* [PATCH 2/2] drm/amdgpu: Add initial support for mmhub v4_2
2025-10-30 20:50 [PATCH 0/2] Add mmhub 4.2.0 support Alex Deucher
@ 2025-10-30 20:50 ` Alex Deucher
0 siblings, 0 replies; 3+ messages in thread
From: Alex Deucher @ 2025-10-30 20:50 UTC (permalink / raw)
To: amd-gfx; +Cc: Likun Gao, Le Ma, Hawking Zhang, Alex Deucher
From: Likun Gao <Likun.Gao@amd.com>
Add initial support for mmhub v4_2_0.
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/amd/amdgpu/Makefile | 3 +-
drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c | 915 ++++++++++++++++++++++
drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h | 28 +
3 files changed, 945 insertions(+), 1 deletion(-)
create mode 100644 drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
create mode 100644 drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 1fc75f46b4273..ad9ec94fcb088 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -104,7 +104,8 @@ amdgpu-y += \
gfxhub_v2_0.o mmhub_v2_0.o gmc_v10_0.o gfxhub_v2_1.o mmhub_v2_3.o \
mmhub_v1_7.o gfxhub_v3_0.o mmhub_v3_0.o mmhub_v3_0_2.o gmc_v11_0.o \
mmhub_v3_0_1.o gfxhub_v3_0_3.o gfxhub_v1_2.o mmhub_v1_8.o mmhub_v3_3.o \
- gfxhub_v11_5_0.o mmhub_v4_1_0.o gfxhub_v12_0.o gmc_v12_0.o
+ gfxhub_v11_5_0.o mmhub_v4_1_0.o gfxhub_v12_0.o gmc_v12_0.o \
+ mmhub_v4_2_0.o
# add UMC block
amdgpu-y += \
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
new file mode 100644
index 0000000000000..5e1d5e986f191
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
@@ -0,0 +1,915 @@
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "amdgpu.h"
+#include "mmhub_v4_2_0.h"
+
+#include "mmhub/mmhub_4_2_0_offset.h"
+#include "mmhub/mmhub_4_2_0_sh_mask.h"
+
+#include "soc15_common.h"
+#include "soc24_enum.h"
+
+#define regMMVM_L2_CNTL3_DEFAULT 0x80100007
+#define regMMVM_L2_CNTL4_DEFAULT 0x000000c1
+#define regMMVM_L2_CNTL5_DEFAULT 0x00003fe0
+
+#define MMHUB_NUM_MID 1
+
+static const char *mmhub_client_ids_v4_2_0[][2] = {
+ [0][0] = "VMC",
+ [4][0] = "DCEDMC",
+ [5][0] = "DCEVGA",
+ [6][0] = "MP0",
+ [7][0] = "MP1",
+ [8][0] = "MPIO",
+ [16][0] = "HDP",
+ [17][0] = "LSDMA",
+ [18][0] = "JPEG",
+ [19][0] = "VCNU0",
+ [21][0] = "VSCH",
+ [22][0] = "VCNU1",
+ [23][0] = "VCN1",
+ [32+20][0] = "VCN0",
+ [2][1] = "DBGUNBIO",
+ [3][1] = "DCEDWB",
+ [4][1] = "DCEDMC",
+ [5][1] = "DCEVGA",
+ [6][1] = "MP0",
+ [7][1] = "MP1",
+ [8][1] = "MPIO",
+ [10][1] = "DBGU0",
+ [11][1] = "DBGU1",
+ [12][1] = "DBGU2",
+ [13][1] = "DBGU3",
+ [14][1] = "XDP",
+ [15][1] = "OSSSYS",
+ [16][1] = "HDP",
+ [17][1] = "LSDMA",
+ [18][1] = "JPEG",
+ [19][1] = "VCNU0",
+ [20][1] = "VCN0",
+ [21][1] = "VSCH",
+ [22][1] = "VCNU1",
+ [23][1] = "VCN1",
+};
+
+static u64 mmhub_v4_2_0_get_fb_location(struct amdgpu_device *adev)
+{
+ u64 base;
+
+ base = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0),
+ regMMMC_VM_FB_LOCATION_BASE_LO32);
+ base &= MMMC_VM_FB_LOCATION_BASE_LO32__FB_BASE_LO32_MASK;
+ base <<= 24;
+
+ base |= ((MMMC_VM_FB_LOCATION_BASE_HI32__FB_BASE_HI1_MASK &
+ RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0),
+ regMMMC_VM_FB_LOCATION_BASE_HI32)) << 56);
+
+ return base;
+}
+
+static u64 mmhub_v4_2_0_get_mc_fb_offset(struct amdgpu_device *adev)
+{
+ return (u64)RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0),
+ regMMMC_VM_FB_OFFSET) << 24;
+}
+
+static void mmhub_v4_2_0_mid_setup_vm_pt_regs(struct amdgpu_device *adev,
+ uint32_t vmid,
+ uint64_t page_table_base,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+ hub->ctx_addr_distance * vmid,
+ lower_32_bits(page_table_base));
+
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+ hub->ctx_addr_distance * vmid,
+ upper_32_bits(page_table_base));
+ }
+}
+
+static void mmhub_v4_2_0_setup_vm_pt_regs(struct amdgpu_device *adev,
+ uint32_t vmid,
+ uint64_t page_table_base)
+{
+ uint32_t mid_mask;
+
+ mid_mask = GENMASK(MMHUB_NUM_MID - 1, 0);
+ mmhub_v4_2_0_mid_setup_vm_pt_regs(adev, vmid,
+ page_table_base,
+ mid_mask);
+}
+
+static void mmhub_v4_2_0_mid_init_gart_aperture_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+ int i;
+
+ if (adev->gmc.pdb0_bo)
+ pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
+ else
+ pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+ mmhub_v4_2_0_mid_setup_vm_pt_regs(adev, 0, pt_base, mid_mask);
+
+ for_each_inst(i, mid_mask) {
+ if (adev->gmc.pdb0_bo) {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+ (u32)(adev->gmc.fb_start >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+ (u32)(adev->gmc.fb_start >> 44));
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+ (u32)(adev->gmc.fb_end >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+ (u32)(adev->gmc.fb_end >> 44));
+ } else {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+ (u32)(adev->gmc.gart_start >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+ (u32)(adev->gmc.gart_start >> 44));
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+ (u32)(adev->gmc.gart_end >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+ (u32)(adev->gmc.gart_end >> 44));
+ }
+ }
+}
+
+static void mmhub_v4_2_0_mid_init_system_aperture_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint64_t value;
+ uint32_t tmp;
+ int i;
+
+ /*
+ * the new L1 policy will block SRIOV guest from writing
+ * these regs, and they will be programed at host.
