diff options
Diffstat (limited to 'drivers/gpu')
55 files changed, 623 insertions, 251 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 5e53a5293935..0d939f07d506 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -703,6 +703,13 @@ err: void amdgpu_amdkfd_set_compute_idle(struct amdgpu_device *adev, bool idle) { + /* Temporary workaround to fix issues observed in some + * compute applications when GFXOFF is enabled on GFX11. + */ + if (IP_VERSION_MAJ(adev->ip_versions[GC_HWIP][0]) == 11) { + pr_debug("GFXOFF is %s\n", idle ? "enabled" : "disabled"); + amdgpu_gfx_off_ctrl(adev, idle); + } amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_COMPUTE, !idle); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c index 0b0a72ca5695..7e80caa05060 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c @@ -111,7 +111,7 @@ static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id) lock_srbm(adev, mec, pipe, 0, 0); - WREG32(SOC15_REG_OFFSET(GC, 0, regCPC_INT_CNTL), + WREG32_SOC15(GC, 0, regCPC_INT_CNTL, CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK | CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 2170db83e41d..93ad00453f4b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -509,13 +509,13 @@ kfd_mem_dmamap_userptr(struct kgd_mem *mem, struct ttm_tt *ttm = bo->tbo.ttm; int ret; + if (WARN_ON(ttm->num_pages != src_ttm->num_pages)) + return -EINVAL; + ttm->sg = kmalloc(sizeof(*ttm->sg), GFP_KERNEL); if (unlikely(!ttm->sg)) return -ENOMEM; - if (WARN_ON(ttm->num_pages != src_ttm->num_pages)) - return -EINVAL; - /* Same sequence as in amdgpu_ttm_tt_pin_userptr */ ret = sg_alloc_table_from_pages(ttm->sg, src_ttm->pages, ttm->num_pages, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 8ee4e8491f39..4048ba1b661e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -327,7 +327,10 @@ static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority, if (r) return r; - ctx->stable_pstate = current_stable_pstate; + if (mgr->adev->pm.stable_pstate_ctx) + ctx->stable_pstate = mgr->adev->pm.stable_pstate_ctx->stable_pstate; + else + ctx->stable_pstate = current_stable_pstate; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 25e1f5ed7ead..9170aeaad93e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2926,6 +2926,14 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev) amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); + /* + * Per PMFW team's suggestion, driver needs to handle gfxoff + * and df cstate features disablement for gpu reset(e.g. Mode1Reset) + * scenario. Add the missing df cstate disablement here. + */ + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + dev_warn(adev->dev, "Failed to disallow df cstate"); + for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.valid) continue; @@ -3200,6 +3208,15 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev) return r; } adev->ip_blocks[i].status.hw = true; + + if (adev->in_s0ix && adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { + /* disable gfxoff for IP resume. The gfxoff will be re-enabled in + * amdgpu_device_resume() after IP resume. + */ + amdgpu_gfx_off_ctrl(adev, false); + DRM_DEBUG("will disable gfxoff for re-initializing other blocks\n"); + } + } return 0; @@ -4172,6 +4189,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) /* Make sure IB tests flushed */ flush_delayed_work(&adev->delayed_init_work); + if (adev->in_s0ix) { + /* re-enable gfxoff after IP resume. This re-enables gfxoff after + * it was disabled for IP resume in amdgpu_device_ip_resume_phase2(). + */ + amdgpu_gfx_off_ctrl(adev, true); + DRM_DEBUG("will enable gfxoff for the mission mode\n"); + } if (fbcon) drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, false); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 134575a3893c..9e6c23266a1a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -424,8 +424,9 @@ error: static bool amdgpu_mem_visible(struct amdgpu_device *adev, struct ttm_resource *mem) { - uint64_t mem_size = (u64)mem->num_pages << PAGE_SHIFT; + u64 mem_size = (u64)mem->num_pages << PAGE_SHIFT; struct amdgpu_res_cursor cursor; + u64 end; if (mem->mem_type == TTM_PL_SYSTEM || mem->mem_type == TTM_PL_TT) @@ -434,12 +435,21 @@ static bool amdgpu_mem_visible(struct amdgpu_device *adev, return false; amdgpu_res_first(mem, 0, mem_size, &cursor); + end = cursor.start + cursor.size; + while (cursor.remaining) { + amdgpu_res_next(&cursor, cursor.size); - /* ttm_resource_ioremap only supports contiguous memory */ - if (cursor.size != mem_size) - return false; + if (!cursor.remaining) + break; + + /* ttm_resource_ioremap only supports contiguous memory */ + if (end != cursor.start) + return false; + + end = cursor.start + cursor.size; + } - return cursor.start + cursor.size <= adev->gmc.visible_vram_size; + return end <= adev->gmc.visible_vram_size; } /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 9be57389301b..af5aeb0ec2e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -726,6 +726,12 @@ void amdgpu_detect_virtualization(struct amdgpu_device *adev) adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; } + if (amdgpu_sriov_vf(adev) && adev->asic_type == CHIP_SIENNA_CICHLID) + /* VF MMIO access (except mailbox range) from CPU + * will be blocked during sriov runtime + */ + adev->virt.caps |= AMDGPU_VF_MMIO_ACCESS_PROTECT; + /* we have the ability to check now */ if (amdgpu_sriov_vf(adev)) { switch (adev->asic_type) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 239f232f9c02..617d072275eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -31,6 +31,7 @@ #define AMDGPU_SRIOV_CAPS_IS_VF (1 << 2) /* this GPU is a virtual function */ #define AMDGPU_PASSTHROUGH_MODE (1 << 3) /* thw whole GPU is pass through for VM */ #define AMDGPU_SRIOV_CAPS_RUNTIME (1 << 4) /* is out of full access mode */ +#define AMDGPU_VF_MMIO_ACCESS_PROTECT (1 << 5) /* MMIO write access is not allowed in sriov runtime */ /* flags for indirect register access path supported by rlcg for sriov */ #define AMDGPU_RLCG_GC_WRITE_LEGACY (0x8 << 28) @@ -294,6 +295,9 @@ struct amdgpu_video_codec_info; #define amdgpu_passthrough(adev) \ ((adev)->virt.caps & AMDGPU_PASSTHROUGH_MODE) +#define amdgpu_sriov_vf_mmio_access_protection(adev) \ +((adev)->virt.caps & AMDGPU_VF_MMIO_ACCESS_PROTECT) + static inline bool is_virtual_machine(void) { #if defined(CONFIG_X86) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 690fd4f639f1..04130f8813ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -2301,7 +2301,11 @@ void amdgpu_vm_manager_init(struct amdgpu_device *adev) */ #ifdef CONFIG_X86_64 if (amdgpu_vm_update_mode == -1) { - if (amdgpu_gmc_vram_full_visible(&adev->gmc)) + /* For asic with VF MMIO access protection + * avoid using CPU for VM table updates + */ + if (amdgpu_gmc_vram_full_visible(&adev->gmc) && + !amdgpu_sriov_vf_mmio_access_protection(adev)) adev->vm_manager.vm_update_mode = AMDGPU_VM_USE_CPU_FOR_COMPUTE; else diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index daf8ba8235cd..03775e0a8100 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -1729,7 +1729,7 @@ static void gfx_v11_0_init_compute_vmid(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, regSH_MEM_BASES, sh_mem_bases); /* Enable trap for each kfd vmid. */ - data = RREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_PER_VMID_CNTL)); + data = RREG32_SOC15(GC, 0, regSPI_GDBG_PER_VMID_CNTL); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); } soc21_grbm_select(adev, 0, 0, 0, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index 1471bfb9ae38..2475fdbe8010 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -185,6 +185,10 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, /* Use register 17 for GART */ const unsigned eng = 17; unsigned int i; + unsigned char hub_ip = 0; + + hub_ip = (vmhub == AMDGPU_GFXHUB_0) ? + GC_HWIP : MMHUB_HWIP; spin_lock(&adev->gmc.invalidate_lock); /* @@ -198,8 +202,8 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, if (use_semaphore) { for (i = 0; i < adev->usec_timeout; i++) { /* a read return value of 1 means semaphore acuqire */ - tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_sem + - hub->eng_distance * eng); + tmp = RREG32_RLC_NO_KIQ(hub->vm_inv_eng0_sem + + hub->eng_distance * eng, hub_ip); if (tmp & 0x1) break; udelay(1); @@ -209,12 +213,12 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, DRM_ERROR("Timeout waiting for sem acquire in VM flush!