Merge tag 'amd-drm-next-5.14-2021-05-19' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-5.14-2021-05-19:

amdgpu:
- Aldebaran updates
- More LTTPR display work
- Vangogh updates
- SDMA 5.x GCR fixes
- RAS fixes
- PCIe ASPM support
- Modifier fixes
- Enable TMZ on Renoir
- Buffer object code cleanup
- Display overlay fixes
- Initial support for multiple eDP panels
- Initial SR-IOV support for Aldebaran
- DP link training refactor
- Misc code cleanups and bug fixes
- SMU regression fixes for variable sized arrays
- MAINTAINERS fixes for amdgpu

amdkfd:
- Initial SR-IOV support for Aldebaran
- Topology fixes
- Initial HMM SVM support
- Misc code cleanups and bug fixes

radeon:
- Misc code cleanups and bug fixes
- SMU regression fixes for variable sized arrays
- Flickering fix for Oland with multiple 4K displays

UAPI:
- amdgpu: Drop AMDGPU_GEM_CREATE_SHADOW flag.
  This was always a kernel internal flag and userspace use of it has always been blocked.
  It's no longer needed so remove it.
- amdkgd: HMM SVM support
  Overview: https://patchwork.freedesktop.org/series/85562/
  Porposed userspace: https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/tree/fxkamd/hmm-wip

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210520031258.231896-1-alexander.deucher@amd.com

1 files changed, 28 insertions, 12 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index 920fc6d4a127..04c68a79eca4 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -405,18 +405,6 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring,
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	uint64_t csa_mc_addr = amdgpu_sdma_get_csa_mc_addr(ring, vmid);
 
-	/* Invalidate L2, because if we don't do it, we might get stale cache
-	 * lines from previous IBs.
-	 */
-	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_GCR_REQ));
-	amdgpu_ring_write(ring, 0);
-	amdgpu_ring_write(ring, (SDMA_GCR_GL2_INV |
-				 SDMA_GCR_GL2_WB |
-				 SDMA_GCR_GLM_INV |
-				 SDMA_GCR_GLM_WB) << 16);
-	amdgpu_ring_write(ring, 0xffffff80);
-	amdgpu_ring_write(ring, 0xffff);
-
 	/* An IB packet must end on a 8 DW boundary--the next dword
 	 * must be on a 8-dword boundary. Our IB packet below is 6
 	 * dwords long, thus add x number of NOPs, such that, in
@@ -438,6 +426,33 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring,
 }
 
 /**
+ * sdma_v5_0_ring_emit_mem_sync - flush the IB by graphics cache rinse
+ *
+ * @ring: amdgpu ring pointer
+ * @job: job to retrieve vmid from
+ * @ib: IB object to schedule
+ *
+ * flush the IB by graphics cache rinse.
+ */
+static void sdma_v5_0_ring_emit_mem_sync(struct amdgpu_ring *ring)
+{
+    uint32_t gcr_cntl =
+		    SDMA_GCR_GL2_INV | SDMA_GCR_GL2_WB | SDMA_GCR_GLM_INV |
+			SDMA_GCR_GL1_INV | SDMA_GCR_GLV_INV | SDMA_GCR_GLK_INV |
+			SDMA_GCR_GLI_INV(1);
+
+	/* flush entire cache L0/L1/L2, this can be optimized by performance requirement */
+	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_GCR_REQ));
+	amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD1_BASE_VA_31_7(0));
+	amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD2_GCR_CONTROL_15_0(gcr_cntl) |
+			SDMA_PKT_GCR_REQ_PAYLOAD2_BASE_VA_47_32(0));
+	amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD3_LIMIT_VA_31_7(0) |
+			SDMA_PKT_GCR_REQ_PAYLOAD3_GCR_CONTROL_18_16(gcr_cntl >> 16));
+	amdgpu_ring_write(ring, SDMA_PKT_GCR_REQ_PAYLOAD4_LIMIT_VA_47_32(0) |
+			SDMA_PKT_GCR_REQ_PAYLOAD4_VMID(0));
+}
+
+/**
  * sdma_v5_0_ring_emit_hdp_flush - emit an hdp flush on the DMA ring
  *
  * @ring: amdgpu ring pointer
@@ -1643,6 +1658,7 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = {
 		10 + 10 + 10, /* sdma_v5_0_ring_emit_fence x3 for user fence, vm fence */
 	.emit_ib_size = 5 + 7 + 6, /* sdma_v5_0_ring_emit_ib */
 	.emit_ib = sdma_v5_0_ring_emit_ib,
+	.emit_mem_sync = sdma_v5_0_ring_emit_mem_sync,
 	.emit_fence = sdma_v5_0_ring_emit_fence,
 	.emit_pipeline_sync = sdma_v5_0_ring_emit_pipeline_sync,
 	.emit_vm_flush = sdma_v5_0_ring_emit_vm_flush,