diff options
Diffstat (limited to 'drivers/gpu/drm/amd')
47 files changed, 475 insertions, 301 deletions
diff --git a/drivers/gpu/drm/amd/acp/Kconfig b/drivers/gpu/drm/amd/acp/Kconfig index ca77ec10147c..e503e3d6d920 100644 --- a/drivers/gpu/drm/amd/acp/Kconfig +++ b/drivers/gpu/drm/amd/acp/Kconfig @@ -2,6 +2,7 @@ menu "ACP (Audio CoProcessor) Configuration" config DRM_AMD_ACP bool "Enable AMD Audio CoProcessor IP support" + depends on DRM_AMDGPU select MFD_CORE select PM_GENERIC_DOMAINS if PM help diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 2a009c398dcb..01c36b8d6222 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -602,6 +602,8 @@ int amdgpu_sync_wait(struct amdgpu_sync *sync); void amdgpu_sync_free(struct amdgpu_sync *sync); int amdgpu_sync_init(void); void amdgpu_sync_fini(void); +int amdgpu_fence_slab_init(void); +void amdgpu_fence_slab_fini(void); /* * GART structures, functions & helpers @@ -797,6 +799,7 @@ struct amdgpu_ring { unsigned cond_exe_offs; u64 cond_exe_gpu_addr; volatile u32 *cond_exe_cpu_addr; + int vmid; }; /* @@ -934,7 +937,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr, uint32_t gds_base, uint32_t gds_size, uint32_t gws_base, uint32_t gws_size, - uint32_t oa_base, uint32_t oa_size); + uint32_t oa_base, uint32_t oa_size, + bool vmid_switch); void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id); uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr); int amdgpu_vm_update_page_directory(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 199f76baf22c..8943099eb135 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -696,6 +696,17 @@ static uint32_t fw_type_convert(struct cgs_device *cgs_device, uint32_t fw_type) return result; } +static int amdgpu_cgs_rel_firmware(struct cgs_device *cgs_device, enum cgs_ucode_id type) +{ + CGS_FUNC_ADEV; + if ((CGS_UCODE_ID_SMU == type) || (CGS_UCODE_ID_SMU_SK == type)) { + release_firmware(adev->pm.fw); + return 0; + } + /* cannot release other firmware because they are not created by cgs */ + return -EINVAL; +} + static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, enum cgs_ucode_id type, struct cgs_firmware_info *info) @@ -1125,6 +1136,7 @@ static const struct cgs_ops amdgpu_cgs_ops = { amdgpu_cgs_pm_query_clock_limits, amdgpu_cgs_set_camera_voltages, amdgpu_cgs_get_firmware_info, + amdgpu_cgs_rel_firmware, amdgpu_cgs_set_powergating_state, amdgpu_cgs_set_clockgating_state, amdgpu_cgs_get_active_displays_info, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 60a0c9ac11b2..cb07da41152b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -194,12 +194,12 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) bpc = 8; DRM_DEBUG("%s: HDMI deep color 10 bpc exceeds max tmds clock. Using %d bpc.\n", connector->name, bpc); - } else if (bpc > 8) { - /* max_tmds_clock missing, but hdmi spec mandates it for deep color. */ - DRM_DEBUG("%s: Required max tmds clock for HDMI deep color missing. Using 8 bpc.\n", - connector->name); - bpc = 8; } + } else if (bpc > 8) { + /* max_tmds_clock missing, but hdmi spec mandates it for deep color. */ + DRM_DEBUG("%s: Required max tmds clock for HDMI deep color missing. Using 8 bpc.\n", + connector->name); + bpc = 8; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index bb8b149786d7..964f31404f17 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -827,8 +827,10 @@ static uint32_t cail_ioreg_read(struct card_info *info, uint32_t reg) */ static void amdgpu_atombios_fini(struct amdgpu_device *adev) { - if (adev->mode_info.atom_context) + if (adev->mode_info.atom_context) { kfree(adev->mode_info.atom_context->scratch); + kfree(adev->mode_info.atom_context->iio); + } kfree(adev->mode_info.atom_context); adev->mode_info.atom_context = NULL; kfree(adev->mode_info.atom_card_info); @@ -1325,6 +1327,11 @@ static int amdgpu_fini(struct amdgpu_device *adev) adev->ip_block_status[i].valid = false; } + for (i = adev->num_ip_blocks - 1; i >= 0; i--) { + if (adev->ip_blocks[i].funcs->late_fini) + adev->ip_blocks[i].funcs->late_fini((void *)adev); + } + return 0; } @@ -1513,8 +1520,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, amdgpu_atombios_has_gpu_virtualization_table(adev); /* Post card if necessary */ - if (!amdgpu_card_posted(adev) || - adev->virtualization.supports_sr_iov) { + if (!amdgpu_card_posted(adev)) { if (!adev->bios) { dev_err(adev->dev, "Card not posted and no BIOS - ignoring\n"); return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 1dab5f2b725b..f888c015f76c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -50,9 +50,11 @@ * KMS wrapper. * - 3.0.0 - initial driver * - 3.1.0 - allow reading more status registers (GRBM, SRBM, SDMA, CP) + * - 3.2.0 - GFX8: Uses EOP_TC_WB_ACTION_EN, so UMDs don't have to do the same + * at the end of IBs. */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 1 +#define KMS_DRIVER_MINOR 2 #define KMS_DRIVER_PATCHLEVEL 0 int amdgpu_vram_limit = 0; @@ -279,14 +281,26 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x98E4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_STONEY|AMD_IS_APU}, /* Polaris11 */ {0x1002, 0x67E0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, - {0x1002, 0x67E1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, + {0x1002, 0x67E3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, {0x1002, 0x67E8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, - {0x1002, 0x67E9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, {0x1002, 0x67EB, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, + {0x1002, 0x67EF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, {0x1002, 0x67FF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, + {0x1002, 0x67E1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, + {0x1002, 0x67E7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, + {0x1002, 0x67E9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS11}, /* Polaris10 */ {0x1002, 0x67C0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67C1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67C2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67C4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67C7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, {0x1002, 0x67DF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67C8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67C9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67CA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67CC, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, + {0x1002, 0x67CF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS10}, {0, 0, 0} }; @@ -563,9 +577,12 @@ static struct pci_driver amdgpu_kms_pci_driver = { .driver.pm = &amdgpu_pm_ops, }; + + static int __init amdgpu_init(void) { amdgpu_sync_init(); + amdgpu_fence_slab_init(); if (vgacon_text_force()) { DRM_ERROR("VGACON disables amdgpu kernel modesetting.\n"); return -EINVAL; @@ -576,7 +593,6 @@ static int __init amdgpu_init(void) driver->driver_features |= DRIVER_MODESET; driver->num_ioctls = amdgpu_max_kms_ioctl; amdgpu_register_atpx_handler(); - /* let modprobe override vga console setting */ return drm_pci_init(driver, pdriver); } @@ -587,6 +603,7 @@ static void __exit amdgpu_exit(void) drm_pci_exit(driver, pdriver); amdgpu_unregister_atpx_handler(); amdgpu_sync_fini(); + amdgpu_fence_slab_fini(); } module_init(amdgpu_init); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index ba9c04283d01..d1558768cfb7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -55,8 +55,21 @@ struct amdgpu_fence { }; static struct kmem_cache *amdgpu_fence_slab; -static atomic_t amdgpu_fence_slab_ref = ATOMIC_INIT(0); +int amdgpu_fence_slab_init(void) +{ + amdgpu_fence_slab = kmem_cache_create( + "amdgpu_fence", sizeof(struct amdgpu_fence), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!amdgpu_fence_slab) + return -ENOMEM; + return 0; +} + +void amdgpu_fence_slab_fini(void) +{ + kmem_cache_destroy(amdgpu_fence_slab); +} /* * Cast helper */ @@ -396,13 +409,6 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring, */ int amdgpu_fence_driver_init(struct amdgpu_device *adev) { - if (atomic_inc_return(&amdgpu_fence_slab_ref) == 1) { - amdgpu_fence_slab = kmem_cache_create( - "amdgpu_fence", sizeof(struct amdgpu_fence), 