\n"); } - WREG32_NO_KIQ(hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req); + WREG32_RLC_NO_KIQ(hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req, hub_ip); /* Wait for ACK with a delay.*/ for (i = 0; i < adev->usec_timeout; i++) { - tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + - hub->eng_distance * eng); + tmp = RREG32_RLC_NO_KIQ(hub->vm_inv_eng0_ack + + hub->eng_distance * eng, hub_ip); tmp &= 1 << vmid; if (tmp) break; @@ -228,8 +232,8 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, * add semaphore release after invalidation, * write with 0 means semaphore release */ - WREG32_NO_KIQ(hub->vm_inv_eng0_sem + - hub->eng_distance * eng, 0); + WREG32_RLC_NO_KIQ(hub->vm_inv_eng0_sem + + hub->eng_distance * eng, 0, hub_ip); /* Issue additional private vm invalidation to MMHUB */ if ((vmhub != AMDGPU_GFXHUB_0) && diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index f92744b8d79d..2dd827472d6e 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -1145,6 +1145,42 @@ static int mes_v11_0_sw_fini(void *handle) return 0; } +static void mes_v11_0_kiq_dequeue_sched(struct amdgpu_device *adev) +{ + uint32_t data; + int i; + + mutex_lock(&adev->srbm_mutex); + soc21_grbm_select(adev, 3, AMDGPU_MES_SCHED_PIPE, 0, 0); + + /* disable the queue if it's active */ + if (RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1) { + WREG32_SOC15(GC, 0, regCP_HQD_DEQUEUE_REQUEST, 1); + for (i = 0; i < adev->usec_timeout; i++) { + if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1)) + break; + udelay(1); + } + } + data = RREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL); + data = REG_SET_FIELD(data, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_EN, 0); + data = REG_SET_FIELD(data, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_HIT, 1); + WREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, data); + + WREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, 0); + + WREG32_SOC15(GC, 0, regCP_HQD_PQ_WPTR_LO, 0); + WREG32_SOC15(GC, 0, regCP_HQD_PQ_WPTR_HI, 0); + WREG32_SOC15(GC, 0, regCP_HQD_PQ_RPTR, 0); + + soc21_grbm_select(adev, 0, 0, 0, 0); + mutex_unlock(&adev->srbm_mutex); + + adev->mes.ring.sched.ready = false; +} + static void mes_v11_0_kiq_setting(struct amdgpu_ring *ring) { uint32_t tmp; @@ -1196,6 +1232,9 @@ failure: static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev) { + if (adev->mes.ring.sched.ready) + mes_v11_0_kiq_dequeue_sched(adev); + mes_v11_0_enable(adev, false); return 0; } @@ -1251,9 +1290,6 @@ failure: static int mes_v11_0_hw_fini(void *handle) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - adev->mes.ring.sched.ready = false; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index 4d304f22889e..998b5d17b271 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -32,8 +32,6 @@ #include "gc/gc_10_1_0_offset.h" #include "soc15_common.h" -#define mmMM_ATC_L2_MISC_CG_Sienna_Cichlid 0x064d -#define mmMM_ATC_L2_MISC_CG_Sienna_Cichlid_BASE_IDX 0 #define mmDAGB0_CNTL_MISC2_Sienna_Cichlid 0x0070 #define mmDAGB0_CNTL_MISC2_Sienna_Cichlid_BASE_IDX 0 @@ -574,7 +572,6 @@ static void mmhub_v2_0_update_medium_grain_clock_gating(struct amdgpu_device *ad case IP_VERSION(2, 1, 0): case IP_VERSION(2, 1, 1): case IP_VERSION(2, 1, 2): - def = data = RREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG_Sienna_Cichlid); def1 = data1 = RREG32_SOC15(MMHUB, 0, mmDAGB0_CNTL_MISC2_Sienna_Cichlid); break; default: @@ -608,8 +605,6 @@ static void mmhub_v2_0_update_medium_grain_clock_gating(struct amdgpu_device *ad case IP_VERSION(2, 1, 0): case IP_VERSION(2, 1, 1): case IP_VERSION(2, 1, 2): - if (def != data) - WREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG_Sienna_Cichlid, data); if (def1 != data1) WREG32_SOC15(MMHUB, 0, mmDAGB0_CNTL_MISC2_Sienna_Cichlid, data1); break; @@ -634,8 +629,8 @@ static void mmhub_v2_0_update_medium_grain_light_sleep(struct amdgpu_device *ade case IP_VERSION(2, 1, 0): case IP_VERSION(2, 1, 1): case IP_VERSION(2, 1, 2): - def = data = RREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG_Sienna_Cichlid); - break; + /* There is no ATCL2 in MMHUB for 2.1.x */ + return; default: def = data = RREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG); break; @@ -646,18 +641,8 @@ static void mmhub_v2_0_update_medium_grain_light_sleep(struct amdgpu_device *ade else data &= ~MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK; - if (def != data) { - switch (adev->ip_versions[MMHUB_HWIP][0]) { - case IP_VERSION(2, 1, 0): - case IP_VERSION(2, 1, 1): - case IP_VERSION(2, 1, 2): - WREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG_Sienna_Cichlid, data); - break; - default: - WREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG, data); - break; - } - } + if (def != data) + WREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG, data); } static int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev, @@ -695,7 +680,10 @@ static void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u64 *flags) case IP_VERSION(2, 1, 0): case IP_VERSION(2, 1, 1): case IP_VERSION(2, 1, 2): - data = RREG32_SOC15(MMHUB, 0, mmMM_ATC_L2_MISC_CG_Sienna_Cichlid); + /* There is no ATCL2 in MMHUB for 2.1.x. Keep the status + * based on DAGB + */ + data = MM_ATC_L2_MISC_CG__ENABLE_MASK; data1 = RREG32_SOC15(MMHUB, 0, mmDAGB0_CNTL_MISC2_Sienna_Cichlid); break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 56424f75dd2c..65181efba50e 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -1504,11 +1504,6 @@ static int sdma_v4_0_start(struct amdgpu_device *adev) WREG32_SDMA(i, mmSDMA0_CNTL, temp); if (!amdgpu_sriov_vf(adev)) { - ring = &adev->sdma.instance[i].ring; - adev->nbio.funcs->sdma_doorbell_range(adev, i, - ring->use_doorbell, ring->doorbell_index, - adev->doorbell_index.sdma_doorbell_range); - /* unhalt engine */ temp = RREG32_SDMA(i, mmSDMA0_F32_CNTL); temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 183024d7c184..e3b2b6b4f1a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -1211,6 +1211,20 @@ static int soc15_common_sw_fini(void *handle) return 0; } +static void soc15_sdma_doorbell_range_init(struct amdgpu_device *adev) +{ + int i; + + /* sdma doorbell range is programed by hypervisor */ + if (!amdgpu_sriov_vf(adev)) { + for (i = 0; i < adev->sdma.num_instances; i++) { + adev->nbio.funcs->sdma_doorbell_range(adev, i, + true, adev->doorbell_index.sdma_engine[i] << 1, + adev->doorbell_index.sdma_doorbell_range); + } + } +} + static int soc15_common_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -1230,6 +1244,13 @@ static int soc15_common_hw_init(void *handle) /* enable the doorbell aperture */ soc15_enable_doorbell_aperture(adev, true); + /* HW doorbell routing policy: doorbell writing not + * in SDMA/IH/MM/ACV range will be routed to CP. So + * we need to init SDMA doorbell range prior + * to CP ip block init and ring test. IH already + * happens before CP. + */ + soc15_sdma_doorbell_range_init(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index a5409531a2fd..64f241281262 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -795,6 +795,102 @@ static struct kfd_gpu_cache_info