+ * so skip programing these regs.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ /* Program the AGP BAR */
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BASE_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BASE_HI32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_LO32,
+ lower_32_bits(adev->gmc.agp_start >> 24));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_HI32,
+ upper_32_bits(adev->gmc.agp_start >> 24));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_LO32,
+ lower_32_bits(adev->gmc.agp_end >> 24));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_HI32,
+ upper_32_bits(adev->gmc.agp_end >> 24));
+
+ /* Program the system aperture low logical page number. */
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_LO32,
+ lower_32_bits(min(adev->gmc.fb_start,
+ adev->gmc.agp_start) >> 18));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_HI32,
+ upper_32_bits(min(adev->gmc.fb_start,
+ adev->gmc.agp_start) >> 18));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_LO32,
+ lower_32_bits(max(adev->gmc.fb_end,
+ adev->gmc.agp_end) >> 18));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_HI32,
+ upper_32_bits(max(adev->gmc.fb_end,
+ adev->gmc.agp_end) >> 18));
+
+ /* Set default page address. */
+ value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+ (u32)(value >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+ (u32)(value >> 44));
+
+ /* Program "protection fault". */
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+ (u32)(adev->dummy_page_addr >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+ (u32)((u64)adev->dummy_page_addr >> 44));
+
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL2);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL2,
+ ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+ }
+
+ /* In the case squeezing vram into GART aperture, we don't use
+ * FB aperture and AGP aperture. Disable them.
+ */
+ if (adev->gmc.pdb0_bo) {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_TOP_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_TOP_HI32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_BASE_LO32, 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_BASE_HI32, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_HI32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_LO32, 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_HI32, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_LO32,
+ 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_HI32,
+ 0x7F);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_HI32, 0);
+ }
+}
+
+static void mmhub_v4_2_0_mid_init_tlb_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint32_t tmp;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ /* Setup TLB control */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_MX_L1_TLB_CNTL);
+
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ ENABLE_ADVANCED_DRIVER_MODEL, 1);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ MTYPE, MTYPE_UC); /* UC, uncached */
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_MX_L1_TLB_CNTL, tmp);
+ }
+}
+
+static void mmhub_v4_2_0_mid_init_cache_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint32_t tmp;
+ int i;
+
+ /* These registers are not accessible to VF-SRIOV.
+ * The PF will program them instead.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ /* Setup L2 cache */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, ENABLE_L2_CACHE, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ ENABLE_DEFAULT_PAGE_OUT_TO_SYSTEM_MEMORY, 1);
+ /* XXX for emulation, Refer to closed source code.*/
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ L2_PDE0_CACHE_TAG_GENERATION_MODE, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ PDE_FAULT_CLASSIFICATION, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ CONTEXT1_IDENTITY_ACCESS_MODE, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ IDENTITY_MODE_FRAGMENT_SIZE, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL, tmp);
+
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL2);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL2,
+ INVALIDATE_ALL_L1_TLBS, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL2,
+ INVALIDATE_L2_CACHE, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL2, tmp);
+
+ tmp = regMMVM_L2_CNTL3_DEFAULT;
+ if (adev->gmc.translate_further) {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3, BANK_SELECT, 12);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3,
+ L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+ } else {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3, BANK_SELECT, 9);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3,
+ L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+ }
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL3, tmp);
+
+ tmp = regMMVM_L2_CNTL4_DEFAULT;
+ /* For AMD APP APUs setup WC memory */
+ if (adev->gmc.xgmi.connected_to_cpu || adev->gmc.is_app_apu) {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
+ } else {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+ }
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL4, tmp);
+
+ tmp = regMMVM_L2_CNTL5_DEFAULT;
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL5,
+ L2_CACHE_SMALLK_FRAGMENT_SIZE, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL5, tmp);
+ }
+}
+
+static void mmhub_v4_2_0_mid_enable_system_domain(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint32_t tmp;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT0_CNTL,
+ ENABLE_CONTEXT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT0_CNTL,
+ PAGE_TABLE_DEPTH, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT0_CNTL,
+ RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_CNTL, tmp);
+ }
+}
+
+static void mmhub_v4_2_0_mid_disable_identity_aperture(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ int i;
+
+ /* These registers are not accessible to VF-SRIOV.
+ * The PF will program them instead.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
+ 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
+ 0x00001FFF);
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
+ 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
+ 0);
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32,
+ 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32,
+ 0);
+ }
+}
+
+static void mmhub_v4_2_0_mid_setup_vmid_config(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ uint32_t tmp;
+ int i, j;
+
+ for_each_inst(j, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+ for (i = 0; i <= 14; i++) {
+ tmp = RREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_CONTEXT1_CNTL,
+ i * hub->ctx_distance);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
+ adev->vm_manager.num_level);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+ 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ PAGE_TABLE_BLOCK_SIZE,