0, - SLAB_HWCACHE_ALIGN, NULL); - if (!amdgpu_fence_slab) - return -ENOMEM; - } if (amdgpu_debugfs_fence_init(adev)) dev_err(adev->dev, "fence debugfs file creation failed\n"); @@ -437,13 +443,10 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev) amd_sched_fini(&ring->sched); del_timer_sync(&ring->fence_drv.fallback_timer); for (j = 0; j <= ring->fence_drv.num_fences_mask; ++j) - fence_put(ring->fence_drv.fences[i]); + fence_put(ring->fence_drv.fences[j]); kfree(ring->fence_drv.fences); ring->fence_drv.initialized = false; } - - if (atomic_dec_and_test(&amdgpu_fence_slab_ref)) - kmem_cache_destroy(amdgpu_fence_slab); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 34e35423b78e..7a0b1e50f293 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -122,6 +122,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, bool skip_preamble, need_ctx_switch; unsigned patch_offset = ~0; struct amdgpu_vm *vm; + int vmid = 0, old_vmid = ring->vmid; struct fence *hwf; uint64_t ctx; @@ -135,9 +136,11 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, if (job) { vm = job->vm; ctx = job->ctx; + vmid = job->vm_id; } else { vm = NULL; ctx = 0; + vmid = 0; } if (!ring->ready) { @@ -163,7 +166,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, r = amdgpu_vm_flush(ring, job->vm_id, job->vm_pd_addr, job->gds_base, job->gds_size, job->gws_base, job->gws_size, - job->oa_base, job->oa_size); + job->oa_base, job->oa_size, + (ring->current_ctx == ctx) && (old_vmid != vmid)); if (r) { amdgpu_ring_undo(ring); return r; @@ -180,7 +184,6 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, need_ctx_switch = ring->current_ctx != ctx; for (i = 0; i < num_ibs; ++i) { ib = &ibs[i]; - /* drop preamble IBs if we don't have a context switch */ if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && skip_preamble) continue; @@ -188,6 +191,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, amdgpu_ring_emit_ib(ring, ib, job ? job->vm_id : 0, need_ctx_switch); need_ctx_switch = false; + ring->vmid = vmid; } if (ring->funcs->emit_hdp_invalidate) @@ -198,6 +202,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, dev_err(adev->dev, "failed to emit fence (%d)\n", r); if (job && job->vm_id) amdgpu_vm_reset_id(adev, job->vm_id); + ring->vmid = old_vmid; amdgpu_ring_undo(ring); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c index 6bd961fb43dc..82256558e0f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c @@ -183,13 +183,6 @@ static int amdgpu_pp_sw_fini(void *handle) if (ret) return ret; -#ifdef CONFIG_DRM_AMD_POWERPLAY - if (adev->pp_enabled) { - amdgpu_pm_sysfs_fini(adev); - amd_powerplay_fini(adev->powerplay.pp_handle); - } -#endif - return ret; } @@ -223,6 +216,22 @@ static int amdgpu_pp_hw_fini(void *handle) return ret; } +static void amdgpu_pp_late_fini(void *handle) +{ +#ifdef CONFIG_DRM_AMD_POWERPLAY + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + if (adev->pp_enabled) { + amdgpu_pm_sysfs_fini(adev); + amd_powerplay_fini(adev->powerplay.pp_handle); + } + + if (adev->powerplay.ip_funcs->late_fini) + adev->powerplay.ip_funcs->late_fini( + adev->powerplay.pp_handle); +#endif +} + static int amdgpu_pp_suspend(void *handle) { int ret = 0; @@ -311,6 +320,7 @@ const struct amd_ip_funcs amdgpu_pp_ip_funcs = { .sw_fini = amdgpu_pp_sw_fini, .hw_init = amdgpu_pp_hw_init, .hw_fini = amdgpu_pp_hw_fini, + .late_fini = amdgpu_pp_late_fini, .suspend = amdgpu_pp_suspend, .resume = amdgpu_pp_resume, .is_idle = amdgpu_pp_is_idle, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 3b02272db678..870f9494252c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -343,6 +343,7 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring) ring->ring = NULL; ring->ring_obj = NULL; + amdgpu_wb_free(ring->adev, ring->cond_exe_offs); amdgpu_wb_free(ring->adev, ring->fence_offs); amdgpu_wb_free(ring->adev, ring->rptr_offs); amdgpu_wb_free(ring->adev, ring->wptr_offs); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c index 8bf84efafb04..48618ee324eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c @@ -115,6 +115,7 @@ int amdgpu_sa_bo_manager_start(struct amdgpu_device *adev, return r; } r = amdgpu_bo_kmap(sa_manager->bo, &sa_manager->cpu_ptr); + memset(sa_manager->cpu_ptr, 0, sa_manager->size); amdgpu_bo_unreserve(sa_manager->bo); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 01abfc21b4a2..e19520c4b4b6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -253,19 +253,20 @@ int amdgpu_uvd_sw_fini(struct amdgpu_device *adev) { int r; - if (adev->uvd.vcpu_bo == NULL) - return 0; + kfree(adev->uvd.saved_bo); amd_sched_entity_fini(&adev->uvd.ring.sched, &adev->uvd.entity); - r = amdgpu_bo_reserve(adev->uvd.vcpu_bo, false); - if (!r) { - amdgpu_bo_kunmap(adev->uvd.vcpu_bo); - amdgpu_bo_unpin(adev->uvd.vcpu_bo); - amdgpu_bo_unreserve(adev->uvd.vcpu_bo); - } + if (adev->uvd.vcpu_bo) { + r = amdgpu_bo_reserve(adev->uvd.vcpu_bo, false); + if (!r) { + amdgpu_bo_kunmap(adev->uvd.vcpu_bo); + amdgpu_bo_unpin(adev->uvd.vcpu_bo); + amdgpu_bo_unreserve(adev->uvd.vcpu_bo); + } - amdgpu_bo_unref(&adev->uvd.vcpu_bo); + amdgpu_bo_unref(&adev->uvd.vcpu_bo); + } amdgpu_ring_fini(&adev->uvd.ring); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index ea708cb94862..62a4c127620f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -53,6 +53,18 @@ /* Special value that no flush is necessary */ #define AMDGPU_VM_NO_FLUSH (~0ll) +/* Local structure. Encapsulate some VM table update parameters to reduce + * the number of function parameters + */ +struct amdgpu_vm_update_params { + /* address where to copy page table entries from */ + uint64_t src; + /* DMA addresses to use for mapping */ + dma_addr_t *pages_addr; + /* indirect buffer to fill with commands */ + struct amdgpu_ib *ib; +}; + /** * amdgpu_vm_num_pde - return the number of page directory entries * @@ -286,7 +298,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr, uint32_t gds_base, uint32_t gds_size, uint32_t gws_base, uint32_t gws_size, - uint32_t oa_base, uint32_t oa_size) + uint32_t oa_base, uint32_t oa_size, + bool vmid_switch) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vm_id *id = &adev->vm_manager.ids[vm_id]; @@ -300,8 +313,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, int r; if (ring->funcs->emit_pipeline_sync && ( - pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed || - ring->type == AMDGPU_RING_TYPE_COMPUTE)) + pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed || vmid_switch)) amdgpu_ring_emit_pipeline_sync(ring); if (ring->funcs->emit_vm_flush && @@ -389,9 +401,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm, * amdgpu_vm_update_pages - helper to call the right asic function * * @adev: amdgpu_device pointer - * @src: address where to copy page table entries from - * @pages_addr: DMA addresses to use for mapping - * @ib: indirect buffer to fill with commands + * @vm_update_params: see amdgpu_vm_update_params definition * @pe: addr of the page entry * @addr: dst addr to write into pe * @count: number of page entries to update @@ -402,29 +412,29 @@ struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm, * to setup the page table using the DMA. */ static void amdgpu_vm_update_pages(struct amdgpu_device *adev, - uint64_t src, - dma_addr_t *pages_addr, - struct amdgpu_ib *ib, + struct amdgpu_vm_update_params + *vm_update_params, uint64_t pe, uint64_t addr, unsigned count, uint32_t incr, uint32_t flags) { trace_amdgpu_vm_set_page(pe, addr, count, incr, flags); - if (src) { - src += (addr >> 12) * 8; - amdgpu_vm_copy_pte(adev, ib, pe, src, count); + if (vm_update_params->src) { + amdgpu_vm_copy_pte(adev, vm_update_params->ib, + pe, (vm_update_params->src + (addr >> 12) * 8), count); - } else if (pages_addr) { - amdgpu_vm_write_pte(adev, ib, pages_addr, pe, addr, - count, incr, flags); + } else if (vm_update_params->pages_addr) { + amdgpu_vm_write_pte(adev, vm_update_params->ib, + vm_update_params->pages_addr, + pe, addr, count, incr, flags); } else if (count < 3) { - amdgpu_vm_write_pte(adev, ib, NULL, pe, addr, + amdgpu_vm_write_pte(adev, vm_update_params->ib, NULL, pe, addr, count, incr, flags); } else { - amdgpu_vm_set_pte_pde(adev, ib, pe, addr, + amdgpu_vm_set_pte_pde(adev, vm_update_params->ib, pe, addr, count, incr, flags); } } @@ -444,10 +454,12 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, struct amdgpu_ring *ring; struct fence *fence = NULL; struct amdgpu_job *job; + struct amdgpu_vm_update_params vm_update_params; unsigned entries; uint64_t addr; int r; + memset(&vm_update_params, 0, sizeof(vm_update_params)); ring = container_of(vm->entity.sched, struct amdgpu_ring, sched); r = reservation_object_reserve_shared(bo->tbo.resv); @@ -465,7 +477,8 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, if (r) goto error; - amdgpu_vm_update_pages(adev, 0, NULL, &job->ibs[0], addr, 0, entries, + vm_update_params.ib = &job->ibs[0]; + amdgpu_vm_update_pages(adev, &vm_update_params, addr, 0, entries, 0, 0); amdgpu_ring_pad_ib(ring, &job->ibs[0]); @@ -538,11 +551,12 @@ int amdgpu_vm_update_page_directory(struct amdgpu_device *adev, uint64_t last_pde = ~0, last_pt = ~0; unsigned count = 0, pt_idx, ndw; struct amdgpu_job *job; - struct amdgpu_ib *ib; + struct amdgpu_vm_update_params vm_update_params; struct fence *fence = NULL; int r; + memset(&vm_update_params, 0, sizeof(vm_update_params)); ring = container_of(vm->entity.sched, struct amdgpu_ring, sched); /* padding, etc. */ @@ -555,7 +569,7 @@ int amdgpu_vm_update_page_directory(struct amdgpu_device *adev, if (r) return r; - ib = &job->ibs[0]; + vm_update_params.ib = &job->ibs[0]; /* walk over the address space and update the page directory */ for (pt_idx = 0; pt_idx <= vm->max_pde_used; ++pt_idx) { @@ -575,7 +589,7 @@ int amdgpu_vm_update_page_directory(struct amdgpu_device *adev, ((last_pt + incr * count) != pt)) { if (count) { - amdgpu_vm_update_pages(adev, 0, NULL, ib, + amdgpu_vm_update_pages(adev, &vm_update_params, last_pde, last_pt, count, incr, AMDGPU_PTE_VALID); @@ -590,14 +604,15 @@ int amdgpu_vm_update_page_directory(struct amdgpu_device *adev, } if (count) - amdgpu_vm_update_pages(adev, 0, NULL, ib, last_pde, last_pt, - count, incr, AMDGPU_PTE_VALID); + amdgpu_vm_update_pages(adev, &vm_update_params, + last_pde, last_pt, + count, incr, AMDGPU_PTE_VALID); - if (ib->length_dw != 0) { - amdgpu_ring_pad_ib(ring, ib); + if (vm_update_params.ib->length_dw != 0) { + amdgpu_ring_pad_ib(ring, vm_update_params.ib); amdgpu_sync_resv(adev, &job->sync, pd->tbo.resv, AMDGPU_FENCE_OWNER_VM); - WARN_ON(ib->length_dw > ndw); + WARN_ON(vm_update_params.ib->length_dw > ndw); r = amdgpu_job_submit(job, ring, &vm->entity, AMDGPU_FENCE_OWNER_VM, &fence); if (r) @@ -623,18 +638,15 @@ error_free: * amdgpu_vm_frag_ptes - add fragment information to PTEs * * @adev: amdgpu_device pointer - * @src: address where to copy page table entries from - * @pages_addr: DMA addresses to use for mapping - * @ib: IB for the update + * @vm_update_params: see amdgpu_vm_update_params definition * @pe_start: first PTE to handle * @pe_end: last PTE to handle * @addr: addr those PTEs should point to * @flags: hw mapping flags */ static void amdgpu_vm_frag_ptes(struct amdgpu_device *adev, - uint64_t src, - dma_addr_t *pages_addr, - struct amdgpu_ib *ib, + struct amdgpu_vm_update_params + *vm_update_params, uint64_t pe_start, uint64_t pe_end, uint64_t addr, uint32_t flags) { @@ -671,11 +683,11 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_device *adev, return; /* system pages are non continuously */ - if (src || pages_addr || !(flags & AMDGPU_PTE_VALID) || - (frag_start >= frag_end)) { + if (vm_update_params->src || vm_update_params->pages_addr || + !(flags & AMDGPU_PTE_VALID) || (frag_start >= frag_end)) { count = (pe_end - pe_start) / 8; - amdgpu_vm_update_pages(adev, src, pages_addr, ib, pe_start, + amdgpu_vm_update_pages(adev, vm_update_params, pe_start, addr, count, AMDGPU_GPU_PAGE_SIZE, flags); return; @@ -684,21 +696,21 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_device *adev, /* handle the 4K area at the beginning */ if (pe_start != frag_start) { count = (frag_start - pe_start) / 8; - amdgpu_vm_update_pages(adev, 0, NULL, ib, pe_start, addr, + amdgpu_vm_update_pages(adev, vm_update_params, pe_start, addr, count, AMDGPU_GPU_PAGE_SIZE, flags); addr += AMDGPU_GPU_PAGE_SIZE * count; } /* handle the area in the middle */ count = (frag_end - frag_start) / 8; - amdgpu_vm_update_pages(adev, 0, NULL, ib, frag_start, addr, count, + amdgpu_vm_update_pages(adev, vm_update_params, frag_start, addr, count, AMDGPU_GPU_PAGE_SIZE, flags | frag_flags); /* handle the 4K area at the end */ if (frag_end != pe_end) { addr += AMDGPU_GPU_PAGE_SIZE * count; count = (pe_end - frag_end) / 8; - amdgpu_vm_update_pages(adev, 0, NULL, ib, frag_end, addr, + amdgpu_vm_update_pages(adev, vm_update_params, frag_end, addr, count, AMDGPU_GPU_PAGE_SIZE, flags); } } @@ -707,8 +719,7 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_device *adev, * amdgpu_vm_update_ptes - make sure that page tables are valid * * @adev: amdgpu_device pointer - * @src: address where to copy page table entries from - * @pages_addr: DMA addresses to use for mapping + * @vm_update_params: see amdgpu_vm_update_params definition * @vm: requested vm * @start: start of GPU address range * @end: end of GPU address range @@ -718,10 +729,9 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_device *adev, * Update the page tables in the range @start - @end. */ static void amdgpu_vm_update_ptes(struct amdgpu_device *adev, - uint64_t src, - dma_addr_t *pages_addr, + struct amdgpu_vm_update_params + *vm_update_params, struct amdgpu_vm *vm, - struct amdgpu_ib *ib, uint64_t start, uint64_t end, uint64_t dst, uint32_t flags) { @@ -747,7 +757,7 @@ static void amdgpu_vm_update_ptes(struct amdgpu_device *adev, if (last_pe_end != pe_start) { - amdgpu_vm_frag_ptes(adev, src, pages_addr, ib, + amdgpu_vm_frag_ptes(adev, vm_update_params, last_pe_start, last_pe_end, last_dst, flags); @@ -762,7 +772,7 @@ static void amdgpu_vm_update_ptes(struct amdgpu_device *adev, dst += nptes * AMDGPU_GPU_PAGE_SIZE; } - amdgpu_vm_frag_ptes(adev, src, pages_addr, ib, last_pe_start, + amdgpu_vm_frag_ptes(adev, vm_update_params, last_pe_start, last_pe_end, last_dst, flags); } @@ -794,11 +804,14 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, void *owner = AMDGPU_FENCE_OWNER_VM; unsigned nptes, ncmds, ndw; struct amdgpu_job *job; - struct amdgpu_ib *ib; + struct amdgpu_vm_update_params vm_update_params; struct fence *f = NULL; int r; ring = container_of(vm->entity.sched, struct amdgpu_ring, sched); + memset(&vm_update_params, 0, sizeof(vm_update_params)); + vm_update_params.src = src; + vm_update_params.pages_addr = pages_addr; /* sync to everything on unmapping */ if (!(flags & AMDGPU_PTE_VALID)) @@ -815,11 +828,11 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, /* padding, etc. */ ndw = 64; - if (src) { + if (vm_update_params.src) { /* only copy commands needed */ ndw += ncmds * 7; - } else if (pages_addr) { + } else if (vm_update_params.pages_addr) { /* header for write data commands */ ndw += ncmds * 4; @@ -838,7 +851,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, if (r) return r; - ib = &job->ibs[0]; + vm_update_params.ib = &job->ibs[0]; r = amdgpu_sync_resv(adev, &job->sync, vm->page_directory->tbo.resv, owner); @@ -849,11 +862,11 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, if (r) goto error_free; - amdgpu_vm_update_ptes(adev, src, pages_addr, vm, ib, start, + amdgpu_vm_update_ptes(adev, &vm_update_params, vm, start, last + 1, addr, flags); - amdgpu_ring_pad_ib(ring, ib); - WARN_ON(ib->length_dw > ndw); + amdgpu_ring_pad_ib(ring, vm_update_params.ib); + WARN_ON(vm_update_params.ib->length_dw > ndw); r = amdgpu_job_submit(job, ring, &vm->entity, AMDGPU_FENCE_OWNER_VM, &f); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c index ea407db1fbcf..5ec1f1e9c983 100644 --- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c @@ -6221,6 +6221,9 @@ static int ci_dpm_sw_fini(void *handle) ci_dpm_fini(adev); mutex_unlock(&adev->pm.mutex); + release_firmware(adev->pm.fw); + adev->pm.fw = NULL; + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/cik_ih.c b/drivers/gpu/drm/amd/amdgpu/cik_ih.c index 845c21b1b2ee..be3d6f79a864 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_ih.c @@ -103,7 +103,6 @@ static void cik_ih_disable_interrupts(struct amdgpu_device *adev) */ static int