yellow_carp_cache_info[] = { }, }; +static struct kfd_gpu_cache_info gfx1037_cache_info[] = { + { + /* TCP L1 Cache per CU */ + .cache_size = 16, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 1, + }, + { + /* Scalar L1 Instruction Cache per SQC */ + .cache_size = 32, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_INST_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, + { + /* Scalar L1 Data Cache per SQC */ + .cache_size = 16, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, + { + /* GL1 Data Cache per SA */ + .cache_size = 128, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, + { + /* L2 Data Cache per GPU (Total Tex Cache) */ + .cache_size = 256, + .cache_level = 2, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, +}; + +static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = { + { + /* TCP L1 Cache per CU */ + .cache_size = 16, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 1, + }, + { + /* Scalar L1 Instruction Cache per SQC */ + .cache_size = 32, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_INST_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, + { + /* Scalar L1 Data Cache per SQC */ + .cache_size = 16, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, + { + /* GL1 Data Cache per SA */ + .cache_size = 128, + .cache_level = 1, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, + { + /* L2 Data Cache per GPU (Total Tex Cache) */ + .cache_size = 256, + .cache_level = 2, + .flags = (CRAT_CACHE_FLAGS_ENABLED | + CRAT_CACHE_FLAGS_DATA_CACHE | + CRAT_CACHE_FLAGS_SIMD_CACHE), + .num_cu_shared = 2, + }, +}; + static void kfd_populated_cu_info_cpu(struct kfd_topology_device *dev, struct crat_subtype_computeunit *cu) { @@ -1514,11 +1610,17 @@ static int kfd_fill_gpu_cache_info(struct kfd_dev *kdev, num_of_cache_types = ARRAY_SIZE(beige_goby_cache_info); break; case IP_VERSION(10, 3, 3): - case IP_VERSION(10, 3, 6): /* TODO: Double check these on production silicon */ - case IP_VERSION(10, 3, 7): /* TODO: Double check these on production silicon */ pcache_info = yellow_carp_cache_info; num_of_cache_types = ARRAY_SIZE(yellow_carp_cache_info); break; + case IP_VERSION(10, 3, 6): + pcache_info = gc_10_3_6_cache_info; + num_of_cache_types = ARRAY_SIZE(gc_10_3_6_cache_info); + break; + case IP_VERSION(10, 3, 7): + pcache_info = gfx1037_cache_info; + num_of_cache_types = ARRAY_SIZE(gfx1037_cache_info); + break; case IP_VERSION(11, 0, 0): case IP_VERSION(11, 0, 1): case IP_VERSION(11, 0, 2): diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c index 987bde4dca3d..af239b5565ae 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c @@ -1369,7 +1369,7 @@ static bool dm_plane_format_mod_supported(struct drm_plane *plane, { struct amdgpu_device *adev = drm_to_adev(plane->dev); const struct drm_format_info *info = drm_format_info(format); - struct hw_asic_id asic_id = adev->dm.dc->ctx->asic_id; + int i; enum dm_micro_swizzle microtile = modifier_gfx9_swizzle_mode(modifier) & 3; @@ -1386,49 +1386,13 @@ static bool dm_plane_format_mod_supported(struct drm_plane *plane, return true; } - /* check if swizzle mode is supported by this version of DCN */ - switch (asic_id.chip_family) { - case FAMILY_SI: - case FAMILY_CI: - case FAMILY_KV: - case FAMILY_CZ: - case FAMILY_VI: - /* asics before AI does not have modifier support */ - return false; - case FAMILY_AI: - case FAMILY_RV: - case FAMILY_NV: - case FAMILY_VGH: - case FAMILY_YELLOW_CARP: - case AMDGPU_FAMILY_GC_10_3_6: - case AMDGPU_FAMILY_GC_10_3_7: - switch (AMD_FMT_MOD_GET(TILE, modifier)) { - case AMD_FMT_MOD_TILE_GFX9_64K_R_X: - case AMD_FMT_MOD_TILE_GFX9_64K_D_X: - case AMD_FMT_MOD_TILE_GFX9_64K_S_X: - case AMD_FMT_MOD_TILE_GFX9_64K_D: - return true; - default: - return false; - } - break; - case AMDGPU_FAMILY_GC_11_0_0: - case AMDGPU_FAMILY_GC_11_0_1: - switch (AMD_FMT_MOD_GET(TILE, modifier)) { - case AMD_FMT_MOD_TILE_GFX11_256K_R_X: - case AMD_FMT_MOD_TILE_GFX9_64K_R_X: - case AMD_FMT_MOD_TILE_GFX9_64K_D_X: - case AMD_FMT_MOD_TILE_GFX9_64K_S_X: - case AMD_FMT_MOD_TILE_GFX9_64K_D: - return true; - default: - return false; - } - break; - default: - ASSERT(0); /* Unknown asic */ - break; + /* Check that the modifier is on the list of the plane's supported modifiers. */ + for (i = 0; i < plane->modifier_count; i++) { + if (modifier == plane->modifiers[i]) + break; } + if (i == plane->modifier_count) + return false; /* * For D swizzle the canonical modifier depends on the bpp, so check diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c index 6d721fadcbee..7f5e039893e8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c @@ -847,7 +847,7 @@ static const struct resource_caps res_cap_dcn314 = { .num_ddc = 5, .num_vmid = 16, .num_mpc_3dlut = 2, - .num_dsc = 3, + .num_dsc = 4, }; static const struct dc_plane_cap plane_cap = { diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index cb81ed2fbd53..d0c6cf61c676 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -77,7 +77,7 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/dcn32_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_32.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_rq_dlg_calc_32.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_util_32.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_util_32.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn321/dcn321_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/dcn31_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_ccflags) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c index b6369758b491..aa976fe4d426 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c @@ -119,15 +119,15 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = { }, }, .num_states = 1, - .sr_exit_time_us = 12.36, - .sr_enter_plus_exit_time_us = 16.72, + .sr_exit_time_us = 19.95, + .sr_enter_plus_exit_time_us = 24.36, .sr_exit_z8_time_us = 285.0, .sr_enter_plus_exit_z8_time_us = 320, .writeback_latency_us = 12.0, .round_trip_ping_latency_dcfclk_cycles = 263, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, + .urgent_latency_pixel_data_only_us = 9.35, + .urgent_latency_pixel_mixed_with_vm_data_us = 9.35, + .urgent_latency_vm_data_only_us = 9.35, .fclk_change_latency_us = 20, .usr_retraining_latency_us = 2, .smn_latency_us = 2, diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index e85364dff4e0..5cb3e8634739 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -262,8 +262,9 @@ struct kfd2kgd_calls { uint32_t queue_id); int (*hqd_destroy)(struct amdgpu_device *adev, void *mqd, - uint32_t reset_type, unsigned int timeout, - uint32_t pipe_id, uint32_t queue_id); + enum kfd_preempt_type reset_type, + unsigned int timeout, uint32_t pipe_id, + uint32_t queue_id); bool (*hqd_sdma_is_occupied)(struct amdgpu_device *adev, void *mqd); diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h index ae2d337158f3..f77401709d83 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h @@ -27,7 +27,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define PMFW_DRIVER_IF_VERSION 5 +#define PMFW_DRIVER_IF_VERSION 7 typedef struct { int32_t value; @@ -163,8 +163,8 @@ typedef struct { uint16_t DclkFrequency; //[MHz] uint16_t MemclkFrequency; //[MHz] uint16_t spare; //[centi] - uint16_t UvdActivity; //[centi] uint16_t GfxActivity; //[centi] + uint16_t UvdActivity; //[centi] uint16_t Voltage[2]; //[mV] indices: VDDCR_VDD, VDDCR_SOC uint16_t Current[2]; //[mA] indices: VDDCR_VDD, VDDCR_SOC @@ -199,6 +199,19 @@ typedef struct { uint16_t DeviceState; uint16_t CurTemp; //[centi-Celsius] uint16_t spare2; + + uint16_t AverageGfxclkFrequency; + uint16_t AverageFclkFrequency; + uint16_t AverageGfxActivity; + uint16_t AverageSocclkFrequency; + uint16_t AverageVclkFrequency; + uint16_t AverageVcnActivity; + uint16_t