/* * 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 #include #include #include #include #include "amdgpu.h" #include "amdgpu_gfx.h" #include "amdgpu_psp.h" #include "amdgpu_smu.h" #include "amdgpu_atomfirmware.h" #include "amdgpu_userq_fence.h" #include "imu_v12_1.h" #include "soc_v1_0.h" #include "gfx_v12_1_pkt.h" #include "gc/gc_12_1_0_offset.h" #include "gc/gc_12_1_0_sh_mask.h" #include "soc24_enum.h" #include "ivsrcid/gfx/irqsrcs_gfx_12_1_0.h" #include "soc15.h" #include "clearstate_gfx12.h" #include "v12_structs.h" #include "gfx_v12_1.h" #include "mes_v12_1.h" #define GFX12_MEC_HPD_SIZE 2048 #define NUM_SIMD_PER_CU_GFX12_1 4 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L #define regCP_HQD_EOP_CONTROL_DEFAULT 0x00000000 #define regCP_HQD_PQ_DOORBELL_CONTROL_DEFAULT 0x00000000 #define regCP_MQD_CONTROL_DEFAULT 0x00000100 #define regCP_HQD_PQ_CONTROL_DEFAULT 0x00308509 #define regCP_HQD_PQ_RPTR_DEFAULT 0x00000000 #define regCP_HQD_PERSISTENT_STATE_DEFAULT 0x0ae06301 #define regCP_HQD_IB_CONTROL_DEFAULT 0x00100000 MODULE_FIRMWARE("amdgpu/gc_12_1_0_mec.bin"); MODULE_FIRMWARE("amdgpu/gc_12_1_0_rlc.bin"); #define SH_MEM_ALIGNMENT_MODE_UNALIGNED_GFX12_1_0 0x00000001 #define DEFAULT_SH_MEM_CONFIG \ ((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \ (SH_MEM_ALIGNMENT_MODE_UNALIGNED_GFX12_1_0 << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ (3 << SH_MEM_CONFIG__INITIAL_INST_PREFETCH__SHIFT)) static void gfx_v12_1_xcc_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id); static void gfx_v12_1_set_ring_funcs(struct amdgpu_device *adev); static void gfx_v12_1_set_irq_funcs(struct amdgpu_device *adev); static void gfx_v12_1_set_rlc_funcs(struct amdgpu_device *adev); static void gfx_v12_1_set_mqd_funcs(struct amdgpu_device *adev); static void gfx_v12_1_set_imu_funcs(struct amdgpu_device *adev); static int gfx_v12_1_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info); static uint64_t gfx_v12_1_get_gpu_clock_counter(struct amdgpu_device *adev); static void gfx_v12_1_xcc_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num, u32 instance, int xcc_id); static void gfx_v12_1_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val); static int gfx_v12_1_wait_for_rlc_autoload_complete(struct amdgpu_device *adev); static void gfx_v12_1_ring_invalidate_tlbs(struct amdgpu_ring *ring, uint16_t pasid, uint32_t flush_type, bool all_hub, uint8_t dst_sel); static void gfx_v12_1_xcc_set_safe_mode(struct amdgpu_device *adev, int xcc_id); static void gfx_v12_1_xcc_unset_safe_mode(struct amdgpu_device *adev, int xcc_id); static void gfx_v12_1_update_perf_clk(struct amdgpu_device *adev, bool enable); static void gfx_v12_1_xcc_update_perf_clk(struct amdgpu_device *adev, bool enable, int xcc_id); static int gfx_v12_1_init_cp_compute_microcode_bo(struct amdgpu_device *adev); static void gfx_v12_1_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) { amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_SET_RESOURCES, 6)); amdgpu_ring_write(kiq_ring, PACKET3_SET_RESOURCES_VMID_MASK(0) | PACKET3_SET_RESOURCES_QUEUE_TYPE(0)); /* vmid_mask:0 queue_type:0 (KIQ) */ amdgpu_ring_write(kiq_ring, lower_32_bits(queue_mask)); /* queue mask lo */ amdgpu_ring_write(kiq_ring, upper_32_bits(queue_mask)); /* queue mask hi */ amdgpu_ring_write(kiq_ring, 0); /* gws mask lo */ amdgpu_ring_write(kiq_ring, 0); /* gws mask hi */ amdgpu_ring_write(kiq_ring, 0); /* oac mask */ amdgpu_ring_write(kiq_ring, 0); } static void gfx_v12_1_kiq_map_queues(struct amdgpu_ring *kiq_ring, struct amdgpu_ring *ring) { uint64_t mqd_addr = amdgpu_bo_gpu_offset(ring->mqd_obj); uint64_t wptr_addr = ring->wptr_gpu_addr; uint32_t me = 0, eng_sel = 0; switch (ring->funcs->type) { case AMDGPU_RING_TYPE_COMPUTE: me = 1; eng_sel = 0; break; case AMDGPU_RING_TYPE_MES: me = 2; eng_sel = 5; break; default: WARN_ON(1); } amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_MAP_QUEUES, 5)); /* Q_sel:0, vmid:0, vidmem: 1, engine:0, num_Q:1*/ amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */ PACKET3_MAP_QUEUES_QUEUE_SEL(0) | /* Queue_Sel */ PACKET3_MAP_QUEUES_VMID(0) | /* VMID */ PACKET3_MAP_QUEUES_QUEUE(ring->queue) | PACKET3_MAP_QUEUES_PIPE(ring->pipe) | PACKET3_MAP_QUEUES_ME((me)) | PACKET3_MAP_QUEUES_QUEUE_TYPE(0) | /*queue_type: normal compute queue */ PACKET3_MAP_QUEUES_ALLOC_FORMAT(0) | /* alloc format: all_on_one_pipe */ PACKET3_MAP_QUEUES_ENGINE_SEL(eng_sel) | PACKET3_MAP_QUEUES_NUM_QUEUES(1)); /* num_queues: must be 1 */ amdgpu_ring_write(kiq_ring, PACKET3_MAP_QUEUES_DOORBELL_OFFSET(ring->doorbell_index)); amdgpu_ring_write(kiq_ring, lower_32_bits(mqd_addr)); amdgpu_ring_write(kiq_ring, upper_32_bits(mqd_addr)); amdgpu_ring_write(kiq_ring, lower_32_bits(wptr_addr)); amdgpu_ring_write(kiq_ring, upper_32_bits(wptr_addr)); } static void gfx_v12_1_kiq_unmap_queues(struct amdgpu_ring *kiq_ring, struct amdgpu_ring *ring, enum amdgpu_unmap_queues_action action, u64 gpu_addr, u64 seq) { struct amdgpu_device *adev = kiq_ring->adev; uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; if (adev->enable_mes && !adev->gfx.kiq[0].ring.sched.ready) { amdgpu_mes_unmap_legacy_queue(adev, ring, action, gpu_addr, seq, kiq_ring->xcc_id); return; } amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_UNMAP_QUEUES, 4)); amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */ PACKET3_UNMAP_QUEUES_ACTION(action) | PACKET3_UNMAP_QUEUES_QUEUE_SEL(0) | PACKET3_UNMAP_QUEUES_ENGINE_SEL(eng_sel) | PACKET3_UNMAP_QUEUES_NUM_QUEUES(1)); amdgpu_ring_write(kiq_ring, PACKET3_UNMAP_QUEUES_DOORBELL_OFFSET0(ring->doorbell_index)); if (action == PREEMPT_QUEUES_NO_UNMAP) { amdgpu_ring_write(kiq_ring, lower_32_bits(gpu_addr)); amdgpu_ring_write(kiq_ring, upper_32_bits(gpu_addr)); amdgpu_ring_write(kiq_ring, seq); } else { amdgpu_ring_write(kiq_ring, 0); amdgpu_ring_write(kiq_ring, 0); amdgpu_ring_write(kiq_ring, 0); } } static void gfx_v12_1_kiq_query_status(struct amdgpu_ring *kiq_ring, struct amdgpu_ring *ring, u64 addr, u64 seq) { uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_QUERY_STATUS, 5)); amdgpu_ring_write(kiq_ring, PACKET3_QUERY_STATUS_CONTEXT_ID(0) | PACKET3_QUERY_STATUS_INTERRUPT_SEL(0) | PACKET3_QUERY_STATUS_COMMAND(2)); amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */ PACKET3_QUERY_STATUS_DOORBELL_OFFSET(ring->doorbell_index) | PACKET3_QUERY_STATUS_ENG_SEL(eng_sel)); amdgpu_ring_write(kiq_ring, lower_32_bits(addr)); amdgpu_ring_write(kiq_ring, upper_32_bits(addr)); amdgpu_ring_write(kiq_ring, lower_32_bits(seq)); amdgpu_ring_write(kiq_ring, upper_32_bits(seq)); } static void gfx_v12_1_kiq_invalidate_tlbs(struct amdgpu_ring *kiq_ring, uint16_t pasid, uint32_t flush_type, bool all_hub) { gfx_v12_1_ring_invalidate_tlbs(kiq_ring, pasid, flush_type, all_hub, 1); } static const struct kiq_pm4_funcs gfx_v12_1_kiq_pm4_funcs = { .kiq_set_resources = gfx_v12_1_kiq_set_resources, .kiq_map_queues = gfx_v12_1_kiq_map_queues, .kiq_unmap_queues = gfx_v12_1_kiq_unmap_queues, .kiq_query_status = gfx_v12_1_kiq_query_status, .kiq_invalidate_tlbs = gfx_v12_1_kiq_invalidate_tlbs, .set_resources_size = 8, .map_queues_size = 7, .unmap_queues_size = 6, .query_status_size = 7, .invalidate_tlbs_size = 2, }; static void gfx_v12_1_set_kiq_pm4_funcs(struct amdgpu_device *adev) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i =0; i < num_xcc; i++) adev->gfx.kiq[i].pmf = &gfx_v12_1_kiq_pm4_funcs; } static void gfx_v12_1_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, int mem_space, int opt, uint32_t addr0, uint32_t addr1, uint32_t ref, uint32_t mask, uint32_t inv) { if (mem_space == 0) { addr0 = soc_v1_0_normalize_xcc_reg_offset(addr0); addr1 = soc_v1_0_normalize_xcc_reg_offset(addr1); } amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); amdgpu_ring_write(ring, /* memory (1) or register (0) */ (WAIT_REG_MEM_MEM_SPACE(mem_space) | WAIT_REG_MEM_OPERATION(opt) | /* wait */ WAIT_REG_MEM_FUNCTION(3) | /* equal */ WAIT_REG_MEM_ENGINE(eng_sel))); if (mem_space) BUG_ON(addr0 & 0x3); /* Dword align */ amdgpu_ring_write(ring, addr0); amdgpu_ring_write(ring, addr1); amdgpu_ring_write(ring, ref); amdgpu_ring_write(ring, mask); amdgpu_ring_write(ring, inv); /* poll interval */ } static int gfx_v12_1_ring_test_ring(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; uint32_t scratch_reg0_offset, xcc_offset; uint32_t tmp = 0; unsigned i; int r; /* Use register offset which is local to XCC in the packet */ xcc_offset = SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG0); scratch_reg0_offset = SOC15_REG_OFFSET(GC, GET_INST(GC, ring->xcc_id), regSCRATCH_REG0); WREG32(scratch_reg0_offset, 0xCAFEDEAD); tmp = RREG32(scratch_reg0_offset); r = amdgpu_ring_alloc(ring, 5); if (r) { dev_err(adev->dev, "amdgpu: cp failed to lock ring %d (%d).\n", ring->idx, r); return r; } if (ring->funcs->type == AMDGPU_RING_TYPE_KIQ) { gfx_v12_1_ring_emit_wreg(ring, xcc_offset, 0xDEADBEEF); } else { amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1)); amdgpu_ring_write(ring, xcc_offset - PACKET3_SET_UCONFIG_REG_START); amdgpu_ring_write(ring, 0xDEADBEEF); } amdgpu_ring_commit(ring); for (i = 0; i < adev->usec_timeout; i++) { tmp = RREG32(scratch_reg0_offset); if (tmp == 0xDEADBEEF) break; if (amdgpu_emu_mode == 1) msleep(1); else udelay(1); } if (i >= adev->usec_timeout) r = -ETIMEDOUT; return r; } static int gfx_v12_1_ring_test_ib(struct amdgpu_ring *ring, long timeout) { struct amdgpu_device *adev = ring->adev; struct amdgpu_ib ib; struct dma_fence *f = NULL; unsigned index; uint64_t gpu_addr; volatile uint32_t *cpu_ptr; long r; /* MES KIQ fw hasn't indirect buffer support for now */ if (adev->enable_mes_kiq && ring->funcs->type == AMDGPU_RING_TYPE_KIQ) return 0; memset(&ib, 0, sizeof(ib)); r = amdgpu_device_wb_get(adev, &index); if (r) return r; gpu_addr = adev->wb.gpu_addr + (index * 4); adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); cpu_ptr = &adev->wb.wb[index]; r = amdgpu_ib_get(adev, NULL, 16, AMDGPU_IB_POOL_DIRECT, &ib); if (r) { dev_err(adev->dev, "amdgpu: failed to get ib (%ld).\n", r); goto err1; } ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3); ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM; ib.ptr[2] = lower_32_bits(gpu_addr); ib.ptr[3] = upper_32_bits(gpu_addr); ib.ptr[4] = 0xDEADBEEF; ib.length_dw = 5; r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f); if (r) goto err2; r = dma_fence_wait_timeout(f, false, timeout); if (r == 0) { r = -ETIMEDOUT; goto err2; } else if (r < 0) { goto err2; } if (le32_to_cpu(*cpu_ptr) == 0xDEADBEEF) r = 0; else r = -EINVAL; err2: amdgpu_ib_free(&ib, NULL); dma_fence_put(f); err1: amdgpu_device_wb_free(adev, index); return r; } static void gfx_v12_1_free_microcode(struct amdgpu_device *adev) { amdgpu_ucode_release(&adev->gfx.rlc_fw); amdgpu_ucode_release(&adev->gfx.mec_fw); kfree(adev->gfx.rlc.register_list_format); } static int gfx_v12_1_init_toc_microcode(struct amdgpu_device *adev, const char *ucode_prefix) { const struct psp_firmware_header_v1_0 *toc_hdr; int err = 0; err = amdgpu_ucode_request(adev, &adev->psp.toc_fw, AMDGPU_UCODE_REQUIRED, "amdgpu/%s_toc.bin", ucode_prefix); if (err) goto out; toc_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.toc_fw->data; adev->psp.toc.fw_version = le32_to_cpu(toc_hdr->header.ucode_version); adev->psp.toc.feature_version = le32_to_cpu(toc_hdr->sos.fw_version); adev->psp.toc.size_bytes = le32_to_cpu(toc_hdr->header.ucode_size_bytes); adev->psp.toc.start_addr = (uint8_t *)toc_hdr + le32_to_cpu(toc_hdr->header.ucode_array_offset_bytes); return 0; out: amdgpu_ucode_release(&adev->psp.toc_fw); return err; } static int gfx_v12_1_init_microcode(struct amdgpu_device *adev) { char ucode_prefix[15]; int err; const struct rlc_firmware_header_v2_0 *rlc_hdr; uint16_t version_major; uint16_t version_minor; DRM_DEBUG("\n"); amdgpu_ucode_ip_version_decode(adev, GC_HWIP, ucode_prefix, sizeof(ucode_prefix)); if (!amdgpu_sriov_vf(adev)) { err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, AMDGPU_UCODE_REQUIRED, "amdgpu/%s_rlc.bin", ucode_prefix); if (err) goto out; rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; version_major = le16_to_cpu(rlc_hdr->header.header_version_major); version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor); err = amdgpu_gfx_rlc_init_microcode(adev, version_major, version_minor); if (err) goto out; } err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, AMDGPU_UCODE_REQUIRED, "amdgpu/%s_mec.bin", ucode_prefix); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_RS64_MEC); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_RS64_MEC_P0_STACK); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_RS64_MEC_P1_STACK); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_RS64_MEC_P2_STACK); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_RS64_MEC_P3_STACK); if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) err = gfx_v12_1_init_toc_microcode(adev, ucode_prefix); /* only one MEC for gfx 12 */ adev->gfx.mec2_fw = NULL; if (adev->gfx.imu.funcs) { if (adev->gfx.imu.funcs->init_microcode) { err = adev->gfx.imu.funcs->init_microcode(adev); if (err) dev_err(adev->dev, "Failed to load imu firmware!