cik_ih_irq_init(struct amdgpu_device *adev) { - int ret = 0; int rb_bufsz; u32 interrupt_cntl, ih_cntl, ih_rb_cntl; u64 wptr_off; @@ -156,7 +155,7 @@ static int cik_ih_irq_init(struct amdgpu_device *adev) /* enable irqs */ cik_ih_enable_interrupts(adev); - return ret; + return 0; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c index 518dca43b133..9dc4e24e31e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c @@ -66,6 +66,16 @@ MODULE_FIRMWARE("radeon/mullins_sdma1.bin"); u32 amdgpu_cik_gpu_check_soft_reset(struct amdgpu_device *adev); + +static void cik_sdma_free_microcode(struct amdgpu_device *adev) +{ + int i; + for (i = 0; i < adev->sdma.num_instances; i++) { + release_firmware(adev->sdma.instance[i].fw); + adev->sdma.instance[i].fw = NULL; + } +} + /* * sDMA - System DMA * Starting with CIK, the GPU has new asynchronous @@ -419,6 +429,8 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev) /* Initialize the ring buffer's read and write pointers */ WREG32(mmSDMA0_GFX_RB_RPTR + sdma_offsets[i], 0); WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], 0); + WREG32(mmSDMA0_GFX_IB_RPTR + sdma_offsets[i], 0); + WREG32(mmSDMA0_GFX_IB_OFFSET + sdma_offsets[i], 0); /* set the wb address whether it's enabled or not */ WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i], @@ -446,7 +458,12 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev) WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); ring->ready = true; + } + + cik_sdma_enable(adev, true); + for (i = 0; i < adev->sdma.num_instances; i++) { + ring = &adev->sdma.instance[i].ring; r = amdgpu_ring_test_ring(ring); if (r) { ring->ready = false; @@ -529,8 +546,8 @@ static int cik_sdma_start(struct amdgpu_device *adev) if (r) return r; - /* unhalt the MEs */ - cik_sdma_enable(adev, true); + /* halt the engine before programing */ + cik_sdma_enable(adev, false); /* start the gfx rings and rlc compute queues */ r = cik_sdma_gfx_resume(adev); @@ -998,6 +1015,7 @@ static int cik_sdma_sw_fini(void *handle) for (i = 0; i < adev->sdma.num_instances; i++) amdgpu_ring_fini(&adev->sdma.instance[i].ring); + cik_sdma_free_microcode(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c index fa4449e126e6..933e425a8154 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c @@ -1579,7 +1579,6 @@ static int cz_dpm_update_sclk_limit(struct amdgpu_device *adev) static int cz_dpm_set_deep_sleep_sclk_threshold(struct amdgpu_device *adev) { - int ret = 0; struct cz_power_info *pi = cz_get_pi(adev); if (pi->caps_sclk_ds) { @@ -1588,20 +1587,19 @@ static int cz_dpm_set_deep_sleep_sclk_threshold(struct amdgpu_device *adev) CZ_MIN_DEEP_SLEEP_SCLK); } - return ret; + return 0; } /* ?? without dal support, is this still needed in setpowerstate list*/ static int cz_dpm_set_watermark_threshold(struct amdgpu_device *adev) { - int ret = 0; struct cz_power_info *pi = cz_get_pi(adev); cz_send_msg_to_smc_with_parameter(adev, PPSMC_MSG_SetWatermarkFrequency, pi->sclk_dpm.soft_max_clk); - return ret; + return 0; } static int cz_dpm_enable_nbdpm(struct amdgpu_device *adev) @@ -1636,7 +1634,6 @@ static void cz_dpm_nbdpm_lm_pstate_enable(struct amdgpu_device *adev, static int cz_dpm_update_low_memory_pstate(struct amdgpu_device *adev) { - int ret = 0; struct cz_power_info *pi = cz_get_pi(adev); struct cz_ps *ps = &pi->requested_ps; @@ -1647,21 +1644,19 @@ static int cz_dpm_update_low_memory_pstate(struct amdgpu_device *adev) cz_dpm_nbdpm_lm_pstate_enable(adev, true); } - return ret; + return 0; } /* with dpm enabled */ static int cz_dpm_set_power_state(struct amdgpu_device *adev) { - int ret = 0; - cz_dpm_update_sclk_limit(adev); cz_dpm_set_deep_sleep_sclk_threshold(adev); cz_dpm_set_watermark_threshold(adev); cz_dpm_enable_nbdpm(adev); cz_dpm_update_low_memory_pstate(adev); - return ret; + return 0; } static void cz_dpm_post_set_power_state(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/cz_ih.c b/drivers/gpu/drm/amd/amdgpu/cz_ih.c index 863cb16f6126..3d23a70b6432 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_ih.c @@ -103,7 +103,6 @@ static void cz_ih_disable_interrupts(struct amdgpu_device *adev) */ static int cz_ih_irq_init(struct amdgpu_device *adev) { - int ret = 0; int rb_bufsz; u32 interrupt_cntl, ih_cntl, ih_rb_cntl; u64 wptr_off; @@ -157,7 +156,7 @@ static int cz_ih_irq_init(struct amdgpu_device *adev) /* enable interrupts */ cz_ih_enable_interrupts(adev); - return ret; + return 0; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index c11b6007af80..af26ec0bc59d 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -137,7 +137,7 @@ static const u32 polaris11_golden_settings_a11[] = mmDCI_CLK_CNTL, 0x00000080, 0x00000000, mmFBC_DEBUG_COMP, 0x000000f0, 0x00000070, mmFBC_DEBUG1, 0xffffffff, 0x00000008, - mmFBC_MISC, 0x9f313fff, 0x14300008, + mmFBC_MISC, 0x9f313fff, 0x14302008, mmHDMI_CONTROL, 0x313f031f, 0x00000011, }; @@ -145,7 +145,7 @@ static const u32 polaris10_golden_settings_a11[] = { mmDCI_CLK_CNTL, 0x00000080, 0x00000000, mmFBC_DEBUG_COMP, 0x000000f0, 0x00000070, - mmFBC_MISC, 0x9f313fff, 0x14300008, + mmFBC_MISC, 0x9f313fff, 0x14302008, mmHDMI_CONTROL, 0x313f031f, 0x00000011, }; diff --git a/drivers/gpu/drm/amd/amdgpu/fiji_dpm.c b/drivers/gpu/drm/amd/amdgpu/fiji_dpm.c index 245cabf06575..ed03b75175d4 100644 --- a/drivers/gpu/drm/amd/amdgpu/fiji_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/fiji_dpm.c @@ -72,6 +72,11 @@ static int fiji_dpm_sw_init(void *handle) static int fiji_dpm_sw_fini(void *handle) { + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + release_firmware(adev->pm.fw); + adev->pm.fw = NULL; + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 7f18a53ab53a..8c6ad1e72f02 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -991,6 +991,22 @@ out: return err; } +static void gfx_v7_0_free_microcode(struct amdgpu_device *adev) +{ + release_firmware(adev->gfx.pfp_fw); + adev->gfx.pfp_fw = NULL; + release_firmware(adev->gfx.me_fw); + adev->gfx.me_fw = NULL; + release_firmware(adev->gfx.ce_fw); + adev->gfx.ce_fw = NULL; + release_firmware(adev->gfx.mec_fw); + adev->gfx.mec_fw = NULL; + release_firmware(adev->gfx.mec2_fw); + adev->gfx.mec2_fw = NULL; + release_firmware(adev->gfx.rlc_fw); + adev->gfx.rlc_fw = NULL; +} + /** * gfx_v7_0_tiling_mode_table_init - init the hw tiling table * @@ -4489,6 +4505,7 @@ static int gfx_v7_0_sw_fini(void *handle) gfx_v7_0_cp_compute_fini(adev); gfx_v7_0_rlc_fini(adev); gfx_v7_0_mec_fini(adev); + gfx_v7_0_free_microcode(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 92647fbf5b8b..9f6f8669edc3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -267,10 +267,13 @@ static const u32 tonga_mgcg_cgcg_init[] = static const u32 golden_settings_polaris11_a11[] = { + mmCB_HW_CONTROL, 0xfffdf3cf, 0x00006208, mmCB_HW_CONTROL_3, 0x000001ff, 0x00000040, mmDB_DEBUG2, 0xf00fffff, 0x00000400, mmPA_SC_ENHANCE, 0xffffffff, 0x20000001, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000, + mmPA_SC_RASTER_CONFIG, 0x3f3fffff, 0x16000012, + mmPA_SC_RASTER_CONFIG_1, 0x0000003f, 0x00000000, mmRLC_CGCG_CGLS_CTRL, 0x00000003, 0x0001003c, mmRLC_CGCG_CGLS_CTRL_3D, 0xffffffff, 0x0001003c, mmSQ_CONFIG, 0x07f80000, 0x07180000, @@ -284,8 +287,6 @@ static const u32 golden_settings_polaris11_a11[] = static const u32 polaris11_golden_common_all[] = { mmGRBM_GFX_INDEX, 0xffffffff, 0xe0000000, - mmPA_SC_RASTER_CONFIG, 0xffffffff, 0x16000012, - mmPA_SC_RASTER_CONFIG_1, 0xffffffff, 0x00000000, mmGB_ADDR_CONFIG, 0xffffffff, 0x22011002, mmSPI_RESOURCE_RESERVE_CU_0, 0xffffffff, 0x00000800, mmSPI_RESOURCE_RESERVE_CU_1, 0xffffffff, 0x00000800, @@ -296,6 +297,7 @@ static const u32 polaris11_golden_common_all[] = static const u32 golden_settings_polaris10_a11[] = { mmATC_MISC_CG, 0x000c0fc0, 0x000c0200, + mmCB_HW_CONTROL, 0xfffdf3cf, 0x00006208, mmCB_HW_CONTROL_3, 0x000001ff, 0x00000040, mmDB_DEBUG2, 0xf00fffff, 0x00000400, mmPA_SC_ENHANCE, 0xffffffff, 0x20000001, @@ -834,6 +836,26 @@ err1: return r; } + +static void gfx_v8_0_free_microcode(struct amdgpu_device *adev) { + release_firmware(adev->gfx.pfp_fw); + adev->gfx.pfp_fw = NULL; + release_firmware(adev->gfx.me_fw); + adev->gfx.me_fw = NULL; + release_firmware(adev->gfx.ce_fw); + adev->gfx.ce_fw = NULL; + release_firmware(adev->gfx.rlc_fw); + adev->gfx.rlc_fw = NULL; + release_firmware(adev->gfx.mec_fw); + adev->gfx.mec_fw = NULL; + if ((adev->asic_type != CHIP_STONEY) && + (adev->asic_type != CHIP_TOPAZ)) + release_firmware(adev->gfx.mec2_fw); + adev->gfx.mec2_fw = NULL; + + kfree(adev->gfx.rlc.register_list_format); +} + static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; @@ -1981,7 +2003,7 @@ static int gfx_v8_0_sw_fini(void *handle) gfx_v8_0_rlc_fini(adev); - kfree(adev->gfx.rlc.register_list_format); + gfx_v8_0_free_microcode(adev); return 0; } @@ -3972,11 +3994,15 @@ static int gfx_v8_0_cp_gfx_start(struct amdgpu_device *adev) amdgpu_ring_write(ring, 0x3a00161a); amdgpu_ring_write(ring, 0x0000002e); break; - case CHIP_TOPAZ: case CHIP_CARRIZO: amdgpu_ring_write(ring, 0x00000002); amdgpu_ring_write(ring, 0x00000000); break; + case CHIP_TOPAZ: + amdgpu_ring_write(ring, adev->gfx.config.num_rbs == 1 ? + 0x00000000 : 0x00000002); + amdgpu_ring_write(ring, 0x00000000); + break; case CHIP_STONEY: amdgpu_ring_write(ring, 0x00000000); amdgpu_ring_write(ring, 0x00000000); @@ -5725,6 +5751,7 @@ static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr, amdgpu_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4)); amdgpu_ring_write(ring, (EOP_TCL1_ACTION_EN | EOP_TC_ACTION_EN | + EOP_TC_WB_ACTION_EN | EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | EVENT_INDEX(5))); amdgpu_ring_write(ring, addr & 0xfffffffc); diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_dpm.c b/drivers/gpu/drm/amd/amdgpu/iceland_dpm.c index 460bc8ad37e6..825ccd63f2dc 100644 --- a/drivers/gpu/drm/amd/amdgpu/iceland_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/iceland_dpm.c @@ -72,6 +72,11 @@ static int iceland_dpm_sw_init(void *handle) static int iceland_dpm_sw_fini(void *handle) { + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + release_firmware(adev->pm.fw); + adev->pm.fw = NULL; + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c index 39bfc52d0b42..3b8906ce3511 100644 --- a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c @@ -103,7 +103,6 @@ static void iceland_ih_disable_interrupts(struct amdgpu_device *adev) */ static int iceland_ih_irq_init(struct amdgpu_device *adev) { - int ret = 0; int rb_bufsz; u32 interrupt_cntl, ih_cntl, ih_rb_cntl; u64 wptr_off; @@ -157,7 +156,7 @@ static int iceland_ih_irq_init(struct amdgpu_device *adev) /* enable interrupts */ iceland_ih_enable_interrupts(adev); - return ret; + return 0; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c index b45f54714574..a789a863d677 100644 --- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c @@ -2252,7 +2252,7 @@ static void kv_apply_state_adjust_rules(struct amdgpu_device *adev, if (pi->caps_stable_p_state) { stable_p_state_sclk = (max_limits->sclk * 75) / 100; - for (i = table->count - 1; i >= 0; i++) { + for (i = table->count - 1; i >= 0; i--) { if (stable_p_state_sclk >= table->entries[i].clk) { stable_p_state_sclk = table->entries[i].clk; break; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index f4c3130d3fdb..b556bd0a8797 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -105,6 +105,15 @@ static void sdma_v2_4_init_golden_registers(struct amdgpu_device *adev) } } +static void sdma_v2_4_free_microcode(struct amdgpu_device *adev) +{ + int i; + for (i = 0; i < adev->sdma.num_instances; i++) { + release_firmware(adev->sdma.instance[i].fw); + adev->sdma.instance[i].fw = NULL; + } +} + /** * sdma_v2_4_init_microcode - load ucode images from disk * @@ -461,6 +470,8 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) /* Initialize the ring buffer's read and write pointers */ WREG32(mmSDMA0_GFX_RB_RPTR + sdma_offsets[i], 0); WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], 0); + WREG32(mmSDMA0_GFX_IB_RPTR + sdma_offsets[i], 0); + WREG32(mmSDMA0_GFX_IB_OFFSET + sdma_offsets[i], 0); /* set the wb address whether it's enabled or not */ WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i], @@ -489,7 +500,11 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); ring->ready = true; + } + sdma_v2_4_enable(adev, true); + for (i = 0; i < adev->sdma.num_instances; i++) { + ring = &adev->sdma.instance[i].ring; r = amdgpu_ring_test_ring(ring); if (r) { ring->ready = false; @@ -580,8 +595,8 @@ static int sdma_v2_4_start(struct amdgpu_device *adev) return -EINVAL; } - /* unhalt the MEs */ - sdma_v2_4_enable(adev, true); + /* halt the engine before programing */ + sdma_v2_4_enable(adev, false); /* start the gfx rings and rlc compute queues */ r = sdma_v2_4_gfx_resume(adev); @@ -1012,6 +1027,7 @@ static int sdma_v2_4_sw_fini(void *handle) for (i = 0; i < adev->sdma.num_instances; i++) amdgpu_ring_fini(&adev->sdma.instance[i].ring); + sdma_v2_4_free_microcode(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index 063f08a9957a..532ea88da66a 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -109,10 +109,12 @@ static const u32 fiji_mgcg_cgcg_init[] = static const u32 golden_settings_polaris11_a11[] = { mmSDMA0_CHICKEN_BITS, 0xfc910007, 0x00810007, + mmSDMA0_CLK_CTRL, 0xff000fff, 0x00000000, mmSDMA0_GFX_IB_CNTL, 0x800f0111, 0x00000100, mmSDMA0_RLC0_IB_CNTL, 0x800f0111, 0x00000100, mmSDMA0_RLC1_IB_CNTL, 0x800f0111, 0x00000100, mmSDMA1_CHICKEN_BITS, 0xfc910007, 0x00810007, + mmSDMA1_CLK_CTRL, 0xff000fff, 0x00000000, mmSDMA1_GFX_IB_CNTL, 0x800f0111, 0x00000100, mmSDMA1_RLC0_IB_CNTL, 0x800f0111, 0x00000100, mmSDMA1_RLC1_IB_CNTL, 0x800f0111, 0x00000100, @@ -234,6 +236,15 @@ static void sdma_v3_0_init_golden_registers(struct amdgpu_device *adev) } } +static void sdma_v3_0_free_microcode(struct amdgpu_device *adev) +{ + int i; + for (i = 0; i < adev->sdma.num_instances; i++) { + release_firmware(adev->sdma.instance[i].fw); + adev->sdma.instance[i].fw = NULL; + } +} + /** * sdma_v3_0_init_microcode - load ucode images from disk * @@ -670,6 +681,8 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev) /* Initialize the ring buffer's read and write pointers */ WREG32(mmSDMA0_GFX_RB_RPTR + sdma_offsets[i], 0); WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], 0); + WREG32(mmSDMA0_GFX_IB_RPTR + sdma_offsets[i], 0); + WREG32(mmSDMA0_GFX_IB_OFFSET + sdma_offsets[i], 0); /* set the wb address whether it's enabled or not */ WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i], @@ -709,7 +722,15 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev) WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); ring->ready = true; + } + + /* unhalt the MEs */ + sdma_v3_0_enable(adev, true); + /* enable sdma ring preemption */ + sdma_v3_0_ctx_switch_enable(adev, true); + for (i = 0; i < adev->sdma.num_instances; i++) { + ring = &adev->sdma.instance[i].ring; r = amdgpu_ring_test_ring(ring); if (r) { ring->ready = false; @@ -802,10 +823,9 @@ static int sdma_v3_0_start(struct amdgpu_device *adev) } } - /* unhalt the MEs */ - sdma_v3_0_enable(adev, true); - /* enable sdma ring preemption */ - sdma_v3_0_ctx_switch_enable(adev, true); + /* disble sdma engine before programing it */ + sdma_v3_0_ctx_switch_enable(adev, false); + sdma_v3_0_enable(adev, false); /* start the gfx rings and rlc compute queues */ r = sdma_v3_0_gfx_resume(adev); @@ -1245,6 +1265,7 @@ static int sdma_v3_0_sw_fini(void *handle) for (i = 0; i < adev->sdma.num_instances; i++) amdgpu_ring_fini(&adev->sdma.instance[i].ring); + sdma_v3_0_free_microcode(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_dpm.c b/drivers/gpu/drm/amd/amdgpu/tonga_dpm.c index b7615cefcac4..f06f6f4dc3a8 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_dpm.c @@ -71,6 +71,11 @@ static int tonga_dpm_sw_init(void *handle) static int tonga_dpm_sw_fini(void *handle) { + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + release_firmware(adev->pm.fw); + adev->pm.fw = NULL; + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c index f036af937fbc..c92055805a45 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c @@ -99,7 +99,6 @@ static void tonga_ih_disable_interrupts(struct amdgpu_device *adev) */ static int tonga_ih_irq_init(struct amdgpu_device *adev) { - int ret = 0; int rb_bufsz; u32 interrupt_cntl, ih_rb_cntl, ih_doorbell_rtpr; u64 wptr_off; @@ -165,7 +164,7 @@ static int tonga_ih_irq_init(struct amdgpu_device *adev) /* enable interrupts */ tonga_ih_enable_interrupts(adev); - return