AverageDRAMReads; //Filtered DF Bandwidth::DRAM Reads + uint16_t AverageDRAMWrites; //Filtered DF Bandwidth::DRAM Writes + uint16_t AverageSocketPower; //Filtered value of CurrentSocketPower + uint16_t AverageCorePower; //Filtered of [sum of CorePower[8]]) + uint16_t AverageCoreC0Residency[8]; //Filtered of [average C0 residency % per core] + uint32_t MetricsCounter; //Counts the # of metrics table parameter reads per update to the metrics table, i.e. if the metrics table update happens every 1 second, this value could be up to 1000 if the smu collected metrics data every cycle, or as low as 0 if the smu was asleep the whole time. Reset to 0 after writing. } SmuMetrics_t; typedef struct { diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index f442bf085a31..f75b9688f512 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -28,7 +28,7 @@ #define SMU13_DRIVER_IF_VERSION_INV 0xFFFFFFFF #define SMU13_DRIVER_IF_VERSION_YELLOW_CARP 0x04 #define SMU13_DRIVER_IF_VERSION_ALDE 0x08 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x05 +#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x07 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_5 0x04 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0 0x30 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_7 0x2C diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 445005571f76..9cd005131f56 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -2242,9 +2242,17 @@ static void arcturus_get_unique_id(struct smu_context *smu) static int arcturus_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; uint32_t smu_version; int ret; + /* + * Arcturus does not need the cstate disablement + * prerequisite for gpu reset. + */ + if (amdgpu_in_reset(adev) || adev->in_suspend) + return 0; + ret = smu_cmn_get_smc_version(smu, NULL, &smu_version); if (ret) { dev_err(smu->adev->dev, "Failed to get smu version!\n"); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index 619aee51b123..d30ec3005ea1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -1640,6 +1640,15 @@ static bool aldebaran_is_baco_supported(struct smu_context *smu) static int aldebaran_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; + + /* + * Aldebaran does not need the cstate disablement + * prerequisite for gpu reset. + */ + if (amdgpu_in_reset(adev) || adev->in_suspend) + return 0; + return smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 93f9b8377539..750d8da84fac 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -210,7 +210,8 @@ int smu_v13_0_init_pptable_microcode(struct smu_context *smu) return 0; if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 7)) || - (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 0))) + (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 0)) || + (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 10))) return 0; /* override pptable_id from driver parameter */ diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 1d454485e0d9..29529328152d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -119,6 +119,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(NotifyPowerSource, PPSMC_MSG_NotifyPowerSource, 0), MSG_MAP(Mode1Reset, PPSMC_MSG_Mode1Reset, 0), MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareMp1ForUnload, 0), + MSG_MAP(DFCstateControl, PPSMC_MSG_SetExternalClientDfCstateAllow, 0), }; static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = { @@ -1753,6 +1754,15 @@ static int smu_v13_0_0_set_mp1_state(struct smu_context *smu, return ret; } +static int smu_v13_0_0_set_df_cstate(struct smu_context *smu, + enum pp_df_cstate state) +{ + return smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_DFCstateControl, + state, + NULL); +} + static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_0_set_default_dpm_table, @@ -1822,6 +1832,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .mode1_reset_is_support = smu_v13_0_0_is_mode1_reset_supported, .mode1_reset = smu_v13_0_mode1_reset, .set_mp1_state = smu_v13_0_0_set_mp1_state, + .set_df_cstate = smu_v13_0_0_set_df_cstate, }; void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index c422bf8a09b1..c4102cfb734c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -121,6 +121,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_7_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(Mode1Reset, PPSMC_MSG_Mode1Reset, 0), MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareMp1ForUnload, 0), MSG_MAP(SetMGpuFanBoostLimitRpm, PPSMC_MSG_SetMGpuFanBoostLimitRpm, 0), + MSG_MAP(DFCstateControl, PPSMC_MSG_SetExternalClientDfCstateAllow, 0), }; static struct cmn2asic_mapping smu_v13_0_7_clk_map[SMU_CLK_COUNT] = { @@ -1587,6 +1588,16 @@ static bool smu_v13_0_7_is_mode1_reset_supported(struct smu_context *smu) return true; } + +static int smu_v13_0_7_set_df_cstate(struct smu_context *smu, + enum pp_df_cstate state) +{ + return smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_DFCstateControl, + state, + NULL); +} + static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, @@ -1649,6 +1660,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .mode1_reset_is_support = smu_v13_0_7_is_mode1_reset_supported, .mode1_reset = smu_v13_0_mode1_reset, .set_mp1_state = smu_v13_0_7_set_mp1_state, + .set_df_cstate = smu_v13_0_7_set_df_cstate, }; void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c index 49107a6cdac1..0dacbdb49a30 100644 --- a/drivers/gpu/drm/bridge/parade-ps8640.c +++ b/drivers/gpu/drm/bridge/parade-ps8640.c @@ -105,6 +105,7 @@ struct ps8640 { struct gpio_desc *gpio_powerdown; struct device_link *link; bool pre_enabled; + bool need_post_hpd_delay; }; static const struct regmap_config ps8640_regmap_config[] = { @@ -173,14 +174,31 @@ static int _ps8640_wait_hpd_asserted(struct ps8640 *ps_bridge, unsigned long wai { struct regmap *map = ps_bridge->regmap[PAGE2_TOP_CNTL]; int status; + int ret; /* * Apparently something about the firmware in the chip signals that * HPD goes high by reporting GPIO9 as high (even though HPD isn't * actually connected to GPIO9). */ - return regmap_read_poll_timeout(map, PAGE2_GPIO_H, status, - status & PS_GPIO9, wait_us / 10, wait_us); + ret = regmap_read_poll_timeout(map, PAGE2_GPIO_H, status, + status & PS_GPIO9, wait_us / 10, wait_us); + + /* + * The first time we see HPD go high after a reset we delay an extra + * 50 ms. The best guess is that the MCU is doing "stuff" during this + * time (maybe talking to the panel) and we don't want to interrupt it. + * + * No locking is done around "need_post_hpd_delay". If we're here we + * know we're holding a PM Runtime reference and the only other place + * that touches this is PM Runtime resume. + */ + if (!ret && ps_bridge->need_post_hpd_delay) { + ps_bridge->need_post_hpd_delay = false; + msleep(50); + } + + return ret; } static int ps8640_wait_hpd_asserted(struct drm_dp_aux *aux, unsigned long wait_us) @@ -376,6 +394,9 @@ static int __maybe_unused ps8640_resume(struct device *dev) usleep_range(2000, 2500); gpiod_set_value(ps_bridge->gpio_reset, 0); + /* We just reset things, so we need a delay after the first HPD */ + ps_bridge->need_post_hpd_delay = true; + /* * Mystery 200 ms delay for the "MCU to be ready". It's unclear if * this is truly necessary since the MCU will already signal that diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 7d6eb9ad7a02..459571e2cc57 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -135,18 +135,6 @@ static u32 raw_block_offset(const void *bdb, enum bdb_block_id section_id) return block - bdb; } -/* size of the block excluding the header */ -static u32 raw_block_size(const void *bdb, enum bdb_block_id section_id) -{ - const void *block; - - block = find_raw_section(bdb, section_id); - if (!block) - return 0; - - return get_blocksize(block); -} - struct