@ -2142,7 +2172,7 @@ static const struct amdgpu_ring_funcs vcn_v2_0_dec_ring_vm_funcs = {
.emit_ib = vcn_v2_0_dec_ring_emit_ib,
.emit_fence = vcn_v2_0_dec_ring_emit_fence,
.emit_vm_flush = vcn_v2_0_dec_ring_emit_vm_flush,
- .test_ring = amdgpu_vcn_dec_ring_test_ring,
+ .test_ring = vcn_v2_0_dec_ring_test_ring,
.test_ib = amdgpu_vcn_dec_ring_test_ib,
.insert_nop = vcn_v2_0_dec_ring_insert_nop,
.insert_start = vcn_v2_0_dec_ring_insert_start,
Powered by cgit, Gruvboxed by Blaster4385.