+ adev->vm_manager.block_size - 9);
+ /* Send no-retry XNACK on fault to suppress VM fault storm. */
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
+ !amdgpu_noretry);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_CNTL,
+ i * hub->ctx_distance, tmp);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
+ i * hub->ctx_addr_distance, 0);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
+ i * hub->ctx_addr_distance, 0);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
+ i * hub->ctx_addr_distance,
+ lower_32_bits(adev->vm_manager.max_pfn - 1));
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
+ i * hub->ctx_addr_distance,
+ upper_32_bits(adev->vm_manager.max_pfn - 1));
+ }
+ }
+
+ hub->vm_cntx_cntl = tmp;
+}
+
+static void mmhub_v4_2_0_mid_program_invalidation(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ unsigned int i, j;
+
+ for_each_inst(j, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+
+ for (i = 0; i < 18; ++i) {
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+ i * hub->eng_addr_distance, 0xffffffff);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+ i * hub->eng_addr_distance, 0x3fff);
+ }
+ }
+}
+
+static int mmhub_v4_2_0_mid_gart_enable(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ /* GART Enable. */
+ mmhub_v4_2_0_mid_init_gart_aperture_regs(adev, mid_mask);
+ mmhub_v4_2_0_mid_init_system_aperture_regs(adev, mid_mask);
+ mmhub_v4_2_0_mid_init_tlb_regs(adev, mid_mask);
+ mmhub_v4_2_0_mid_init_cache_regs(adev, mid_mask);
+
+ mmhub_v4_2_0_mid_enable_system_domain(adev, mid_mask);
+ mmhub_v4_2_0_mid_disable_identity_aperture(adev, mid_mask);
+ mmhub_v4_2_0_mid_setup_vmid_config(adev, mid_mask);
+ mmhub_v4_2_0_mid_program_invalidation(adev, mid_mask);
+
+ return 0;
+}
+static int mmhub_v4_2_0_gart_enable(struct amdgpu_device *adev)
+{
+ uint32_t mid_mask;
+
+ mid_mask = GENMASK(MMHUB_NUM_MID - 1, 0);
+ return mmhub_v4_2_0_mid_gart_enable(adev, mid_mask);
+}
+
+static void mmhub_v4_2_0_mid_gart_disable(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ u32 tmp;
+ u32 i, j;
+
+ for_each_inst(j, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+ /* Disable all tables */
+ for (i = 0; i < 16; i++)
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_CONTEXT0_CNTL,
+ i * hub->ctx_distance, 0);
+
+ /* Setup TLB control */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, j),
+ regMMMC_VM_MX_L1_TLB_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ ENABLE_L1_TLB, 0);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ ENABLE_ADVANCED_DRIVER_MODEL, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, j),
+ regMMMC_VM_MX_L1_TLB_CNTL, tmp);
+
+ /* Setup L2 cache */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, j), regMMVM_L2_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, ENABLE_L2_CACHE, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, j), regMMVM_L2_CNTL, tmp);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, j), regMMVM_L2_CNTL3, 0);
+ }
+}
+
+static void mmhub_v4_2_0_gart_disable(struct amdgpu_device *adev)
+{
+ uint32_t mid_mask;
+
+ mid_mask = GENMASK(MMHUB_NUM_MID - 1, 0);
+ mmhub_v4_2_0_mid_gart_disable(adev, mid_mask);
+}
+
+static void
+mmhub_v4_2_0_mid_set_fault_enable_default(struct amdgpu_device *adev,
+ bool value, uint32_t mid_mask)
+{
+ u32 tmp;
+ int i;
+
+ /* These registers are not accessible to VF-SRIOV.
+ * The PF will program them instead.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL_LO32);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
+ value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ if (!value) {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ CRASH_ON_NO_RETRY_FAULT, 1);
+ }
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL_LO32, tmp);
+ }
+}
+
+
+/**
+ * mmhub_v4_2_0_set_fault_enable_default - update GART/VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void
+mmhub_v4_2_0_set_fault_enable_default(struct amdgpu_device *adev,
+ bool value)
+{
+ uint32_t mid_mask;
+
+ mid_mask = GENMASK(MMHUB_NUM_MID - 1, 0);
+ mmhub_v4_2_0_mid_set_fault_enable_default(adev, value, mid_mask);
+}
+
+static uint32_t mmhub_v4_2_0_get_invalidate_req(unsigned int vmid,
+ uint32_t flush_type)
+{
+ u32 req = 0;
+
+ /* invalidate using legacy mode on vmid*/
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ,
+ PER_VMID_INVALIDATE_REQ, 1 << vmid);
+ /* Only use legacy inv on mmhub side */
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, 0);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ,
+ CLEAR_PROTECTION_FAULT_STATUS_ADDR, 0);
+
+ return req;
+}
+
+/*TODO: l2 protection fault status is increased to 64bits.
+ * some critical fields like FED are moved to STATUS_HI32 */
+static void
+mmhub_v4_2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,
+ uint32_t status)
+{
+ uint32_t cid, rw;
+ const char *mmhub_cid = NULL;
+
+ cid = REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, CID);
+ rw = REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, RW);
+
+ dev_err(adev->dev,
+ "MMVM_L2_PROTECTION_FAULT_STATUS_LO32:0x%08X\n",
+ status);
+ switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) {
+ case IP_VERSION(4, 2, 0):
+ mmhub_cid = mmhub_client_ids_v4_2_0[cid][rw];
+ break;
+ default:
+ mmhub_cid = NULL;
+ break;
+ }
+ dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+ mmhub_cid ? mmhub_cid : "unknown", cid);
+ dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, MORE_FAULTS));
+ dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, WALKER_ERROR));
+ dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, PERMISSION_FAULTS));
+ dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, MAPPING_ERROR));
+ dev_err(adev->dev, "\t RW: 0x%x\n", rw);
+}
+
+
+static const struct amdgpu_vmhub_funcs mmhub_v4_2_0_vmhub_funcs = {
+ .print_l2_protection_fault_status = mmhub_v4_2_0_print_l2_protection_fault_status,
+ .get_invalidate_req = mmhub_v4_2_0_get_invalidate_req,
+};
+
+static void mmhub_v4_2_0_mid_init(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
+
+ hub->ctx0_ptb_addr_lo32 =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
+ hub->ctx0_ptb_addr_hi32 =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
+ hub->vm_inv_eng0_sem =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_INVALIDATE_ENG0_SEM);
+ hub->vm_inv_eng0_req =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_INVALIDATE_ENG0_REQ);
+ hub->vm_inv_eng0_ack =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_INVALIDATE_ENG0_ACK);
+ hub->vm_context0_cntl =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_CNTL);
+ /* TODO: add a new member to accomandate additional fault status/cntl reg */
+ hub->vm_l2_pro_fault_status =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_STATUS_LO32);
+ hub->vm_l2_pro_fault_cntl =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL_LO32);
+
