diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu')
80 files changed, 674 insertions, 657 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 35faa286bb1c..7dab4768cee6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1441,6 +1441,7 @@ u32 amdgpu_device_pcie_port_rreg(struct amdgpu_device *adev, u32 reg); void amdgpu_device_pcie_port_wreg(struct amdgpu_device *adev, u32 reg, u32 v); +struct dma_fence *amdgpu_device_get_gang(struct amdgpu_device *adev); struct dma_fence *amdgpu_device_switch_gang(struct amdgpu_device *adev, struct dma_fence *gang); bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev); @@ -1567,6 +1568,7 @@ static inline int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, u8 dev_state, bool drv_state) { return 0; } static inline int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_state) { return 0; } +static inline void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps) { } #endif #if defined(CONFIG_ACPI) && defined(CONFIG_SUSPEND) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c index 01d50ad603d3..7945173321a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -222,9 +222,9 @@ static struct aca_bank_error *new_bank_error(struct aca_error *aerr, struct aca_ INIT_LIST_HEAD(&bank_error->node); memcpy(&bank_error->info, info, sizeof(*info)); - spin_lock(&aerr->lock); + mutex_lock(&aerr->lock); list_add_tail(&bank_error->node, &aerr->list); - spin_unlock(&aerr->lock); + mutex_unlock(&aerr->lock); return bank_error; } @@ -235,7 +235,7 @@ static struct aca_bank_error *find_bank_error(struct aca_error *aerr, struct aca struct aca_bank_info *tmp_info; bool found = false; - spin_lock(&aerr->lock); + mutex_lock(&aerr->lock); list_for_each_entry(bank_error, &aerr->list, node) { tmp_info = &bank_error->info; if (tmp_info->socket_id == info->socket_id && @@ -246,7 +246,7 @@ static struct aca_bank_error *find_bank_error(struct aca_error *aerr, struct aca } out_unlock: - spin_unlock(&aerr->lock); + mutex_unlock(&aerr->lock); return found ? bank_error : NULL; } @@ -474,7 +474,7 @@ static int aca_log_aca_error(struct aca_handle *handle, enum aca_error_type type struct aca_error *aerr = &error_cache->errors[type]; struct aca_bank_error *bank_error, *tmp; - spin_lock(&aerr->lock); + mutex_lock(&aerr->lock); if (list_empty(&aerr->list)) goto out_unlock; @@ -485,7 +485,7 @@ static int aca_log_aca_error(struct aca_handle *handle, enum aca_error_type type } out_unlock: - spin_unlock(&aerr->lock); + mutex_unlock(&aerr->lock); return 0; } @@ -542,7 +542,7 @@ int amdgpu_aca_get_error_data(struct amdgpu_device *adev, struct aca_handle *han static void aca_error_init(struct aca_error *aerr, enum aca_error_type type) { - spin_lock_init(&aerr->lock); + mutex_init(&aerr->lock); INIT_LIST_HEAD(&aerr->list); aerr->type = type; aerr->nr_errors = 0; @@ -561,10 +561,11 @@ static void aca_error_fini(struct aca_error *aerr) { struct aca_bank_error *bank_error, *tmp; - spin_lock(&aerr->lock); + mutex_lock(&aerr->lock); list_for_each_entry_safe(bank_error, tmp, &aerr->list, node) aca_bank_error_remove(aerr, bank_error); - spin_unlock(&aerr->lock); + + mutex_destroy(&aerr->lock); } static void aca_fini_error_cache(struct aca_handle *handle) @@ -712,6 +713,15 @@ void amdgpu_aca_fini(struct amdgpu_device *adev) atomic_set(&aca->ue_update_flag, 0); } +int amdgpu_aca_reset(struct amdgpu_device *adev) +{ + struct amdgpu_aca *aca = &adev->aca; + + atomic_set(&aca->ue_update_flag, 0); + + return 0; +} + void amdgpu_aca_set_smu_funcs(struct amdgpu_device *adev, const struct aca_smu_funcs *smu_funcs) { struct amdgpu_aca *aca = &adev->aca; @@ -885,9 +895,7 @@ DEFINE_DEBUGFS_ATTRIBUTE(aca_debug_mode_fops, NULL, amdgpu_aca_smu_debug_mode_se void amdgpu_aca_smu_debugfs_init(struct amdgpu_device *adev, struct dentry *root) { #if defined(CONFIG_DEBUG_FS) - if (!root || - (adev->ip_versions[MP1_HWIP][0] != IP_VERSION(13, 0, 6) && - adev->ip_versions[MP1_HWIP][0] != IP_VERSION(13, 0, 14))) + if (!root) return; debugfs_create_file("aca_debug_mode", 0200, root, adev, &aca_debug_mode_fops); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h index 4327ce1ceacf..5ef6b745f222 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h @@ -25,7 +25,6 @@ #define __AMDGPU_ACA_H__ #include <linux/list.h> -#include <linux/spinlock.h> struct ras_err_data; struct ras_query_context; @@ -134,7 +133,7 @@ struct aca_bank_error { struct aca_error { struct list_head list; - spinlock_t lock; + struct mutex lock; enum aca_error_type type; int nr_errors; }; @@ -192,6 +191,7 @@ struct aca_info { int amdgpu_aca_init(struct amdgpu_device *adev); void amdgpu_aca_fini(struct amdgpu_device *adev); +int amdgpu_aca_reset(struct amdgpu_device *adev); void amdgpu_aca_set_smu_funcs(struct amdgpu_device *adev, const struct aca_smu_funcs *smu_funcs); bool amdgpu_aca_is_enabled(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 7099ff9cf8c5..f85ace0384d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -383,6 +383,8 @@ static int amdgpu_atif_query_backlight_caps(struct amdgpu_atif *atif) characteristics.min_input_signal; atif->backlight_caps.max_input_signal = characteristics.max_input_signal; + atif->backlight_caps.ac_level = characteristics.ac_level; + atif->backlight_caps.dc_level = characteristics.dc_level; out: kfree(info); return err; @@ -1268,6 +1270,8 @@ void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps) caps->caps_valid = atif->backlight_caps.caps_valid; caps->min_input_signal = atif->backlight_caps.min_input_signal; caps->max_input_signal = atif->backlight_caps.max_input_signal; + caps->ac_level = atif->backlight_caps.ac_level; + caps->dc_level = atif->backlight_caps.dc_level; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 4096cb3e937e..2de36883fc9f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2350,7 +2350,6 @@ void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev) static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[40]; int err; const struct gpu_info_firmware_header_v1_0 *hdr; @@ -2384,12 +2383,12 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) break; } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_gpu_info.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->firmware.gpu_info_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->firmware.gpu_info_fw, + "amdgpu/%s_gpu_info.bin", chip_name); if (err) { dev_err(adev->dev, - "Failed to get gpu_info firmware \"%s\"\n", - fw_name); + "Failed to get gpu_info firmware \"%s_gpu_info.bin\"\n", + chip_name); goto out; } @@ -5070,6 +5069,8 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev, struct amdgpu_hive_info *hive = NULL; if (test_bit(AMDGPU_HOST_FLR, &reset_context->flags)) { + amdgpu_virt_ready_to_reset(adev); + amdgpu_virt_wait_reset(adev); clear_bit(AMDGPU_HOST_FLR, &reset_context->flags); r = amdgpu_virt_request_full_gpu(adev, true); } else { @@ -6275,20 +6276,11 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) struct amdgpu_reset_context reset_context; u32 memsize; struct list_head device_list; - struct amdgpu_hive_info *hive; - int hive_ras_recovery = 0; - struct amdgpu_ras *ras; /* PCI error slot reset should be skipped During RAS recovery */ - hive = amdgpu_get_xgmi_hive(adev); - if (hive) { - hive_ras_recovery = atomic_read(&hive->ras_recovery); - amdgpu_put_xgmi_hive(hive); - } - ras = amdgpu_ras_get_context(adev); if ((amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) || - amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) && - ras && (atomic_read(&ras->in_recovery) || hive_ras_recovery)) + amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) && + amdgpu_ras_in_recovery(adev)) return PCI_ERS_RESULT_RECOVERED; DRM_INFO("PCI error: slot reset callback!!\n"); @@ -6531,6 +6523,22 @@ void amdgpu_device_pcie_port_wreg(struct amdgpu_device *adev, } /** + * amdgpu_device_get_gang - return a reference to the current gang + * @adev: amdgpu_device pointer + * + * Returns: A new reference to the current gang leader. + */ +struct dma_fence *amdgpu_device_get_gang(struct amdgpu_device *adev) +{ + struct dma_fence *fence; + + rcu_read_lock(); + fence = dma_fence_get_rcu_safe(&adev->gang_submit); + rcu_read_unlock(); + return fence; +} + +/** * amdgpu_device_switch_gang - switch to a new gang * @adev: amdgpu_device pointer * @gang: the gang to switch to @@ -6546,10 +6554,7 @@ struct dma_fence *amdgpu_device_switch_gang(struct amdgpu_device *adev, do { dma_fence_put(old); - rcu_read_lock(); - old = dma_fence_get_rcu_safe(&adev->gang_submit); - rcu_read_unlock(); - + old = amdgpu_device_get_gang(adev); if (old == gang) break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index c71356cb393d..0cb8aea93a70 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -260,19 +260,21 @@ static int amdgpu_discovery_read_binary_from_mem(struct amdgpu_device *adev, u32 msg; int i, ret = 0; - /* It can take up to a second for IFWI init to complete on some dGPUs, - * but generally it should be in the 60-100ms range. Normally this starts - * as soon as the device gets power so by the time the OS loads this has long - * completed. However, when a card is hotplugged via e.g., USB4, we need to - * wait for this to complete. Once the C2PMSG is updated, we can - * continue. - */ + if (!amdgpu_sriov_vf(adev)) { + /* It can take up to a second for IFWI init to complete on some dGPUs, + * but generally it should be in the 60-100ms range. Normally this starts + * as soon as the device gets power so by the time the OS loads this has long + * completed. However, when a card is hotplugged via e.g., USB4, we need to + * wait for this to complete. Once the C2PMSG is updated, we can + * continue. + */ - for (i = 0; i < 1000; i++) { - msg = RREG32(mmMP0_SMN_C2PMSG_33); - if (msg & 0x80000000) - break; - usleep_range(1000, 1100); + for (i = 0; i < 1000; i++) { + msg = RREG32(mmMP0_SMN_C2PMSG_33); + if (msg & 0x80000000) + break; + usleep_range(1000, 1100); + } } vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 0b3b10d21952..8e81a83d37d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -41,8 +41,6 @@ #include <linux/dma-buf.h> #include <linux/dma-fence-array.h> #include <linux/pci-p2pdma.h> -#include <linux/pm_runtime.h> -#include "amdgpu_trace.h" /** * amdgpu_dma_buf_attach - &dma_buf_ops.attach