ret; + return 0; } /** diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index ac005796b71c..7708d90b9da9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -242,13 +242,19 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn, pqm_uninit(&p->pqm); /* Iterate over all process device data structure and check - * if we should reset all wavefronts */ - list_for_each_entry(pdd, &p->per_device_data, per_device_list) + * if we should delete debug managers and reset all wavefronts + */ + list_for_each_entry(pdd, &p->per_device_data, per_device_list) { + if ((pdd->dev->dbgmgr) && + (pdd->dev->dbgmgr->pasid == p->pasid)) + kfd_dbgmgr_destroy(pdd->dev->dbgmgr); + if (pdd->reset_wavefronts) { pr_warn("amdkfd: Resetting all wave fronts\n"); dbgdev_wave_reset_wavefronts(pdd->dev, p); pdd->reset_wavefronts = false; } + } mutex_unlock(&p->mutex); @@ -404,42 +410,52 @@ void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid) idx = srcu_read_lock(&kfd_processes_srcu); + /* + * Look for the process that matches the pasid. If there is no such + * process, we either released it in amdkfd's own notifier, or there + * is a bug. Unfortunately, there is no way to tell... + */ hash_for_each_rcu(kfd_processes_table, i, p, kfd_processes) - if (p->pasid == pasid) - break; + if (p->pasid == pasid) { - srcu_read_unlock(&kfd_processes_srcu, idx); + srcu_read_unlock(&kfd_processes_srcu, idx); - BUG_ON(p->pasid != pasid); + pr_debug("Unbinding process %d from IOMMU\n", pasid); - mutex_lock(&p->mutex); + mutex_lock(&p->mutex); - if ((dev->dbgmgr) && (dev->dbgmgr->pasid == p->pasid)) - kfd_dbgmgr_destroy(dev->dbgmgr); + if ((dev->dbgmgr) && (dev->dbgmgr->pasid == p->pasid)) + kfd_dbgmgr_destroy(dev->dbgmgr); - pqm_uninit(&p->pqm); + pqm_uninit(&p->pqm); - pdd = kfd_get_process_device_data(dev, p); + pdd = kfd_get_process_device_data(dev, p); - if (!pdd) { - mutex_unlock(&p->mutex); - return; - } + if (!pdd) { + mutex_unlock(&p->mutex); + return; + } - if (pdd->reset_wavefronts) { - dbgdev_wave_reset_wavefronts(pdd->dev, p); - pdd->reset_wavefronts = false; - } + if (pdd->reset_wavefronts) { + dbgdev_wave_reset_wavefronts(pdd->dev, p); + pdd->reset_wavefronts = false; + } - /* - * Just mark pdd as unbound, because we still need it to call - * amd_iommu_unbind_pasid() in when the process exits. - * We don't call amd_iommu_unbind_pasid() here - * because the IOMMU called us. - */ - pdd->bound = false; + /* + * Just mark pdd as unbound, because we still need it + * to call amd_iommu_unbind_pasid() in when the + * process exits. + * We don't call amd_iommu_unbind_pasid() here + * because the IOMMU called us. + */ + pdd->bound = false; - mutex_unlock(&p->mutex); + mutex_unlock(&p->mutex); + + return; + } + + srcu_read_unlock(&kfd_processes_srcu, idx); } struct kfd_process_device *kfd_get_first_process_device_data(struct kfd_process *p) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 74909e72a009..884c96f50c3d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -666,7 +666,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.simd_count); if (dev->mem_bank_count < dev->node_props.mem_banks_count) { - pr_warn("kfd: mem_banks_count truncated from %d to %d\n", + pr_info_once("kfd: mem_banks_count truncated from %d to %d\n", dev->node_props.mem_banks_count, dev->mem_bank_count); sysfs_show_32bit_prop(buffer, "mem_banks_count", diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 6080951d539d..afce1edbe250 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -157,6 +157,7 @@ struct amd_ip_funcs { int (*hw_init)(void *handle); /* tears down the hw state */ int (*hw_fini)(void *handle); + void (*late_fini)(void *handle); /* handles IP specific hw/sw changes for suspend */ int (*suspend)(void *handle); /* handles IP specific hw/sw changes for resume */ diff --git a/drivers/gpu/drm/amd/include/cgs_common.h b/drivers/gpu/drm/amd/include/cgs_common.h index a461e155a160..7464daf89ca1 100644 --- a/drivers/gpu/drm/amd/include/cgs_common.h +++ b/drivers/gpu/drm/amd/include/cgs_common.h @@ -581,6 +581,9 @@ typedef int (*cgs_get_firmware_info)(struct cgs_device *cgs_device, enum cgs_ucode_id type, struct cgs_firmware_info *info); +typedef int (*cgs_rel_firmware)(struct cgs_device *cgs_device, + enum cgs_ucode_id type); + typedef int(*cgs_set_powergating_state)(struct cgs_device *cgs_device, enum amd_ip_block_type block_type, enum amd_powergating_state state); @@ -645,6 +648,7 @@ struct cgs_ops { cgs_set_camera_voltages_t set_camera_voltages; /* Firmware Info */ cgs_get_firmware_info get_firmware_info; + cgs_rel_firmware rel_firmware; /* cg pg interface*/ cgs_set_powergating_state set_powergating_state; cgs_set_clockgating_state set_clockgating_state; @@ -738,6 +742,8 @@ struct cgs_device CGS_CALL(set_camera_voltages,dev,mask,voltages) #define cgs_get_firmware_info(dev, type, info) \ CGS_CALL(get_firmware_info, dev, type, info) +#define cgs_rel_firmware(dev, type) \ + CGS_CALL(rel_firmware, dev, type) #define cgs_set_powergating_state(dev, block_type, state) \ CGS_CALL(set_powergating_state, dev, block_type, state) #define cgs_set_clockgating_state(dev, block_type, state) \ diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c index 8e345bfddb69..e629f8a9fe93 100644 --- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c @@ -73,11 +73,14 @@ static int pp_sw_init(void *handle) ret = hwmgr->hwmgr_func->backend_init(hwmgr); if (ret) - goto err; + goto err1; pr_info("amdgpu: powerplay initialized\n"); return 0; +err1: + if (hwmgr->pptable_func->pptable_fini) + hwmgr->pptable_func->pptable_fini(hwmgr); err: pr_err("amdgpu: powerplay initialization failed\n"); return ret; @@ -100,6 +103,9 @@ static int pp_sw_fini(void *handle) if (hwmgr->hwmgr_func->backend_fini != NULL) ret = hwmgr->hwmgr_func->backend_fini(hwmgr); + if (hwmgr->pptable_func->pptable_fini) + hwmgr->pptable_func->pptable_fini(hwmgr); + return ret; } diff --git a/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c b/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c index 46410e3c7349..fb88e4e5d625 100644 --- a/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c +++ b/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c @@ -58,9 +58,6 @@ static void pem_fini(struct pp_eventmgr *eventmgr) pem_unregister_interrupts(eventmgr); pem_handle_event(eventmgr, AMD_PP_EVENT_UNINITIALIZE, &event_data); - - if (eventmgr != NULL) - kfree(eventmgr); } int eventmgr_init(struct pp_instance *handle) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c index c94f9faa220a..586f73276226 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c @@ -1830,7 +1830,7 @@ static uint16_t fiji_find_closest_vddci(struct pp_hwmgr *hwmgr, uint16_t vddci) PP_ASSERT_WITH_CODE(false, "VDDCI is larger than max VDDCI in VDDCI Voltage Table!", - return vddci_table->entries[i].value); + return vddci_table->entries[i-1].value); } static int fiji_get_dependency_volt_by_clk(struct pp_hwmgr *hwmgr, @@ -3573,46 +3573,11 @@ static int fiji_force_dpm_highest(struct pp_hwmgr *hwmgr) return 0; } -static void fiji_apply_dal_min_voltage_request(struct pp_hwmgr *hwmgr) -{ - struct phm_ppt_v1_information *table_info = - (struct phm_ppt_v1_information *)hwmgr->pptable; - struct phm_clock_voltage_dependency_table *table = - table_info->vddc_dep_on_dal_pwrl; - struct phm_ppt_v1_clock_voltage_dependency_table *vddc_table; - enum PP_DAL_POWERLEVEL dal_power_level = hwmgr->dal_power_level; - uint32_t req_vddc = 0, req_volt, i; - - if (!table && !