+ hub->ctx_distance = regMMVM_CONTEXT1_CNTL - regMMVM_CONTEXT0_CNTL;
+ hub->ctx_addr_distance = regMMVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+ hub->eng_distance = regMMVM_INVALIDATE_ENG1_REQ -
+ regMMVM_INVALIDATE_ENG0_REQ;
+ hub->eng_addr_distance = regMMVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
+ regMMVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+
+ hub->vm_cntx_cntl_vm_fault = MMVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
+
+ hub->vm_l2_bank_select_reserved_cid2 =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_BANK_SELECT_RESERVED_CID2);
+
+ hub->vm_contexts_disable =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i), regMMVM_CONTEXTS_DISABLE);
+
+ hub->vmhub_funcs = &mmhub_v4_2_0_vmhub_funcs;
+ }
+}
+
+static void mmhub_v4_2_0_init(struct amdgpu_device *adev)
+{
+ uint32_t mid_mask;
+
+ mid_mask = GENMASK(MMHUB_NUM_MID - 1, 0);
+ mmhub_v4_2_0_mid_init(adev, mid_mask);
+}
+
+static void
+mmhub_v4_2_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+ bool enable)
+{
+ uint32_t def, data;
+ uint32_t def1, data1, def2 = 0, data2 = 0;
+ def = data = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG);
+ def1 = data1 = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB0_CNTL_MISC2);
+ def2 = data2 = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB1_CNTL_MISC2);
+
+ if (enable) {
+ data |= MM_ATC_L2_MISC_CG__ENABLE_MASK;
+ data1 &= ~(DAGB0_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB0_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+
+ data2 &= ~(DAGB1_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB1_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+ } else {
+ data &= ~MM_ATC_L2_MISC_CG__ENABLE_MASK;
+ data1 |= (DAGB0_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB0_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+
+ data2 |= (DAGB1_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB1_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+ }
+
+ if (def != data)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG, data);
+ if (def1 != data1)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB0_CNTL_MISC2, data1);
+
+ if (def2 != data2)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB1_CNTL_MISC2, data2);
+}
+
+static void
+mmhub_v4_2_0_update_medium_grain_light_sleep(struct amdgpu_device *adev,
+ bool enable)
+{
+ uint32_t def, data;
+
+ def = data = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG);
+
+ if (enable)
+ data |= MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK;
+ else
+ data &= ~MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK;
+
+ if (def != data)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG, data);
+}
+
+static int mmhub_v4_2_0_set_clockgating(struct amdgpu_device *adev,
+ enum amd_clockgating_state state)
+{
+ if (amdgpu_sriov_vf(adev))
+ return 0;
+
+ if (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)
+ mmhub_v4_2_0_update_medium_grain_clock_gating(adev,
+ state == AMD_CG_STATE_GATE);
+
+ if (adev->cg_flags & AMD_CG_SUPPORT_MC_LS)
+ mmhub_v4_2_0_update_medium_grain_light_sleep(adev,
+ state == AMD_CG_STATE_GATE);
+
+ return 0;
+}
+
+static void mmhub_v4_2_0_get_clockgating(struct amdgpu_device *adev, u64 *flags)
+{
+ int data;
+
+ if (amdgpu_sriov_vf(adev))
+ *flags = 0;
+
+ data = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG);
+
+ /* AMD_CG_SUPPORT_MC_MGCG */
+ if (data & MM_ATC_L2_MISC_CG__ENABLE_MASK)
+ *flags |= AMD_CG_SUPPORT_MC_MGCG;
+
+ /* AMD_CG_SUPPORT_MC_LS */
+ if (data & MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK)
+ *flags |= AMD_CG_SUPPORT_MC_LS;
+}
+
+const struct amdgpu_mmhub_funcs mmhub_v4_2_0_funcs = {
+ .init = mmhub_v4_2_0_init,
+ .get_fb_location = mmhub_v4_2_0_get_fb_location,
+ .get_mc_fb_offset = mmhub_v4_2_0_get_mc_fb_offset,
+ .setup_vm_pt_regs = mmhub_v4_2_0_setup_vm_pt_regs,
+ .gart_enable = mmhub_v4_2_0_gart_enable,
+ .gart_disable = mmhub_v4_2_0_gart_disable,
+ .set_fault_enable_default = mmhub_v4_2_0_set_fault_enable_default,
+ .set_clockgating = mmhub_v4_2_0_set_clockgating,
+ .get_clockgating = mmhub_v4_2_0_get_clockgating,
+};
+
+static int mmhub_v4_2_0_xcp_resume(void *handle, uint32_t inst_mask)
+{
+ struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+ bool value;
+
+ if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
+ value = false;
+ else
+ value = true;
+
+ mmhub_v4_2_0_mid_set_fault_enable_default(adev, value, inst_mask);
+
+ if (!amdgpu_sriov_vf(adev))
+ return mmhub_v4_2_0_mid_gart_enable(adev, inst_mask);
+
+ return 0;
+}
+
+static int mmhub_v4_2_0_xcp_suspend(void *handle, uint32_t inst_mask)
+{
+ struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+ if (!amdgpu_sriov_vf(adev))
+ mmhub_v4_2_0_mid_gart_disable(adev, inst_mask);
+
+ return 0;
+}
+
+struct amdgpu_xcp_ip_funcs mmhub_v4_2_0_xcp_funcs = {
+ .suspend = &mmhub_v4_2_0_xcp_suspend,
+ .resume = &mmhub_v4_2_0_xcp_resume
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
new file mode 100644
index 0000000000000..4ea6de7ac7ae7
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __MMHUB_V4_2_0_H__
+#define __MMHUB_V4_2_0_H__
+
+extern const struct amdgpu_mmhub_funcs mmhub_v4_2_0_funcs;
+
+#endif
--
2.51.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [PATCH 0/2] Add MMHUB 4.2
@ 2025-11-19 14:50 Alex Deucher
2025-11-19 14:50 ` [PATCH 2/2] drm/amdgpu: Add initial support for mmhub v4_2 Alex Deucher
0 siblings, 1 reply; 3+ messages in thread
From: Alex Deucher @ 2025-11-19 14:50 UTC (permalink / raw)
To: amd-gfx; +Cc: Alex Deucher
Add MMHUB 4.2 support. The first patch
adds register headers and is large for the list.
Hawking Zhang (1):
drm/amdgpu: Add mmhub v4_2_0 ip headers v5
Likun Gao (1):
drm/amdgpu: Add initial support for mmhub v4_2
drivers/gpu/drm/amd/amdgpu/Makefile | 3 +-
drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c | 913 +++++
drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h | 28 +
.../asic_reg/mmhub/mmhub_4_2_0_offset.h | 802 +++++
.../asic_reg/mmhub/mmhub_4_2_0_sh_mask.h | 3013 +++++++++++++++++
5 files changed, 4758 insertions(+), 1 deletion(-)
create mode 100644 drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
create mode 100644 drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
create mode 100644 drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_4_2_0_offset.h
create mode 100644 drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_4_2_0_sh_mask.h
--
2.51.1
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH 2/2] drm/amdgpu: Add initial support for mmhub v4_2
2025-11-19 14:50 [PATCH 0/2] Add MMHUB 4.2 Alex Deucher
@ 2025-11-19 14:50 ` Alex Deucher
0 siblings, 0 replies; 3+ messages in thread
From: Alex Deucher @ 2025-11-19 14:50 UTC (permalink / raw)
To: amd-gfx; +Cc: Likun Gao, Le Ma, Hawking Zhang, Alex Deucher
From: Likun Gao <Likun.Gao@amd.com>
Add initial support for mmhub v4_2_0.