bdb_block_entry { struct list_head node; enum bdb_block_id section_id; @@ -231,9 +219,14 @@ static bool validate_lfp_data_ptrs(const void *bdb, { int fp_timing_size, dvo_timing_size, panel_pnp_id_size, panel_name_size; int data_block_size, lfp_data_size; + const void *data_block; int i; - data_block_size = raw_block_size(bdb, BDB_LVDS_LFP_DATA); + data_block = find_raw_section(bdb, BDB_LVDS_LFP_DATA); + if (!data_block) + return false; + + data_block_size = get_blocksize(data_block); if (data_block_size == 0) return false; @@ -261,21 +254,6 @@ static bool validate_lfp_data_ptrs(const void *bdb, if (16 * lfp_data_size > data_block_size) return false; - /* - * Except for vlv/chv machines all real VBTs seem to have 6 - * unaccounted bytes in the fp_timing table. And it doesn't - * appear to be a really intentional hole as the fp_timing - * 0xffff terminator is always within those 6 missing bytes. - */ - if (fp_timing_size + dvo_timing_size + panel_pnp_id_size != lfp_data_size && - fp_timing_size + 6 + dvo_timing_size + panel_pnp_id_size != lfp_data_size) - return false; - - if (ptrs->ptr[0].fp_timing.offset + fp_timing_size > ptrs->ptr[0].dvo_timing.offset || - ptrs->ptr[0].dvo_timing.offset + dvo_timing_size != ptrs->ptr[0].panel_pnp_id.offset || - ptrs->ptr[0].panel_pnp_id.offset + panel_pnp_id_size != lfp_data_size) - return false; - /* make sure the table entries have uniform size */ for (i = 1; i < 16; i++) { if (ptrs->ptr[i].fp_timing.table_size != fp_timing_size || @@ -289,6 +267,23 @@ static bool validate_lfp_data_ptrs(const void *bdb, return false; } + /* + * Except for vlv/chv machines all real VBTs seem to have 6 + * unaccounted bytes in the fp_timing table. And it doesn't + * appear to be a really intentional hole as the fp_timing + * 0xffff terminator is always within those 6 missing bytes. + */ + if (fp_timing_size + 6 + dvo_timing_size + panel_pnp_id_size == lfp_data_size) + fp_timing_size += 6; + + if (fp_timing_size + dvo_timing_size + panel_pnp_id_size != lfp_data_size) + return false; + + if (ptrs->ptr[0].fp_timing.offset + fp_timing_size != ptrs->ptr[0].dvo_timing.offset || + ptrs->ptr[0].dvo_timing.offset + dvo_timing_size != ptrs->ptr[0].panel_pnp_id.offset || + ptrs->ptr[0].panel_pnp_id.offset + panel_pnp_id_size != lfp_data_size) + return false; + /* make sure the tables fit inside the data block */ for (i = 0; i < 16; i++) { if (ptrs->ptr[i].fp_timing.offset + fp_timing_size > data_block_size || @@ -300,6 +295,15 @@ static bool validate_lfp_data_ptrs(const void *bdb, if (ptrs->panel_name.offset + 16 * panel_name_size > data_block_size) return false; + /* make sure fp_timing terminators are present at expected locations */ + for (i = 0; i < 16; i++) { + const u16 *t = data_block + ptrs->ptr[i].fp_timing.offset + + fp_timing_size - 2; + + if (*t != 0xffff) + return false; + } + return true; } @@ -333,18 +337,6 @@ static bool fixup_lfp_data_ptrs(const void *bdb, void *ptrs_block) return validate_lfp_data_ptrs(bdb, ptrs); } -static const void *find_fp_timing_terminator(const u8 *data, int size) -{ - int i; - - for (i = 0; i < size - 1; i++) { - if (data[i] == 0xff && data[i+1] == 0xff) - return &data[i]; - } - - return NULL; -} - static int make_lfp_data_ptr(struct lvds_lfp_data_ptr_table *table, int table_size, int total_size) { @@ -368,11 +360,22 @@ static void next_lfp_data_ptr(struct lvds_lfp_data_ptr_table *next, static void *generate_lfp_data_ptrs(struct drm_i915_private *i915, const void *bdb) { - int i, size, table_size, block_size, offset; - const void *t0, *t1, *block; + int i, size, table_size, block_size, offset, fp_timing_size; struct bdb_lvds_lfp_data_ptrs *ptrs; + const void *block; void *ptrs_block; + /* + * The hardcoded fp_timing_size is only valid for + * modernish VBTs. All older VBTs definitely should + * include block 41 and thus we don't need to + * generate one. + */ + if (i915->vbt.version < 155) + return NULL; + + fp_timing_size = 38; + block = find_raw_section(bdb, BDB_LVDS_LFP_DATA); if (!block) return NULL; @@ -381,17 +384,8 @@ static void *generate_lfp_data_ptrs(struct drm_i915_private *i915, block_size = get_blocksize(block); - size = block_size; - t0 = find_fp_timing_terminator(block, size); - if (!t0) - return NULL; - - size -= t0 - block - 2; - t1 = find_fp_timing_terminator(t0 + 2, size); - if (!t1) - return NULL; - - size = t1 - t0; + size = fp_timing_size + sizeof(struct lvds_dvo_timing) + + sizeof(struct lvds_pnp_id); if (size * 16 > block_size) return NULL; @@ -409,7 +403,7 @@ static void *generate_lfp_data_ptrs(struct drm_i915_private *i915, table_size = sizeof(struct lvds_dvo_timing); size = make_lfp_data_ptr(&ptrs->ptr[0].dvo_timing, table_size, size); - table_size = t0 - block + 2; + table_size = fp_timing_size; size = make_lfp_data_ptr(&ptrs->ptr[0].fp_timing, table_size, size); if (ptrs->ptr[0].fp_timing.table_size) @@ -424,14 +418,14 @@ static void *generate_lfp_data_ptrs(struct drm_i915_private *i915, return NULL; } - size = t1 - t0; + size = fp_timing_size + sizeof(struct lvds_dvo_timing) + + sizeof(struct lvds_pnp_id); for (i = 1; i < 16; i++) { next_lfp_data_ptr(&ptrs->ptr[i].fp_timing, &ptrs->ptr[i-1].fp_timing, size); next_lfp_data_ptr(&ptrs->ptr[i].dvo_timing, &ptrs->ptr[i-1].dvo_timing, size); next_lfp_data_ptr(&ptrs->ptr[i].panel_pnp_id, &ptrs->ptr[i-1].panel_pnp_id, size); } - size = t1 - t0; table_size = sizeof(struct lvds_lfp_panel_name); if (16 * (size + table_size) <= block_size) { diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 3ed7eeacc706..d4492b6d23d2 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -3923,6 +3923,8 @@ intel_dp_handle_hdmi_link_status_change(struct intel_dp *intel_dp) drm_dp_pcon_hdmi_frl_link_error_count(&intel_dp->aux, &intel_dp->attached_connector->base); + intel_dp->frl.is_trained = false; + /* Restart FRL training or fall back to TMDS mode */ intel_dp_check_frl_training(intel_dp); } diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index 19122bc6d2ab..b5f65b093c10 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -2747,13 +2747,10 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device) if (!intel_sdvo_connector) return false; - if (device == 0) { - intel_sdvo->controlled_output |= SDVO_OUTPUT_TMDS0; + if (device == 0) intel_sdvo_connector->output_flag = SDVO_OUTPUT_TMDS0; - } else if (device == 1) { - intel_sdvo->controlled_output |= SDVO_OUTPUT_TMDS1; + else if (device == 1) intel_sdvo_connector->output_flag = SDVO_OUTPUT_TMDS1; - } intel_connector = &intel_sdvo_connector->base; connector = &intel_connector->base; @@ -2808,7 +2805,6 @@ intel_sdvo_tv_init(struct intel_sdvo *intel_sdvo, int type) encoder->encoder_type = DRM_MODE_ENCODER_TVDAC; connector->connector_type = DRM_MODE_CONNECTOR_SVIDEO; - intel_sdvo->controlled_output |= type; intel_sdvo_connector->output_flag = type; if (intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo) < 0) { @@ -2849,13 +2845,10 @@ intel_sdvo_analog_init(struct intel_sdvo *intel_sdvo, int device) encoder->encoder_type = DRM_MODE_ENCODER_DAC; connector->connector_type = DRM_MODE_CONNECTOR_VGA; - if (device == 0) { - intel_sdvo->controlled_output |= SDVO_OUTPUT_RGB0; + if (device == 0) intel_sdvo_connector->output_flag = SDVO_OUTPUT_RGB0; - } else if (device == 1) { - intel_sdvo->controlled_output |= SDVO_OUTPUT_RGB1; + else if (device == 1) intel_sdvo_connector->output_flag = SDVO_OUTPUT_RGB1; - } if (intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo) < 0) { kfree(intel_sdvo_connector); @@ -2885,13 +2878,10 @@ intel_sdvo_lvds_init(struct intel_sdvo *intel_sdvo, int device) encoder->encoder_type = DRM_MODE_ENCODER_LVDS; connector->connector_type = DRM_MODE_CONNECTOR_LVDS; - if (device == 0) { - intel_sdvo->controlled_output |= SDVO_OUTPUT_LVDS0; + if (device == 0) intel_sdvo_connector->output_flag = SDVO_OUTPUT_LVDS0; - } else if (device == 1) { - intel_sdvo->controlled_output |= SDVO_OUTPUT_LVDS1; + else if (device == 1) intel_sdvo_connector->output_flag = SDVO_OUTPUT_LVDS1; - } if (intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo) < 0) { kfree(intel_sdvo_connector); @@ -2926,16 +2916,39 @@ err: return false; } +static u16 intel_sdvo_filter_output_flags(u16 flags) +{ + flags &= SDVO_OUTPUT_MASK; + + /* SDVO requires XXX1 function may not exist unless it has XXX0 function.