drm/amdgpu/vcn3.0: add dec software ring vm functions to support

Add dec software ring vm functions to support.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index c5e0a53..a94dce4 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -48,6 +48,7 @@
 #define mmUVD_RBC_IB_SIZE_INTERNAL_OFFSET                      0x25c
 
 #define VCN_INSTANCES_SIENNA_CICHLID                           2
+#define DEC_SW_RING_ENABLED                                    FALSE
 
 static int amdgpu_ih_clientid_vcns[] = {
        SOC15_IH_CLIENTID_VCN,
@@ -1673,6 +1674,98 @@ static void vcn_v3_0_dec_ring_set_wptr(struct amdgpu_ring *ring)
        }
 }
 
+void vcn_v3_0_dec_sw_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
+                               u64 seq, uint32_t flags)
+{
+       WARN_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
+
+       amdgpu_ring_write(ring, VCN_DEC_SW_CMD_FENCE);
+       amdgpu_ring_write(ring, addr);
+       amdgpu_ring_write(ring, upper_32_bits(addr));
+       amdgpu_ring_write(ring, seq);
+       amdgpu_ring_write(ring, VCN_DEC_SW_CMD_TRAP);
+}
+
+void vcn_v3_0_dec_sw_ring_insert_end(struct amdgpu_ring *ring)
+{
+       amdgpu_ring_write(ring, VCN_DEC_SW_CMD_END);
+}
+
+void vcn_v3_0_dec_sw_ring_emit_ib(struct amdgpu_ring *ring,
+                              struct amdgpu_job *job,
+                              struct amdgpu_ib *ib,
+                              uint32_t flags)
+{
+       uint32_t vmid = AMDGPU_JOB_GET_VMID(job);
+
+       amdgpu_ring_write(ring, VCN_DEC_SW_CMD_IB);
+       amdgpu_ring_write(ring, vmid);
+       amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
+       amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
+       amdgpu_ring_write(ring, ib->length_dw);
+}
+
+void vcn_v3_0_dec_sw_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
+                               uint32_t val, uint32_t mask)
+{
+       amdgpu_ring_write(ring, VCN_DEC_SW_CMD_REG_WAIT);
+       amdgpu_ring_write(ring, reg << 2);
+       amdgpu_ring_write(ring, mask);
+       amdgpu_ring_write(ring, val);
+}
+
+void vcn_v3_0_dec_sw_ring_emit_vm_flush(struct amdgpu_ring *ring,
+                               uint32_t vmid, uint64_t pd_addr)
+{
+       struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->funcs->vmhub];
+       uint32_t data0, data1, mask;
+
+       pd_addr = amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
+
+       /* wait for register write */
+       data0 = hub->ctx0_ptb_addr_lo32 + vmid * hub->ctx_addr_distance;
+       data1 = lower_32_bits(pd_addr);
+       mask = 0xffffffff;
+       vcn_v3_0_dec_sw_ring_emit_reg_wait(ring, data0, data1, mask);
+}
+
+void vcn_v3_0_dec_sw_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val)
+{
+       amdgpu_ring_write(ring, VCN_DEC_SW_CMD_REG_WRITE);
+       amdgpu_ring_write(ring, reg << 2);
+       amdgpu_ring_write(ring, val);
+}
+
+static const struct amdgpu_ring_funcs vcn_v3_0_dec_sw_ring_vm_funcs = {
+       .type = AMDGPU_RING_TYPE_VCN_DEC,
+       .align_mask = 0x3f,
+       .nop = VCN_DEC_SW_CMD_NO_OP,
+       .vmhub = AMDGPU_MMHUB_0,
+       .get_rptr = vcn_v3_0_dec_ring_get_rptr,
+       .get_wptr = vcn_v3_0_dec_ring_get_wptr,
+       .set_wptr = vcn_v3_0_dec_ring_set_wptr,
+       .emit_frame_size =
+               SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
+               SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 4 +
+               4 + /* vcn_v3_0_dec_sw_ring_emit_vm_flush */
+               5 + 5 + /* vcn_v3_0_dec_sw_ring_emit_fdec_swe x2 vm fdec_swe */
+               1, /* vcn_v3_0_dec_sw_ring_insert_end */
+       .emit_ib_size = 5, /* vcn_v3_0_dec_sw_ring_emit_ib */
+       .emit_ib = vcn_v3_0_dec_sw_ring_emit_ib,
+       .emit_fence = vcn_v3_0_dec_sw_ring_emit_fence,
+       .emit_vm_flush = vcn_v3_0_dec_sw_ring_emit_vm_flush,
+       .test_ring = amdgpu_vcn_dec_sw_ring_test_ring,
+       .test_ib = NULL,//amdgpu_vcn_dec_sw_ring_test_ib,
+       .insert_nop = amdgpu_ring_insert_nop,
+       .insert_end = vcn_v3_0_dec_sw_ring_insert_end,
+       .pad_ib = amdgpu_ring_generic_pad_ib,
+       .begin_use = amdgpu_vcn_ring_begin_use,
+       .end_use = amdgpu_vcn_ring_end_use,
+       .emit_wreg = vcn_v3_0_dec_sw_ring_emit_wreg,
+       .emit_reg_wait = vcn_v3_0_dec_sw_ring_emit_reg_wait,
+       .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+};
+
 static const struct amdgpu_ring_funcs vcn_v3_0_dec_ring_vm_funcs = {
        .type = AMDGPU_RING_TYPE_VCN_DEC,
        .align_mask = 0xf,
@@ -1810,9 +1903,13 @@ static void vcn_v3_0_set_dec_ring_funcs(struct amdgpu_device *adev)
                if (adev->vcn.harvest_config & (1 << i))
                        continue;
 
-               adev->vcn.inst[i].ring_dec.funcs = &vcn_v3_0_dec_ring_vm_funcs;
+               if (!DEC_SW_RING_ENABLED)
+                       adev->vcn.inst[i].ring_dec.funcs = &vcn_v3_0_dec_ring_vm_funcs;
+               else
+                       adev->vcn.inst[i].ring_dec.funcs = &vcn_v3_0_dec_sw_ring_vm_funcs;
                adev->vcn.inst[i].ring_dec.me = i;
-               DRM_INFO("VCN(%d) decode is enabled in VM mode\n", i);
+               DRM_INFO("VCN(%d) decode%s is enabled in VM mode\n", i,
+                         DEC_SW_RING_ENABLED?"(Software Ring)":"");
        }
 }