\n"); } } out: if (err) { amdgpu_ucode_release(&adev->gfx.rlc_fw); amdgpu_ucode_release(&adev->gfx.mec_fw); } return err; } static u32 gfx_v12_1_get_csb_size(struct amdgpu_device *adev) { u32 count = 0; const struct cs_section_def *sect = NULL; const struct cs_extent_def *ext = NULL; count += 1; for (sect = gfx12_cs_data; sect->section != NULL; ++sect) { if (sect->id == SECT_CONTEXT) { for (ext = sect->section; ext->extent != NULL; ++ext) count += 2 + ext->reg_count; } else return 0; } return count; } static void gfx_v12_1_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer) { u32 count = 0, clustercount = 0, i; const struct cs_section_def *sect = NULL; const struct cs_extent_def *ext = NULL; if (adev->gfx.rlc.cs_data == NULL) return; if (buffer == NULL) return; count += 1; for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) { if (sect->id == SECT_CONTEXT) { for (ext = sect->section; ext->extent != NULL; ++ext) { clustercount++; buffer[count++] = ext->reg_count; buffer[count++] = ext->reg_index; for (i = 0; i < ext->reg_count; i++) buffer[count++] = cpu_to_le32(ext->extent[i]); } } else return; } buffer[0] = clustercount; } static void gfx_v12_1_rlc_fini(struct amdgpu_device *adev) { /* clear state block */ amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, &adev->gfx.rlc.clear_state_gpu_addr, (void **)&adev->gfx.rlc.cs_ptr); /* jump table block */ amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, &adev->gfx.rlc.cp_table_gpu_addr, (void **)&adev->gfx.rlc.cp_table_ptr); } static void gfx_v12_1_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) { int xcc_id, num_xcc; struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) { reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[GET_INST(GC, xcc_id)]; reg_access_ctrl->grbm_cntl = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regGRBM_GFX_CNTL); reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX); reg_access_ctrl->vfi_cmd = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_VFI_CMD); reg_access_ctrl->vfi_stat = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_VFI_STAT); reg_access_ctrl->vfi_addr = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_VFI_ADDR); reg_access_ctrl->vfi_data = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_VFI_DATA); reg_access_ctrl->vfi_grbm_cntl = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_VFI_GRBM_GFX_CNTL); reg_access_ctrl->vfi_grbm_idx = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_VFI_GRBM_GFX_INDEX); reg_access_ctrl->vfi_grbm_cntl_data = 0; reg_access_ctrl->vfi_grbm_idx_data = 0; } adev->gfx.rlc.rlcg_reg_access_supported = true; } static int gfx_v12_1_rlc_init(struct amdgpu_device *adev) { const struct cs_section_def *cs_data; int r, i, num_xcc; adev->gfx.rlc.cs_data = gfx12_cs_data; cs_data = adev->gfx.rlc.cs_data; if (cs_data) { /* init clear state block */ r = amdgpu_gfx_rlc_init_csb(adev); if (r) return r; } /* init spm vmid with 0xf */ num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { if (adev->gfx.rlc.funcs->update_spm_vmid) adev->gfx.rlc.funcs->update_spm_vmid(adev, i, NULL, 0xf); } return 0; } static void gfx_v12_1_mec_fini(struct amdgpu_device *adev) { amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL); amdgpu_bo_free_kernel(&adev->gfx.mec.mec_fw_obj, NULL, NULL); amdgpu_bo_free_kernel(&adev->gfx.mec.mec_fw_data_obj, NULL, NULL); } static int gfx_v12_1_mec_init(struct amdgpu_device *adev) { int r, i, num_xcc; u32 *hpd; size_t mec_hpd_size; bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) bitmap_zero(adev->gfx.mec_bitmap[i].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); /* take ownership of the relevant compute queues */ amdgpu_gfx_compute_queue_acquire(adev); mec_hpd_size = adev->gfx.num_compute_rings * GFX12_MEC_HPD_SIZE * num_xcc; if (mec_hpd_size) { r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, &adev->gfx.mec.hpd_eop_obj, &adev->gfx.mec.hpd_eop_gpu_addr, (void **)&hpd); if (r) { dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r); gfx_v12_1_mec_fini(adev); return r; } memset(hpd, 0, mec_hpd_size); amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj); amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj); } return 0; } static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t wave, uint32_t address) { WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (address << SQ_IND_INDEX__INDEX__SHIFT)); return RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_DATA); } static void wave_read_regs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t wave, uint32_t thread, uint32_t regno, uint32_t num, uint32_t *out) { WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (regno << SQ_IND_INDEX__INDEX__SHIFT) | (thread << SQ_IND_INDEX__WORKITEM_ID__SHIFT) | (SQ_IND_INDEX__AUTO_INCR_MASK)); while (num--) *(out++) = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_DATA); } static void gfx_v12_1_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* in gfx12 the SIMD_ID is specified as part of the INSTANCE * field when performing a select_se_sh so it should be * zero here */ WARN_ON(simd != 0); /* type 4 wave data */ dst[(*no_fields)++] = 4; dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_STATUS); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_PC_LO); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_PC_HI); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_EXEC_LO); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_EXEC_HI); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_HW_ID1); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_HW_ID2); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_GPR_ALLOC); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_LDS_ALLOC); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_IB_STS); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_IB_STS2); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_IB_DBG1); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_M0); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_MODE); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_STATE_PRIV); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_EXCP_FLAG_PRIV); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_EXCP_FLAG_USER); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_TRAP_CTRL); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_ACTIVE); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_VALID_AND_IDLE); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_DVGPR_ALLOC_LO); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_DVGPR_ALLOC_HI); dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, wave, ixSQ_WAVE_SCHED_MODE); } static void gfx_v12_1_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { WARN_ON(simd != 0); wave_read_regs(adev, xcc_id, wave, 0, start + SQIND_WAVE_SGPRS_OFFSET, size, dst); } static void gfx_v12_1_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst) { wave_read_regs(adev, xcc_id, wave, thread, start + SQIND_WAVE_VGPRS_OFFSET, size, dst); } static void gfx_v12_1_select_me_pipe_q(struct amdgpu_device *adev, u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { soc_v1_0_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, xcc_id)); } static int gfx_v12_1_get_xccs_per_xcp(struct amdgpu_device *adev) { /* Fill this in when the interface is ready */ return 1; } static int gfx_v12_1_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node) { int logic_xcc; int xcc = (ih_node & 0x7) - 2 + (ih_node >> 3) * 4; for (logic_xcc = 0; logic_xcc < NUM_XCC(adev->gfx.xcc_mask); logic_xcc++) { if (xcc == GET_INST(GC, logic_xcc)) return logic_xcc; } dev_err(adev->dev, "Couldn't find xcc mapping from IH node"); return -EINVAL; } static const struct amdgpu_gfx_funcs gfx_v12_1_gfx_funcs = { .get_gpu_clock_counter = &gfx_v12_1_get_gpu_clock_counter, .select_se_sh = &gfx_v12_1_xcc_select_se_sh, .read_wave_data = &gfx_v12_1_read_wave_data, .read_wave_sgprs = &gfx_v12_1_read_wave_sgprs, .read_wave_vgprs = &gfx_v12_1_read_wave_vgprs, .select_me_pipe_q = &gfx_v12_1_select_me_pipe_q, .update_perfmon_mgcg = &gfx_v12_1_update_perf_clk, .get_xccs_per_xcp = &gfx_v12_1_get_xccs_per_xcp, .ih_node_to_logical_xcc = &gfx_v12_1_ih_to_xcc_inst, }; static int gfx_v12_1_gpu_early_init(struct amdgpu_device *adev) { switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(12, 1, 0): adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; adev->gfx.config.sc_prim_fifo_size_backend = 0x100; adev->gfx.config.sc_hiz_tile_fifo_size = 0; adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0; break; default: BUG(); break; } return 0; } static int gfx_v12_1_compute_ring_init(struct amdgpu_device *adev, int ring_id, int xcc_id, int mec, int pipe, int queue) { int r; unsigned irq_type; struct amdgpu_ring *ring; unsigned int hw_prio; uint32_t xcc_doorbell_start; ring = &adev->gfx.compute_ring[xcc_id * adev->gfx.num_compute_rings + ring_id]; /* mec0 is me1 */ ring->xcc_id = xcc_id; ring->me = mec + 1; ring->pipe = pipe; ring->queue = queue; ring->ring_obj = NULL; ring->use_doorbell = true; xcc_doorbell_start = adev->doorbell_index.mec_ring0 + xcc_id * adev->doorbell_index.xcc_doorbell_range; ring->doorbell_index = (xcc_doorbell_start + ring_id) << 1; ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + (ring_id + xcc_id * adev->gfx.num_compute_rings) * GFX12_MEC_HPD_SIZE; ring->vm_hub = AMDGPU_GFXHUB(xcc_id); sprintf(ring->name, "comp_%d.%d.%d.%d", ring->xcc_id, ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP + ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec) + ring->pipe; hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring) ? AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL; /* type-2 packets are deprecated on MEC, use type-3 instead */ r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, irq_type, hw_prio, NULL); if (r) return r; return 0; } static struct { SOC24_FIRMWARE_ID id; unsigned int offset; unsigned int size; unsigned int size_x16; unsigned int num_inst; } rlc_autoload_info[SOC24_FIRMWARE_ID_MAX]; #define RLC_TOC_OFFSET_DWUNIT 8 #define RLC_SIZE_MULTIPLE 1024 #define RLC_TOC_UMF_SIZE_inM 23ULL #define RLC_TOC_FORMAT_API 165ULL #define RLC_NUM_INS_CODE0 1 #define RLC_NUM_INS_CODE1 8 #define RLC_NUM_INS_CODE2 2 #define RLC_NUM_INS_CODE3 16 static void gfx_v12_1_parse_rlc_toc(struct amdgpu_device *adev, void *rlc_toc) { RLC_TABLE_OF_CONTENT_V2 *ucode = rlc_toc; while (ucode && (ucode->id > SOC24_FIRMWARE_ID_INVALID)) { rlc_autoload_info[ucode->id].id = ucode->id; rlc_autoload_info[ucode->id].offset = ucode->offset * RLC_TOC_OFFSET_DWUNIT * 4; rlc_autoload_info[ucode->id].size = ucode->size_x16 ? ucode->size * RLC_SIZE_MULTIPLE * 4 : ucode->size * 4; switch (ucode->vfflr_image_code) { case 0: rlc_autoload_info[ucode->id].num_inst = RLC_NUM_INS_CODE0; break; case 1: rlc_autoload_info[ucode->id].num_inst = RLC_NUM_INS_CODE1; break; case 2: rlc_autoload_info[ucode->id].num_inst = RLC_NUM_INS_CODE2; break; case 3: rlc_autoload_info[ucode->id].num_inst = RLC_NUM_INS_CODE3; break; default: dev_err(adev->dev, "Invalid Instance number detected\n"); break; } ucode++; } } static uint32_t gfx_v12_1_calc_toc_total_size(struct amdgpu_device *adev) { uint32_t total_size = 0; SOC24_FIRMWARE_ID id; gfx_v12_1_parse_rlc_toc(adev, adev->psp.toc.start_addr); for (id = SOC24_FIRMWARE_ID_RLC_G_UCODE; id < SOC24_FIRMWARE_ID_MAX; id++) total_size += rlc_autoload_info[id].size; /* In case the offset in rlc toc ucode is aligned */ if (total_size < rlc_autoload_info[SOC24_FIRMWARE_ID_MAX-1].offset) total_size = rlc_autoload_info[SOC24_FIRMWARE_ID_MAX-1].offset + rlc_autoload_info[SOC24_FIRMWARE_ID_MAX-1].size; if (total_size < (RLC_TOC_UMF_SIZE_inM << 20)) total_size = RLC_TOC_UMF_SIZE_inM << 20; return total_size; } static int gfx_v12_1_rlc_autoload_buffer_init(struct amdgpu_device *adev) { int r; uint32_t total_size; total_size = gfx_v12_1_calc_toc_total_size(adev); r = amdgpu_bo_create_reserved(adev, total_size, 64 * 1024, AMDGPU_GEM_DOMAIN_VRAM, &adev->gfx.rlc.rlc_autoload_bo, &adev->gfx.rlc.rlc_autoload_gpu_addr, (void **)&adev->gfx.rlc.rlc_autoload_ptr); if (r) { dev_err(adev->dev, "(%d) failed to create fw autoload bo\n", r); return r; } return 0; } static void gfx_v12_1_rlc_backdoor_autoload_copy_ucode(struct amdgpu_device *adev, SOC24_FIRMWARE_ID id, const void *fw_data, uint32_t fw_size) { uint32_t toc_offset; uint32_t toc_fw_size, toc_fw_inst_size; char *ptr = adev->gfx.rlc.rlc_autoload_ptr; int i, num_inst; if (id <= SOC24_FIRMWARE_ID_INVALID || id >= SOC24_FIRMWARE_ID_MAX) return; toc_offset = rlc_autoload_info[id].offset; toc_fw_size = rlc_autoload_info[id].size; num_inst = rlc_autoload_info[id].num_inst; toc_fw_inst_size = toc_fw_size / num_inst; if (fw_size == 0) fw_size = toc_fw_inst_size; if (fw_size > toc_fw_inst_size) fw_size = toc_fw_inst_size; for (i = 0; i < num_inst; i++) { if ((num_inst == RLC_NUM_INS_CODE0) || ((1 << (i / 2)) & adev->gfx.xcc_mask)) { memcpy(ptr + toc_offset + i * toc_fw_inst_size, fw_data, fw_size); if (fw_size < toc_fw_inst_size) memset(ptr + toc_offset + fw_size + i * toc_fw_inst_size, 0, toc_fw_inst_size - fw_size); } } } static void gfx_v12_1_rlc_backdoor_autoload_copy_toc_ucode(struct amdgpu_device *adev) { void *data; uint32_t size; uint32_t *toc_ptr; data = adev->psp.toc.start_addr; size = rlc_autoload_info[SOC24_FIRMWARE_ID_RLC_TOC].size; toc_ptr = (uint32_t *)data + size / 4 - 2; *toc_ptr = (RLC_TOC_FORMAT_API << 24) | 0x1; gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RLC_TOC, data, size); } static void gfx_v12_1_rlc_backdoor_autoload_copy_gfx_ucode(struct amdgpu_device *adev) { const __le32 *fw_data; uint32_t fw_size; const struct gfx_firmware_header_v2_0 *cpv2_hdr; const struct rlc_firmware_header_v2_0 *rlc_hdr; const struct rlc_firmware_header_v2_1 *rlcv21_hdr; const struct rlc_firmware_header_v2_2 *rlcv22_hdr; uint16_t version_major, version_minor; /* mec ucode */ cpv2_hdr = (const struct gfx_firmware_header_v2_0 *) adev->gfx.mec_fw->data; /* instruction */ fw_data = (const __le32 *) (adev->gfx.mec_fw->data + le32_to_cpu(cpv2_hdr->ucode_offset_bytes)); fw_size = le32_to_cpu(cpv2_hdr->ucode_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RS64_MEC, fw_data, fw_size); /* data */ fw_data = (const __le32 *) (adev->gfx.mec_fw->data + le32_to_cpu(cpv2_hdr->data_offset_bytes)); fw_size = le32_to_cpu(cpv2_hdr->data_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RS64_MEC_P0_STACK, fw_data, fw_size); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RS64_MEC_P1_STACK, fw_data, fw_size); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RS64_MEC_P2_STACK, fw_data, fw_size); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RS64_MEC_P3_STACK, fw_data, fw_size); /* rlc ucode */ rlc_hdr = (const struct