summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c86
1 files changed, 31 insertions, 55 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 4e694ae9f308..27943e57681c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -3796,7 +3796,7 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, ALIGNMENT_MODE,
SH_MEM_ALIGNMENT_MODE_UNALIGNED);
WREG32(mmSH_MEM_CONFIG, tmp);
- tmp = adev->mc.shared_aperture_start >> 48;
+ tmp = adev->gmc.shared_aperture_start >> 48;
WREG32(mmSH_MEM_BASES, tmp);
}
@@ -4847,6 +4847,9 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring)
/* reset MQD to a clean status */
if (adev->gfx.mec.mqd_backup[mqd_idx])
memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation));
+ /* reset ring buffer */
+ ring->wptr = 0;
+ amdgpu_ring_clear_ring(ring);
} else {
amdgpu_ring_clear_ring(ring);
}
@@ -4921,13 +4924,6 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
/* Test KCQs */
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
ring = &adev->gfx.compute_ring[i];
- if (adev->in_gpu_reset) {
- /* move reset ring buffer to here to workaround
- * compute ring test failed
- */
- ring->wptr = 0;
- amdgpu_ring_clear_ring(ring);
- }
ring->ready = true;
r = amdgpu_ring_test_ring(ring);
if (r)
@@ -6230,19 +6226,6 @@ static void gfx_v8_0_ring_emit_vgt_flush(struct amdgpu_ring *ring)
EVENT_INDEX(0));
}
-
-static void gfx_v8_0_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
-{
- amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
- amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
- WRITE_DATA_DST_SEL(0) |
- WR_CONFIRM));
- amdgpu_ring_write(ring, mmHDP_DEBUG0);
- amdgpu_ring_write(ring, 0);
- amdgpu_ring_write(ring, 1);
-
-}
-
static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
struct amdgpu_ib *ib,
unsigned vmid, bool ctx_switch)
@@ -6332,28 +6315,7 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
{
int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX);
- amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
- amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
- WRITE_DATA_DST_SEL(0)) |
- WR_CONFIRM);
- if (vmid < 8) {
- amdgpu_ring_write(ring,
- (mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR + vmid));
- } else {
- amdgpu_ring_write(ring,
- (mmVM_CONTEXT8_PAGE_TABLE_BASE_ADDR + vmid - 8));
- }
- amdgpu_ring_write(ring, 0);
- amdgpu_ring_write(ring, pd_addr >> 12);
-
- /* bits 0-15 are the VM contexts0-15 */
- /* invalidate the cache */
- amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
- amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
- WRITE_DATA_DST_SEL(0)));
- amdgpu_ring_write(ring, mmVM_INVALIDATE_REQUEST);
- amdgpu_ring_write(ring, 0);
- amdgpu_ring_write(ring, 1 << vmid);
+ amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
/* wait for the invalidate to complete */
amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5));
@@ -6617,8 +6579,22 @@ static void gfx_v8_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg)
static void gfx_v8_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg,
uint32_t val)
{
+ uint32_t cmd;
+
+ switch (ring->funcs->type) {
+ case AMDGPU_RING_TYPE_GFX:
+ cmd = WRITE_DATA_ENGINE_SEL(1) | WR_CONFIRM;
+ break;
+ case AMDGPU_RING_TYPE_KIQ:
+ cmd = 1 << 16; /* no inc addr */
+ break;
+ default:
+ cmd = WR_CONFIRM;
+ break;
+ }
+
amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
- amdgpu_ring_write(ring, (1 << 16)); /* no inc addr */
+ amdgpu_ring_write(ring, cmd);
amdgpu_ring_write(ring, reg);
amdgpu_ring_write(ring, 0);
amdgpu_ring_write(ring, val);
@@ -6871,7 +6847,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
