summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c115
1 files changed, 100 insertions, 15 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index ae09fc1cfe6b7..a674c8a58dc23 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -755,7 +755,7 @@ static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev);
static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev,
struct amdgpu_cu_info *cu_info);
static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev);
-static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume);
+static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bool usegds);
static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring);
static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
void *ras_error_status);
@@ -1995,6 +1995,7 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
+ (ring_id * GFX9_MEC_HPD_SIZE);
+ ring->vm_hub = AMDGPU_GFXHUB_0;
sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue);
irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
@@ -2094,6 +2095,7 @@ static int gfx_v9_0_sw_init(void *handle)
/* disable scheduler on the real ring */
ring->no_scheduler = true;
+ ring->vm_hub = AMDGPU_GFXHUB_0;
r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq,
AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP,
AMDGPU_RING_PRIO_DEFAULT, NULL);
@@ -2111,6 +2113,7 @@ static int gfx_v9_0_sw_init(void *handle)
ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1;
ring->is_sw_ring = true;
hw_prio = amdgpu_sw_ring_priority(i);
+ ring->vm_hub = AMDGPU_GFXHUB_0;
r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq,
AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, hw_prio,
NULL);
@@ -3604,8 +3607,10 @@ static int gfx_v9_0_kiq_resume(struct amdgpu_device *adev)
return r;
r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
- if (unlikely(r != 0))
+ if (unlikely(r != 0)) {
+ amdgpu_bo_unreserve(ring->mqd_obj);
return r;
+ }
gfx_v9_0_kiq_init_queue(ring);
amdgpu_bo_kunmap(ring->mqd_obj);
@@ -3751,7 +3756,8 @@ static int gfx_v9_0_hw_fini(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
- amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
+ if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX))
+ amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
@@ -5121,7 +5127,8 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
gfx_v9_0_ring_emit_de_meta(ring,
(!amdgpu_sriov_vf(ring->adev) &&
flags & AMDGPU_IB_PREEMPTED) ?
- true : false);
+ true : false,
+ job->gds_size > 0 && job->gds_base != 0);
}
amdgpu_ring_write(ring, header);
@@ -5132,9 +5139,83 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
#endif
lower_32_bits(ib->gpu_addr));
amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
+ amdgpu_ring_ib_on_emit_cntl(ring);
amdgpu_ring_write(ring, control);
}
+static void gfx_v9_0_ring_patch_cntl(struct amdgpu_ring *ring,
+ unsigned offset)
+{
+ u32 control = ring->ring[offset];
+
+ control |= INDIRECT_BUFFER_PRE_RESUME(1);
+ ring->ring[offset] = control;
+}
+
+static void gfx_v9_0_ring_patch_ce_meta(struct amdgpu_ring *ring,
+ unsigned offset)
+{
+ struct amdgpu_device *adev = ring->adev;
+ void *ce_payload_cpu_addr;
+ uint64_t payload_offset, payload_size;
+
+ payload_size = sizeof(struct v9_ce_ib_state);
+
+ if (ring->is_mes_queue) {
+ payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data,
+ gfx[0].gfx_meta_data) +
+ offsetof(struct v9_gfx_meta_data, ce_payload);
+ ce_payload_cpu_addr =
+ amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset);
+ } else {
+ payload_offset = offsetof(struct v9_gfx_meta_data, ce_payload);
+ ce_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
+ }
+
+ if (offset + (payload_size >> 2) <= ring->buf_mask + 1) {
+ memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr, payload_size);
+ } else {
+ memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr,
+ (ring->buf_mask + 1 - offset) << 2);
+ payload_size -= (ring->buf_mask + 1 - offset) << 2;
+ memcpy((void *)&ring->ring[0],
+ ce_payload_cpu_addr + ((ring->buf_mask + 1 - offset) << 2),
+ payload_size);
+ }
+}
+
+static void gfx_v9_0_ring_patch_de_meta(struct amdgpu_ring *ring,
+ unsigned offset)
+{
+ struct amdgpu_device *adev = ring->adev;
+ void *de_payload_cpu_addr;
+ uint64_t payload_offset, payload_size;
+
+ payload_size = sizeof(struct v9_de_ib_state);
+
+ if (ring->is_mes_queue) {
+ payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data,
+ gfx[0].gfx_meta_data) +
+ offsetof(struct v9_gfx_meta_data, de_payload);
+ de_payload_cpu_addr =
+ amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset);
+ } else {
+ payload_offset = offsetof(struct v9_gfx_meta_data, de_payload);