*/ + if (!(flags & SDVO_OUTPUT_TMDS0)) + flags &= ~SDVO_OUTPUT_TMDS1; + + if (!(flags & SDVO_OUTPUT_RGB0)) + flags &= ~SDVO_OUTPUT_RGB1; + + if (!(flags & SDVO_OUTPUT_LVDS0)) + flags &= ~SDVO_OUTPUT_LVDS1; + + return flags; +} + static bool intel_sdvo_output_setup(struct intel_sdvo *intel_sdvo, u16 flags) { - /* SDVO requires XXX1 function may not exist unless it has XXX0 function.*/ + struct drm_i915_private *i915 = to_i915(intel_sdvo->base.base.dev); + + flags = intel_sdvo_filter_output_flags(flags); + + intel_sdvo->controlled_output = flags; + + intel_sdvo_select_ddc_bus(i915, intel_sdvo); if (flags & SDVO_OUTPUT_TMDS0) if (!intel_sdvo_dvi_init(intel_sdvo, 0)) return false; - if ((flags & SDVO_TMDS_MASK) == SDVO_TMDS_MASK) + if (flags & SDVO_OUTPUT_TMDS1) if (!intel_sdvo_dvi_init(intel_sdvo, 1)) return false; @@ -2956,7 +2969,7 @@ intel_sdvo_output_setup(struct intel_sdvo *intel_sdvo, u16 flags) if (!intel_sdvo_analog_init(intel_sdvo, 0)) return false; - if ((flags & SDVO_RGB_MASK) == SDVO_RGB_MASK) + if (flags & SDVO_OUTPUT_RGB1) if (!intel_sdvo_analog_init(intel_sdvo, 1)) return false; @@ -2964,14 +2977,13 @@ intel_sdvo_output_setup(struct intel_sdvo *intel_sdvo, u16 flags) if (!intel_sdvo_lvds_init(intel_sdvo, 0)) return false; - if ((flags & SDVO_LVDS_MASK) == SDVO_LVDS_MASK) + if (flags & SDVO_OUTPUT_LVDS1) if (!intel_sdvo_lvds_init(intel_sdvo, 1)) return false; - if ((flags & SDVO_OUTPUT_MASK) == 0) { + if (flags == 0) { unsigned char bytes[2]; - intel_sdvo->controlled_output = 0; memcpy(bytes, &intel_sdvo->caps.output_flags, 2); DRM_DEBUG_KMS("%s: Unknown SDVO output type (0x%02x%02x)\n", SDVO_NAME(intel_sdvo), @@ -3383,8 +3395,6 @@ bool intel_sdvo_init(struct drm_i915_private *dev_priv, */ intel_sdvo->base.cloneable = 0; - intel_sdvo_select_ddc_bus(dev_priv, intel_sdvo); - /* Set the input timing to the screen. Assume always input 0. */ if (!intel_sdvo_set_target_input(intel_sdvo)) goto err_output; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.c b/drivers/gpu/drm/i915/gem/i915_gem_internal.c index c698f95af15f..629acb403a2c 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_internal.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.c @@ -6,7 +6,6 @@ #include <linux/scatterlist.h> #include <linux/slab.h> -#include <linux/swiotlb.h> #include "i915_drv.h" #include "i915_gem.h" @@ -38,22 +37,12 @@ static int i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj) struct scatterlist *sg; unsigned int sg_page_sizes; unsigned int npages; - int max_order; + int max_order = MAX_ORDER; + unsigned int max_segment; gfp_t gfp; - max_order = MAX_ORDER; -#ifdef CONFIG_SWIOTLB - if (is_swiotlb_active(obj->base.dev->dev)) { - unsigned int max_segment; - - max_segment = swiotlb_max_segment(); - if (max_segment) { - max_segment = max_t(unsigned int, max_segment, - PAGE_SIZE) >> PAGE_SHIFT; - max_order = min(max_order, ilog2(max_segment)); - } - } -#endif + max_segment = i915_sg_segment_size(i915->drm.dev) >> PAGE_SHIFT; + max_order = min(max_order, get_order(max_segment)); gfp = GFP_KERNEL | __GFP_HIGHMEM | __GFP_RECLAIMABLE; if (IS_I965GM(i915) || IS_I965G(i915)) { diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index 4eed3dd90ba8..34b9c76cd8e6 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -194,7 +194,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj) struct intel_memory_region *mem = obj->mm.region; struct address_space *mapping = obj->base.filp->f_mapping; const unsigned long page_count = obj->base.size / PAGE_SIZE; - unsigned int max_segment = i915_sg_segment_size(); + unsigned int max_segment = i915_sg_segment_size(i915->drm.dev); struct sg_table *st; struct sgt_iter sgt_iter; struct page *page; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index 6f3ab7ade41a..e85cfc36359a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -189,7 +189,7 @@ static int i915_ttm_tt_shmem_populate(struct ttm_device *bdev, struct drm_i915_private *i915 = container_of(bdev, typeof(*i915), bdev); struct intel_memory_region *mr = i915->mm.regions[INTEL_MEMORY_SYSTEM]; struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm); - const unsigned int max_segment = i915_sg_segment_size(); + const unsigned int max_segment = i915_sg_segment_size(i915->drm.dev); const size_t size = (size_t)ttm->num_pages << PAGE_SHIFT; struct file *filp = i915_tt->filp; struct sgt_iter sgt_iter; @@ -568,7 +568,7 @@ static struct i915_refct_sgt *i915_ttm_tt_get_st(struct ttm_tt *ttm) ret = sg_alloc_table_from_pages_segment(st, ttm->pages, ttm->num_pages, 0, (unsigned long)ttm->num_pages << PAGE_SHIFT, - i915_sg_segment_size(), GFP_KERNEL); + i915_sg_segment_size(i915_tt->dev), GFP_KERNEL); if (ret) { st->sgl = NULL; return ERR_PTR(ret); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 8423df021b71..e4515d6acd43 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -129,7 +129,7 @@ static void i915_gem_object_userptr_drop_ref(struct drm_i915_gem_object *obj) static int i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj) { const unsigned long num_pages = obj->base.size >> PAGE_SHIFT; - unsigned int max_segment = i915_sg_segment_size(); + unsigned int max_segment = i915_sg_segment_size(obj->base.dev->dev); struct sg_table *st; unsigned int sg_page_sizes; struct page **pvec; diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index e8111fce56d0..dfa9a7c4ebc6 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -2301,11 +2301,11 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) } if (IS_DG1_GRAPHICS_STEP(i915, STEP_A0, STEP_B0) || - IS_ROCKETLAKE(i915) || IS_TIGERLAKE(i915)) { + IS_ROCKETLAKE(i915) || IS_TIGERLAKE(i915) || IS_ALDERLAKE_P(i915)) { /* * Wa_1607030317:tgl * Wa_1607186500:tgl - * Wa_1607297627:tgl,rkl,dg1[a0] + * Wa_1607297627:tgl,rkl,dg1[a0],adlp * * On TGL and RKL there are multiple entries for this WA in the * BSpec; some indicate this is an A0-only WA, others indicate diff --git a/drivers/gpu/drm/i915/i915_scatterlist.h b/drivers/gpu/drm/i915/i915_scatterlist.h index 9ddb3e743a3e..b0a1db44f895 100644 --- a/drivers/gpu/drm/i915/i915_scatterlist.h +++ b/drivers/gpu/drm/i915/i915_scatterlist.h @@ -9,7 +9,8 @@ #include <linux/pfn.h> #include <linux/scatterlist.h> -#include <linux/swiotlb.h> +#include <linux/dma-mapping.h> +#include <xen/xen.h> #include "i915_gem.h" @@ -127,19 +128,26 @@ static inline unsigned int i915_sg_dma_sizes(struct scatterlist *sg) return page_sizes; } -static inline unsigned int i915_sg_segment_size(void) +static inline unsigned int i915_sg_segment_size(struct device *dev) { - unsigned int size = swiotlb_max_segment(); - - if (size == 0) - size = UINT_MAX; - - size = rounddown(size, PAGE_SIZE); - /* swiotlb_max_segment_size can return 1 byte when it means one page. */ - if (size < PAGE_SIZE) - size = PAGE_SIZE; - - return size; + size_t max = min_t(size_t, UINT_MAX, dma_max_mapping_size(dev)); + + /* + * For Xen PV guests pages aren't contiguous in DMA (machine) address + * space. The DMA API takes care of that both in dma_alloc_* (by + * calling into the hypervisor to make the pages contiguous) and in + * dma_map_* (by bounce buffering). But i915 abuses ignores the + * coherency aspects of the DMA API and thus can't cope with bounce + * buffering