implementation @@ -58,42 +56,11 @@ static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf, struct drm_gem_object *obj = dmabuf->priv; struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); - int r; if (pci_p2pdma_distance(adev->pdev, attach->dev, false) < 0) attach->peer2peer = false; - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - trace_amdgpu_runpm_reference_dumps(1, __func__); - if (r < 0) - goto out; - return 0; - -out: - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - trace_amdgpu_runpm_reference_dumps(0, __func__); - return r; -} - -/** - * amdgpu_dma_buf_detach - &dma_buf_ops.detach implementation - * - * @dmabuf: DMA-buf where we remove the attachment from - * @attach: the attachment to remove - * - * Called when an attachment is removed from the DMA-buf. - */ -static void amdgpu_dma_buf_detach(struct dma_buf *dmabuf, - struct dma_buf_attachment *attach) -{ - struct drm_gem_object *obj = dmabuf->priv; - struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); - struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - trace_amdgpu_runpm_reference_dumps(0, __func__); } /** @@ -266,7 +233,6 @@ static int amdgpu_dma_buf_begin_cpu_access(struct dma_buf *dma_buf, const struct dma_buf_ops amdgpu_dmabuf_ops = { .attach = amdgpu_dma_buf_attach, - .detach = amdgpu_dma_buf_detach, .pin = amdgpu_dma_buf_pin, .unpin = amdgpu_dma_buf_unpin, .map_dma_buf = amdgpu_dma_buf_map, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 10832b470448..2f24a6aa13bf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -181,7 +181,6 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amd amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr, seq, flags | AMDGPU_FENCE_FLAG_INT); pm_runtime_get_noresume(adev_to_drm(adev)->dev); - trace_amdgpu_runpm_reference_dumps(1, __func__); ptr = &ring->fence_drv.fences[seq & ring->fence_drv.num_fences_mask]; if (unlikely(rcu_dereference_protected(*ptr, 1))) { struct dma_fence *old; @@ -309,7 +308,6 @@ bool amdgpu_fence_process(struct amdgpu_ring *ring) dma_fence_put(fence); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - trace_amdgpu_runpm_reference_dumps(0, __func__); } while (last_seq != seq); return true; @@ -980,7 +978,9 @@ static void amdgpu_debugfs_reset_work(struct work_struct *work) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; + reset_context.src = AMDGPU_RESET_SRC_USER; set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); + set_bit(AMDGPU_SKIP_COREDUMP, &reset_context.flags); amdgpu_device_gpu_recover(adev, NULL, &reset_context); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index c623e23049d1..256b95232de5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -34,6 +34,7 @@ #include <asm/set_memory.h> #endif #include "amdgpu.h" +#include "amdgpu_reset.h" #include <drm/drm_drv.h> #include <drm/ttm/ttm_tt.h> @@ -325,10 +326,7 @@ void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, page_base += AMDGPU_GPU_PAGE_SIZE; } } - mb(); - amdgpu_device_flush_hdp(adev, NULL); - for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) - amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); + amdgpu_gart_invalidate_tlb(adev); drm_dev_exit(idx); } @@ -408,7 +406,10 @@ void amdgpu_gart_invalidate_tlb(struct amdgpu_device *adev) return; mb(); - amdgpu_device_flush_hdp(adev, NULL); + if (down_read_trylock(&adev->reset_domain->sem)) { + amdgpu_device_flush_hdp(adev, NULL); + up_read(&adev->reset_domain->sem); + } for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 3b67809a59bc..1f22b4208729 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -334,6 +334,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, AMDGPU_GEM_CREATE_VM_ALWAYS_VALID | AMDGPU_GEM_CREATE_EXPLICIT_SYNC | AMDGPU_GEM_CREATE_ENCRYPTED | + AMDGPU_GEM_CREATE_GFX12_DCC | AMDGPU_GEM_CREATE_DISCARDABLE)) return -EINVAL; @@ -341,6 +342,11 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, if (args->in.domains & ~AMDGPU_GEM_DOMAIN_MASK) return -EINVAL; + if ((flags & AMDGPU_GEM_CREATE_GFX12_DCC) && + ((amdgpu_ip_version(adev, GC_HWIP, 0) < IP_VERSION(12, 0, 0)) || + !(args->in.domains & AMDGPU_GEM_DOMAIN_VRAM))) + return -EINVAL; + if (!amdgpu_is_tmz(adev) && (flags & AMDGPU_GEM_CREATE_ENCRYPTED)) { DRM_NOTE_ONCE("Cannot allocate secure buffer since TMZ is disabled\n"); return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 19b1817b55d7..82452606ae6c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -506,9 +506,6 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id) { struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; struct amdgpu_ring *kiq_ring = &kiq->ring; - struct amdgpu_hive_info *hive; - struct amdgpu_ras *ras; - int hive_ras_recovery = 0; int i, r = 0; int j; @@ -533,16 +530,9 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id) * This is workaround: only skip kiq_ring test * during ras recovery in suspend stage for gfx9.4.3 */ - hive = amdgpu_get_xgmi_hive(adev); - if (hive) { - hive_ras_recovery = atomic_read(&hive->ras_recovery); - amdgpu_put_xgmi_hive(hive); - } - - ras = amdgpu_ras_get_context(adev); if ((amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) || - amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) && - ras && (atomic_read(&ras->in_recovery) || hive_ras_recovery)) { + amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) && + amdgpu_ras_in_recovery(adev)) { spin_unlock(&kiq->ring_lock); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 603c0738fd03..322b8ff67cde 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -620,10 +620,8 @@ void amdgpu_gmc_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, int r; if (!hub->sdma_invalidation_workaround || vmid || - !adev->mman.buffer_funcs_enabled || - !adev->ib_pool_ready || amdgpu_in_reset(adev) || + !adev->mman.buffer_funcs_enabled || !adev->ib_pool_ready || !ring->sched.ready) { - /* * A GPU reset should flush all TLBs anyway, so no need to do * this while one is ongoing. @@ -684,12 +682,17 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring; struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst]; unsigned int ndw; - signed long r; + int r; uint32_t seq; - if (!adev->gmc.flush_pasid_uses_kiq || !ring->sched.ready || - !down_read_trylock(&adev->reset_domain->sem)) { + /* + * A GPU reset should flush all TLBs anyway, so no need to do + * this while one is ongoing. + */ + if (!down_read_trylock(&adev->reset_domain->sem)) + return 0; + if (!adev->gmc.flush_pasid_uses_kiq || !ring->sched.ready) { if (adev->gmc.flush_tlb_needs_extra_type_2) adev->gmc.gmc_funcs->flush_gpu_tlb_pasid(adev, pasid, 2, all_hub, @@ -703,43 +706,40 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, adev->gmc.gmc_funcs->flush_gpu_tlb_pasid(adev, pasid, flush_type, all_hub, inst); - return 0; - } + r = 0; + } else { + /* 2 dwords flush + 8 dwords fence */ + ndw = kiq->pmf->invalidate_tlbs_size + 8; - /* 2 dwords flush + 8 dwords fence */ - ndw = kiq->pmf->invalidate_tlbs_size + 8; + if (adev->gmc.flush_tlb_needs_extra_type_2) + ndw += kiq->pmf->invalidate_tlbs_size; - if (adev->gmc.flush_tlb_needs_extra_type_2) - ndw += kiq->pmf->invalidate_tlbs_size; + if (adev->gmc.flush_tlb_needs_extra_type_0) + ndw += kiq->pmf->invalidate_tlbs_size; - if (adev->gmc.flush_tlb_needs_extra_type_0) - ndw += kiq->pmf->invalidate_tlbs_size; + spin_lock(&adev->gfx.kiq[inst].ring_lock); + amdgpu_ring_alloc(ring, ndw); + if (adev->gmc.flush_tlb_needs_extra_type_2) + kiq->pmf->kiq_invalidate_tlbs(ring, pasid, 2, all_hub); - spin_lock(&adev->gfx.kiq[inst].ring_lock); - amdgpu_ring_alloc(ring, ndw); - if (adev->gmc.flush_tlb_needs_extra_type_2) - kiq->pmf->kiq_invalidate_tlbs(ring, pasid, 2, all_hub); + if (flush_type == 2 && adev->gmc.flush_tlb_needs_extra_type_0) + kiq->pmf->kiq_invalidate_tlbs(ring, pasid, 0, all_hub); - if (flush_type == 2 && adev->gmc.flush_tlb_needs_extra_type_0) - kiq->pmf->kiq_invalidate_tlbs(ring, pasid, 0, all_hub); + kiq->pmf->kiq_invalidate_tlbs(ring, pasid, flush_type, all_hub); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) { + amdgpu_ring_undo(ring); + spin_unlock(&adev->gfx.kiq[inst].ring_lock); + goto error_unlock_reset; + } - kiq->pmf->kiq_invalidate_tlbs(ring, pasid, flush_type, all_hub); - r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); - if (r) { - amdgpu_ring_undo(ring); + amdgpu_ring_commit(ring); spin_unlock(&adev->gfx.kiq[inst].ring_lock); - goto error_unlock_reset; - } - - amdgpu_ring_commit(ring); - spin_unlock(&adev->gfx.kiq[inst].ring_lock); - r = amdgpu_fence_wait_polling(ring, seq, usec_timeout); - if (r < 1) { - dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r); - r = -ETIME; - goto error_unlock_reset; + if (amdgpu_fence_wait_polling(ring, seq, usec_timeout) < 1) { + dev_err(adev->dev, "timeout waiting for kiq fence\n"); + r = -ETIME; + } } - r = 0; error_unlock_reset: up_read(&adev->reset_domain->sem); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c index 44367f03316f..0760e70402ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c @@ -200,8 +200,6 @@ void amdgpu_gtt_mgr_recover(struct amdgpu_gtt_mgr *mgr) amdgpu_ttm_recover_gart(node->base.bo); } spin_unlock(&mgr->lock); - - amdgpu_gart_invalidate_tlb(adev); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index 3d7fcdeaf8cf..b5b9d4f40f53 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -290,18 +290,36 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm, !dma_fence_is_signaled((*id)->last_flush))) { struct dma_fence *tmp; - /* Don't use per engine and per process VMID at the same time */ - if (adev->vm_manager.concurrent_flush) - ring = NULL; - - /* to prevent one context starved by another context */ - (*id)->pd_gpu_addr = 0; - tmp = amdgpu_sync_peek_fence(&(*id)->active, ring); - if (tmp) { + /* Wait for the gang to be assembled before using a + * reserved VMID or otherwise the gang could deadlock. + */ + tmp = amdgpu_device_get_gang(adev); + if (!dma_fence_is_signaled(tmp) && tmp != job->gang_submit) { *id = NULL; - *fence = dma_fence_get(tmp); + *fence = tmp; return 0; } + dma_fence_put(tmp); + + /* Make sure the id is owned by the gang before proceeding */ + if (!job->gang_submit || + (*id)->owner != vm->immediate.fence_context) { + + /* Don't use per engine and per process VMID at the + * same time + */ + if (adev->vm_manager.concurrent_flush) + ring = NULL; + + /* to prevent one context starved by another context */ + (*id)->pd_gpu_addr = 0; + tmp = amdgpu_sync_peek_fence(&(*id)->active, ring); + if (tmp) { + *id = NULL; + *fence = dma_fence_get(tmp); + return 0; + } + } needs_flush = true; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index e4742b65032d..cf0c4470ab9c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -77,6 +77,7 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; + reset_context.src = AMDGPU_RESET_SRC_JOB; clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); r = amdgpu_device_gpu_recover(ring->adev, job, &reset_context); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c index da40c2d97df8..9d3a3c778504 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c @@ -193,27 +193,26 @@ static int amdgpu_mca_bank_set_merge(struct mca_bank_set *mca_set, struct mca_ba return 0; } -static int amdgpu_mca_bank_set_remove_node(struct mca_bank_set *mca_set, struct mca_bank_node *node) +static void amdgpu_mca_bank_set_remove_node(struct mca_bank_set *mca_set, struct mca_bank_node *node) { if (!node) - return -EINVAL; + return; list_del(&node->node); kvfree(node); mca_set->nr_entries--; - - return 0; } static void amdgpu_mca_bank_set_release(struct mca_bank_set *mca_set) { struct mca_bank_node *node, *tmp; - list_for_each_entry_safe(node, tmp, &mca_set->list, node) { - list_del(&node->node); - kvfree(node); - } + if (list_empty(&mca_set->list)) + return; + + list_for_each_entry_safe(node, tmp, &mca_set->list, node) + amdgpu_mca_bank_set_remove_node(mca_set, node); } void amdgpu_mca_smu_init_funcs(struct amdgpu_device *adev, const struct amdgpu_mca_smu_funcs *mca_funcs) @@ -233,7 +232,7 @@ int amdgpu_mca_init(struct amdgpu_device *adev) for (i = 0; i < ARRAY_SIZE(mca->mca_caches); i++) { mca_cache = &mca->mca_caches[i]; - spin_lock_init(&mca_cache->lock); + mutex_init(&mca_cache->lock); amdgpu_mca_bank_set_init(&mca_cache->mca_set); } @@ -251,6 +250,7 @@ void amdgpu_mca_fini(struct amdgpu_device *adev) for (i = 0; i < ARRAY_SIZE(mca->mca_caches); i++) { mca_cache = &mca->mca_caches[i]; amdgpu_mca_bank_set_release(&mca_cache->mca_set); + mutex_destroy(&mca_cache->lock); } } @@ -455,9 +455,9 @@ static int amdgpu_mca_add_mca_set_to_cache(struct amdgpu_device *adev, enum amdg struct mca_bank_cache *mca_cache = &adev->mca.mca_caches[type]; int ret; - spin_lock(&mca_cache->lock); + mutex_lock(&mca_cache->lock); ret = amdgpu_mca_bank_set_merge(&mca_cache->mca_set, new); - spin_unlock(&mca_cache->lock); + mutex_unlock(&mca_cache->lock); return ret; } @@ -487,10 +487,10 @@ int amdgpu_mca_smu_log_ras_error(struct amdgpu_device *adev, enum amdgpu_ras_blo } /* dispatch mca set again if mca cache has valid data */ - spin_lock(&mca_cache->lock); + mutex_lock(&mca_cache->lock); if (mca_cache->mca_set.nr_entries) ret = amdgpu_mca_dispatch_mca_set(adev, blk, type, &mca_cache->mca_set, err_data); - spin_unlock(&mca_cache->lock); + mutex_unlock(&mca_cache->lock); out_mca_release: amdgpu_mca_bank_set_release(&mca_set); @@ -608,9 +608,7 @@ DEFINE_DEBUGFS_ATTRIBUTE(mca_debug_mode_fops, NULL, amdgpu_mca_smu_debug_mode_se void amdgpu_mca_smu_debugfs_init(struct amdgpu_device *adev, struct dentry *root) { #if defined(CONFIG_DEBUG_FS) - if (!root || - (amdgpu_ip_version(adev, MP1_HWIP, 0) != IP_VERSION(13, 0, 6) && - amdgpu_ip_version(adev, MP1_HWIP, 0) != IP_VERSION(13, 0, 14))) + if (!root) return; debugfs_create_file("mca_debug_mode", 0200, root, adev, &mca_debug_mode_fops); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h index c3c184c88dad..e80323ff90c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h @@ -84,7 +84,7 @@ struct mca_bank_set { struct mca_bank_cache { struct mca_bank_set mca_set; - spinlock_t lock; + struct mutex lock; }; struct amdgpu_mca { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index 62edf6328566..e499d6ba306b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -32,18 +32,6 @@ #define AMDGPU_MES_MAX_NUM_OF_QUEUES_PER_PROCESS 1024 #define AMDGPU_ONE_DOORBELL_SIZE 8 -signed long amdgpu_mes_fence_wait_polling(u64 *fence, - u64 wait_seq, - signed long timeout) -{ - - while ((s64)(wait_seq - *fence) > 0 && timeout > 0) { - udelay(2); - timeout -= 2; - } - return timeout > 0 ? timeout : 0; -} - int amdgpu_mes_doorbell_process_slice(struct amdgpu_device *adev) { return roundup(AMDGPU_ONE_DOORBELL_SIZE * @@ -1528,11 +1516,9 @@ int amdgpu_mes_init_microcode(struct amdgpu_device *adev, int pipe) r = amdgpu_ucode_request(adev, &adev->mes.fw[pipe], fw_name); if (r && need_retry && pipe == AMDGPU_MES_SCHED_PIPE) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mes.bin", - ucode_prefix); - DRM_INFO("try to fall back to %s\n", fw_name); + dev_info(adev->dev, "try to fall back to %s_mes.bin\n", ucode_prefix); r = amdgpu_ucode_request(adev, &adev->mes.fw[pipe], - fw_name); + "amdgpu/%s_mes.bin", ucode_prefix); } if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index df9f0404d842..e11051271f71 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -352,10 +352,6 @@ struct amdgpu_mes_funcs { #define