rlc_firmware_header_v2_0 *) adev->gfx.rlc_fw->data; fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(rlc_hdr->header.ucode_array_offset_bytes)); fw_size = le32_to_cpu(rlc_hdr->header.ucode_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RLC_G_UCODE, fw_data, fw_size); version_major = le16_to_cpu(rlc_hdr->header.header_version_major); version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor); if (version_major == 2) { if (version_minor >= 1) { rlcv21_hdr = (const struct rlc_firmware_header_v2_1 *)adev->gfx.rlc_fw->data; fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(rlcv21_hdr->save_restore_list_gpm_offset_bytes)); fw_size = le32_to_cpu(rlcv21_hdr->save_restore_list_gpm_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RLCG_SCRATCH, fw_data, fw_size); fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(rlcv21_hdr->save_restore_list_srm_offset_bytes)); fw_size = le32_to_cpu(rlcv21_hdr->save_restore_list_srm_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RLC_SRM_ARAM, fw_data, fw_size); } if (version_minor >= 2) { rlcv22_hdr = (const struct rlc_firmware_header_v2_2 *)adev->gfx.rlc_fw->data; fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(rlcv22_hdr->rlc_iram_ucode_offset_bytes)); fw_size = le32_to_cpu(rlcv22_hdr->rlc_iram_ucode_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RLX6_UCODE, fw_data, fw_size); fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(rlcv22_hdr->rlc_dram_ucode_offset_bytes)); fw_size = le32_to_cpu(rlcv22_hdr->rlc_dram_ucode_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_RLX6_DRAM_BOOT, fw_data, fw_size); } } } static void gfx_v12_1_rlc_backdoor_autoload_copy_sdma_ucode(struct amdgpu_device *adev) { const __le32 *fw_data; uint32_t fw_size; const struct sdma_firmware_header_v3_0 *sdma_hdr; if (adev->sdma.instance[0].fw) { sdma_hdr = (const struct sdma_firmware_header_v3_0 *) adev->sdma.instance[0].fw->data; fw_data = (const __le32 *) (adev->sdma.instance[0].fw->data + le32_to_cpu(sdma_hdr->ucode_offset_bytes)); fw_size = le32_to_cpu(sdma_hdr->ucode_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, SOC24_FIRMWARE_ID_SDMA_UCODE_TH0, fw_data, fw_size); } } static void gfx_v12_1_rlc_backdoor_autoload_copy_mes_ucode(struct amdgpu_device *adev) { const __le32 *fw_data; unsigned fw_size; const struct mes_firmware_header_v1_0 *mes_hdr; int pipe, ucode_id, data_id; for (pipe = 0; pipe < 2; pipe++) { if (pipe == 0) { ucode_id = SOC24_FIRMWARE_ID_RS64_MES_P0; data_id = SOC24_FIRMWARE_ID_RS64_MES_P0_STACK; } else { ucode_id = SOC24_FIRMWARE_ID_RS64_MES_P1; data_id = SOC24_FIRMWARE_ID_RS64_MES_P1_STACK; } mes_hdr = (const struct mes_firmware_header_v1_0 *) adev->mes.fw[pipe]->data; fw_data = (const __le32 *)(adev->mes.fw[pipe]->data + le32_to_cpu(mes_hdr->mes_ucode_offset_bytes)); fw_size = le32_to_cpu(mes_hdr->mes_ucode_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, ucode_id, fw_data, fw_size); fw_data = (const __le32 *)(adev->mes.fw[pipe]->data + le32_to_cpu(mes_hdr->mes_ucode_data_offset_bytes)); fw_size = le32_to_cpu(mes_hdr->mes_ucode_data_size_bytes); gfx_v12_1_rlc_backdoor_autoload_copy_ucode(adev, data_id, fw_data, fw_size); } } static int gfx_v12_1_rlc_backdoor_autoload_enable(struct amdgpu_device *adev) { uint32_t rlc_g_offset, rlc_g_size; uint64_t gpu_addr; uint32_t data; int i, num_xcc; /* RLC autoload sequence 2: copy ucode */ gfx_v12_1_rlc_backdoor_autoload_copy_sdma_ucode(adev); gfx_v12_1_rlc_backdoor_autoload_copy_gfx_ucode(adev); gfx_v12_1_rlc_backdoor_autoload_copy_mes_ucode(adev); gfx_v12_1_rlc_backdoor_autoload_copy_toc_ucode(adev); rlc_g_offset = rlc_autoload_info[SOC24_FIRMWARE_ID_RLC_G_UCODE].offset; rlc_g_size = rlc_autoload_info[SOC24_FIRMWARE_ID_RLC_G_UCODE].size; gpu_addr = adev->gfx.rlc.rlc_autoload_gpu_addr + rlc_g_offset - adev->gmc.vram_start; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { WREG32_SOC15(GC, GET_INST(GC, i), regGFX_IMU_RLC_BOOTLOADER_ADDR_HI, upper_32_bits(gpu_addr)); WREG32_SOC15(GC, GET_INST(GC, i), regGFX_IMU_RLC_BOOTLOADER_ADDR_LO, lower_32_bits(gpu_addr)); WREG32_SOC15(GC, GET_INST(GC, i), regGFX_IMU_RLC_BOOTLOADER_SIZE, rlc_g_size); } if (adev->gfx.imu.funcs) { /* RLC autoload sequence 3: load IMU fw */ if (adev->gfx.imu.funcs->load_microcode) adev->gfx.imu.funcs->load_microcode(adev); } /* unhalt rlc to start autoload */ for (i = 0; i < num_xcc; i++) { data = RREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_THREAD_ENABLE); data = REG_SET_FIELD(data, RLC_GPM_THREAD_ENABLE, THREAD0_ENABLE, 1); data = REG_SET_FIELD(data, RLC_GPM_THREAD_ENABLE, THREAD1_ENABLE, 1); WREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_THREAD_ENABLE, data); WREG32_SOC15(GC, GET_INST(GC, i), regRLC_CNTL, RLC_CNTL__RLC_ENABLE_F32_MASK); } return 0; } static int gfx_v12_1_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id = 0; unsigned num_compute_rings; int xcc_id, num_xcc; struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(12, 1, 0): adev->gfx.mec.num_mec = 1; adev->gfx.mec.num_pipe_per_mec = 4; adev->gfx.mec.num_queue_per_pipe = 8; break; default: adev->gfx.mec.num_mec = 2; adev->gfx.mec.num_pipe_per_mec = 2; adev->gfx.mec.num_queue_per_pipe = 4; break; } /* recalculate compute rings to use based on hardware configuration */ num_compute_rings = (adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_queue_per_pipe) / 2; adev->gfx.num_compute_rings = min(adev->gfx.num_compute_rings, num_compute_rings); num_xcc = NUM_XCC(adev->gfx.xcc_mask); /* EOP Event */ r = amdgpu_irq_add_id(adev, SOC_V1_0_IH_CLIENTID_GRBM_CP, GFX_12_1_0__SRCID__CP_EOP_INTERRUPT, &adev->gfx.eop_irq); if (r) return r; /* Privileged reg */ r = amdgpu_irq_add_id(adev, SOC_V1_0_IH_CLIENTID_GRBM_CP, GFX_12_1_0__SRCID__CP_PRIV_REG_FAULT, &adev->gfx.priv_reg_irq); if (r) return r; /* Privileged inst */ r = amdgpu_irq_add_id(adev, SOC_V1_0_IH_CLIENTID_GRBM_CP, GFX_12_1_0__SRCID__CP_PRIV_INSTR_FAULT, &adev->gfx.priv_inst_irq); if (r) return r; adev->gfx.gfx_current_status = AMDGPU_GFX_NORMAL_MODE; r = gfx_v12_1_rlc_init(adev); if (r) { dev_err(adev->dev, "Failed to init rlc BOs!\n"); return r; } r = gfx_v12_1_mec_init(adev); if (r) { dev_err(adev->dev, "Failed to init MEC BOs!\n"); return r; } /* set up the compute queues - allocate horizontally across pipes */ for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) { ring_id = 0; for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) { if (!amdgpu_gfx_is_mec_queue_enabled(adev, xcc_id, i, k, j)) continue; r = gfx_v12_1_compute_ring_init(adev, ring_id, xcc_id, i, k, j); if (r) return r; ring_id++; } } } if (!adev->enable_mes_kiq) { r = amdgpu_gfx_kiq_init(adev, GFX12_MEC_HPD_SIZE, xcc_id); if (r) { dev_err(adev->dev, "Failed to init KIQ BOs!\n"); return r; } r = amdgpu_gfx_kiq_init_ring(adev, xcc_id); if (r) return r; } r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v12_1_compute_mqd), xcc_id); if (r) return r; } /* allocate visible FB for rlc auto-loading fw */ if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) { r = gfx_v12_1_rlc_autoload_buffer_init(adev); if (r) return r; } else if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { r = gfx_v12_1_init_cp_compute_microcode_bo(adev); if (r) return r; } r = gfx_v12_1_gpu_early_init(adev); if (r) return r; r = amdgpu_gfx_sysfs_init(adev); if (r) return r; return 0; } static void gfx_v12_1_rlc_autoload_buffer_fini(struct amdgpu_device *adev) { amdgpu_bo_free_kernel(&adev->gfx.rlc.rlc_autoload_bo, &adev->gfx.rlc.rlc_autoload_gpu_addr, (void **)&adev->gfx.rlc.rlc_autoload_ptr); } static int gfx_v12_1_sw_fini(struct amdgpu_ip_block *ip_block) { int i, num_xcc; struct amdgpu_device *adev = ip_block->adev; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < adev->gfx.num_compute_rings * num_xcc; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); for (i = 0; i < num_xcc; i++) { amdgpu_gfx_mqd_sw_fini(adev, i); if (!adev->enable_mes_kiq) { amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[i].ring); amdgpu_gfx_kiq_fini(adev, i); } } gfx_v12_1_rlc_fini(adev); gfx_v12_1_mec_fini(adev); if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) gfx_v12_1_rlc_autoload_buffer_fini(adev); gfx_v12_1_free_microcode(adev); amdgpu_gfx_sysfs_fini(adev); return 0; } static void gfx_v12_1_xcc_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num, u32 instance, int xcc_id) { u32 data; if (instance == 0xffffffff) data = REG_SET_FIELD(0, GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES, 1); else data = REG_SET_FIELD(0, GRBM_GFX_INDEX, INSTANCE_INDEX, instance); if (se_num == 0xffffffff) data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SE_BROADCAST_WRITES, 1); else data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SE_INDEX, se_num); if (sh_num == 0xffffffff) data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SA_BROADCAST_WRITES, 1); else data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SA_INDEX, sh_num); WREG32_SOC15_RLC_SHADOW_EX(reg, GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX, data); } static u32 gfx_v12_1_get_sa_active_bitmap(struct amdgpu_device *adev, int xcc_id) { u32 gc_disabled_sa_mask, gc_user_disabled_sa_mask, sa_mask; gc_disabled_sa_mask = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCC_GC_SA_UNIT_DISABLE); gc_disabled_sa_mask = REG_GET_FIELD(gc_disabled_sa_mask, CC_GC_SA_UNIT_DISABLE, SA_DISABLE); gc_user_disabled_sa_mask = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regGC_USER_SA_UNIT_DISABLE); gc_user_disabled_sa_mask = REG_GET_FIELD(gc_user_disabled_sa_mask, GC_USER_SA_UNIT_DISABLE, SA_DISABLE); sa_mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_sh_per_se * adev->gfx.config.max_shader_engines); return sa_mask & (~(gc_disabled_sa_mask | gc_user_disabled_sa_mask)); } static u32 gfx_v12_1_get_rb_active_bitmap(struct amdgpu_device *adev, int xcc_id) { u32 gc_disabled_rb_mask, gc_user_disabled_rb_mask; u32 rb_mask; gc_disabled_rb_mask = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCC_RB_BACKEND_DISABLE); gc_disabled_rb_mask = REG_GET_FIELD(gc_disabled_rb_mask, CC_RB_BACKEND_DISABLE, BACKEND_DISABLE); gc_user_disabled_rb_mask = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regGC_USER_RB_BACKEND_DISABLE); gc_user_disabled_rb_mask = REG_GET_FIELD(gc_user_disabled_rb_mask, GC_USER_RB_BACKEND_DISABLE, BACKEND_DISABLE); rb_mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_backends_per_se * adev->gfx.config.max_shader_engines); return rb_mask & (~(gc_disabled_rb_mask | gc_user_disabled_rb_mask)); } static void gfx_v12_1_setup_rb(struct amdgpu_device *adev) { u32 rb_bitmap_width_per_sa; u32 max_sa; u32 active_sa_bitmap; u32 global_active_rb_bitmap; u32 active_rb_bitmap = 0; u32 i; int xcc_id; for (xcc_id = 0; xcc_id < NUM_XCC(adev->gfx.xcc_mask); xcc_id++) { /* query sa bitmap from SA_UNIT_DISABLE registers */ active_sa_bitmap = gfx_v12_1_get_sa_active_bitmap(adev, xcc_id); /* query rb bitmap from RB_BACKEND_DISABLE registers */ global_active_rb_bitmap = gfx_v12_1_get_rb_active_bitmap(adev, xcc_id); /* generate active rb bitmap according to active sa bitmap */ max_sa = adev->gfx.config.max_shader_engines * adev->gfx.config.max_sh_per_se; rb_bitmap_width_per_sa = adev->gfx.config.max_backends_per_se / adev->gfx.config.max_sh_per_se; for (i = 0; i < max_sa; i++) { if (active_sa_bitmap & (1 << i)) active_rb_bitmap |= (0x3 << (i * rb_bitmap_width_per_sa)); } active_rb_bitmap |= global_active_rb_bitmap; } adev->gfx.config.backend_enable_mask = active_rb_bitmap; adev->gfx.config.num_rbs = hweight32(active_rb_bitmap); } static void gfx_v12_1_xcc_init_compute_vmid(struct amdgpu_device *adev, int xcc_id) { int i; uint32_t sh_mem_bases; uint32_t data; /* * Configure apertures: * LDS: 0x20000000'00000000 - 0x20000001'00000000 (4GB) * Scratch: 0x10000000'00000000 - 0x10000001'00000000 (4GB) */ sh_mem_bases = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE, (adev->gmc.private_aperture_start >> 58)); sh_mem_bases = REG_SET_FIELD(sh_mem_bases, SH_MEM_BASES, SHARED_BASE, (adev->gmc.shared_aperture_start >> 48)); mutex_lock(&adev->srbm_mutex); for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { soc_v1_0_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, xcc_id)); /* CP and shaders */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSH_MEM_BASES, sh_mem_bases); /* Enable trap for each kfd vmid. */ data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSPI_GDBG_PER_VMID_CNTL); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSPI_GDBG_PER_VMID_CNTL, data); /* Disable VGPR deallocation instruction for each KFD vmid. */ data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_DEBUG); data = REG_SET_FIELD(data, SQ_DEBUG, DISABLE_VGPR_DEALLOC, 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_DEBUG, data); } soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); } static void gfx_v12_1_tcp_harvest(struct amdgpu_device *adev) { /* TODO: harvest feature to be added later. */ } static void gfx_v12_1_get_tcc_info(struct amdgpu_device *adev) { } static void gfx_v12_1_xcc_constants_init(struct amdgpu_device *adev, int xcc_id) { u32 tmp; int i; /* XXX SH_MEM regs */ /* where to put LDS, scratch, GPUVM in FSA64 space */ mutex_lock(&adev->srbm_mutex); for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) { soc_v1_0_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, xcc_id)); /* CP and shaders */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG); if (i != 0) { tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE, (adev->gmc.private_aperture_start >> 58)); tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE, (adev->gmc.shared_aperture_start >> 48)); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSH_MEM_BASES, tmp); } } soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); gfx_v12_1_xcc_init_compute_vmid(adev, xcc_id); } static void gfx_v12_1_constants_init(struct amdgpu_device *adev) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); gfx_v12_1_setup_rb(adev); gfx_v12_1_get_cu_info(adev, &adev->gfx.cu_info); gfx_v12_1_get_tcc_info(adev); adev->gfx.config.pa_sc_tile_steering_override = 0; for (i = 0; i < num_xcc; i++) gfx_v12_1_xcc_constants_init(adev, i); } static void gfx_v12_1_xcc_enable_gui_idle_interrupt(struct amdgpu_device *adev, bool enable, int xcc_id) { u32 tmp; if (amdgpu_sriov_vf(adev)) return; tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0); tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE, enable ? 