.emit_frame_size = /* maximum 215dw if count 16 IBs in */
5 + /* COND_EXEC */
7 + /* PIPELINE_SYNC */
- 19 + /* VM_FLUSH */
+ VI_FLUSH_GPU_TLB_NUM_WREG * 5 + 9 + /* VM_FLUSH */
8 + /* FENCE for VM_FLUSH */
20 + /* GDS switch */
4 + /* double SWITCH_BUFFER,
@@ -6893,7 +6869,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
.emit_gds_switch = gfx_v8_0_ring_emit_gds_switch,
.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
- .emit_hdp_invalidate = gfx_v8_0_ring_emit_hdp_invalidate,
.test_ring = gfx_v8_0_ring_test_ring,
.test_ib = gfx_v8_0_ring_test_ib,
.insert_nop = amdgpu_ring_insert_nop,
@@ -6902,6 +6877,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
.emit_cntxcntl = gfx_v8_ring_emit_cntxcntl,
.init_cond_exec = gfx_v8_0_ring_emit_init_cond_exec,
.patch_cond_exec = gfx_v8_0_ring_emit_patch_cond_exec,
+ .emit_wreg = gfx_v8_0_ring_emit_wreg,
};
static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
@@ -6915,9 +6891,9 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
.emit_frame_size =
20 + /* gfx_v8_0_ring_emit_gds_switch */
7 + /* gfx_v8_0_ring_emit_hdp_flush */
- 5 + /* gfx_v8_0_ring_emit_hdp_invalidate */
+ 5 + /* hdp_invalidate */
7 + /* gfx_v8_0_ring_emit_pipeline_sync */
- 17 + /* gfx_v8_0_ring_emit_vm_flush */
+ VI_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + /* gfx_v8_0_ring_emit_vm_flush */
7 + 7 + 7, /* gfx_v8_0_ring_emit_fence_compute x3 for user fence, vm fence */
.emit_ib_size = 4, /* gfx_v8_0_ring_emit_ib_compute */
.emit_ib = gfx_v8_0_ring_emit_ib_compute,
@@ -6926,12 +6902,12 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
.emit_gds_switch = gfx_v8_0_ring_emit_gds_switch,
.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
- .emit_hdp_invalidate = gfx_v8_0_ring_emit_hdp_invalidate,
.test_ring = gfx_v8_0_ring_test_ring,
.test_ib = gfx_v8_0_ring_test_ib,
.insert_nop = amdgpu_ring_insert_nop,
.pad_ib = amdgpu_ring_generic_pad_ib,
.set_priority = gfx_v8_0_ring_set_priority_compute,
+ .emit_wreg = gfx_v8_0_ring_emit_wreg,
};
static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_kiq = {
@@ -6945,7 +6921,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_kiq = {
.emit_frame_size =
20 + /* gfx_v8_0_ring_emit_gds_switch */
7 + /* gfx_v8_0_ring_emit_hdp_flush */
- 5 + /* gfx_v8_0_ring_emit_hdp_invalidate */
+ 5 + /* hdp_invalidate */
7 + /* gfx_v8_0_ring_emit_pipeline_sync */
17 + /* gfx_v8_0_ring_emit_vm_flush */
7 + 7 + 7, /* gfx_v8_0_ring_emit_fence_kiq x3 for user fence, vm fence */
@@ -7151,12 +7127,12 @@ static void gfx_v8_0_ring_emit_ce_meta(struct amdgpu_ring *ring)
} ce_payload = {};
if (ring->adev->virt.chained_ib_support) {
- ce_payload_addr = AMDGPU_VA_RESERVED_SIZE - 2 * 4096 +
- offsetof(struct vi_gfx_meta_data_chained_ib, ce_payload);
+ ce_payload_addr = amdgpu_csa_vaddr(ring->adev) +
+ offsetof(struct vi_gfx_meta_data_chained_ib, ce_payload);
cnt_ce = (sizeof(ce_payload.chained) >> 2) + 4 - 2;
} else {
- ce_payload_addr = AMDGPU_VA_RESERVED_SIZE - 2 * 4096 +
- offsetof(struct vi_gfx_meta_data, ce_payload);
+ ce_payload_addr = amdgpu_csa_vaddr(ring->adev) +
+ offsetof(struct vi_gfx_meta_data, ce_payload);
cnt_ce = (sizeof(ce_payload.regular) >> 2) + 4 - 2;
}
@@ -7179,7 +7155,7 @@ static void gfx_v8_0_ring_emit_de_meta(struct amdgpu_ring *ring)
struct vi_de_ib_state_chained_ib chained;
} de_payload = {};
- csa_addr = AMDGPU_VA_RESERVED_SIZE - 2 * 4096;
+ csa_addr = amdgpu_csa_vaddr(ring->adev);
gds_addr = csa_addr + 4096;
if (ring->adev->virt.chained_ib_support) {
de_payload.chained.gds_backup_addrlo = lower_32_bits(gds_addr);
generated by cgit v1.2.3 (git 2.39.0) at 2024-11-26 01:10:08 +0300