Merge tag 'amd-drm-next-6.11-2024-06-07' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.11-2024-06-07:

amdgpu:
- DCN 4.0.x support
- DCN 3.5 updates
- GC 12.0 support
- DP MST fixes
- Cursor fixes
- MES11 updates
- MMHUB 4.1 support
- DML2 Updates
- DCN 3.1.5 fixes
- IPS fixes
- Various code cleanups
- GMC 12.0 support
- SDMA 7.0 support
- SMU 13 updates
- SR-IOV fixes
- VCN 5.x fixes
- MES12 support
- SMU 14.x updates
- Devcoredump improvements
- Fixes for HDP flush on platforms with >4k pages
- GC 9.4.3 fixes
- RAS ACA updates
- Silence UBSAN flex array warnings
- MMHUB 3.3 updates

amdkfd:
- Contiguous VRAM allocations
- GC 12.0 support
- SDMA 7.0 support
- SR-IOV fixes

radeon:
- Backlight workaround for iMac
- Silence UBSAN flex array warnings

UAPI:
- GFX12 modifier and DCC support
  Proposed Mesa changes:
  https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29510
- KFD GFX ALU exceptions
  Proposed ROCdebugger changes:
  https://github.com/ROCm/ROCdbgapi/commit/08c760622b6601abf906f75abbc5e21d9fd425df
  https://github.com/ROCm/ROCgdb/commit/944fe1c1414a68700414e86e32273b6bfa62ba6f
- KFD Contiguous VRAM allocation flag
  Proposed ROCr/HIP changes:
  https://github.com/ROCm/ROCT-Thunk-Interface/commit/f7b4a269914a3ab4f1e2453c2879adb97b5cc9e5
  https://github.com/ROCm/ROCR-Runtime/pull/214/commits/26e8530d05a775872cb06dde6693db72be0c454a
  https://github.com/ROCm/clr/commit/1d48f2a1ab38b632919c4b7274899b3faf4279ff

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240607195900.902537-1-alexander.deucher@amd.com

1 files changed, 53 insertions, 0 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c
index 238ea40c2450..8928f9160822 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c
@@ -33,6 +33,10 @@
 #define regMMVM_L2_CNTL3_DEFAULT				0x80100007
 #define regMMVM_L2_CNTL4_DEFAULT				0x000000c1
 #define regMMVM_L2_CNTL5_DEFAULT				0x00003fe0
+#define regDAGB0_L1TLB_REG_RW_3_3                   0x00a4
+#define regDAGB0_L1TLB_REG_RW_3_3_BASE_IDX          1
+#define regDAGB1_L1TLB_REG_RW_3_3                   0x0163
+#define regDAGB1_L1TLB_REG_RW_3_3_BASE_IDX          1
 
 static const char *mmhub_client_ids_v3_3[][2] = {
 	[0][0] = "VMC",
@@ -359,6 +363,49 @@ static void mmhub_v3_3_program_invalidation(struct amdgpu_device *adev)
 	}
 }
 
+static void mmhub_v3_3_init_saw_regs(struct amdgpu_device *adev)
+{
+	uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+	uint32_t tmp;
+
+	/* Program page table base, gart start, gart end */
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+			lower_32_bits(pt_base >> 12));
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+			upper_32_bits(pt_base >> 12));
+
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+		     (u32)(adev->gmc.gart_start >> 12));
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+		     (u32)(adev->gmc.gart_start >> 44));
+
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+		     (u32)(adev->gmc.gart_end >> 12));
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+		     (u32)(adev->gmc.gart_end >> 44));
+
+	tmp = RREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_CNTL);
+	tmp = REG_SET_FIELD(tmp, MMVM_L2_SAW_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
+	tmp = REG_SET_FIELD(tmp, MMVM_L2_SAW_CONTEXT0_CNTL, PAGE_TABLE_DEPTH, 0);
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXT0_CNTL, tmp);
+
+	/* Disable all contexts except context 0 */
+	tmp = 0xfffe;
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CONTEXTS_DISABLE, tmp);
+
+	/* Program saw cntl4 */
+	tmp = RREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CNTL4);
+	tmp = REG_SET_FIELD(tmp, MMVM_L2_SAW_CNTL4, VMC_TAP_CONTEXT0_PDE_REQUEST_SNOOP, 1);
+	tmp = REG_SET_FIELD(tmp, MMVM_L2_SAW_CNTL4, VMC_TAP_CONTEXT0_PTE_REQUEST_SNOOP, 1);
+	WREG32_SOC15(MMHUB, 0, regMMVM_L2_SAW_CNTL4, tmp);
+}
+
+static void mmhub_v3_3_enable_tls(struct amdgpu_device *adev)
+{
+	WREG32_SOC15(MMHUB, 0, regDAGB0_L1TLB_REG_RW_3_3, 0);
+	WREG32_SOC15(MMHUB, 0, regDAGB1_L1TLB_REG_RW_3_3, 3);
+}
+
 static int mmhub_v3_3_gart_enable(struct amdgpu_device *adev)
 {
 	/* GART Enable. */
@@ -372,6 +419,12 @@ static int mmhub_v3_3_gart_enable(struct amdgpu_device *adev)
 	mmhub_v3_3_setup_vmid_config(adev);
 	mmhub_v3_3_program_invalidation(adev);
 
+	/* standalone alone walker init */
+	mmhub_v3_3_init_saw_regs(adev);
+
+	/* enable mmhub tls */
+	mmhub_v3_3_enable_tls(adev);
+
 	return 0;
 }