+ de_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
+ }
+
+ if (offset + (payload_size >> 2) <= ring->buf_mask + 1) {
+ memcpy((void *)&ring->ring[offset], de_payload_cpu_addr, payload_size);
+ } else {
+ memcpy((void *)&ring->ring[offset], de_payload_cpu_addr,
+ (ring->buf_mask + 1 - offset) << 2);
+ payload_size -= (ring->buf_mask + 1 - offset) << 2;
+ memcpy((void *)&ring->ring[0],
+ de_payload_cpu_addr + ((ring->buf_mask + 1 - offset) << 2),
+ payload_size);
+ }
+}
+
static void gfx_v9_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
struct amdgpu_job *job,
struct amdgpu_ib *ib,
@@ -5330,6 +5411,8 @@ static void gfx_v9_0_ring_emit_ce_meta(struct amdgpu_ring *ring, bool resume)
amdgpu_ring_write(ring, lower_32_bits(ce_payload_gpu_addr));
amdgpu_ring_write(ring, upper_32_bits(ce_payload_gpu_addr));
+ amdgpu_ring_ib_on_emit_ce(ring);
+
if (resume)
amdgpu_ring_write_multiple(ring, ce_payload_cpu_addr,
sizeof(ce_payload) >> 2);
@@ -5363,10 +5446,6 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
amdgpu_ring_alloc(ring, 13);
gfx_v9_0_ring_emit_fence(ring, ring->trail_fence_gpu_addr,
ring->trail_seq, AMDGPU_FENCE_FLAG_EXEC | AMDGPU_FENCE_FLAG_INT);
- /*reset the CP_VMID_PREEMPT after trailing fence*/
- amdgpu_ring_emit_wreg(ring,
- SOC15_REG_OFFSET(GC, 0, mmCP_VMID_PREEMPT),
- 0x0);
/* assert IB preemption, emit the trailing fence */
kiq->pmf->kiq_unmap_queues(kiq_ring, ring, PREEMPT_QUEUES_NO_UNMAP,
@@ -5389,6 +5468,10 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
DRM_WARN("ring %d timeout to preempt ib\n", ring->idx);
}
+ /*reset the CP_VMID_PREEMPT after trailing fence*/
+ amdgpu_ring_emit_wreg(ring,
+ SOC15_REG_OFFSET(GC, 0, mmCP_VMID_PREEMPT),
+ 0x0);
amdgpu_ring_commit(ring);
/* deassert preemption condition */
@@ -5396,7 +5479,7 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
return r;
}
-static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
+static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bool usegds)
{
struct amdgpu_device *adev = ring->adev;
struct v9_de_ib_state de_payload = {0};
@@ -5427,8 +5510,10 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
PAGE_SIZE);
}
- de_payload.gds_backup_addrlo = lower_32_bits(gds_addr);
- de_payload.gds_backup_addrhi = upper_32_bits(gds_addr);
+ if (usegds) {
+ de_payload.gds_backup_addrlo = lower_32_bits(gds_addr);
+ de_payload.gds_backup_addrhi = upper_32_bits(gds_addr);
+ }
cnt = (sizeof(de_payload) >> 2) + 4 - 2;
amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, cnt));
@@ -5439,6 +5524,7 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
amdgpu_ring_write(ring, lower_32_bits(de_payload_gpu_addr));
amdgpu_ring_write(ring, upper_32_bits(de_payload_gpu_addr));
+ amdgpu_ring_ib_on_emit_de(ring);
if (resume)
amdgpu_ring_write_multiple(ring, de_payload_cpu_addr,
sizeof(de_payload) >> 2);
@@ -6750,7 +6836,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = {
.nop = PACKET3(PACKET3_NOP, 0x3FFF),
.support_64bit_ptrs = true,
.secure_submission_supported = true,
- .vmhub = AMDGPU_GFXHUB_0,
.get_rptr = gfx_v9_0_ring_get_rptr_gfx,
.get_wptr = gfx_v9_0_ring_get_wptr_gfx,
.set_wptr = gfx_v9_0_ring_set_wptr_gfx,
@@ -6804,7 +6889,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_sw_ring_funcs_gfx = {
.nop = PACKET3(PACKET3_NOP, 0x3FFF),
.support_64bit_ptrs = true,
.secure_submission_supported = true,
- .vmhub = AMDGPU_GFXHUB_0,
.get_rptr = amdgpu_sw_ring_get_rptr_gfx,
.get_wptr = amdgpu_sw_ring_get_wptr_gfx,
.set_wptr = amdgpu_sw_ring_set_wptr_gfx,
@@ -6851,6 +6935,9 @@ static const struct amdgpu_ring_funcs gfx_v9_0_sw_ring_funcs_gfx = {
.emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait,
.soft_recovery = gfx_v9_0_ring_soft_recovery,
.emit_mem_sync = gfx_v9_0_emit_mem_sync,
+ .patch_cntl = gfx_v9_0_ring_patch_cntl,
+ .patch_de = gfx_v9_0_ring_patch_de_meta,
+ .patch_ce = gfx_v9_0_ring_patch_ce_meta,
};
static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = {
@@ -6858,7 +6945,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = {
.align_mask = 0xff,
.nop = PACKET3(PACKET3_NOP, 0x3FFF),
.support_64bit_ptrs = true,
- .vmhub = AMDGPU_GFXHUB_0,
.get_rptr = gfx_v9_0_ring_get_rptr_compute,
.get_wptr = gfx_v9_0_ring_get_wptr_compute,
.set_wptr = gfx_v9_0_ring_set_wptr_compute,
@@ -6897,7 +6983,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_kiq = {
.align_mask = 0xff,
.nop = PACKET3(PACKET3_NOP, 0x3FFF),
.support_64bit_ptrs = true,
- .vmhub = AMDGPU_GFXHUB_0,
.get_rptr = gfx_v9_0_ring_get_rptr_compute,
.get_wptr = gfx_v9_0_ring_get_wptr_compute,
.set_wptr = gfx_v9_0_ring_set_wptr_compute,
generated by cgit v1.2.3 (git 2.39.1) at 2025-08-09 21:05:25 +0000