1 : 0); tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_EMPTY_INT_ENABLE, enable ? 1 : 0); tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CMP_BUSY_INT_ENABLE, enable ? 1 : 0); tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, GFX_IDLE_INT_ENABLE, enable ? 1 : 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0, tmp); } static int gfx_v12_1_xcc_init_csb(struct amdgpu_device *adev, int xcc_id) { adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CSIB_ADDR_HI, adev->gfx.rlc.clear_state_gpu_addr >> 32); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CSIB_ADDR_LO, adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size); return 0; } static void gfx_v12_1_xcc_rlc_stop(struct amdgpu_device *adev, int xcc_id) { u32 tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CNTL); tmp = REG_SET_FIELD(tmp, RLC_CNTL, RLC_ENABLE_F32, 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CNTL, tmp); } static void gfx_v12_1_rlc_stop(struct amdgpu_device *adev) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) gfx_v12_1_xcc_rlc_stop(adev, i); } static void gfx_v12_1_xcc_rlc_reset(struct amdgpu_device *adev, int xcc_id) { WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), GRBM_SOFT_RESET, SOFT_RESET_RLC, 1); udelay(50); WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), GRBM_SOFT_RESET, SOFT_RESET_RLC, 0); udelay(50); } static void gfx_v12_1_rlc_reset(struct amdgpu_device *adev) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) gfx_v12_1_xcc_rlc_reset(adev, i); } static void gfx_v12_1_xcc_rlc_smu_handshake_cntl(struct amdgpu_device *adev, bool enable, int xcc_id) { uint32_t rlc_pg_cntl; rlc_pg_cntl = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_PG_CNTL); if (!enable) { /* RLC_PG_CNTL[23] = 0 (default) * RLC will wait for handshake acks with SMU * GFXOFF will be enabled * RLC_PG_CNTL[23] = 1 * RLC will not issue any message to SMU * hence no handshake between SMU & RLC * GFXOFF will be disabled */ rlc_pg_cntl |= RLC_PG_CNTL__SMU_HANDSHAKE_DISABLE_MASK; } else rlc_pg_cntl &= ~RLC_PG_CNTL__SMU_HANDSHAKE_DISABLE_MASK; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_PG_CNTL, rlc_pg_cntl); } static void gfx_v12_1_xcc_rlc_start(struct amdgpu_device *adev, int xcc_id) { /* TODO: enable rlc & smu handshake until smu * and gfxoff feature works as expected */ if (!(amdgpu_pp_feature_mask & PP_GFXOFF_MASK)) gfx_v12_1_xcc_rlc_smu_handshake_cntl(adev, false, xcc_id); WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_CNTL, RLC_ENABLE_F32, 1); udelay(50); } static void gfx_v12_1_rlc_start(struct amdgpu_device *adev) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { gfx_v12_1_xcc_rlc_start(adev, i); } } static void gfx_v12_1_xcc_rlc_enable_srm(struct amdgpu_device *adev, int xcc_id) { uint32_t tmp; /* enable Save Restore Machine */ tmp = RREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_SRM_CNTL)); tmp |= RLC_SRM_CNTL__AUTO_INCR_ADDR_MASK; tmp |= RLC_SRM_CNTL__SRM_ENABLE_MASK; WREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_SRM_CNTL), tmp); } static void gfx_v12_1_xcc_load_rlcg_microcode(struct amdgpu_device *adev, int xcc_id) { const struct rlc_firmware_header_v2_0 *hdr; const __le32 *fw_data; unsigned i, fw_size; hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_ADDR, RLCG_UCODE_LOADING_START_ADDRESS); for (i = 0; i < fw_size; i++) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++)); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version); } static void gfx_v12_1_xcc_load_rlc_iram_dram_microcode(struct amdgpu_device *adev, int xcc_id) { const struct rlc_firmware_header_v2_2 *hdr; const __le32 *fw_data; unsigned i, fw_size; u32 tmp; hdr = (const struct rlc_firmware_header_v2_2 *)adev->gfx.rlc_fw->data; fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(hdr->rlc_iram_ucode_offset_bytes)); fw_size = le32_to_cpu(hdr->rlc_iram_ucode_size_bytes) / 4; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_IRAM_ADDR, 0); for (i = 0; i < fw_size; i++) { if ((amdgpu_emu_mode == 1) && (i % 100 == 99)) msleep(1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_IRAM_DATA, le32_to_cpup(fw_data++)); } WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_IRAM_ADDR, adev->gfx.rlc_fw_version); fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(hdr->rlc_dram_ucode_offset_bytes)); fw_size = le32_to_cpu(hdr->rlc_dram_ucode_size_bytes) / 4; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_DRAM_ADDR, 0); for (i = 0; i < fw_size; i++) { if ((amdgpu_emu_mode == 1) && (i % 100 == 99)) msleep(1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_DRAM_DATA, le32_to_cpup(fw_data++)); } WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_IRAM_ADDR, adev->gfx.rlc_fw_version); tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_CNTL); tmp = REG_SET_FIELD(tmp, RLC_LX6_CNTL, PDEBUG_ENABLE, 1); tmp = REG_SET_FIELD(tmp, RLC_LX6_CNTL, BRESET, 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_LX6_CNTL, tmp); } static int gfx_v12_1_xcc_rlc_load_microcode(struct amdgpu_device *adev, int xcc_id) { const struct rlc_firmware_header_v2_0 *hdr; uint16_t version_major; uint16_t version_minor; if (!adev->gfx.rlc_fw) return -EINVAL; hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; amdgpu_ucode_print_rlc_hdr(&hdr->header); version_major = le16_to_cpu(hdr->header.header_version_major); version_minor = le16_to_cpu(hdr->header.header_version_minor); if (version_major == 2) { gfx_v12_1_xcc_load_rlcg_microcode(adev, xcc_id); if (amdgpu_dpm == 1) { if (version_minor >= 2) gfx_v12_1_xcc_load_rlc_iram_dram_microcode(adev, xcc_id); } return 0; } return -EINVAL; } static int gfx_v12_1_xcc_rlc_resume(struct amdgpu_device *adev, int xcc_id) { int r; if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { gfx_v12_1_xcc_init_csb(adev, xcc_id); if (!amdgpu_sriov_vf(adev)) /* enable RLC SRM */ gfx_v12_1_xcc_rlc_enable_srm(adev, xcc_id); } else { if (amdgpu_sriov_vf(adev)) { gfx_v12_1_xcc_init_csb(adev, xcc_id); return 0; } gfx_v12_1_xcc_rlc_stop(adev, xcc_id); /* disable CG */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, 0); /* disable PG */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_PG_CNTL, 0); if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { /* legacy rlc firmware loading */ r = gfx_v12_1_xcc_rlc_load_microcode(adev, xcc_id); if (r) return r; } gfx_v12_1_xcc_init_csb(adev, xcc_id); gfx_v12_1_xcc_rlc_start(adev, xcc_id); } return 0; } static int gfx_v12_1_rlc_resume(struct amdgpu_device *adev) { int r, i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { r = gfx_v12_1_xcc_rlc_resume(adev, i); if (r) return r; } return 0; } static void gfx_v12_1_xcc_config_gfx_rs64(struct amdgpu_device *adev, int xcc_id) { const struct gfx_firmware_header_v2_0 *mec_hdr; uint32_t pipe_id, tmp; mec_hdr = (const struct gfx_firmware_header_v2_0 *) adev->gfx.mec_fw->data; /* config mec program start addr */ for (pipe_id = 0; pipe_id < 4; pipe_id++) { soc_v1_0_grbm_select(adev, 1, pipe_id, 0, 0, GET_INST(GC, xcc_id)); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_PRGRM_CNTR_START, mec_hdr->ucode_start_addr_lo >> 2 | mec_hdr->ucode_start_addr_hi << 30); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_PRGRM_CNTR_START_HI, mec_hdr->ucode_start_addr_hi >> 2); } soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); /* reset mec pipe */ tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_CNTL); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE0_RESET, 1); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE1_RESET, 1); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE2_RESET, 1); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE3_RESET, 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_CNTL, tmp); /* clear mec pipe reset */ tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE0_RESET, 0); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE1_RESET, 0); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE2_RESET, 0); tmp = REG_SET_FIELD(tmp, CP_MEC_RS64_CNTL, MEC_PIPE3_RESET, 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_CNTL, tmp); } static void gfx_v12_1_config_gfx_rs64(struct amdgpu_device *adev) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) gfx_v12_1_xcc_config_gfx_rs64(adev, i); } static void gfx_v12_1_xcc_set_mec_ucode_start_addr(struct amdgpu_device *adev, int xcc_id) { const struct gfx_firmware_header_v2_0 *cp_hdr; unsigned pipe_id; cp_hdr = (const struct gfx_firmware_header_v2_0 *) adev->gfx.mec_fw->data; mutex_lock(&adev->srbm_mutex); for (pipe_id = 0; pipe_id < adev->gfx.mec.num_pipe_per_mec; pipe_id++) { soc_v1_0_grbm_select(adev, 1, pipe_id, 0, 0, GET_INST(GC, xcc_id)); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_PRGRM_CNTR_START, cp_hdr->ucode_start_addr_lo >> 2 | cp_hdr->ucode_start_addr_hi << 30); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_PRGRM_CNTR_START_HI, cp_hdr->ucode_start_addr_hi >> 2); } soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); } static int gfx_v12_1_xcc_wait_for_rlc_autoload_complete(struct amdgpu_device *adev, int xcc_id) { uint32_t cp_status; uint32_t bootload_status; int i; for (i = 0; i < adev->usec_timeout; i++) { cp_status = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_STAT); bootload_status = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_RLCS_BOOTLOAD_STATUS); if ((cp_status == 0) && (REG_GET_FIELD(bootload_status, RLC_RLCS_BOOTLOAD_STATUS, BOOTLOAD_COMPLETE) == 1)) { break; } udelay(1); if (amdgpu_emu_mode) msleep(10); } if (i >= adev->usec_timeout) { dev_err(adev->dev, "rlc autoload: xcc%d gc ucode autoload timeout\n", xcc_id); return -ETIMEDOUT; } if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) { gfx_v12_1_xcc_set_mec_ucode_start_addr(adev, xcc_id); } return 0; } static int gfx_v12_1_wait_for_rlc_autoload_complete(struct amdgpu_device *adev) { int xcc_id; for (xcc_id = 0; xcc_id < NUM_XCC(adev->gfx.xcc_mask); xcc_id++) gfx_v12_1_xcc_wait_for_rlc_autoload_complete(adev, xcc_id); return 0; } static void gfx_v12_1_xcc_cp_compute_enable(struct amdgpu_device *adev, bool enable, int xcc_id) { u32 data; data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_CNTL); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_INVALIDATE_ICACHE, enable ? 0 : 1); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE0_RESET, enable ? 0 : 1); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE1_RESET, enable ? 0 : 1); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE2_RESET, enable ? 0 : 1); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE3_RESET, enable ? 0 : 1); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE0_ACTIVE, enable ? 1 : 0); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE1_ACTIVE, enable ? 1 : 0); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE2_ACTIVE, enable ? 1 : 0); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_PIPE3_ACTIVE, enable ? 1 : 0); data = REG_SET_FIELD(data, CP_MEC_RS64_CNTL, MEC_HALT, enable ? 0 : 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_RS64_CNTL, data); adev->gfx.kiq[xcc_id].ring.sched.ready = enable; udelay(50); } static int gfx_v12_1_init_cp_compute_microcode_bo(struct amdgpu_device *adev) { const struct gfx_firmware_header_v2_0 *mec_hdr; const __le32 *fw_ucode, *fw_data; u32 fw_ucode_size, fw_data_size; u32 *fw_ucode_ptr, *fw_data_ptr; int i, r, xcc_id; if (!adev->gfx.mec_fw) return -EINVAL; mec_hdr = (const struct gfx_firmware_header_v2_0 *)adev->gfx.mec_fw->data; amdgpu_ucode_print_gfx_hdr(&mec_hdr->header); fw_ucode = (const __le32 *) (adev->gfx.mec_fw->data + le32_to_cpu(mec_hdr->ucode_offset_bytes)); fw_ucode_size = le32_to_cpu(mec_hdr->ucode_size_bytes); fw_data = (const __le32 *) (adev->gfx.mec_fw->data + le32_to_cpu(mec_hdr->data_offset_bytes)); fw_data_size = le32_to_cpu(mec_hdr->data_size_bytes); if (adev->gfx.mec.mec_fw_obj == NULL) { r = amdgpu_bo_create_reserved(adev, fw_ucode_size, 64 * 1024, AMDGPU_GEM_DOMAIN_VRAM, &adev->gfx.mec.mec_fw_obj, &adev->gfx.mec.mec_fw_gpu_addr, (void **)&fw_ucode_ptr); if (r) { dev_err(adev->dev, "(%d) failed to create mec fw ucode bo\n", r); gfx_v12_1_mec_fini(adev); return r; } memcpy(fw_ucode_ptr, fw_ucode, fw_ucode_size); amdgpu_bo_kunmap(adev->gfx.mec.mec_fw_obj); amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_obj); } if (adev->gfx.mec.mec_fw_data_obj == NULL) { r = amdgpu_bo_create_reserved(adev, ALIGN(fw_data_size, 64 * 1024) * adev->gfx.mec.num_pipe_per_mec * NUM_XCC(adev->gfx.xcc_mask), 64 * 1024, AMDGPU_GEM_DOMAIN_VRAM, &adev->gfx.mec.mec_fw_data_obj, &adev->gfx.mec.mec_fw_data_gpu_addr, (void **)&fw_data_ptr); if (r) { dev_err(adev->dev, "(%d) failed to create mec fw data bo\n", r); gfx_v12_1_mec_fini(adev); return r; } for (xcc_id = 0; xcc_id < NUM_XCC(adev->gfx.xcc_mask); xcc_id++) { for (i = 0; i < adev->gfx.mec.num_pipe_per_mec; i++) { u32 offset = (xcc_id * adev->gfx.mec.num_pipe_per_mec + i) * ALIGN(fw_data_size, 64 * 1024) / 4; memcpy(fw_data_ptr + offset, fw_data, fw_data_size); } } amdgpu_bo_kunmap(adev->gfx.mec.mec_fw_data_obj); amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_data_obj); } return 0; } static int gfx_v12_1_xcc_cp_compute_load_microcode_rs64(struct amdgpu_device *adev, int xcc_id) { const struct gfx_firmware_header_v2_0 *mec_hdr; u32 fw_data_size; u32 tmp, i, usec_timeout = 50000; /* Wait for 50 ms */ if (!adev->gfx.mec_fw) return -EINVAL; mec_hdr = (const struct gfx_firmware_header_v2_0 *)adev->gfx.mec_fw->data; fw_data_size = le32_to_cpu(mec_hdr->data_size_bytes); gfx_v12_1_xcc_cp_compute_enable(adev, false, xcc_id); tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, VMID, 0); tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, EXE_DISABLE, 0); tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_CNTL, tmp); tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_MEC_DC_BASE_CNTL, VMID, 0); tmp = REG_SET_FIELD(tmp, CP_MEC_DC_BASE_CNTL, CACHE_POLICY, 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DC_BASE_CNTL, tmp); mutex_lock(&adev->srbm_mutex); for (i = 0; i < adev->gfx.mec.num_pipe_per_mec; i++) { soc_v1_0_grbm_select(adev, 1, i, 0, 0, GET_INST(GC, xcc_id)); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_MDBASE_LO, lower_32_bits(adev->gfx.mec.mec_fw_data_gpu_addr + (xcc_id * adev->gfx.mec.num_pipe_per_mec + i) * ALIGN(fw_data_size, 64 * 1024))); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_MDBASE_HI, upper_32_bits(adev->gfx.mec.mec_fw_data_gpu_addr + (xcc_id * adev->gfx.mec.num_pipe_per_mec + i) * ALIGN(fw_data_size, 64 * 1024))); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_LO, lower_32_bits(adev->gfx.mec.mec_fw_gpu_addr)); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_HI, upper_32_bits(adev->gfx.mec.mec_fw_gpu_addr)); } mutex_unlock(&adev->srbm_mutex); soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); /* Trigger an invalidation of the L1 instruction caches */ tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DC_OP_CNTL); tmp = REG_SET_FIELD(tmp, CP_MEC_DC_OP_CNTL, INVALIDATE_DCACHE, 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DC_OP_CNTL, tmp); /* Wait for invalidation complete */ for (i = 0; i < usec_timeout; i++) { tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DC_OP_CNTL); if (1 == REG_GET_FIELD(tmp, CP_MEC_DC_OP_CNTL, INVALIDATE_DCACHE_COMPLETE)) break; udelay(1); } if (i >= usec_timeout) { dev_err(adev->dev, "failed to invalidate data cache\n"); return -EINVAL; } /* Trigger an invalidation of the L1 instruction caches */ tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_OP_CNTL); tmp = REG_SET_FIELD(tmp, CP_CPC_IC_OP_CNTL, INVALIDATE_CACHE, 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_OP_CNTL, tmp); /* Wait for invalidation complete */ for (i = 0; i < usec_timeout; i++) { tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_OP_CNTL); if (1 == REG_GET_FIELD(tmp, CP_CPC_IC_OP_CNTL, INVALIDATE_CACHE_COMPLETE)) break; udelay(1); } if (i >= usec_timeout) { dev_err(adev->dev, "failed to invalidate instruction cache\n"); return -EINVAL; } gfx_v12_1_xcc_set_mec_ucode_start_addr(adev, xcc_id); return 0; } static void gfx_v12_1_xcc_kiq_setting(struct amdgpu_ring *ring, int xcc_id) { uint32_t tmp; struct amdgpu_device *adev = ring->adev; /* tell RLC which is KIQ queue */ tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS); tmp &= 0xffffff00; tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp); tmp |= 0x80; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp); } static void gfx_v12_1_xcc_cp_set_doorbell_range(struct amdgpu_device *adev, int xcc_id) { /* disable gfx engine doorbell range */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_DOORBELL_RANGE_LOWER, 0); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_DOORBELL_RANGE_UPPER, 0); /* set compute engine doorbell range */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DOORBELL_RANGE_LOWER, ((adev->doorbell_index.kiq + xcc_id * adev->doorbell_index.xcc_doorbell_range) * 2) << 2); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DOORBELL_RANGE_UPPER, ((adev->doorbell_index.userqueue_end + xcc_id * adev->doorbell_index.xcc_doorbell_range) * 2) << 2); } static int gfx_v12_1_compute_mqd_init(struct amdgpu_device *adev, void *m, struct amdgpu_mqd_prop *prop) { struct v12_1_compute_mqd *mqd = m; uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr; uint32_t tmp; mqd->header = 0xC0310800; mqd->compute_pipelinestat_enable = 0x00000001; mqd->compute_static_thread_mgmt_se0 = 0xffffffff; mqd->compute_static_thread_mgmt_se1 = 0xffffffff; mqd->compute_static_thread_mgmt_se2 = 0xffffffff; mqd->compute_static_thread_mgmt_se3 = 0xffffffff; mqd->compute_misc_reserved = 0x00000007; eop_base_addr = prop->eop_gpu_addr >> 8; mqd->cp_hqd_eop_base_addr_lo = eop_base_addr; mqd->cp_hqd_eop_base_addr_hi = upper_32_bits(eop_base_addr); /* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */ tmp = regCP_HQD_EOP_CONTROL_DEFAULT; tmp = REG_SET_FIELD(tmp, CP_HQD_EOP_CONTROL, EOP_SIZE, (order_base_2(GFX12_MEC_HPD_SIZE / 4) - 1)); mqd->cp_hqd_eop_control = tmp; /* enable doorbell? */ tmp = regCP_HQD_PQ_DOORBELL_CONTROL_DEFAULT; if (prop->use_doorbell) { tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_OFFSET, prop->doorbell_index); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_SOURCE, 0); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_HIT, 0); } else { tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 0); } mqd->cp_hqd_pq_doorbell_control = tmp; /* disable the queue if it's active */ mqd->cp_hqd_dequeue_request = 0; mqd->cp_hqd_pq_rptr = 0; mqd->cp_hqd_pq_wptr_lo = 0; mqd->cp_hqd_pq_wptr_hi = 0; /* set the pointer to the MQD */ mqd->cp_mqd_base_addr_lo = prop->mqd_gpu_addr & 0xfffffffc; mqd->cp_mqd_base_addr_hi = upper_32_bits(prop->mqd_gpu_addr); /* set MQD vmid to 0 */ tmp = regCP_MQD_CONTROL_DEFAULT; tmp = REG_SET_FIELD(tmp, CP_MQD_CONTROL, VMID, 0); mqd->cp_mqd_control = tmp; /* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */ hqd_gpu_addr = prop->hqd_base_gpu_addr >> 8; mqd->cp_hqd_pq_base_lo = hqd_gpu_addr; mqd->cp_hqd_pq_base_hi = upper_32_bits(hqd_gpu_addr); /* set up the HQD, this is similar to CP_RB0_CNTL */ tmp = regCP_HQD_PQ_CONTROL_DEFAULT; tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, QUEUE_SIZE, (order_base_2(prop->queue_size / 4) - 1)); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, RPTR_BLOCK_SIZE, (order_base_2(AMDGPU_GPU_PAGE_SIZE / 4) - 1)); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 0); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, 0); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1); mqd->cp_hqd_pq_control = tmp; /* set the wb address whether it's enabled or not */ wb_gpu_addr = prop->rptr_gpu_addr; mqd->cp_hqd_pq_rptr_report_addr_lo = wb_gpu_addr & 0xfffffffc; mqd->cp_hqd_pq_rptr_report_addr_hi = upper_32_bits(wb_gpu_addr) & 0xffff; /* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */ wb_gpu_addr = prop->wptr_gpu_addr; mqd->cp_hqd_pq_wptr_poll_addr_lo = wb_gpu_addr & 0xfffffffc; mqd->cp_hqd_pq_wptr_poll_addr_hi = upper_32_bits(wb_gpu_addr) & 0xffff; tmp = 0; /* enable the doorbell if requested */ if (prop->use_doorbell) { tmp = regCP_HQD_PQ_DOORBELL_CONTROL_DEFAULT; tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_OFFSET, prop->doorbell_index); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_SOURCE, 0); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_HIT, 0); } mqd->cp_hqd_pq_doorbell_control = tmp; /* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */ mqd->cp_hqd_pq_rptr = regCP_HQD_PQ_RPTR_DEFAULT; /* set the vmid for the queue */ mqd->cp_hqd_vmid = 0; tmp = regCP_HQD_PERSISTENT_STATE_DEFAULT; tmp = REG_SET_FIELD(tmp, CP_HQD_PERSISTENT_STATE, PRELOAD_SIZE, 0x63); mqd->cp_hqd_persistent_state = tmp; /* set MIN_IB_AVAIL_SIZE */ tmp = regCP_HQD_IB_CONTROL_DEFAULT; tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 1); mqd->cp_hqd_ib_control = tmp; /* set static priority for a compute queue/ring */ mqd->cp_hqd_pipe_priority = prop->hqd_pipe_priority; mqd->cp_hqd_queue_priority = prop->hqd_queue_priority; mqd->cp_mqd_stride_size = prop->mqd_stride_size ? prop->mqd_stride_size : AMDGPU_MQD_SIZE_ALIGN(adev->mqds[AMDGPU_HW_IP_COMPUTE].mqd_size); mqd->cp_hqd_active = prop->hqd_active; return 0; } static int gfx_v12_1_xcc_kiq_init_register(struct amdgpu_ring *ring, int xcc_id) { struct amdgpu_device *adev = ring->adev; struct v12_1_compute_mqd *mqd = ring->mqd_ptr; int j; /* inactivate the queue */ if (amdgpu_sriov_vf(adev)) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE, 0); /* disable wptr polling */ WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), CP_PQ_WPTR_POLL_CNTL, EN, 0); /* write the EOP addr */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR, mqd->cp_hqd_eop_base_addr_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR_HI, mqd->cp_hqd_eop_base_addr_hi); /* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_CONTROL, mqd->cp_hqd_eop_control); /* enable doorbell? */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, mqd->cp_hqd_pq_doorbell_control); /* disable the queue if it's active */ if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1) { WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, 1); for (j = 0; j < adev->usec_timeout; j++) { if (!(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1)) break; udelay(1); } WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, mqd->cp_hqd_dequeue_request); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR, mqd->cp_hqd_pq_rptr); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO, mqd->cp_hqd_pq_wptr_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI, mqd->cp_hqd_pq_wptr_hi); } /* set the pointer to the MQD */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MQD_BASE_ADDR, mqd->cp_mqd_base_addr_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MQD_BASE_ADDR_HI, mqd->cp_mqd_base_addr_hi); /* set MQD vmid to 0 */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MQD_CONTROL, mqd->cp_mqd_control); /* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_BASE, mqd->cp_hqd_pq_base_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_BASE_HI, mqd->cp_hqd_pq_base_hi); /* set up the HQD, this is similar to CP_RB0_CNTL */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_CONTROL, mqd->cp_hqd_pq_control); /* set the wb address whether it's enabled or not */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR_REPORT_ADDR, mqd->cp_hqd_pq_rptr_report_addr_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR_REPORT_ADDR_HI, mqd->cp_hqd_pq_rptr_report_addr_hi); /* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_POLL_ADDR, mqd->cp_hqd_pq_wptr_poll_addr_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_POLL_ADDR_HI, mqd->cp_hqd_pq_wptr_poll_addr_hi); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, mqd->cp_hqd_pq_doorbell_control); /* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO, mqd->cp_hqd_pq_wptr_lo); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI, mqd->cp_hqd_pq_wptr_hi); /* set the vmid for the queue */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_VMID, mqd->cp_hqd_vmid); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE, mqd->cp_hqd_persistent_state); /* activate the queue */ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE, mqd->cp_hqd_active); if (ring->use_doorbell) WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), CP_PQ_STATUS, DOORBELL_ENABLE, 1); return 0; } static int gfx_v12_1_xcc_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id) { struct amdgpu_device *adev = ring->adev; struct v12_1_compute_mqd *mqd = ring->mqd_ptr; gfx_v12_1_xcc_kiq_setting(ring, xcc_id); if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ /* reset MQD to a clean status */ if (adev->gfx.kiq[xcc_id].mqd_backup) memcpy(mqd, adev->gfx.kiq[xcc_id].mqd_backup, sizeof(*mqd)); /* reset ring buffer */ ring->wptr = 0; amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); soc_v1_0_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id)); gfx_v12_1_xcc_kiq_init_register(ring, xcc_id); soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); } else { memset((void *)mqd, 0, sizeof(*mqd)); if (amdgpu_sriov_vf(adev) && adev->in_suspend) amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); soc_v1_0_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id)); amdgpu_ring_init_mqd(ring); gfx_v12_1_xcc_kiq_init_register(ring, xcc_id); soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); if (adev->gfx.kiq[xcc_id].mqd_backup) memcpy(adev->gfx.kiq[xcc_id].mqd_backup, mqd, sizeof(*mqd)); } return 0; } static int gfx_v12_1_xcc_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id) { struct amdgpu_device *adev = ring->adev; struct v12_1_compute_mqd *mqd = ring->mqd_ptr; int mqd_idx = ring - &adev->gfx.compute_ring[0]; if (!amdgpu_in_reset(adev) && !adev->in_suspend) { memset((void *)mqd, 0, sizeof(*mqd)); mutex_lock(&adev->srbm_mutex); soc_v1_0_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id)); amdgpu_ring_init_mqd(ring); soc_v1_0_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy_fromio(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); } else { /* restore MQD to a clean status */ if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy_toio(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd)); /* reset ring buffer */ ring->wptr = 0; atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); } return 0; } static int gfx_v12_1_xcc_kiq_resume(struct amdgpu_device *adev, int xcc_id) { struct amdgpu_ring *ring; int r; ring = &adev->gfx.kiq[xcc_id].ring; r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) return r; r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); if (unlikely(r != 0)) { amdgpu_bo_unreserve(ring->mqd_obj); return r; } gfx_v12_1_xcc_kiq_init_queue(ring, xcc_id); amdgpu_bo_kunmap(ring->mqd_obj); ring->mqd_ptr = NULL; amdgpu_bo_unreserve(ring->mqd_obj); ring->sched.ready = true; return 0; } static int gfx_v12_1_xcc_kcq_resume(struct amdgpu_device *adev, int xcc_id) { struct amdgpu_ring *ring = NULL; int r = 0, i; if (!amdgpu_async_gfx_ring) gfx_v12_1_xcc_cp_compute_enable(adev, true, xcc_id); for (i = 0; i < adev->gfx.num_compute_rings; i++) { ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings]; r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) goto done; r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); if (!r) { r = gfx_v12_1_xcc_kcq_init_queue(ring, xcc_id); amdgpu_bo_kunmap(ring->mqd_obj); ring->mqd_ptr = NULL; } amdgpu_bo_unreserve(ring->mqd_obj); if (r) goto done; } r = amdgpu_gfx_enable_kcq(adev, xcc_id); done: return r; } static int gfx_v12_1_xcc_cp_resume(struct amdgpu_device *adev, uint16_t xcc_mask) { int r, i, xcc_id; struct amdgpu_ring *ring; for_each_inst(xcc_id, xcc_mask) { if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { /* legacy firmware loading */ r = gfx_v12_1_xcc_cp_compute_load_microcode_rs64(adev, xcc_id); if (r) return r; } /* GFX CGCG and LS is set by default */ if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) gfx_v12_1_xcc_enable_gui_idle_interrupt(adev, true, xcc_id); gfx_v12_1_xcc_cp_set_doorbell_range(adev, xcc_id); gfx_v12_1_xcc_cp_compute_enable(adev, true, xcc_id); if (adev->enable_mes_kiq && adev->mes.kiq_hw_init) r = amdgpu_mes_kiq_hw_init(adev, xcc_id); else r = gfx_v12_1_xcc_kiq_resume(adev, xcc_id); if (r) return r; r = gfx_v12_1_xcc_kcq_resume(adev, xcc_id); if (r) return r; for (i = 0; i < adev->gfx.num_compute_rings; i++) { ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings]; r = amdgpu_ring_test_helper(ring); if (r) return r; } } return 0; } static int gfx_v12_1_cp_resume(struct amdgpu_device *adev) { int num_xcc, num_xcp, num_xcc_per_xcp; uint16_t xcc_mask; int r = 0; num_xcc = NUM_XCC(adev->gfx.xcc_mask); if (amdgpu_sriov_vf(adev)) { enum amdgpu_gfx_partition mode; mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr, AMDGPU_XCP_FL_NONE); if (mode == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE) return -EINVAL; if (adev->gfx.funcs && adev->gfx.funcs->get_xccs_per_xcp) { num_xcc_per_xcp = adev->gfx.funcs->get_xccs_per_xcp(adev); adev->gfx.num_xcc_per_xcp = num_xcc_per_xcp; num_xcp = num_xcc / num_xcc_per_xcp; } else { return -EINVAL; } r = amdgpu_xcp_init(adev->xcp_mgr, num_xcp, mode); } else { if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr, AMDGPU_XCP_FL_NONE) == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE) r = amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, amdgpu_user_partt_mode); } if (r) return r; xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0); return gfx_v12_1_xcc_cp_resume(adev, xcc_mask); } static int gfx_v12_1_gfxhub_enable(struct amdgpu_device *adev) { int r, i; bool value; r = adev->gfxhub.funcs->gart_enable(adev); if (r) return r; value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ? false : true; adev->gfxhub.funcs->set_fault_enable_default(adev, value); /* TODO investigate why TLB flush is needed, * are we missing a flush somewhere else? */ for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) { if (AMDGPU_IS_GFXHUB(i)) adev->gmc.gmc_funcs->flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(i), 0); } return 0; } static int get_gb_addr_config(struct amdgpu_device *adev) { u32 gb_addr_config; gb_addr_config = RREG32_SOC15(GC, GET_INST(GC, 0), regGB_ADDR_CONFIG_READ); if (gb_addr_config == 0) return -EINVAL; adev->gfx.config.gb_addr_config_fields.num_pkrs = 1 << REG_GET_FIELD(gb_addr_config, GB_ADDR_CONFIG_READ, NUM_PKRS); adev->gfx.config.gb_addr_config = gb_addr_config; adev->gfx.config.gb_addr_config_fields.num_pipes = 1 << REG_GET_FIELD(adev->gfx.config.gb_addr_config, GB_ADDR_CONFIG_READ, NUM_PIPES); adev->gfx.config.max_tile_pipes = adev->gfx.config.gb_addr_config_fields.num_pipes; adev->gfx.config.gb_addr_config_fields.max_compress_frags = 1 << REG_GET_FIELD(adev->gfx.config.gb_addr_config, GB_ADDR_CONFIG_READ, MAX_COMPRESSED_FRAGS); adev->gfx.config.gb_addr_config_fields.num_rb_per_se = 1 << REG_GET_FIELD(adev->gfx.config.gb_addr_config, GB_ADDR_CONFIG_READ, NUM_RB_PER_SE); adev->gfx.config.gb_addr_config_fields.num_se = 1 << REG_GET_FIELD(adev->gfx.config.gb_addr_config, GB_ADDR_CONFIG_READ, NUM_SHADER_ENGINES); adev->gfx.config.gb_addr_config_fields.pipe_interleave_size = 1 << (8 + REG_GET_FIELD(adev->gfx.config.gb_addr_config, GB_ADDR_CONFIG_READ, PIPE_INTERLEAVE_SIZE)); return 0; } static void gfx_v12_1_xcc_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG); data |= CPC_PSP_DEBUG__GPA_OVERRIDE_MASK; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG, data); data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPG_PSP_DEBUG); data |= CPG_PSP_DEBUG__GPA_OVERRIDE_MASK; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPG_PSP_DEBUG, data); } static void gfx_v12_1_xcc_setup_tcp_thrashing_ctrl(struct amdgpu_device *adev, int xcc_id) { uint32_t val; /* Set the TCP UTCL0 register to enable atomics */ val = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_UTCL0_THRASHING_CTRL); val = REG_SET_FIELD(val, TCP_UTCL0_THRASHING_CTRL, THRASHING_EN, 0x2); val = REG_SET_FIELD(val, TCP_UTCL0_THRASHING_CTRL, RETRY_FRAGMENT_THRESHOLD_UP_EN, 0x1); val = REG_SET_FIELD(val, TCP_UTCL0_THRASHING_CTRL, RETRY_FRAGMENT_THRESHOLD_DOWN_EN, 0x1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_UTCL0_THRASHING_CTRL, val); } static void gfx_v12_1_xcc_enable_atomics(struct amdgpu_device *adev, int xcc_id) { uint32_t data; /* Set the TCP UTCL0 register to enable atomics */ data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_UTCL0_CNTL1); data = REG_SET_FIELD(data, TCP_UTCL0_CNTL1, ATOMIC_REQUESTER_EN, 0x1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_UTCL0_CNTL1, data); } static void gfx_v12_1_xcc_disable_burst(struct amdgpu_device *adev, int xcc_id) { WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGL1_DRAM_BURST_CTRL, 0xf); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGLARB_DRAM_BURST_CTRL, 0xf); } static void gfx_v12_1_xcc_disable_early_write_ack(struct amdgpu_device *adev, int xcc_id) { uint32_t data; data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_CNTL3); data = REG_SET_FIELD(data, TCP_CNTL3, DISABLE_EARLY_WRITE_ACK, 0x1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_CNTL3, data); } static void gfx_v12_1_xcc_disable_tcp_spill_cache(struct amdgpu_device *adev, int xcc_id) { uint32_t data; data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_CNTL); data = REG_SET_FIELD(data, TCP_CNTL, TCP_SPILL_CACHE_DISABLE, 0x1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regTCP_CNTL, data); } static void gfx_v12_1_init_golden_registers(struct amdgpu_device *adev) { int i; for (i = 0; i < NUM_XCC(adev->gfx.xcc_mask); i++) { gfx_v12_1_xcc_disable_burst(adev, i); gfx_v12_1_xcc_enable_atomics(adev, i); gfx_v12_1_xcc_setup_tcp_thrashing_ctrl(adev, i); gfx_v12_1_xcc_disable_early_write_ack(adev, i); gfx_v12_1_xcc_disable_tcp_spill_cache(adev, i); } } static int gfx_v12_1_hw_init(struct amdgpu_ip_block *ip_block) { int r, i, num_xcc; struct amdgpu_device *adev = ip_block->adev; if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) { /* rlc autoload firmware */ r = gfx_v12_1_rlc_backdoor_autoload_enable(adev); if (r) return r; } else { if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { num_xcc = NUM_XCC(adev->gfx.xcc_mask); if (adev->gfx.imu.funcs) { if (adev->gfx.imu.funcs->load_microcode) adev->gfx.imu.funcs->load_microcode(adev); } for (i = 0; i < num_xcc; i++) { /* disable gpa mode in backdoor loading */ gfx_v12_1_xcc_disable_gpa_mode(adev, i); } } } if ((adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) || (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP)) { r = gfx_v12_1_wait_for_rlc_autoload_complete(adev); if (r) { dev_err(adev->dev, "(%d) failed to wait rlc autoload complete\n", r); return r; } } adev->gfx.is_poweron = true; if (get_gb_addr_config(adev)) DRM_WARN("Invalid gb_addr_config !\n"); if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) gfx_v12_1_config_gfx_rs64(adev); r = gfx_v12_1_gfxhub_enable(adev); if (r) return r; gfx_v12_1_init_golden_registers(adev); gfx_v12_1_constants_init(adev); if (adev->nbio.funcs->gc_doorbell_init) adev->nbio.funcs->gc_doorbell_init(adev); r = gfx_v12_1_rlc_resume(adev); if (r) return r; /* * init golden registers and rlc resume may override some registers, * reconfig them here */ gfx_v12_1_tcp_harvest(adev); r = gfx_v12_1_cp_resume(adev); if (r) return r; return r; } static void gfx_v12_1_xcc_fini(struct amdgpu_device *adev, int xcc_id) { uint32_t tmp; if (!adev->no_hw_access) { if (amdgpu_gfx_disable_kcq(adev, xcc_id)) DRM_ERROR("KCQ disable failed\n"); amdgpu_mes_kiq_hw_fini(adev, xcc_id); } if (amdgpu_sriov_vf(adev)) { /* Program KIQ position of RLC_CP_SCHEDULERS during destroy */ tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS); tmp &= 0xffffff00; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp); } gfx_v12_1_xcc_cp_compute_enable(adev, false, xcc_id); gfx_v12_1_xcc_enable_gui_idle_interrupt(adev, false, xcc_id); } static int gfx_v12_1_hw_fini(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; int i, num_xcc; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { gfx_v12_1_xcc_fini(adev, i); } adev->gfxhub.funcs->gart_disable(adev); adev->gfx.is_poweron = false; return 0; } static int gfx_v12_1_suspend(struct amdgpu_ip_block *ip_block) { return gfx_v12_1_hw_fini(ip_block); } static int gfx_v12_1_resume(struct amdgpu_ip_block *ip_block) { return gfx_v12_1_hw_init(ip_block); } static bool gfx_v12_1_is_idle(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { if (REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, i), regGRBM_STATUS), GRBM_STATUS, GUI_ACTIVE)) return false; } return true; } static int gfx_v12_1_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { if (gfx_v12_1_is_idle(ip_block)) return 0; udelay(1); } return -ETIMEDOUT; } static uint64_t gfx_v12_1_get_gpu_clock_counter(struct amdgpu_device *adev) { uint64_t clock = 0; if (adev->smuio.funcs && adev->smuio.funcs->get_gpu_clock_counter) clock = adev->smuio.funcs->get_gpu_clock_counter(adev); else dev_warn(adev->dev, "query gpu clock counter is not supported\n"); return clock; } static int gfx_v12_1_early_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; adev->gfx.funcs = &gfx_v12_1_gfx_funcs; adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), AMDGPU_MAX_COMPUTE_RINGS); gfx_v12_1_set_kiq_pm4_funcs(adev); gfx_v12_1_set_ring_funcs(adev); gfx_v12_1_set_irq_funcs(adev); gfx_v12_1_set_rlc_funcs(adev); gfx_v12_1_set_mqd_funcs(adev); gfx_v12_1_set_imu_funcs(adev); gfx_v12_1_init_rlcg_reg_access_ctrl(adev); return gfx_v12_1_init_microcode(adev); } static int gfx_v12_1_late_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); if (r) return r; r = amdgpu_irq_get(adev, &adev->gfx.priv_inst_irq, 0); if (r) return r; return 0; } static bool gfx_v12_1_is_rlc_enabled(struct amdgpu_device *adev) { uint32_t rlc_cntl; /* if RLC is not enabled, do nothing */ rlc_cntl = RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CNTL); return (REG_GET_FIELD(rlc_cntl, RLC_CNTL, RLC_ENABLE_F32)) ? true : false; } static void gfx_v12_1_xcc_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; unsigned i; data = RLC_SAFE_MODE__CMD_MASK; data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, data); /* wait for RLC_SAFE_MODE */ for (i = 0; i < adev->usec_timeout; i++) { if (!REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD)) break; udelay(1); } } static void gfx_v12_1_xcc_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) { WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, RLC_SAFE_MODE__CMD_MASK); } static void gfx_v12_1_update_perf_clk(struct amdgpu_device *adev, bool enable) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) gfx_v12_1_xcc_update_perf_clk(adev, enable, i); } static void gfx_v12_1_update_spm_vmid(struct amdgpu_device *adev, int xcc_id, struct amdgpu_ring *ring, unsigned vmid) { u32 reg, data; reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_SPM_MC_CNTL); if (amdgpu_sriov_is_pp_one_vf(adev)) data = RREG32_NO_KIQ(reg); else data = RREG32(reg); data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; if (amdgpu_sriov_is_pp_one_vf(adev)) WREG32_SOC15_NO_KIQ(GC, GET_INST(GC, xcc_id), regRLC_SPM_MC_CNTL, data); else WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SPM_MC_CNTL, data); if (ring && amdgpu_sriov_is_pp_one_vf(adev) && ((ring->funcs->type == AMDGPU_RING_TYPE_GFX) || (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE))) { uint32_t reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_SPM_MC_CNTL); amdgpu_ring_emit_wreg(ring, reg, data); } } static const struct amdgpu_rlc_funcs gfx_v12_1_rlc_funcs = { .is_rlc_enabled = gfx_v12_1_is_rlc_enabled, .set_safe_mode = gfx_v12_1_xcc_set_safe_mode, .unset_safe_mode = gfx_v12_1_xcc_unset_safe_mode, .init = gfx_v12_1_rlc_init, .get_csb_size = gfx_v12_1_get_csb_size, .get_csb_buffer = gfx_v12_1_get_csb_buffer, .resume = gfx_v12_1_rlc_resume, .stop = gfx_v12_1_rlc_stop, .reset = gfx_v12_1_rlc_reset, .start = gfx_v12_1_rlc_start, .update_spm_vmid = gfx_v12_1_update_spm_vmid, }; #if 0 static void gfx_v12_cntl_power_gating(struct amdgpu_device *adev, bool enable) { /* TODO */ } static void gfx_v12_cntl_pg(struct amdgpu_device *adev, bool enable) { /* TODO */ } #endif static int gfx_v12_1_set_powergating_state(struct amdgpu_ip_block *ip_block, enum amd_powergating_state state) { struct amdgpu_device *adev = ip_block->adev; bool enable = (state == AMD_PG_STATE_GATE); if (amdgpu_sriov_vf(adev)) return 0; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(12, 1, 0): amdgpu_gfx_off_ctrl(adev, enable); break; default: break; } return 0; } static void gfx_v12_1_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev, bool enable, int xcc_id) { uint32_t def, data; if (!