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/amd/amdgpu/Makefile | 3 +-
drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c | 913 ++++++++++++++++++++++
drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h | 28 +
3 files changed, 943 insertions(+), 1 deletion(-)
create mode 100644 drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
create mode 100644 drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index ad57ade4001c5..226b022ab40be 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -104,7 +104,8 @@ amdgpu-y += \
gfxhub_v2_0.o mmhub_v2_0.o gmc_v10_0.o gfxhub_v2_1.o mmhub_v2_3.o \
mmhub_v1_7.o gfxhub_v3_0.o mmhub_v3_0.o mmhub_v3_0_2.o gmc_v11_0.o \
mmhub_v3_0_1.o gfxhub_v3_0_3.o gfxhub_v1_2.o mmhub_v1_8.o mmhub_v3_3.o \
- gfxhub_v11_5_0.o mmhub_v4_1_0.o gfxhub_v12_0.o gmc_v12_0.o
+ gfxhub_v11_5_0.o mmhub_v4_1_0.o gfxhub_v12_0.o gmc_v12_0.o \
+ mmhub_v4_2_0.o
# add UMC block
amdgpu-y += \
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
new file mode 100644
index 0000000000000..23f0042125008
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c
@@ -0,0 +1,913 @@
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "amdgpu.h"
+#include "mmhub_v4_2_0.h"
+
+#include "mmhub/mmhub_4_2_0_offset.h"
+#include "mmhub/mmhub_4_2_0_sh_mask.h"
+
+#include "soc15_common.h"
+#include "soc24_enum.h"
+
+#define regMMVM_L2_CNTL3_DEFAULT 0x80100007
+#define regMMVM_L2_CNTL4_DEFAULT 0x000000c1
+#define regMMVM_L2_CNTL5_DEFAULT 0x00003fe0
+
+static const char *mmhub_client_ids_v4_2_0[][2] = {
+ [0][0] = "VMC",
+ [4][0] = "DCEDMC",
+ [5][0] = "DCEVGA",
+ [6][0] = "MP0",
+ [7][0] = "MP1",
+ [8][0] = "MPIO",
+ [16][0] = "HDP",
+ [17][0] = "LSDMA",
+ [18][0] = "JPEG",
+ [19][0] = "VCNU0",
+ [21][0] = "VSCH",
+ [22][0] = "VCNU1",
+ [23][0] = "VCN1",
+ [32+20][0] = "VCN0",
+ [2][1] = "DBGUNBIO",
+ [3][1] = "DCEDWB",
+ [4][1] = "DCEDMC",
+ [5][1] = "DCEVGA",
+ [6][1] = "MP0",
+ [7][1] = "MP1",
+ [8][1] = "MPIO",
+ [10][1] = "DBGU0",
+ [11][1] = "DBGU1",
+ [12][1] = "DBGU2",
+ [13][1] = "DBGU3",
+ [14][1] = "XDP",
+ [15][1] = "OSSSYS",
+ [16][1] = "HDP",
+ [17][1] = "LSDMA",
+ [18][1] = "JPEG",
+ [19][1] = "VCNU0",
+ [20][1] = "VCN0",
+ [21][1] = "VSCH",
+ [22][1] = "VCNU1",
+ [23][1] = "VCN1",
+};
+
+static u64 mmhub_v4_2_0_get_fb_location(struct amdgpu_device *adev)
+{
+ u64 base;
+
+ base = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0),
+ regMMMC_VM_FB_LOCATION_BASE_LO32);
+ base &= MMMC_VM_FB_LOCATION_BASE_LO32__FB_BASE_LO32_MASK;
+ base <<= 24;
+
+ base |= ((MMMC_VM_FB_LOCATION_BASE_HI32__FB_BASE_HI1_MASK &
+ RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0),
+ regMMMC_VM_FB_LOCATION_BASE_HI32)) << 56);
+
+ return base;
+}
+
+static u64 mmhub_v4_2_0_get_mc_fb_offset(struct amdgpu_device *adev)
+{
+ return (u64)RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0),
+ regMMMC_VM_FB_OFFSET) << 24;
+}
+
+static void mmhub_v4_2_0_mid_setup_vm_pt_regs(struct amdgpu_device *adev,
+ uint32_t vmid,
+ uint64_t page_table_base,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+ hub->ctx_addr_distance * vmid,
+ lower_32_bits(page_table_base));
+
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+ hub->ctx_addr_distance * vmid,
+ upper_32_bits(page_table_base));
+ }
+}
+
+static void mmhub_v4_2_0_setup_vm_pt_regs(struct amdgpu_device *adev,
+ uint32_t vmid,
+ uint64_t page_table_base)
+{
+ uint32_t mid_mask;
+
+ mid_mask = adev->aid_mask;
+ mmhub_v4_2_0_mid_setup_vm_pt_regs(adev, vmid,
+ page_table_base,
+ mid_mask);
+}
+
+static void mmhub_v4_2_0_mid_init_gart_aperture_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+ int i;
+
+ if (adev->gmc.pdb0_bo)
+ pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
+ else
+ pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+ mmhub_v4_2_0_mid_setup_vm_pt_regs(adev, 0, pt_base, mid_mask);
+
+ for_each_inst(i, mid_mask) {
+ if (adev->gmc.pdb0_bo) {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+ (u32)(adev->gmc.fb_start >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+ (u32)(adev->gmc.fb_start >> 44));
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+ (u32)(adev->gmc.fb_end >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+ (u32)(adev->gmc.fb_end >> 44));
+ } else {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+ (u32)(adev->gmc.gart_start >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+ (u32)(adev->gmc.gart_start >> 44));
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+ (u32)(adev->gmc.gart_end >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+ (u32)(adev->gmc.gart_end >> 44));
+ }
+ }
+}
+
+static void mmhub_v4_2_0_mid_init_system_aperture_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint64_t value;
+ uint32_t tmp;
+ int i;
+
+ /*
+ * the new L1 policy will block SRIOV guest from writing
+ * these regs, and they will be programed at host.