(dal_power_level >= PP_DAL_POWERLEVEL_ULTRALOW && - dal_power_level <= PP_DAL_POWERLEVEL_PERFORMANCE)) - return; - - for (i= 0; i < table->count; i++) { - if (dal_power_level == table->entries[i].clk) { - req_vddc = table->entries[i].v; - break; - } - } - - vddc_table = table_info->vdd_dep_on_sclk; - for (i= 0; i < vddc_table->count; i++) { - if (req_vddc <= vddc_table->entries[i].vddc) { - req_volt = (((uint32_t)vddc_table->entries[i].vddc) * VOLTAGE_SCALE) - << VDDC_SHIFT; - smum_send_msg_to_smc_with_parameter(hwmgr->smumgr, - PPSMC_MSG_VddC_Request, req_volt); - return; - } - } - printk(KERN_ERR "DAL requested level can not" - " found a available voltage in VDDC DPM Table \n"); -} - static int fiji_upload_dpmlevel_enable_mask(struct pp_hwmgr *hwmgr) { struct fiji_hwmgr *data = (struct fiji_hwmgr *)(hwmgr->backend); - fiji_apply_dal_min_voltage_request(hwmgr); + phm_apply_dal_min_voltage_request(hwmgr); if (!data->sclk_dpm_key_disabled) { if (data->dpm_level_enable_mask.sclk_dpm_enable_mask) @@ -4349,7 +4314,7 @@ static int fiji_populate_and_upload_sclk_mclk_dpm_levels( if (data->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK + DPMTABLE_UPDATE_SCLK)) { - result = fiji_populate_all_memory_levels(hwmgr); + result = fiji_populate_all_graphic_levels(hwmgr); PP_ASSERT_WITH_CODE((0 == result), "Failed to populate SCLK during PopulateNewDPMClocksStates Function!", return result); @@ -5109,11 +5074,11 @@ static int fiji_get_pp_table(struct pp_hwmgr *hwmgr, char **table) struct fiji_hwmgr *data = (struct fiji_hwmgr *)(hwmgr->backend); if (!data->soft_pp_table) { - data->soft_pp_table = kzalloc(hwmgr->soft_pp_table_size, GFP_KERNEL); + data->soft_pp_table = kmemdup(hwmgr->soft_pp_table, + hwmgr->soft_pp_table_size, + GFP_KERNEL); if (!data->soft_pp_table) return -ENOMEM; - memcpy(data->soft_pp_table, hwmgr->soft_pp_table, - hwmgr->soft_pp_table_size); } *table = (char *)&data->soft_pp_table; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c index 7d69ed635bc2..20f20e075588 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c @@ -30,6 +30,9 @@ #include "pppcielanes.h" #include "pp_debug.h" #include "ppatomctrl.h" +#include "ppsmc.h" + +#define VOLTAGE_SCALE 4 extern int cz_hwmgr_init(struct pp_hwmgr *hwmgr); extern int tonga_hwmgr_init(struct pp_hwmgr *hwmgr); @@ -90,6 +93,13 @@ int hwmgr_fini(struct pp_hwmgr *hwmgr) if (hwmgr == NULL || hwmgr->ps == NULL) return -EINVAL; + /* do hwmgr finish*/ + kfree(hwmgr->backend); + + kfree(hwmgr->start_thermal_controller.function_list); + + kfree(hwmgr->set_temperature_range.function_list); + kfree(hwmgr->ps); kfree(hwmgr); return 0; @@ -459,7 +469,7 @@ uint16_t phm_find_closest_vddci(struct pp_atomctrl_voltage_table *vddci_table, u PP_ASSERT_WITH_CODE(false, "VDDCI is larger than max VDDCI in VDDCI Voltage Table!", - return vddci_table->entries[i].value); + return vddci_table->entries[i-1].value); } int phm_find_boot_level(void *table, @@ -566,3 +576,38 @@ uint32_t phm_get_lowest_enabled_level(struct pp_hwmgr *hwmgr, uint32_t mask) return level; } + +void phm_apply_dal_min_voltage_request(struct pp_hwmgr *hwmgr) +{ + struct phm_ppt_v1_information *table_info = + (struct phm_ppt_v1_information *)hwmgr->pptable; + struct phm_clock_voltage_dependency_table *table = + table_info->vddc_dep_on_dal_pwrl; + struct phm_ppt_v1_clock_voltage_dependency_table *vddc_table; + enum PP_DAL_POWERLEVEL dal_power_level = hwmgr->dal_power_level; + uint32_t req_vddc = 0, req_volt, i; + + if (!table || table->count <= 0 + || dal_power_level < PP_DAL_POWERLEVEL_ULTRALOW + || dal_power_level > PP_DAL_POWERLEVEL_PERFORMANCE) + return; + + for (i = 0; i < table->count; i++) { + if (dal_power_level == table->entries[i].clk) { + req_vddc = table->entries[i].v; + break; + } + } + + vddc_table = table_info->vdd_dep_on_sclk; + for (i = 0; i < vddc_table->count; i++) { + if (req_vddc <= vddc_table->entries[i].vddc) { + req_volt = (((uint32_t)vddc_table->entries[i].vddc) * VOLTAGE_SCALE); + smum_send_msg_to_smc_with_parameter(hwmgr->smumgr, + PPSMC_MSG_VddC_Request, req_volt); + return; + } + } + printk(KERN_ERR "DAL requested level can not" + " found a available voltage in VDDC DPM Table \n"); +} diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c index 93768fa1dcdc..aa6be033f21b 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c @@ -189,41 +189,6 @@ int phm_get_current_pcie_lane_number(struct pp_hwmgr *hwmgr) return decode_pcie_lane_width(link_width); } -void phm_apply_dal_min_voltage_request(struct pp_hwmgr *hwmgr) -{ - struct phm_ppt_v1_information *table_info = - (struct phm_ppt_v1_information *)hwmgr->pptable; - struct phm_clock_voltage_dependency_table *table = - table_info->vddc_dep_on_dal_pwrl; - struct phm_ppt_v1_clock_voltage_dependency_table *vddc_table; - enum PP_DAL_POWERLEVEL dal_power_level = hwmgr->dal_power_level; - uint32_t req_vddc = 0, req_volt, i; - - if (!table && !(dal_power_level >= PP_DAL_POWERLEVEL_ULTRALOW && - dal_power_level <= PP_DAL_POWERLEVEL_PERFORMANCE)) - return; - - for (i = 0; i < table->count; i++) { - if (dal_power_level == table->entries[i].clk) { - req_vddc = table->entries[i].v; - break; - } - } - - vddc_table = table_info->vdd_dep_on_sclk; - for (i = 0; i < vddc_table->count; i++) { - if (req_vddc <= vddc_table->entries[i].vddc) { - req_volt = (((uint32_t)vddc_table->entries[i].vddc) * VOLTAGE_SCALE) - << VDDC_SHIFT; - smum_send_msg_to_smc_with_parameter(hwmgr->smumgr, - PPSMC_MSG_VddC_Request, req_volt); - return; - } - } - printk(KERN_ERR "DAL requested level can not" - " found a available voltage in VDDC DPM Table \n"); -} - /** * Enable voltage control * @@ -2091,7 +2056,7 @@ static int polaris10_init_smc_table(struct pp_hwmgr *hwmgr) "Failed to populate Clock Stretcher Data Table!", return result); } - + table->CurrSclkPllRange = 0xff; table->GraphicsVoltageChangeEnable = 1; table->GraphicsThermThrottleEnable = 1; table->GraphicsInterval = 1; @@ -2184,6 +2149,7 @@ static int polaris10_init_smc_table(struct pp_hwmgr *hwmgr) CONVERT_FROM_HOST_TO_SMC_UL(table->SmioMask1); CONVERT_FROM_HOST_TO_SMC_UL(table->SmioMask2); CONVERT_FROM_HOST_TO_SMC_UL(table->SclkStepSize); + CONVERT_FROM_HOST_TO_SMC_UL(table->CurrSclkPllRange); CONVERT_FROM_HOST_TO_SMC_US(table->TemperatureLimitHigh); CONVERT_FROM_HOST_TO_SMC_US(table->TemperatureLimitLow); CONVERT_FROM_HOST_TO_SMC_US(table->VoltageResponseTime); @@ -4760,11 +4726,11 @@ static int polaris10_get_pp_table(struct pp_hwmgr *hwmgr, char **table) struct polaris10_hwmgr *data = (struct polaris10_hwmgr *)(hwmgr->backend); if (!data->soft_pp_table) { - data->soft_pp_table = kzalloc(hwmgr->soft_pp_table_size, GFP_KERNEL); + data->soft_pp_table = kmemdup(hwmgr->soft_pp_table, + hwmgr->soft_pp_table_size, + GFP_KERNEL); if (!data->soft_pp_table) return -ENOMEM; - memcpy(data->soft_pp_table, hwmgr->soft_pp_table, - hwmgr->soft_pp_table_size); } *table = (char *)&data->soft_pp_table; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_powertune.c index 0b99ab3ba0c5..ae96f14b827c 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_powertune.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_powertune.c @@ -286,7 +286,7 @@ int polaris10_populate_pm_fuses(struct pp_hwmgr *hwmgr) if (polaris10_copy_bytes_to_smc(hwmgr->smumgr, pm_fuse_table_offset, (uint8_t *)&data->power_tune_table, - sizeof(struct SMU74_Discrete_PmFuses), data->sram_end)) + (sizeof(struct SMU74_Discrete_PmFuses) - 92), data->sram_end)) PP_ASSERT_WITH_CODE(false, "Attempt to download PmFuseTable Failed!", return -EINVAL); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c index 1faad92b50d3..d27e8c40602a 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c @@ -2847,27 +2847,6 @@ static int tonga_setup_default_dpm_tables(struct pp_hwmgr *hwmgr) } } - /* Initialize Vddc DPM table based on allow Vddc values. And populate corresponding std values. */ - for (i = 0; i < allowed_vdd_sclk_table->count; i++) { - data->dpm_table.vddc_table.dpm_levels[i].value = allowed_vdd_mclk_table->entries[i].vddc; - /* tonga_hwmgr->dpm_table.VddcTable.dpm_levels[i].param1 = stdVoltageTable->entries[i].Leakage; */ - /* param1 is for corresponding std voltage */ - data->dpm_table.vddc_table.dpm_levels[i].enabled = 1; - } - data->dpm_table.vddc_table.count = allowed_vdd_sclk_table->count; - - if (NULL != allowed_vdd_mclk_table) { - /* Initialize Vddci DPM table based on allow Mclk values */ - for (i = 0; i < allowed_vdd_mclk_table->count; i++) { - data->dpm_table.vdd_ci_table.dpm_levels[i].value = allowed_vdd_mclk_table->entries[i].vddci; - data->dpm_table.vdd_ci_table.dpm_levels[i].enabled = 1; - data->dpm_table.mvdd_table.dpm_levels[i].value = allowed_vdd_mclk_table->entries[i].mvdd; - data->dpm_table.mvdd_table.dpm_levels[i].enabled = 1; - } - data->dpm_table.vdd_ci_table.count = allowed_vdd_mclk_table->count; - data->dpm_table.mvdd_table.count = allowed_vdd_mclk_table->count; - } - /* setup PCIE gen speed levels*/ tonga_setup_default_pcie_tables(hwmgr); @@ -5331,7 +5310,7 @@ static int tonga_freeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) (data->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK + DPMTABLE_UPDATE_SCLK))) { PP_ASSERT_WITH_CODE( - true == tonga_is_dpm_running(hwmgr), + 0 == tonga_is_dpm_running(hwmgr), "Trying to freeze SCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE( @@ -5344,7 +5323,7 @@ static int tonga_freeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) if ((0 == data->mclk_dpm_key_disabled) && (data->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)) { - PP_ASSERT_WITH_CODE(true == tonga_is_dpm_running(hwmgr), + PP_ASSERT_WITH_CODE(0 == tonga_is_dpm_running(hwmgr), "Trying to freeze MCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE( @@ -5445,7 +5424,7 @@ static int tonga_populate_and_upload_sclk_mclk_dpm_levels(struct pp_hwmgr *hwmgr } if (data->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK + DPMTABLE_UPDATE_SCLK)) { - result = tonga_populate_all_memory_levels(hwmgr); + result = tonga_populate_all_graphic_levels(hwmgr); PP_ASSERT_WITH_CODE((0 == result), "Failed to populate SCLK during PopulateNewDPMClocksStates Function!", return result); @@ -5647,7 +5626,7 @@ static int tonga_unfreeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) (data->need_update_smu7_dpm_table & (DPMTABLE_OD_UPDATE_SCLK + DPMTABLE_UPDATE_SCLK))) { - PP_ASSERT_WITH_CODE(true == tonga_is_dpm_running(hwmgr), + PP_ASSERT_WITH_CODE(0 == tonga_is_dpm_running(hwmgr), "Trying to Unfreeze SCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE( @@ -5661,7 +5640,7 @@ static int tonga_unfreeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) (data->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK)) { PP_ASSERT_WITH_CODE( - true == tonga_is_dpm_running(hwmgr), + 0 == tonga_is_dpm_running(hwmgr), "Trying to Unfreeze MCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE( @@ -6056,11 +6035,11 @@ static int tonga_get_pp_table(struct pp_hwmgr *hwmgr, char **table) struct tonga_hwmgr *data = (struct tonga_hwmgr *)(hwmgr->backend); if (!data->soft_pp_table) { - data->soft_pp_table = kzalloc(hwmgr->soft_pp_table_size, GFP_KERNEL); + data->soft_pp_table = kmemdup(hwmgr->soft_pp_table, + hwmgr->soft_pp_table_size, + GFP_KERNEL); if (!data->soft_pp_table) return -ENOMEM; - memcpy(data->soft_pp_table, hwmgr->soft_pp_table, - hwmgr->soft_pp_table_size); } *table = (char *)&data->soft_pp_table; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c index 10e3630ee39d..296ec7ef6d45 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c @@ -1040,48 +1040,44 @@ int tonga_pp_tables_uninitialize(struct pp_hwmgr *hwmgr) struct phm_ppt_v1_information *pp_table_information = (struct phm_ppt_v1_information *)(hwmgr->pptable); - if (NULL != hwmgr->soft_pp_table) { - kfree(hwmgr->soft_pp_table); + if (NULL != hwmgr->soft_pp_table) hwmgr->soft_pp_table = NULL; - } - if (NULL != pp_table_information->vdd_dep_on_sclk) - pp_table_information->vdd_dep_on_sclk = NULL; + kfree(pp_table_information->vdd_dep_on_sclk); + pp_table_information->vdd_dep_on_sclk = NULL; - if (NULL != pp_table_information->vdd_dep_on_mclk) - pp_table_information->vdd_dep_on_mclk = NULL; + kfree(pp_table_information->vdd_dep_on_mclk); + pp_table_information->vdd_dep_on_mclk = NULL; - if (NULL != pp_table_information->valid_mclk_values) - pp_table_information->valid_mclk_values = NULL; + kfree(pp_table_information->valid_mclk_values); + pp_table_information->valid_mclk_values = NULL; - if (NULL != pp_table_information->valid_sclk_values) - pp_table_information->valid_sclk_values = NULL; + kfree(pp_table_information->valid_sclk_values); + pp_table_information->valid_sclk_values = NULL; - if (NULL != pp_table_information->vddc_lookup_table) - pp_table_information->vddc_lookup_table = NULL; + kfree(pp_table_information->vddc_lookup_table); + pp_table_information->vddc_lookup_table = NULL; - if (NULL != pp_table_information->vddgfx_lookup_table) - pp_table_information->vddgfx_lookup_table = NULL; + kfree(pp_table_information->vddgfx_lookup_table); + pp_table_information->vddgfx_lookup_table = NULL; - if (NULL != pp_table_information->mm_dep_table) - pp_table_information->mm_dep_table = NULL; + kfree(pp_table_information->mm_dep_table); + pp_table_information->mm_dep_table = NULL; - if (NULL != pp_table_information->cac_dtp_table) - pp_table_information->cac_dtp_table = NULL; + kfree(pp_table_information->cac_dtp_table); + pp_table_information->cac_dtp_table = NULL; - if (NULL != hwmgr->dyn_state.cac_dtp_table) - hwmgr->dyn_state.cac_dtp_table = NULL; + kfree(hwmgr->dyn_state.cac_dtp_table); + hwmgr->dyn_state.cac_dtp_table = NULL; - if (NULL != pp_table_information->ppm_parameter_table) - pp_table_information->ppm_parameter_table = NULL; + kfree(pp_table_information->ppm_parameter_table); + pp_table_information->ppm_parameter_table = NULL; - if (NULL != pp_table_information->pcie_table) - pp_table_information->pcie_table = NULL; + kfree(pp_table_information->pcie_table); + pp_table_information->pcie_table = NULL; - if (NULL != hwmgr->pptable) { - kfree(hwmgr->pptable); - hwmgr->pptable = NULL; - } + kfree(hwmgr->pptable); + hwmgr->pptable = NULL; return result; } diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h index fd4ce7aaeee9..28f571449495 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h @@ -673,7 +673,7 @@ extern int phm_get_sclk_for_voltage_evv(struct pp_hwmgr *hwmgr, phm_ppt_v1_volta extern int phm_initializa_dynamic_state_adjustment_rule_settings(struct pp_hwmgr *hwmgr); extern int phm_hwmgr_backend_fini(struct pp_hwmgr *hwmgr); extern uint32_t phm_get_lowest_enabled_level(struct pp_hwmgr *hwmgr, uint32_t mask); - +extern void phm_apply_dal_min_voltage_request(struct pp_hwmgr *hwmgr); #define PHM_ENTIRE_REGISTER_MASK 0xFFFFFFFFU diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/cz_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/cz_smumgr.c index da18f44fd1c8..87c023e518ab 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/cz_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/cz_smumgr.c @@ -639,7 +639,7 @@ static int cz_smu_populate_firmware_entries(struct pp_smumgr *smumgr) cz_smu->driver_buffer_length = 0; - for (i = 0; i < sizeof(firmware_list)/sizeof(*firmware_list); i++) { + for (i = 0; i < ARRAY_SIZE(firmware_list); i++) { firmware_type = cz_translate_firmware_enum_to_arg(smumgr, firmware_list[i]); diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c index 673a75c74e18..8e52a2e82db5 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c @@ -1006,10 +1006,16 @@ static int fiji_smu_init(struct pp_smumgr *smumgr) static int fiji_smu_fini(struct pp_smumgr *smumgr) { + struct fiji_smumgr *priv = (struct fiji_smumgr *)(smumgr->backend); + + smu_free_memory(smumgr->device, (void *)priv->header_buffer.handle); + if (smumgr->backend) { kfree(smumgr->backend); smumgr->backend = NULL; } + + cgs_rel_firmware(smumgr->device, CGS_UCODE_ID_SMU); return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c index de618ead9db8..043b6ac09d5f 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c @@ -469,6 +469,7 @@ int polaris10_smu_fini(struct pp_smumgr *smumgr) kfree(smumgr->backend); smumgr->backend = NULL; } + cgs_rel_firmware(smumgr->device, CGS_UCODE_ID_SMU); return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c index c483baf6b4fb..0728c1e3d97a 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c @@ -81,6 +81,7 @@ int smum_init(struct amd_pp_init *pp_init, struct pp_instance *handle) int smum_fini(struct pp_smumgr *smumgr) { + kfree(smumgr->device); kfree(smumgr); return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c index 32820b680d88..b22722eabafc 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c @@ -328,10 +328,17 @@ int tonga_write_smc_sram_dword(struct pp_smumgr *smumgr, static int tonga_smu_fini(struct pp_smumgr *smumgr) { + struct tonga_smumgr *priv = (struct tonga_smumgr *)(smumgr->backend); + + smu_free_memory(smumgr->device, (void *)priv->smu_buffer.handle); + smu_free_memory(smumgr->device, (void *)priv->header_buffer.handle); + if (smumgr->backend != NULL) { kfree(smumgr->backend); smumgr->backend = NULL; } + + cgs_rel_firmware(smumgr->device, CGS_UCODE_ID_SMU); return 0; } |