(adev->cg_flags & (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_GFX_3D_CGCG | AMD_CG_SUPPORT_GFX_3D_CGLS))) return; if (enable) { def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); /* unset CGCG override */ if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG) data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK; if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS) data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK; if (adev->cg_flags & AMD_CG_SUPPORT_GFX_3D_CGCG || adev->cg_flags & AMD_CG_SUPPORT_GFX_3D_CGLS) data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_GFX3D_CG_OVERRIDE_MASK; /* update CGCG override bits */ if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); /* enable cgcg FSM(0x0000363F) */ def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL); if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG) { data &= ~RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD_MASK; data |= (0x36 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) | RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK; } if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS) { data &= ~RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY_MASK; data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) | RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK; } if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data); /* set IDLE_POLL_COUNT(0x00900100) */ def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_WPTR_POLL_CNTL); data &= ~CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY_MASK; data &= ~CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT_MASK; data |= (0x0100 << CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY__SHIFT) | (0x0090 << CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT); if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_WPTR_POLL_CNTL, data); data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL); data = REG_SET_FIELD(data, CP_INT_CNTL, CNTX_BUSY_INT_ENABLE, 1); data = REG_SET_FIELD(data, CP_INT_CNTL, CNTX_EMPTY_INT_ENABLE, 1); data = REG_SET_FIELD(data, CP_INT_CNTL, CMP_BUSY_INT_ENABLE, 1); data = REG_SET_FIELD(data, CP_INT_CNTL, GFX_IDLE_INT_ENABLE, 1); WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL, data); } else { /* Program RLC_CGCG_CGLS_CTRL */ def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL); if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG) data &= ~RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK; if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS) data &= ~RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK; if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data); } } static void gfx_v12_1_xcc_update_medium_grain_clock_gating(struct amdgpu_device *adev, bool enable, int xcc_id) { uint32_t data, def; if (!(adev->cg_flags & (AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS))) return; /* It is disabled by HW by default */ if (enable) { if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG) { /* 1 - RLC_CGTT_MGCG_OVERRIDE */ def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK | RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK | RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK); if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); } } else { if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG) { def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); data |= (RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK | RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK | RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK); if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); } } } static void gfx_v12_1_xcc_update_repeater_fgcg(struct amdgpu_device *adev, bool enable, int xcc_id) { uint32_t def, data; if (!(adev->cg_flags & AMD_CG_SUPPORT_REPEATER_FGCG)) return; def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); if (enable) data &= ~(RLC_CGTT_MGCG_OVERRIDE__GFXIP_REPEATER_FGCG_OVERRIDE_MASK | RLC_CGTT_MGCG_OVERRIDE__RLC_REPEATER_FGCG_OVERRIDE_MASK); else data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_REPEATER_FGCG_OVERRIDE_MASK | RLC_CGTT_MGCG_OVERRIDE__RLC_REPEATER_FGCG_OVERRIDE_MASK; if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); } static void gfx_v12_1_xcc_update_sram_fgcg(struct amdgpu_device *adev, bool enable, int xcc_id) { uint32_t def, data; if (!(adev->cg_flags & AMD_CG_SUPPORT_GFX_FGCG)) return; def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); if (enable) data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK; else data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK; if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); } static void gfx_v12_1_xcc_update_perf_clk(struct amdgpu_device *adev, bool enable, int xcc_id) { uint32_t def, data; if (!(adev->cg_flags & AMD_CG_SUPPORT_GFX_PERF_CLK)) return; def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); if (enable) data &= ~RLC_CGTT_MGCG_OVERRIDE__PERFMON_CLOCK_STATE_MASK; else data |= RLC_CGTT_MGCG_OVERRIDE__PERFMON_CLOCK_STATE_MASK; if (def != data) WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); } static int gfx_v12_1_xcc_update_gfx_clock_gating(struct amdgpu_device *adev, bool enable, int xcc_id) { amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id); gfx_v12_1_xcc_update_coarse_grain_clock_gating(adev, enable, xcc_id); gfx_v12_1_xcc_update_medium_grain_clock_gating(adev, enable, xcc_id); gfx_v12_1_xcc_update_repeater_fgcg(adev, enable, xcc_id); gfx_v12_1_xcc_update_sram_fgcg(adev, enable, xcc_id); gfx_v12_1_xcc_update_perf_clk(adev, enable, xcc_id); if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_3D_CGCG | AMD_CG_SUPPORT_GFX_3D_CGLS)) gfx_v12_1_xcc_enable_gui_idle_interrupt(adev, enable, xcc_id); amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id); return 0; } static int gfx_v12_1_set_clockgating_state(struct amdgpu_ip_block *ip_block, enum amd_clockgating_state state) { struct amdgpu_device *adev = ip_block->adev; int i, num_xcc; if (amdgpu_sriov_vf(adev)) return 0; num_xcc = NUM_XCC(adev->gfx.xcc_mask); switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(12, 1, 0): for (i = 0; i < num_xcc; i++) gfx_v12_1_xcc_update_gfx_clock_gating(adev, state == AMD_CG_STATE_GATE, i); break; default: break; } return 0; } static void gfx_v12_1_get_clockgating_state(struct amdgpu_ip_block *ip_block, u64 *flags) { struct amdgpu_device *adev = ip_block->adev; int data; /* AMD_CG_SUPPORT_GFX_MGCG */ data = RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CGTT_MGCG_OVERRIDE); if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK)) *flags |= AMD_CG_SUPPORT_GFX_MGCG; /* AMD_CG_SUPPORT_REPEATER_FGCG */ if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_REPEATER_FGCG_OVERRIDE_MASK)) *flags |= AMD_CG_SUPPORT_REPEATER_FGCG; /* AMD_CG_SUPPORT_GFX_FGCG */ if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK)) *flags |= AMD_CG_SUPPORT_GFX_FGCG; /* AMD_CG_SUPPORT_GFX_PERF_CLK */ if (!(data & RLC_CGTT_MGCG_OVERRIDE__PERFMON_CLOCK_STATE_MASK)) *flags |= AMD_CG_SUPPORT_GFX_PERF_CLK; /* AMD_CG_SUPPORT_GFX_CGCG */ data = RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CGCG_CGLS_CTRL); if (data & RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK) *flags |= AMD_CG_SUPPORT_GFX_CGCG; /* AMD_CG_SUPPORT_GFX_CGLS */ if (data & RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK) *flags |= AMD_CG_SUPPORT_GFX_CGLS; } static u64 gfx_v12_1_ring_get_rptr_compute(struct amdgpu_ring *ring) { /* gfx12 hardware is 32bit rptr */ return *(uint32_t *)ring->rptr_cpu_addr; } static u64 gfx_v12_1_ring_get_wptr_compute(struct amdgpu_ring *ring) { u64 wptr; /* XXX check if swapping is necessary on BE */ if (ring->use_doorbell) wptr = atomic64_read((atomic64_t *)ring->wptr_cpu_addr); else BUG(); return wptr; } static void gfx_v12_1_ring_set_wptr_compute(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; /* XXX check if swapping is necessary on BE */ if (ring->use_doorbell) { atomic64_set((atomic64_t *)ring->wptr_cpu_addr, ring->wptr); WDOORBELL64(ring->doorbell_index, ring->wptr); } else { BUG(); /* only DOORBELL method supported on gfx12 now */ } } static void gfx_v12_1_ring_emit_ib_compute(struct amdgpu_ring *ring, struct amdgpu_job *job, struct amdgpu_ib *ib, uint32_t flags) { unsigned vmid = AMDGPU_JOB_GET_VMID(job); u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24); /* Currently, there is a high possibility to get wave ID mismatch * between ME and GDS, leading to a hw deadlock, because ME generates * different wave IDs than the GDS expects. This situation happens * randomly when at least 5 compute pipes use GDS ordered append. * The wave IDs generated by ME are also wrong after suspend/resume. * Those are probably bugs somewhere else in the kernel driver. * * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and * GDS to 0 for this ring (me/pipe). */ if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) { amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); amdgpu_ring_write(ring, regGDS_COMPUTE_MAX_WAVE_ID); } amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2)); BUG_ON(ib->gpu_addr & 0x3); /* Dword align */ amdgpu_ring_write(ring, #ifdef __BIG_ENDIAN (2 << 0) | #endif lower_32_bits(ib->gpu_addr)); amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr)); amdgpu_ring_write(ring, control); } static void gfx_v12_1_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) { bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT; bool int_sel = flags & AMDGPU_FENCE_FLAG_INT; /* RELEASE_MEM - flush caches, send int */ amdgpu_ring_write(ring, PACKET3(PACKET3_RELEASE_MEM, 6)); amdgpu_ring_write(ring, (PACKET3_RELEASE_MEM_GCR_SEQ(1) | PACKET3_RELEASE_MEM_GCR_GLV_WB | PACKET3_RELEASE_MEM_GCR_GL2_WB | PACKET3_RELEASE_MEM_GCR_GL2_SCOPE(2) | PACKET3_RELEASE_MEM_TEMPORAL(3) | PACKET3_RELEASE_MEM_EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | PACKET3_RELEASE_MEM_EVENT_INDEX(5))); amdgpu_ring_write(ring, (PACKET3_RELEASE_MEM_DATA_SEL(write64bit ? 2 : 1) | PACKET3_RELEASE_MEM_INT_SEL(int_sel ? 2 : 0))); /* * the address should be Qword aligned if 64bit write, Dword * aligned if only send 32bit data low (discard data high) */ if (write64bit) BUG_ON(addr & 0x7); else BUG_ON(addr & 0x3); amdgpu_ring_write(ring, lower_32_bits(addr)); amdgpu_ring_write(ring, upper_32_bits(addr)); amdgpu_ring_write(ring, lower_32_bits(seq)); amdgpu_ring_write(ring, upper_32_bits(seq)); amdgpu_ring_write(ring, 0); } static void gfx_v12_1_ring_emit_pipeline_sync(struct amdgpu_ring *ring) { int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX); uint32_t seq = ring->fence_drv.sync_seq; uint64_t addr = ring->fence_drv.gpu_addr; gfx_v12_1_wait_reg_mem(ring, usepfp, 1, 0, lower_32_bits(addr), upper_32_bits(addr), seq, 0xffffffff, 4); } static void gfx_v12_1_ring_invalidate_tlbs(struct amdgpu_ring *ring, uint16_t pasid, uint32_t flush_type, bool all_hub, uint8_t dst_sel) { amdgpu_ring_write(ring, PACKET3(PACKET3_INVALIDATE_TLBS, 0)); amdgpu_ring_write(ring, PACKET3_INVALIDATE_TLBS_DST_SEL(dst_sel) | PACKET3_INVALIDATE_TLBS_ALL_HUB(all_hub) | PACKET3_INVALIDATE_TLBS_PASID(pasid) | PACKET3_INVALIDATE_TLBS_FLUSH_TYPE(flush_type)); } static void gfx_v12_1_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr) { amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr); /* compute doesn't have PFP */ if (ring->funcs->type == AMDGPU_RING_TYPE_GFX) { /* sync PFP to ME, otherwise we might get invalid PFP reads */ amdgpu_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0)); amdgpu_ring_write(ring, 0x0); } } static void gfx_v12_1_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned int flags) { struct amdgpu_device *adev = ring->adev; /* we only allocate 32bit for each seq wb address */ BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT); /* write fence seq to the "addr" */ amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | WRITE_DATA_DST_SEL(5) | WR_CONFIRM)); amdgpu_ring_write(ring, lower_32_bits(addr)); amdgpu_ring_write(ring, upper_32_bits(addr)); amdgpu_ring_write(ring, lower_32_bits(seq)); if (flags & AMDGPU_FENCE_FLAG_INT) { /* set register to trigger INT */ 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, SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regCPC_INT_STATUS)); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, 0x20000000); /* src_id is 178 */ } } static void gfx_v12_1_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t reg_val_offs) { struct amdgpu_device *adev = ring->adev; reg = soc_v1_0_normalize_xcc_reg_offset(reg); amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); amdgpu_ring_write(ring, 0 | /* src: register*/ (5 << 8) | /* dst: memory */ (1 << 20)); /* write confirm */ amdgpu_ring_write(ring, reg); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + reg_val_offs * 4)); amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + reg_val_offs * 4)); } static void gfx_v12_1_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val) { uint32_t cmd = 0; reg = soc_v1_0_normalize_xcc_reg_offset(reg); switch (ring->funcs->type) { 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, cmd); amdgpu_ring_write(ring, reg); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, val); } static void gfx_v12_1_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, uint32_t val, uint32_t mask) { gfx_v12_1_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20); } static void gfx_v12_1_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask) { int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX); gfx_v12_1_wait_reg_mem(ring, usepfp, 0, 1, reg0, reg1, ref, mask, 0x20); } static void gfx_v12_1_xcc_set_compute_eop_interrupt_state(struct amdgpu_device *adev, int me, int pipe, enum amdgpu_interrupt_state state, int xcc_id) { u32 mec_int_cntl, mec_int_cntl_reg; /* * amdgpu controls only the first MEC. That's why this function only * handles the setting of interrupts for this specific MEC. All other * pipes' interrupts are set by amdkfd. */ if (me == 1) { switch (pipe) { case 0: mec_int_cntl_reg = SOC15_REG_OFFSET( GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE0_INT_CNTL); break; case 1: mec_int_cntl_reg = SOC15_REG_OFFSET( GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE1_INT_CNTL); break; case 2: mec_int_cntl_reg = SOC15_REG_OFFSET( GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE2_INT_CNTL); break; case 3: mec_int_cntl_reg = SOC15_REG_OFFSET( GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE3_INT_CNTL); break; default: DRM_DEBUG("invalid pipe %d\n", pipe); return; } } else { DRM_DEBUG("invalid me %d\n", me); return; } switch (state) { case AMDGPU_IRQ_STATE_DISABLE: mec_int_cntl = RREG32_XCC(mec_int_cntl_reg, xcc_id); mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, TIME_STAMP_INT_ENABLE, 0); mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, GENERIC0_INT_ENABLE, 0); WREG32_XCC(mec_int_cntl_reg, mec_int_cntl, xcc_id); break; case AMDGPU_IRQ_STATE_ENABLE: mec_int_cntl = RREG32_XCC(mec_int_cntl_reg, xcc_id); mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, TIME_STAMP_INT_ENABLE, 1); mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, GENERIC0_INT_ENABLE, 1); WREG32_XCC(mec_int_cntl_reg, mec_int_cntl, xcc_id); break; default: break; } } static int gfx_v12_1_set_eop_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned type, enum amdgpu_interrupt_state state) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { switch (type) { case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP: gfx_v12_1_xcc_set_compute_eop_interrupt_state( adev, 1, 0, state, i); break; case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP: gfx_v12_1_xcc_set_compute_eop_interrupt_state( adev, 1, 1, state, i); break; case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP: gfx_v12_1_xcc_set_compute_eop_interrupt_state( adev, 1, 2, state, i); break; case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP: gfx_v12_1_xcc_set_compute_eop_interrupt_state( adev, 1, 3, state, i); break; default: break; } } return 0; } static int gfx_v12_1_eop_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { u32 doorbell_offset = entry->src_data[0]; u8 me_id, pipe_id, queue_id; struct amdgpu_ring *ring; int i, xcc_id; DRM_DEBUG("IH: CP EOP\n"); if (adev->enable_mes && doorbell_offset) { struct amdgpu_userq_fence_driver *fence_drv = NULL; struct xarray *xa = &adev->userq_xa; unsigned long flags; xa_lock_irqsave(xa, flags); fence_drv = xa_load(xa, doorbell_offset); if (fence_drv) amdgpu_userq_fence_driver_process(fence_drv); xa_unlock_irqrestore(xa, flags); } else { me_id = (entry->ring_id & 0x0c) >> 2; pipe_id = (entry->ring_id & 0x03) >> 0; queue_id = (entry->ring_id & 0x70) >> 4; xcc_id = gfx_v12_1_ih_to_xcc_inst(adev, entry->node_id); if (xcc_id == -EINVAL) return -EINVAL; switch (me_id) { case 0: if (pipe_id == 0) amdgpu_fence_process(&adev->gfx.gfx_ring[0]); else amdgpu_fence_process(&adev->gfx.gfx_ring[1]); break; case 1: case 2: for (i = 0; i < adev->gfx.num_compute_rings; i++) { ring = &adev->gfx.compute_ring [i + xcc_id * adev->gfx.num_compute_rings]; /* Per-queue interrupt is supported for MEC starting from VI. * The interrupt can only be enabled/disabled per pipe instead * of per queue. */ if ((ring->me == me_id) && (ring->pipe == pipe_id) && (ring->queue == queue_id)) amdgpu_fence_process(ring); } break; } } return 0; } static int gfx_v12_1_set_priv_reg_fault_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, unsigned type, enum amdgpu_interrupt_state state) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); switch (state) { case AMDGPU_IRQ_STATE_DISABLE: case AMDGPU_IRQ_STATE_ENABLE: for (i = 0; i < num_xcc; i++) WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), CP_INT_CNTL_RING0, PRIV_REG_INT_ENABLE, state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); break; default: break; } return 0; } static int gfx_v12_1_set_priv_inst_fault_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, unsigned type, enum amdgpu_interrupt_state state) { int i, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); switch (state) { case AMDGPU_IRQ_STATE_DISABLE: case AMDGPU_IRQ_STATE_ENABLE: for (i = 0; i < num_xcc; i++) WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), CP_INT_CNTL_RING0, PRIV_INSTR_INT_ENABLE, state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); break; default: break; } return 0; } static void gfx_v12_1_handle_priv_fault(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { u8 me_id, pipe_id, queue_id; struct amdgpu_ring *ring; int i, xcc_id; me_id = (entry->ring_id & 0x0c) >> 2; pipe_id = (entry->ring_id & 0x03) >> 0; queue_id = (entry->ring_id & 0x70) >> 4; xcc_id = gfx_v12_1_ih_to_xcc_inst(adev, entry->node_id); if (xcc_id == -EINVAL) return; switch (me_id) { case 0: for (i = 0; i < adev->gfx.num_gfx_rings; i++) { ring = &adev->gfx.gfx_ring[i]; /* we only enabled 1 gfx queue per pipe for now */ if (ring->me == me_id && ring->pipe == pipe_id) drm_sched_fault(&ring->sched); } break; case 1: case 2: for (i = 0; i < adev->gfx.num_compute_rings; i++) { ring = &adev->gfx.compute_ring [i + xcc_id * adev->gfx.num_compute_rings]; if (ring->me == me_id && ring->pipe == pipe_id && ring->queue == queue_id) drm_sched_fault(&ring->sched); } break; default: BUG(); break; } } static int gfx_v12_1_priv_reg_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { DRM_ERROR("Illegal register access in command stream\n"); gfx_v12_1_handle_priv_fault(adev, entry); return 0; } static int gfx_v12_1_priv_inst_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { DRM_ERROR("Illegal instruction in command stream\n"); gfx_v12_1_handle_priv_fault(adev, entry); return 0; } static void gfx_v12_1_emit_mem_sync(struct amdgpu_ring *ring) { const unsigned int gcr_cntl = PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_INV(1) | PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_WB(1) | PACKET3_ACQUIRE_MEM_GCR_CNTL_GLV_INV(1) | PACKET3_ACQUIRE_MEM_GCR_CNTL_GLK_INV(1) | PACKET3_ACQUIRE_MEM_GCR_CNTL_GLI_INV(1) | PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_SCOPE(2); /* ACQUIRE_MEM - make one or more surfaces valid for use by the subsequent operations */ amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 6)); amdgpu_ring_write(ring, 0); /* CP_COHER_CNTL */ amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ amdgpu_ring_write(ring, 0xffffff); /* CP_COHER_SIZE_HI */ amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ amdgpu_ring_write(ring, 0); /* CP_COHER_BASE_HI */ amdgpu_ring_write(ring, 0x0000000A); /* POLL_INTERVAL */ amdgpu_ring_write(ring, gcr_cntl); /* GCR_CNTL */ } static const struct amd_ip_funcs gfx_v12_1_ip_funcs = { .name = "gfx_v12_1", .early_init = gfx_v12_1_early_init, .late_init = gfx_v12_1_late_init, .sw_init = gfx_v12_1_sw_init, .sw_fini = gfx_v12_1_sw_fini, .hw_init = gfx_v12_1_hw_init, .hw_fini = gfx_v12_1_hw_fini, .suspend = gfx_v12_1_suspend, .resume = gfx_v12_1_resume, .is_idle = gfx_v12_1_is_idle, .wait_for_idle = gfx_v12_1_wait_for_idle, .set_clockgating_state = gfx_v12_1_set_clockgating_state, .set_powergating_state = gfx_v12_1_set_powergating_state, .get_clockgating_state = gfx_v12_1_get_clockgating_state, }; static const struct amdgpu_ring_funcs gfx_v12_1_ring_funcs_compute = { .type = AMDGPU_RING_TYPE_COMPUTE, .align_mask = 0xff, .nop = PACKET3(PACKET3_NOP, 0x3FFF), .support_64bit_ptrs = true, .get_rptr = gfx_v12_1_ring_get_rptr_compute, .get_wptr = gfx_v12_1_ring_get_wptr_compute, .set_wptr = gfx_v12_1_ring_set_wptr_compute, .emit_frame_size = 7 + /* gfx_v12_1_ring_emit_pipeline_sync */ SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v12_1_ring_emit_vm_flush */ 8 + 8 + 8 + /* gfx_v12_1_ring_emit_fence x3 for user fence, vm fence */ 8, /* gfx_v12_1_emit_mem_sync */ .emit_ib_size = 7, /* gfx_v12_1_ring_emit_ib_compute */ .emit_ib = gfx_v12_1_ring_emit_ib_compute, .emit_fence = gfx_v12_1_ring_emit_fence, .emit_pipeline_sync = gfx_v12_1_ring_emit_pipeline_sync, .emit_vm_flush = gfx_v12_1_ring_emit_vm_flush, .test_ring = gfx_v12_1_ring_test_ring, .test_ib = gfx_v12_1_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .emit_wreg = gfx_v12_1_ring_emit_wreg, .emit_reg_wait = gfx_v12_1_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v12_1_ring_emit_reg_write_reg_wait, .emit_mem_sync = gfx_v12_1_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v12_1_ring_funcs_kiq = { .type = AMDGPU_RING_TYPE_KIQ, .align_mask = 0xff, .nop = PACKET3(PACKET3_NOP, 0x3FFF), .support_64bit_ptrs = true, .get_rptr = gfx_v12_1_ring_get_rptr_compute, .get_wptr = gfx_v12_1_ring_get_wptr_compute, .set_wptr = gfx_v12_1_ring_set_wptr_compute, .emit_frame_size = 7 + /* gfx_v12_1_ring_emit_pipeline_sync */ SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v12_1_ring_emit_vm_flush */ 8 + 8 + 8, /* gfx_v12_1_ring_emit_fence_kiq x3 for user fence, vm fence */ .emit_ib_size = 7, /* gfx_v12_1_ring_emit_ib_compute */ .emit_ib = gfx_v12_1_ring_emit_ib_compute, .emit_fence = gfx_v12_1_ring_emit_fence_kiq, .test_ring = gfx_v12_1_ring_test_ring, .test_ib = gfx_v12_1_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .emit_rreg = gfx_v12_1_ring_emit_rreg, .emit_wreg = gfx_v12_1_ring_emit_wreg, .emit_reg_wait = gfx_v12_1_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v12_1_ring_emit_reg_write_reg_wait, }; static void gfx_v12_1_set_ring_funcs(struct amdgpu_device *adev) { int i, j, num_xcc; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { adev->gfx.kiq[i].ring.funcs = &gfx_v12_1_ring_funcs_kiq; for (j = 0; j < adev->gfx.num_compute_rings; j++) adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings].funcs = &gfx_v12_1_ring_funcs_compute; } } static const struct amdgpu_irq_src_funcs gfx_v12_1_eop_irq_funcs = { .set = gfx_v12_1_set_eop_interrupt_state, .process = gfx_v12_1_eop_irq, }; static const struct amdgpu_irq_src_funcs gfx_v12_1_priv_reg_irq_funcs = { .set = gfx_v12_1_set_priv_reg_fault_state, .process = gfx_v12_1_priv_reg_irq, }; static const struct amdgpu_irq_src_funcs gfx_v12_1_priv_inst_irq_funcs = { .set = gfx_v12_1_set_priv_inst_fault_state, .process = gfx_v12_1_priv_inst_irq, }; static void gfx_v12_1_set_irq_funcs(struct amdgpu_device *adev) { adev->gfx.eop_irq.num_types = AMDGPU_CP_IRQ_LAST; adev->gfx.eop_irq.funcs = &gfx_v12_1_eop_irq_funcs; adev->gfx.priv_reg_irq.num_types = 1; adev->gfx.priv_reg_irq.funcs = &gfx_v12_1_priv_reg_irq_funcs; adev->gfx.priv_inst_irq.num_types = 1; adev->gfx.priv_inst_irq.funcs = &gfx_v12_1_priv_inst_irq_funcs; } static void gfx_v12_1_set_imu_funcs(struct amdgpu_device *adev) { if (adev->flags & AMD_IS_APU) adev->gfx.imu.mode = MISSION_MODE; else adev->gfx.imu.mode = DEBUG_MODE; if (!amdgpu_sriov_vf(adev)) adev->gfx.imu.funcs = &gfx_v12_1_imu_funcs; } static void gfx_v12_1_set_rlc_funcs(struct amdgpu_device *adev) { adev->gfx.rlc.funcs = &gfx_v12_1_rlc_funcs; } static void gfx_v12_1_set_mqd_funcs(struct amdgpu_device *adev) { /* set compute eng mqd */ adev->mqds[AMDGPU_HW_IP_COMPUTE].mqd_size = sizeof(struct v12_1_compute_mqd); adev->mqds[AMDGPU_HW_IP_COMPUTE].init_mqd = gfx_v12_1_compute_mqd_init; } static void gfx_v12_1_set_user_cu_inactive_bitmap_per_sh(struct amdgpu_device *adev, u32 bitmap, int xcc_id) { u32 data; if (!bitmap) return; data = bitmap << GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_WGPS__SHIFT; data &= GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_WGPS_MASK; WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGC_USER_SHADER_ARRAY_CONFIG, data); } static u32 gfx_v12_1_get_cu_active_bitmap_per_sh(struct amdgpu_device *adev, int xcc_id) { u32 data, mask; data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCC_GC_SHADER_ARRAY_CONFIG); data |= RREG32_SOC15(GC, GET_INST(GC, xcc_id), regGC_USER_SHADER_ARRAY_CONFIG); data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_WGPS_MASK; data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_WGPS__SHIFT; mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_cu_per_sh); return (~data) & mask; } static int gfx_v12_1_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info) { int i, j, k, counter, xcc_id, active_cu_number = 0; u32 mask, bitmap; unsigned int disable_masks[2 * 2]; if (!adev || !cu_info) return -EINVAL; if (adev->gfx.config.max_shader_engines > 2 || adev->gfx.config.max_sh_per_se > 2) { dev_err(adev->dev, "Max SE (%d) and Max SA per SE (%d) is greater than expected\n", adev->gfx.config.max_shader_engines, adev->gfx.config.max_sh_per_se); return -EINVAL; } amdgpu_gfx_parse_disable_cu(adev, disable_masks, adev->gfx.config.max_shader_engines, adev->gfx.config.max_sh_per_se); mutex_lock(&adev->grbm_idx_mutex); for (xcc_id = 0; xcc_id < NUM_XCC(adev->gfx.xcc_mask); xcc_id++) { for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { bitmap = i * adev->gfx.config.max_sh_per_se + j; if (!((gfx_v12_1_get_sa_active_bitmap(adev, xcc_id) >> bitmap) & 1)) continue; mask = 1; counter = 0; gfx_v12_1_xcc_select_se_sh(adev, i, j, 0xffffffff, xcc_id); gfx_v12_1_set_user_cu_inactive_bitmap_per_sh( adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j], xcc_id); bitmap = gfx_v12_1_get_cu_active_bitmap_per_sh(adev, xcc_id); cu_info->bitmap[xcc_id][i][j] = bitmap; for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { if (bitmap & mask) counter++; mask <<= 1; } active_cu_number += counter; } } gfx_v12_1_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, xcc_id); } mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; cu_info->simd_per_cu = NUM_SIMD_PER_CU_GFX12_1; cu_info->lds_size = 320; return 0; } const struct amdgpu_ip_block_version gfx_v12_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_GFX, .major = 12, .minor = 1, .rev = 0, .funcs = &gfx_v12_1_ip_funcs, }; static int gfx_v12_1_xcp_resume(void *handle, uint32_t inst_mask) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; uint32_t tmp_mask; int i, r; /* TODO : Initialize golden regs */ /* gfx_v12_1_init_golden_registers(adev); */ tmp_mask = inst_mask; for_each_inst(i, tmp_mask) gfx_v12_1_xcc_constants_init(adev, i); if (!amdgpu_sriov_vf(adev)) { tmp_mask = inst_mask; for_each_inst(i, tmp_mask) { r = gfx_v12_1_xcc_rlc_resume(adev, i); if (r) return r; } } r = gfx_v12_1_xcc_cp_resume(adev, inst_mask); return r; } static int gfx_v12_1_xcp_suspend(void *handle, uint32_t inst_mask) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; int i; for_each_inst(i, inst_mask) gfx_v12_1_xcc_fini(adev, i); return 0; } struct amdgpu_xcp_ip_funcs gfx_v12_1_xcp_funcs = { .suspend = &gfx_v12_1_xcp_suspend, .resume = &gfx_v12_1_xcp_resume };