+ * so skip programing these regs.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ /* Program the AGP BAR */
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BASE_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BASE_HI32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_LO32,
+ lower_32_bits(adev->gmc.agp_start >> 24));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_HI32,
+ upper_32_bits(adev->gmc.agp_start >> 24));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_LO32,
+ lower_32_bits(adev->gmc.agp_end >> 24));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_HI32,
+ upper_32_bits(adev->gmc.agp_end >> 24));
+
+ /* Program the system aperture low logical page number. */
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_LO32,
+ lower_32_bits(min(adev->gmc.fb_start,
+ adev->gmc.agp_start) >> 18));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_HI32,
+ upper_32_bits(min(adev->gmc.fb_start,
+ adev->gmc.agp_start) >> 18));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_LO32,
+ lower_32_bits(max(adev->gmc.fb_end,
+ adev->gmc.agp_end) >> 18));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_HI32,
+ upper_32_bits(max(adev->gmc.fb_end,
+ adev->gmc.agp_end) >> 18));
+
+ /* Set default page address. */
+ value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+ (u32)(value >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+ (u32)(value >> 44));
+
+ /* Program "protection fault". */
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+ (u32)(adev->dummy_page_addr >> 12));
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+ (u32)((u64)adev->dummy_page_addr >> 44));
+
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL2);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL2,
+ ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+ }
+
+ /* In the case squeezing vram into GART aperture, we don't use
+ * FB aperture and AGP aperture. Disable them.
+ */
+ if (adev->gmc.pdb0_bo) {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_TOP_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_TOP_HI32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_BASE_LO32, 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_FB_LOCATION_BASE_HI32, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_TOP_HI32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_LO32, 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_AGP_BOT_HI32, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_LO32,
+ 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_LOW_ADDR_HI32,
+ 0x7F);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_LO32, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_SYSTEM_APERTURE_HIGH_ADDR_HI32, 0);
+ }
+}
+
+static void mmhub_v4_2_0_mid_init_tlb_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint32_t tmp;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ /* Setup TLB control */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_MX_L1_TLB_CNTL);
+
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ ENABLE_ADVANCED_DRIVER_MODEL, 1);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ MTYPE, MTYPE_UC); /* UC, uncached */
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMMC_VM_MX_L1_TLB_CNTL, tmp);
+ }
+}
+
+static void mmhub_v4_2_0_mid_init_cache_regs(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint32_t tmp;
+ int i;
+
+ /* These registers are not accessible to VF-SRIOV.
+ * The PF will program them instead.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ /* Setup L2 cache */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, ENABLE_L2_CACHE, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ ENABLE_DEFAULT_PAGE_OUT_TO_SYSTEM_MEMORY, 1);
+ /* XXX for emulation, Refer to closed source code.*/
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ L2_PDE0_CACHE_TAG_GENERATION_MODE, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ PDE_FAULT_CLASSIFICATION, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ CONTEXT1_IDENTITY_ACCESS_MODE, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL,
+ IDENTITY_MODE_FRAGMENT_SIZE, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL, tmp);
+
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL2);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL2,
+ INVALIDATE_ALL_L1_TLBS, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL2,
+ INVALIDATE_L2_CACHE, 1);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL2, tmp);
+
+ tmp = regMMVM_L2_CNTL3_DEFAULT;
+ if (adev->gmc.translate_further) {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3, BANK_SELECT, 12);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3,
+ L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+ } else {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3, BANK_SELECT, 9);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL3,
+ L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+ }
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL3, tmp);
+
+ tmp = regMMVM_L2_CNTL4_DEFAULT;
+ /* For AMD APP APUs setup WC memory */
+ if (adev->gmc.xgmi.connected_to_cpu || adev->gmc.is_app_apu) {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
+ } else {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL4,
+ VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+ }
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL4, tmp);
+
+ tmp = regMMVM_L2_CNTL5_DEFAULT;
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL5,
+ L2_CACHE_SMALLK_FRAGMENT_SIZE, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_CNTL5, tmp);
+ }
+}
+
+static void mmhub_v4_2_0_mid_enable_system_domain(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ uint32_t tmp;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT0_CNTL,
+ ENABLE_CONTEXT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT0_CNTL,
+ PAGE_TABLE_DEPTH, 0);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT0_CNTL,
+ RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_CNTL, tmp);
+ }
+}
+
+static void mmhub_v4_2_0_mid_disable_identity_aperture(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ int i;
+
+ /* These registers are not accessible to VF-SRIOV.
+ * The PF will program them instead.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
+ 0xFFFFFFFF);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
+ 0x00001FFF);
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
+ 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
+ 0);
+
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32,
+ 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32,
+ 0);
+ }
+}
+
+static void mmhub_v4_2_0_mid_setup_vmid_config(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ uint32_t tmp;
+ int i, j;
+
+ for_each_inst(j, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+ for (i = 0; i <= 14; i++) {