actually happening, so add a hack here to force small + * allocations and mappings when running in PV mode on Xen. + * + * Note this will still break if bounce buffering is required for other + * reasons, like confidential computing hypervisors or PCIe root ports + * with addressing limitations. + */ + if (xen_pv_domain()) + max = PAGE_SIZE; + return round_down(max, PAGE_SIZE); } bool i915_sg_trim(struct sg_table *orig_st); diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 6ed5786bcd29..744cca507946 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -591,8 +591,15 @@ void intel_runtime_pm_enable(struct intel_runtime_pm *rpm) pm_runtime_use_autosuspend(kdev); } - /* Enable by default */ - pm_runtime_allow(kdev); + /* + * FIXME: Temp hammer to keep autosupend disable on lmem supported platforms. + * As per PCIe specs 5.3.1.4.1, all iomem read write request over a PCIe + * function will be unsupported in case PCIe endpoint function is in D3. + * Let's keep i915 autosuspend control 'on' till we fix all known issue + * with lmem access in D3. + */ + if (!IS_DGFX(i915)) + pm_runtime_allow(kdev); /* * The core calls the driver load handler with an RPM reference held. diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c index 55f443328d8e..730355f9e2d4 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -91,7 +91,7 @@ struct a6xx_state_memobj { static void *state_kcalloc(struct a6xx_gpu_state *a6xx_state, int nr, size_t objsize) { struct a6xx_state_memobj *obj = - kzalloc((nr * objsize) + sizeof(*obj), GFP_KERNEL); + kvzalloc((nr * objsize) + sizeof(*obj), GFP_KERNEL); if (!obj) return NULL; @@ -1040,8 +1040,11 @@ static void a6xx_gpu_state_destroy(struct kref *kref) if (a6xx_state->gmu_hfi) kvfree(a6xx_state->gmu_hfi->data); + if (a6xx_state->gmu_debug) + kvfree(a6xx_state->gmu_debug->data); + list_for_each_entry_safe(obj, tmp, &a6xx_state->objs, node) - kfree(obj); + kvfree(obj); adreno_gpu_state_destroy(state); kfree(a6xx_state); diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 382fb7f9e497..5a0e8491cd3a 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -729,7 +729,12 @@ static char *adreno_gpu_ascii85_encode(u32 *src, size_t len) return buf; } -/* len is expected to be in bytes */ +/* len is expected to be in bytes + * + * WARNING: *ptr should be allocated with kvmalloc or friends. It can be free'd + * with kvfree() and replaced with a newly kvmalloc'd buffer on the first call + * when the unencoded raw data is encoded + */ void adreno_show_object(struct drm_printer *p, void **ptr, int len, bool *encoded) { diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c index 7288041dd86a..7444b75c4215 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c @@ -56,8 +56,9 @@ static int mdp4_lvds_connector_get_modes(struct drm_connector *connector) return ret; } -static int mdp4_lvds_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) +static enum drm_mode_status +mdp4_lvds_connector_mode_valid(struct drm_connector *connector, + struct drm_display_mode *mode) { struct mdp4_lvds_connector *mdp4_lvds_connector = to_mdp4_lvds_connector(connector); diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c index 013ca02e17cb..3ac139a4bbe8 100644 --- a/drivers/gpu/drm/msm/dp/dp_ctrl.c +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c @@ -1245,8 +1245,7 @@ static int dp_ctrl_link_train(struct dp_ctrl_private *ctrl, { int ret = 0; const u8 *dpcd = ctrl->panel->dpcd; - u8 encoding = DP_SET_ANSI_8B10B; - u8 ssc; + u8 encoding[] = { 0, DP_SET_ANSI_8B10B }; u8 assr; struct dp_link_info link_info = {0}; @@ -1258,13 +1257,11 @@ static int dp_ctrl_link_train(struct dp_ctrl_private *ctrl, dp_aux_link_configure(ctrl->aux, &link_info); - if (drm_dp_max_downspread(dpcd)) { - ssc = DP_SPREAD_AMP_0_5; - drm_dp_dpcd_write(ctrl->aux, DP_DOWNSPREAD_CTRL, &ssc, 1); - } + if (drm_dp_max_downspread(dpcd)) + encoding[0] |= DP_SPREAD_AMP_0_5; - drm_dp_dpcd_write(ctrl->aux, DP_MAIN_LINK_CHANNEL_CODING_SET, - &encoding, 1); + /* config DOWNSPREAD_CTRL and MAIN_LINK_CHANNEL_CODING_SET */ + drm_dp_dpcd_write(ctrl->aux, DP_DOWNSPREAD_CTRL, encoding, 2); if (drm_dp_alternate_scrambler_reset_cap(dpcd)) { assr = DP_ALTERNATE_SCRAMBLER_RESET_ENABLE; diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c index bfd0aeff3f0d..a49f6dbbe888 100644 --- a/drivers/gpu/drm/msm/dp/dp_display.c +++ b/drivers/gpu/drm/msm/dp/dp_display.c @@ -1249,7 +1249,7 @@ int dp_display_request_irq(struct msm_dp *dp_display) return -EINVAL; } - rc = devm_request_irq(&dp->pdev->dev, dp->irq, + rc = devm_request_irq(dp_display->drm_dev->dev, dp->irq, dp_display_irq_handler, IRQF_TRIGGER_HIGH, "dp_display_isr", dp); if (rc < 0) { @@ -1528,6 +1528,11 @@ void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor) } } +static void of_dp_aux_depopulate_bus_void(void *data) +{ + of_dp_aux_depopulate_bus(data); +} + static int dp_display_get_next_bridge(struct msm_dp *dp) { int rc; @@ -1552,10 +1557,16 @@ static int dp_display_get_next_bridge(struct msm_dp *dp) * panel driver is probed asynchronously but is the best we * can do without a bigger driver reorganization. */ - rc = devm_of_dp_aux_populate_ep_devices(dp_priv->aux); + rc = of_dp_aux_populate_bus(dp_priv->aux, NULL); of_node_put(aux_bus); if (rc) goto error; + + rc = devm_add_action_or_reset(dp->drm_dev->dev, + of_dp_aux_depopulate_bus_void, + dp_priv->aux); + if (rc) + goto error; } else if (dp->is_edp) { DRM_ERROR("eDP aux_bus not found\n"); return -ENODEV; @@ -1568,7 +1579,7 @@ static int dp_display_get_next_bridge(struct msm_dp *dp) * For DisplayPort interfaces external bridges are optional, so * silently ignore an error if one is not present (-ENODEV). */ - rc = dp_parser_find_next_bridge(dp_priv->parser); + rc = devm_dp_parser_find_next_bridge(dp->drm_dev->dev, dp_priv->parser); if (!dp->is_edp && rc == -ENODEV) return 0; @@ -1597,6 +1608,12 @@ int msm_dp_modeset_init(struct msm_dp *dp_display, struct drm_device *dev, return -EINVAL; priv = dev->dev_private; + + if (priv->num_bridges == ARRAY_SIZE(priv->bridges)) { + DRM_DEV_ERROR(dev->dev, "too many bridges\n"); + return -ENOSPC; + } + dp_display->drm_dev = dev; dp_priv = container_of(dp_display, struct dp_display_private, dp_display); diff --git a/drivers/gpu/drm/msm/dp/dp_drm.c b/drivers/gpu/drm/msm/dp/dp_drm.c index 6df25f7662e7..6db82f9b03af 100644 --- a/drivers/gpu/drm/msm/dp/dp_drm.c +++ b/drivers/gpu/drm/msm/dp/dp_drm.c @@ -31,6 +31,36 @@ static enum drm_connector_status dp_bridge_detect(struct drm_bridge *bridge) connector_status_disconnected; } +static int dp_bridge_atomic_check(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct msm_dp *dp; + + dp = to_dp_bridge(bridge)->dp_display; + + drm_dbg_dp(dp->drm_dev, "is_connected = %s\n", + (dp->is_connected) ? "true" : "false"); + + /* + * There is no protection in the DRM framework to check if the display + * pipeline has been already disabled before trying to disable it again. + * Hence if the sink is unplugged, the pipeline gets disabled, but the + * crtc->active is still true. Any attempt to set the mode or manually + * disable this encoder will result in the crash. + * + * TODO: add support for telling the DRM subsystem that the pipeline is + * disabled by the hardware and thus all access to it should be forbidden. + * After that this piece of code can be removed. + */ + if (bridge->ops & DRM_BRIDGE_OP_HPD) + return (dp->is_connected) ? 