+ tmp = RREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_CONTEXT1_CNTL,
+ i * hub->ctx_distance);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
+ adev->vm_manager.num_level);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+ 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ PAGE_TABLE_BLOCK_SIZE,
+ adev->vm_manager.block_size - 9);
+ /* Send no-retry XNACK on fault to suppress VM fault storm. */
+ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL,
+ RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
+ !amdgpu_noretry);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_CNTL,
+ i * hub->ctx_distance, tmp);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
+ i * hub->ctx_addr_distance, 0);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
+ i * hub->ctx_addr_distance, 0);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
+ i * hub->ctx_addr_distance,
+ lower_32_bits(adev->vm_manager.max_pfn - 1));
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j), regMMVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
+ i * hub->ctx_addr_distance,
+ upper_32_bits(adev->vm_manager.max_pfn - 1));
+ }
+ }
+
+ hub->vm_cntx_cntl = tmp;
+}
+
+static void mmhub_v4_2_0_mid_program_invalidation(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ unsigned int i, j;
+
+ for_each_inst(j, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+
+ for (i = 0; i < 18; ++i) {
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+ i * hub->eng_addr_distance, 0xffffffff);
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+ i * hub->eng_addr_distance, 0x3fff);
+ }
+ }
+}
+
+static int mmhub_v4_2_0_mid_gart_enable(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ /* GART Enable. */
+ mmhub_v4_2_0_mid_init_gart_aperture_regs(adev, mid_mask);
+ mmhub_v4_2_0_mid_init_system_aperture_regs(adev, mid_mask);
+ mmhub_v4_2_0_mid_init_tlb_regs(adev, mid_mask);
+ mmhub_v4_2_0_mid_init_cache_regs(adev, mid_mask);
+
+ mmhub_v4_2_0_mid_enable_system_domain(adev, mid_mask);
+ mmhub_v4_2_0_mid_disable_identity_aperture(adev, mid_mask);
+ mmhub_v4_2_0_mid_setup_vmid_config(adev, mid_mask);
+ mmhub_v4_2_0_mid_program_invalidation(adev, mid_mask);
+
+ return 0;
+}
+static int mmhub_v4_2_0_gart_enable(struct amdgpu_device *adev)
+{
+ uint32_t mid_mask;
+
+ mid_mask = adev->aid_mask;
+ return mmhub_v4_2_0_mid_gart_enable(adev, mid_mask);
+}
+
+static void mmhub_v4_2_0_mid_gart_disable(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ u32 tmp;
+ u32 i, j;
+
+ for_each_inst(j, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+ /* Disable all tables */
+ for (i = 0; i < 16; i++)
+ WREG32_SOC15_OFFSET(MMHUB, GET_INST(MMHUB, j),
+ regMMVM_CONTEXT0_CNTL,
+ i * hub->ctx_distance, 0);
+
+ /* Setup TLB control */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, j),
+ regMMMC_VM_MX_L1_TLB_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ ENABLE_L1_TLB, 0);
+ tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL,
+ ENABLE_ADVANCED_DRIVER_MODEL, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, j),
+ regMMMC_VM_MX_L1_TLB_CNTL, tmp);
+
+ /* Setup L2 cache */
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, j), regMMVM_L2_CNTL);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_CNTL, ENABLE_L2_CACHE, 0);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, j), regMMVM_L2_CNTL, tmp);
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, j), regMMVM_L2_CNTL3, 0);
+ }
+}
+
+static void mmhub_v4_2_0_gart_disable(struct amdgpu_device *adev)
+{
+ uint32_t mid_mask;
+
+ mid_mask = adev->aid_mask;
+ mmhub_v4_2_0_mid_gart_disable(adev, mid_mask);
+}
+
+static void
+mmhub_v4_2_0_mid_set_fault_enable_default(struct amdgpu_device *adev,
+ bool value, uint32_t mid_mask)
+{
+ u32 tmp;
+ int i;
+
+ /* These registers are not accessible to VF-SRIOV.
+ * The PF will program them instead.
+ */
+ if (amdgpu_sriov_vf(adev))
+ return;
+
+ for_each_inst(i, mid_mask) {
+ tmp = RREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL_LO32);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
+ value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+ if (!value) {
+ tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL_LO32,
+ CRASH_ON_NO_RETRY_FAULT, 1);
+ }
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL_LO32, tmp);
+ }
+}
+
+
+/**
+ * mmhub_v4_2_0_set_fault_enable_default - update GART/VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void
+mmhub_v4_2_0_set_fault_enable_default(struct amdgpu_device *adev,
+ bool value)
+{
+ uint32_t mid_mask;
+
+ mid_mask = adev->aid_mask;
+ mmhub_v4_2_0_mid_set_fault_enable_default(adev, value, mid_mask);
+}
+
+static uint32_t mmhub_v4_2_0_get_invalidate_req(unsigned int vmid,
+ uint32_t flush_type)
+{
+ u32 req = 0;
+
+ /* invalidate using legacy mode on vmid*/
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ,
+ PER_VMID_INVALIDATE_REQ, 1 << vmid);
+ /* Only use legacy inv on mmhub side */
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, 0);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
+ req = REG_SET_FIELD(req, MMVM_INVALIDATE_ENG0_REQ,
+ CLEAR_PROTECTION_FAULT_STATUS_ADDR, 0);
+
+ return req;
+}
+
+/*TODO: l2 protection fault status is increased to 64bits.
+ * some critical fields like FED are moved to STATUS_HI32 */
+static void
+mmhub_v4_2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,
+ uint32_t status)
+{
+ uint32_t cid, rw;
+ const char *mmhub_cid = NULL;
+
+ cid = REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, CID);
+ rw = REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, RW);
+
+ dev_err(adev->dev,
+ "MMVM_L2_PROTECTION_FAULT_STATUS_LO32:0x%08X\n",
+ status);
+ switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) {
+ case IP_VERSION(4, 2, 0):
+ mmhub_cid = mmhub_client_ids_v4_2_0[cid][rw];
+ break;
+ default:
+ mmhub_cid = NULL;
+ break;
+ }
+ dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+ mmhub_cid ? mmhub_cid : "unknown", cid);
+ dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, MORE_FAULTS));
+ dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, WALKER_ERROR));
+ dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, PERMISSION_FAULTS));
+ dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
+ REG_GET_FIELD(status,
+ MMVM_L2_PROTECTION_FAULT_STATUS_LO32, MAPPING_ERROR));
+ dev_err(adev->dev, "\t RW: 0x%x\n", rw);
+}
+
+
+static const struct amdgpu_vmhub_funcs mmhub_v4_2_0_vmhub_funcs = {
+ .print_l2_protection_fault_status = mmhub_v4_2_0_print_l2_protection_fault_status,
+ .get_invalidate_req = mmhub_v4_2_0_get_invalidate_req,
+};
+
+static void mmhub_v4_2_0_mid_init(struct amdgpu_device *adev,
+ uint32_t mid_mask)
+{
+ struct amdgpu_vmhub *hub;
+ int i;
+
+ for_each_inst(i, mid_mask) {
+ hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
+
+ hub->ctx0_ptb_addr_lo32 =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
+ hub->ctx0_ptb_addr_hi32 =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
+ hub->vm_inv_eng0_sem =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_INVALIDATE_ENG0_SEM);
+ hub->vm_inv_eng0_req =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_INVALIDATE_ENG0_REQ);
+ hub->vm_inv_eng0_ack =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_INVALIDATE_ENG0_ACK);
+ hub->vm_context0_cntl =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_CONTEXT0_CNTL);
+ /* TODO: add a new member to accomandate additional fault status/cntl reg */
+ hub->vm_l2_pro_fault_status =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_STATUS_LO32);
+ hub->vm_l2_pro_fault_cntl =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i),
+ regMMVM_L2_PROTECTION_FAULT_CNTL_LO32);
+
+ hub->ctx_distance = regMMVM_CONTEXT1_CNTL - regMMVM_CONTEXT0_CNTL;
+ hub->ctx_addr_distance = regMMVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
+ regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+ hub->eng_distance = regMMVM_INVALIDATE_ENG1_REQ -
+ regMMVM_INVALIDATE_ENG0_REQ;
+ hub->eng_addr_distance = regMMVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
+ regMMVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+
+ hub->vm_cntx_cntl_vm_fault = MMVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
+ MMVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
+
+ hub->vm_l2_bank_select_reserved_cid2 =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i), regMMVM_L2_BANK_SELECT_RESERVED_CID2);
+
+ hub->vm_contexts_disable =
+ SOC15_REG_OFFSET(MMHUB, GET_INST(MMHUB, i), regMMVM_CONTEXTS_DISABLE);
+
+ hub->vmhub_funcs = &mmhub_v4_2_0_vmhub_funcs;
+ }
+}
+
+static void mmhub_v4_2_0_init(struct amdgpu_device *adev)
+{
+ uint32_t mid_mask;
+
+ mid_mask = adev->aid_mask;
+ mmhub_v4_2_0_mid_init(adev, mid_mask);
+}
+
+static void
+mmhub_v4_2_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+ bool enable)
+{
+ uint32_t def, data;
+ uint32_t def1, data1, def2 = 0, data2 = 0;
+ def = data = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG);
+ def1 = data1 = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB0_CNTL_MISC2);
+ def2 = data2 = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB1_CNTL_MISC2);
+
+ if (enable) {
+ data |= MM_ATC_L2_MISC_CG__ENABLE_MASK;
+ data1 &= ~(DAGB0_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB0_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+
+ data2 &= ~(DAGB1_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB1_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+ } else {
+ data &= ~MM_ATC_L2_MISC_CG__ENABLE_MASK;
+ data1 |= (DAGB0_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB0_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+
+ data2 |= (DAGB1_CNTL_MISC2__DISABLE_RDRET_TAP_CHAIN_FGCG_MASK |
+ DAGB1_CNTL_MISC2__DISABLE_WRRET_TAP_CHAIN_FGCG_MASK);
+ }
+
+ if (def != data)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG, data);
+ if (def1 != data1)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB0_CNTL_MISC2, data1);
+
+ if (def2 != data2)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regDAGB1_CNTL_MISC2, data2);
+}
+
+static void
+mmhub_v4_2_0_update_medium_grain_light_sleep(struct amdgpu_device *adev,
+ bool enable)
+{
+ uint32_t def, data;
+
+ def = data = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG);
+
+ if (enable)
+ data |= MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK;
+ else
+ data &= ~MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK;
+
+ if (def != data)
+ WREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG, data);
+}
+
+static int mmhub_v4_2_0_set_clockgating(struct amdgpu_device *adev,
+ enum amd_clockgating_state state)
+{
+ if (amdgpu_sriov_vf(adev))
+ return 0;
+
+ if (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)
+ mmhub_v4_2_0_update_medium_grain_clock_gating(adev,
+ state == AMD_CG_STATE_GATE);
+
+ if (adev->cg_flags & AMD_CG_SUPPORT_MC_LS)
+ mmhub_v4_2_0_update_medium_grain_light_sleep(adev,
+ state == AMD_CG_STATE_GATE);
+
+ return 0;
+}
+
+static void mmhub_v4_2_0_get_clockgating(struct amdgpu_device *adev, u64 *flags)
+{
+ int data;
+
+ if (amdgpu_sriov_vf(adev))
+ *flags = 0;
+
+ data = RREG32_SOC15(MMHUB, GET_INST(MMHUB, 0), regMM_ATC_L2_MISC_CG);
+
+ /* AMD_CG_SUPPORT_MC_MGCG */
+ if (data & MM_ATC_L2_MISC_CG__ENABLE_MASK)
+ *flags |= AMD_CG_SUPPORT_MC_MGCG;
+
+ /* AMD_CG_SUPPORT_MC_LS */
+ if (data & MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK)
+ *flags |= AMD_CG_SUPPORT_MC_LS;
+}
+
+const struct amdgpu_mmhub_funcs mmhub_v4_2_0_funcs = {
+ .init = mmhub_v4_2_0_init,
+ .get_fb_location = mmhub_v4_2_0_get_fb_location,
+ .get_mc_fb_offset = mmhub_v4_2_0_get_mc_fb_offset,
+ .setup_vm_pt_regs = mmhub_v4_2_0_setup_vm_pt_regs,
+ .gart_enable = mmhub_v4_2_0_gart_enable,
+ .gart_disable = mmhub_v4_2_0_gart_disable,
+ .set_fault_enable_default = mmhub_v4_2_0_set_fault_enable_default,
+ .set_clockgating = mmhub_v4_2_0_set_clockgating,
+ .get_clockgating = mmhub_v4_2_0_get_clockgating,
+};
+
+static int mmhub_v4_2_0_xcp_resume(void *handle, uint32_t inst_mask)
+{
+ struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+ bool value;
+
+ if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
+ value = false;
+ else
+ value = true;
+
+ mmhub_v4_2_0_mid_set_fault_enable_default(adev, value, inst_mask);
+
+ if (!amdgpu_sriov_vf(adev))
+ return mmhub_v4_2_0_mid_gart_enable(adev, inst_mask);
+
+ return 0;
+}
+
+static int mmhub_v4_2_0_xcp_suspend(void *handle, uint32_t inst_mask)
+{
+ struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+ if (!amdgpu_sriov_vf(adev))
+ mmhub_v4_2_0_mid_gart_disable(adev, inst_mask);
+
+ return 0;
+}
+
+struct amdgpu_xcp_ip_funcs mmhub_v4_2_0_xcp_funcs = {
+ .suspend = &mmhub_v4_2_0_xcp_suspend,
+ .resume = &mmhub_v4_2_0_xcp_resume
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
new file mode 100644
index 0000000000000..4ea6de7ac7ae7
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __MMHUB_V4_2_0_H__
+#define __MMHUB_V4_2_0_H__
+
+extern const struct amdgpu_mmhub_funcs mmhub_v4_2_0_funcs;
+
+#endif
--
2.51.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
end of thread, other threads:[~2025-11-19 14:51 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-11-19 14:50 [PATCH 0/2] Add MMHUB 4.2 Alex Deucher
2025-11-19 14:50 ` [PATCH 2/2] drm/amdgpu: Add initial support for mmhub v4_2 Alex Deucher
-- strict thread matches above, loose matches on Subject: below --
2025-10-30 20:50 [PATCH 0/2] Add mmhub 4.2.0 support Alex Deucher
2025-10-30 20:50 ` [PATCH 2/2] drm/amdgpu: Add initial support for mmhub v4_2 Alex Deucher
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.