0 : -ENOTCONN; + + return 0; +} + + /** * dp_bridge_get_modes - callback to add drm modes via drm_mode_probed_add() * @bridge: Poiner to drm bridge @@ -61,6 +91,9 @@ static int dp_bridge_get_modes(struct drm_bridge *bridge, struct drm_connector * } static const struct drm_bridge_funcs dp_bridge_ops = { + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, .enable = dp_bridge_enable, .disable = dp_bridge_disable, .post_disable = dp_bridge_post_disable, @@ -68,6 +101,7 @@ static const struct drm_bridge_funcs dp_bridge_ops = { .mode_valid = dp_bridge_mode_valid, .get_modes = dp_bridge_get_modes, .detect = dp_bridge_detect, + .atomic_check = dp_bridge_atomic_check, }; struct drm_bridge *dp_bridge_init(struct msm_dp *dp_display, struct drm_device *dev, diff --git a/drivers/gpu/drm/msm/dp/dp_parser.c b/drivers/gpu/drm/msm/dp/dp_parser.c index dd732215d55b..dcbe893d66d7 100644 --- a/drivers/gpu/drm/msm/dp/dp_parser.c +++ b/drivers/gpu/drm/msm/dp/dp_parser.c @@ -240,12 +240,12 @@ static int dp_parser_clock(struct dp_parser *parser) return 0; } -int dp_parser_find_next_bridge(struct dp_parser *parser) +int devm_dp_parser_find_next_bridge(struct device *dev, struct dp_parser *parser) { - struct device *dev = &parser->pdev->dev; + struct platform_device *pdev = parser->pdev; struct drm_bridge *bridge; - bridge = devm_drm_of_get_bridge(dev, dev->of_node, 1, 0); + bridge = devm_drm_of_get_bridge(dev, pdev->dev.of_node, 1, 0); if (IS_ERR(bridge)) return PTR_ERR(bridge); diff --git a/drivers/gpu/drm/msm/dp/dp_parser.h b/drivers/gpu/drm/msm/dp/dp_parser.h index 866c1a82bf1a..d30ab773db46 100644 --- a/drivers/gpu/drm/msm/dp/dp_parser.h +++ b/drivers/gpu/drm/msm/dp/dp_parser.h @@ -138,8 +138,9 @@ struct dp_parser { struct dp_parser *dp_parser_get(struct platform_device *pdev); /** - * dp_parser_find_next_bridge() - find an additional bridge to DP + * devm_dp_parser_find_next_bridge() - find an additional bridge to DP * + * @dev: device to tie bridge lifetime to * @parser: dp_parser data from client * * This function is used to find any additional bridge attached to @@ -147,6 +148,6 @@ struct dp_parser *dp_parser_get(struct platform_device *pdev); * * Return: 0 if able to get the bridge, otherwise negative errno for failure. */ -int dp_parser_find_next_bridge(struct dp_parser *parser); +int devm_dp_parser_find_next_bridge(struct device *dev, struct dp_parser *parser); #endif diff --git a/drivers/gpu/drm/msm/dsi/dsi.c b/drivers/gpu/drm/msm/dsi/dsi.c index 1625328fa430..3db85b5c0feb 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.c +++ b/drivers/gpu/drm/msm/dsi/dsi.c @@ -227,6 +227,12 @@ int msm_dsi_modeset_init(struct msm_dsi *msm_dsi, struct drm_device *dev, return -EINVAL; priv = dev->dev_private; + + if (priv->num_bridges == ARRAY_SIZE(priv->bridges)) { + DRM_DEV_ERROR(dev->dev, "too many bridges\n"); + return -ENOSPC; + } + msm_dsi->dev = dev; ret = msm_dsi_host_modeset_init(msm_dsi->host, dev); diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index 93fe61b86967..f28fb21e3891 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -300,6 +300,11 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, struct platform_device *pdev = hdmi->pdev; int ret; + if (priv->num_bridges == ARRAY_SIZE(priv->bridges)) { + DRM_DEV_ERROR(dev->dev, "too many bridges\n"); + return -ENOSPC; + } + hdmi->dev = dev; hdmi->encoder = encoder; @@ -339,7 +344,7 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, goto fail; } - ret = devm_request_irq(&pdev->dev, hdmi->irq, + ret = devm_request_irq(dev->dev, hdmi->irq, msm_hdmi_irq, IRQF_TRIGGER_HIGH, "hdmi_isr", hdmi); if (ret < 0) { diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 0759e2d99f59..869ba58f0254 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -241,6 +241,7 @@ static int msm_drm_uninit(struct device *dev) for (i = 0; i < priv->num_bridges; i++) drm_bridge_remove(priv->bridges[i]); + priv->num_bridges = 0; pm_runtime_get_sync(dev); msm_irq_uninstall(ddev); diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c index 110e83aad9bb..1aa3700551f4 100644 --- a/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi-rockchip.c @@ -1031,23 +1031,31 @@ static int dw_mipi_dsi_rockchip_host_attach(void *priv_data, if (ret) { DRM_DEV_ERROR(dsi->dev, "Failed to register component: %d\n", ret); - return ret; + goto out; } second = dw_mipi_dsi_rockchip_find_second(dsi); - if (IS_ERR(second)) - return PTR_ERR(second); + if (IS_ERR(second)) { + ret = PTR_ERR(second); + goto out; + } if (second) { ret = component_add(second, &dw_mipi_dsi_rockchip_ops); if (ret) { DRM_DEV_ERROR(second, "Failed to register component: %d\n", ret); - return ret; + goto out; } } return 0; + +out: + mutex_lock(&dsi->usage_mutex); + dsi->usage_mode = DW_DSI_USAGE_IDLE; + mutex_unlock(&dsi->usage_mutex); + return ret; } static int dw_mipi_dsi_rockchip_host_detach(void *priv_data, @@ -1634,5 +1642,11 @@ struct platform_driver dw_mipi_dsi_rockchip_driver = { .of_match_table = dw_mipi_dsi_rockchip_dt_ids, .pm = &dw_mipi_dsi_rockchip_pm_ops, .name = "dw-mipi-dsi-rockchip", + /* + * For dual-DSI display, one DSI pokes at the other DSI's + * drvdata in dw_mipi_dsi_rockchip_find_second(). This is not + * safe for asynchronous probe. + */ + .probe_type = PROBE_FORCE_SYNCHRONOUS, }, }; diff --git a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c index c14f88893868..2f4b8f64cbad 100644 --- a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c @@ -565,7 +565,8 @@ static int dw_hdmi_rockchip_bind(struct device *dev, struct device *master, ret = rockchip_hdmi_parse_dt(hdmi); if (ret) { - DRM_DEV_ERROR(hdmi->dev, "Unable to parse OF data\n"); + if (ret != -EPROBE_DEFER) + DRM_DEV_ERROR(hdmi->dev, "Unable to parse OF data\n"); return ret; } diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c index 985584147da1..cf8322c300bd 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c @@ -364,9 +364,12 @@ rockchip_gem_create_with_handle(struct drm_file *file_priv, { struct rockchip_gem_object *rk_obj; struct drm_gem_object *obj; + bool is_framebuffer; int ret; - rk_obj = rockchip_gem_create_object(drm, size, false); + is_framebuffer = drm->fb_helper && file_priv == drm->fb_helper->client.file; + + rk_obj = rockchip_gem_create_object(drm, size, is_framebuffer); if (IS_ERR(rk_obj)) return ERR_CAST(rk_obj); diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index 6b8dfa1e7650..c186ace7f83b 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -490,6 +490,7 @@ module_init(vc4_drm_register); module_exit(vc4_drm_unregister); MODULE_ALIAS("platform:vc4-drm"); +MODULE_SOFTDEP("pre: snd-soc-hdmi-codec"); MODULE_DESCRIPTION("Broadcom VC4 DRM Driver"); MODULE_AUTHOR("Eric Anholt <eric@anholt.net>"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 1e5f68704d7d..874c6bd787c5 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -2869,12 +2869,37 @@ static int vc4_hdmi_runtime_resume(struct device *dev) struct vc4_hdmi *vc4_hdmi = dev_get_drvdata(dev); unsigned long __maybe_unused flags; u32 __maybe_unused value; + unsigned long rate; int ret; + /* + * The HSM clock is in the HDMI power domain, so we need to set + * its frequency while the power domain is active so that it + * keeps its rate. + */ + ret = clk_set_min_rate(vc4_hdmi->hsm_clock, HSM_MIN_CLOCK_FREQ); + if (ret) + return ret; + ret = clk_prepare_enable(vc4_hdmi->hsm_clock); if (ret) return ret; + /* + * Whenever the RaspberryPi boots without an HDMI monitor + * plugged in, the firmware won't have initialized the HSM clock + * rate and it will be reported as 0. + * + * If we try to access a register of the controller in such a + * case, it will lead to a silent CPU stall. Let's make sure we + * prevent such a case. + */ + rate = clk_get_rate(vc4_hdmi->hsm_clock); + if (!rate) { + ret = -EINVAL; + goto err_disable_clk; + } + if (vc4_hdmi->variant->reset) vc4_hdmi->variant->reset(vc4_hdmi); @@ -2896,6 +2921,10 @@ static int vc4_hdmi_runtime_resume(struct device *dev) #endif return 0; + +err_disable_clk: + clk_disable_unprepare(vc4_hdmi->hsm_clock); + return ret; } static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data) |