amdgpu_mes_kiq_hw_init(adev) (adev)->mes.kiq_hw_init((adev)) #define amdgpu_mes_kiq_hw_fini(adev) (adev)->mes.kiq_hw_fini((adev)) -signed long amdgpu_mes_fence_wait_polling(u64 *fence, - u64 wait_seq, - signed long timeout); - int amdgpu_mes_ctx_get_offs(struct amdgpu_ring *ring, unsigned int id_offs); int amdgpu_mes_init_microcode(struct amdgpu_device *adev, int pipe); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 5a648a657dc6..93025522fbfc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -3181,12 +3181,10 @@ int psp_ring_cmd_submit(struct psp_context *psp, int psp_init_asd_microcode(struct psp_context *psp, const char *chip_name) { struct amdgpu_device *adev = psp->adev; - char fw_name[PSP_FW_NAME_LEN]; const struct psp_firmware_header_v1_0 *asd_hdr; int err = 0; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->psp.asd_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->psp.asd_fw, "amdgpu/%s_asd.bin", chip_name); if (err) goto out; @@ -3205,12 +3203,10 @@ out: int psp_init_toc_microcode(struct psp_context *psp, const char *chip_name) { struct amdgpu_device *adev = psp->adev; - char fw_name[PSP_FW_NAME_LEN]; const struct psp_firmware_header_v1_0 *toc_hdr; int err = 0; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_toc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->psp.toc_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->psp.toc_fw, "amdgpu/%s_toc.bin", chip_name); if (err) goto out; @@ -3362,7 +3358,6 @@ static int psp_init_sos_base_fw(struct amdgpu_device *adev) int psp_init_sos_microcode(struct psp_context *psp, const char *chip_name) { struct amdgpu_device *adev = psp->adev; - char fw_name[PSP_FW_NAME_LEN]; const struct psp_firmware_header_v1_0 *sos_hdr; const struct psp_firmware_header_v1_1 *sos_hdr_v1_1; const struct psp_firmware_header_v1_2 *sos_hdr_v1_2; @@ -3372,8 +3367,7 @@ int psp_init_sos_microcode(struct psp_context *psp, const char *chip_name) uint8_t *ucode_array_start_addr; int fw_index = 0; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sos.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, "amdgpu/%s_sos.bin", chip_name); if (err) goto out; @@ -3598,11 +3592,9 @@ int psp_init_ta_microcode(struct psp_context *psp, const char *chip_name) { const struct common_firmware_header *hdr; struct amdgpu_device *adev = psp->adev; - char fw_name[PSP_FW_NAME_LEN]; int err; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, "amdgpu/%s_ta.bin", chip_name); if (err) return err; @@ -3628,7 +3620,6 @@ int psp_init_ta_microcode(struct psp_context *psp, const char *chip_name) int psp_init_cap_microcode(struct psp_context *psp, const char *chip_name) { struct amdgpu_device *adev = psp->adev; - char fw_name[PSP_FW_NAME_LEN]; const struct psp_firmware_header_v1_0 *cap_hdr_v1_0; struct amdgpu_firmware_info *info = NULL; int err = 0; @@ -3638,8 +3629,7 @@ int psp_init_cap_microcode(struct psp_context *psp, const char *chip_name) return -EINVAL; } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_cap.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->psp.cap_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->psp.cap_fw, "amdgpu/%s_cap.bin", chip_name); if (err) { if (err == -ENODEV) { dev_warn(adev->dev, "cap microcode does not exist, skip\n"); @@ -3713,7 +3703,6 @@ static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev, struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); int ret, idx; - char fw_name[100]; const struct firmware *usbc_pd_fw; struct amdgpu_bo *fw_buf_bo = NULL; uint64_t fw_pri_mc_addr; @@ -3727,8 +3716,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev, if (!drm_dev_enter(ddev, &idx)) return -ENODEV; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s", buf); - ret = request_firmware(&usbc_pd_fw, fw_name, adev->dev); + ret = amdgpu_ucode_request(adev, &usbc_pd_fw, "amdgpu/%s", buf); if (ret) goto fail; @@ -3750,7 +3738,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev, amdgpu_bo_free_kernel(&fw_buf_bo, &fw_pri_mc_addr, &fw_pri_cpu_addr); rel_buf: - release_firmware(usbc_pd_fw); + amdgpu_ucode_release(&usbc_pd_fw); fail: if (ret) { dev_err(adev->dev, "Failed to load USBC PD FW, err = %d", ret); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index b3d11703df04..68e9935028db 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1409,11 +1409,8 @@ int amdgpu_ras_reset_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block block) { struct amdgpu_ras_block_object *block_obj = amdgpu_ras_get_ras_block(adev, block, 0); - struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs; const struct aca_smu_funcs *smu_funcs = adev->aca.smu_funcs; - struct amdgpu_hive_info *hive; - int hive_ras_recovery = 0; if (!block_obj || !block_obj->hw_ops) { dev_dbg_once(adev->dev, "%s doesn't config RAS function\n", @@ -1425,15 +1422,8 @@ int amdgpu_ras_reset_error_count(struct amdgpu_device *adev, !amdgpu_ras_get_aca_debug_mode(adev)) return -EOPNOTSUPP; - hive = amdgpu_get_xgmi_hive(adev); - if (hive) { - hive_ras_recovery = atomic_read(&hive->ras_recovery); - amdgpu_put_xgmi_hive(hive); - } - /* skip ras error reset in gpu reset */ - if ((amdgpu_in_reset(adev) || atomic_read(&ras->in_recovery) || - hive_ras_recovery) && + if ((amdgpu_in_reset(adev) || amdgpu_ras_in_recovery(adev)) && ((smu_funcs && smu_funcs->set_debug_mode) || (mca_funcs && mca_funcs->mca_set_debug_mode))) return -EOPNOTSUPP; @@ -1911,6 +1901,23 @@ static void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, obj, &amdgpu_ras_debugfs_ops); } +static bool amdgpu_ras_aca_is_supported(struct amdgpu_device *adev) +{ + bool ret; + + switch (amdgpu_ip_version(adev, MP0_HWIP, 0)) { + case IP_VERSION(13, 0, 6): + case IP_VERSION(13, 0, 14): + ret = true; + break; + default: + ret = false; + break; + } + + return ret; +} + void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); @@ -1937,10 +1944,12 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) } } - if (amdgpu_aca_is_enabled(adev)) - amdgpu_aca_smu_debugfs_init(adev, dir); - else - amdgpu_mca_smu_debugfs_init(adev, dir); + if (amdgpu_ras_aca_is_supported(adev)) { + if (amdgpu_aca_is_enabled(adev)) + amdgpu_aca_smu_debugfs_init(adev, dir); + else + amdgpu_mca_smu_debugfs_init(adev, dir); + } } /* debugfs end */ @@ -2049,8 +2058,9 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager * struct amdgpu_device *adev = obj->adev; struct amdgpu_ras_block_object *block_obj = amdgpu_ras_get_ras_block(adev, obj->head.block, 0); + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - if (!block_obj) + if (!block_obj || !con) return; /* both query_poison_status and handle_poison_consumption are optional, @@ -2073,14 +2083,17 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager * if (block_obj->hw_ops && block_obj->hw_ops->handle_poison_consumption) poison_stat = block_obj->hw_ops->handle_poison_consumption(adev); - /* gpu reset is fallback for failed and default cases */ - if (poison_stat) { + /* gpu reset is fallback for failed and default cases. + * For RMA case, amdgpu_umc_poison_handler will handle gpu reset. + */ + if (poison_stat && !con->is_rma) { dev_info(adev->dev, "GPU reset for %s RAS poison consumption is issued!\n", block_obj->ras_comm.name); amdgpu_ras_reset_gpu(adev); - } else { - amdgpu_gfx_poison_consumption_handler(adev, entry); } + + if (!poison_stat) + amdgpu_gfx_poison_consumption_handler(adev, entry); } static void amdgpu_ras_interrupt_poison_creation_handler(struct ras_manager *obj, @@ -2118,6 +2131,7 @@ static void amdgpu_ras_interrupt_umc_handler(struct ras_manager *obj, /* Let IP handle its data, maybe we need get the output * from the callback to update the error type/count, etc */ + amdgpu_ras_set_fed(obj->adev, true); ret = data->cb(obj->adev, &err_data, entry); /* ue will trigger an interrupt, and in that case * we need do a reset to recovery the whole system. @@ -2438,6 +2452,23 @@ static void amdgpu_ras_set_fed_all(struct amdgpu_device *adev, } } +bool amdgpu_ras_in_recovery(struct amdgpu_device *adev) +{ + struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev); + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + int hive_ras_recovery = 0; + + if (hive) { + hive_ras_recovery = atomic_read(&hive->ras_recovery); + amdgpu_put_xgmi_hive(hive); + } + + if (ras && (atomic_read(&ras->in_recovery) || hive_ras_recovery)) + return true; + + return false; +} + static void amdgpu_ras_do_recovery(struct work_struct *work) { struct amdgpu_ras *ras = @@ -2487,6 +2518,7 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) reset_context.method = AMD_RESET_METHOD_NONE; reset_context.reset_req_dev = adev; + reset_context.src = AMDGPU_RESET_SRC_RAS; /* Perform full reset in fatal error mode */ if (!amdgpu_ras_is_poison_mode_supported(ras->adev)) @@ -2795,16 +2827,21 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work) page_retirement_dwork.work); struct amdgpu_device *adev = con->adev; struct ras_err_data err_data; + unsigned long err_cnt; - if (amdgpu_in_reset(adev) || atomic_read(&con->in_recovery)) + if (amdgpu_in_reset(adev) || amdgpu_ras_in_recovery(adev)) return; amdgpu_ras_error_data_init(&err_data); amdgpu_umc_handle_bad_pages(adev, &err_data); + err_cnt = err_data.err_addr_cnt; amdgpu_ras_error_data_fini(&err_data); + if (err_cnt && con->is_rma) + amdgpu_ras_reset_gpu(adev); + mutex_lock(&con->umc_ecc_log.lock); if (radix_tree_tagged(&con->umc_ecc_log.de_page_tree, UMC_ECC_NEW_DETECTED_TAG)) @@ -2861,7 +2898,8 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev, if (poison_msg->pasid_fn) poison_msg->pasid_fn(adev, pasid, poison_msg->data); - if (reset) { + /* for RMA, amdgpu_ras_poison_creation_handler will trigger gpu reset */ + if (reset && !con->is_rma) { flush_delayed_work(&con->page_retirement_dwork); con->gpu_reset_flags |= reset; @@ -3427,6 +3465,15 @@ int amdgpu_ras_init(struct amdgpu_device *adev) goto release_con; } + if (amdgpu_ras_aca_is_supported(adev)) { + if (amdgpu_aca_is_enabled(adev)) + r = amdgpu_aca_init(adev); + else + r = amdgpu_mca_init(adev); + if (r) + goto release_con; + } + dev_info(adev->dev, "RAS INFO: ras initialized successfully, " "hardware ability[%x] ras_mask[%x]\n", adev->ras_hw_enabled, adev->ras_enabled); @@ -3635,25 +3682,22 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev) amdgpu_ras_event_mgr_init(adev); - if (amdgpu_aca_is_enabled(adev)) { - if (!amdgpu_in_reset(adev)) { - r = amdgpu_aca_init(adev); + if (amdgpu_ras_aca_is_supported(adev)) { + if (amdgpu_in_reset(adev)) { + if (amdgpu_aca_is_enabled(adev)) + r = amdgpu_aca_reset(adev); + else + r = amdgpu_mca_reset(adev); if (r) return r; } - if (!amdgpu_sriov_vf(adev)) - amdgpu_ras_set_aca_debug_mode(adev, false); - } else { - if (amdgpu_in_reset(adev)) - r = amdgpu_mca_reset(adev); - else - r = amdgpu_mca_init(adev); - if (r) - return r; - - if (!amdgpu_sriov_vf(adev)) - amdgpu_ras_set_mca_debug_mode(adev, false); + if (!amdgpu_sriov_vf(adev)) { + if (amdgpu_aca_is_enabled(adev)) + amdgpu_ras_set_aca_debug_mode(adev, false); + else + amdgpu_ras_set_mca_debug_mode(adev, false); + } } /* Guest side doesn't need init ras feature */ @@ -3727,10 +3771,12 @@ int amdgpu_ras_fini(struct amdgpu_device *adev) amdgpu_ras_fs_fini(adev); amdgpu_ras_interrupt_remove_all(adev); - if (amdgpu_aca_is_enabled(adev)) - amdgpu_aca_fini(adev); - else - amdgpu_mca_fini(adev); + if (amdgpu_ras_aca_is_supported(adev)) { + if (amdgpu_aca_is_enabled(adev)) + amdgpu_aca_fini(adev); + else + amdgpu_mca_fini(adev); + } WARN(AMDGPU_RAS_GET_FEATURES(con->features), "Feature mask is not cleared"); @@ -3982,6 +4028,12 @@ int amdgpu_ras_reset_gpu(struct amdgpu_device *adev) { struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + /* mode1 is the only selection for RMA status */ + if (ras->is_rma) { + ras->gpu_reset_flags = 0; + ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE1_RESET; + } + if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0) amdgpu_reset_domain_schedule(ras->adev->reset_domain, &ras->recovery_work); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index e70c45712ddb..83437fef9df5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -954,6 +954,8 @@ int amdgpu_ras_put_poison_req(struct amdgpu_device *adev, enum amdgpu_ras_block block, uint16_t pasid, pasid_notify pasid_fn, void *data, uint32_t reset); +bool amdgpu_ras_in_recovery(struct amdgpu_device *adev); + __printf(3, 4) void amdgpu_ras_event_log_print(struct amdgpu_device *adev, u64 event_id, const char *fmt, ...); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c index 9deb41d61e8d..66c1a868c0e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c @@ -164,16 +164,14 @@ void amdgpu_device_unlock_reset_domain(struct amdgpu_reset_domain *reset_domain) void amdgpu_reset_get_desc(struct amdgpu_reset_context *rst_ctxt, char *buf, size_t len) { - struct amdgpu_ring *ring; - if (!buf || !len) return; switch (rst_ctxt->src) { case AMDGPU_RESET_SRC_JOB: if (rst_ctxt->job) { - ring = amdgpu_job_ring(rst_ctxt->job); - snprintf(buf, len, "job hang on ring:%s", ring->name); + snprintf(buf, len, "job hang on ring:%s", + rst_ctxt->job->base.sched->name); } else { strscpy(buf, "job hang", len); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 151f83ea803b..183a976ba29d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -215,14 +215,14 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, const struct sdma_firmware_header_v3_0 *sdma_hv3; uint16_t version_major; char ucode_prefix[30]; - char fw_name[52]; amdgpu_ucode_ip_version_decode(adev, SDMA0_HWIP, ucode_prefix, sizeof(ucode_prefix)); if (instance == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s.bin", ucode_prefix); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[instance].fw, + "amdgpu/%s.bin", ucode_prefix); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s%d.bin", ucode_prefix, instance); - err = amdgpu_ucode_request(adev, &adev->sdma.instance[instance].fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[instance].fw, + "amdgpu/%s%d.bin", ucode_prefix, instance); if (err) goto out; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 8de2d05d0735..58906bf7448e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -295,8 +295,8 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, struct amdgpu_res_cursor src_mm, dst_mm; struct dma_fence *fence = NULL; int r = 0; - uint32_t copy_flags = 0; + struct amdgpu_bo *abo_src, *abo_dst; if (!adev->mman.buffer_funcs_enabled) { DRM_ERROR("Trying to move memory with ring turned off.\n"); @@ -325,8 +325,14 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, if (r) goto error; + abo_src = ttm_to_amdgpu_bo(src->bo); + abo_dst = ttm_to_amdgpu_bo(dst->bo); if (tmz) copy_flags |= AMDGPU_COPY_FLAGS_TMZ; + if (abo_src->flags & AMDGPU_GEM_CREATE_GFX12_DCC) + copy_flags |= AMDGPU_COPY_FLAGS_READ_DECOMPRESSED; + if (abo_dst->flags & AMDGPU_GEM_CREATE_GFX12_DCC) + copy_flags |= AMDGPU_COPY_FLAGS_WRITE_COMPRESSED; r = amdgpu_copy_buffer(ring, from, to, cur_size, resv, &next, false, true, copy_flags); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index b6f53129dea3..f2eb1cf364c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -110,6 +110,8 @@ struct amdgpu_copy_mem { }; #define AMDGPU_COPY_FLAGS_TMZ (1 << 0) +#define AMDGPU_COPY_FLAGS_READ_DECOMPRESSED (1 << 1) +#define AMDGPU_COPY_FLAGS_WRITE_COMPRESSED (1 << 2) int amdgpu_gtt_mgr_init(struct amdgpu_device *adev, uint64_t gtt_size); void amdgpu_gtt_mgr_fini(struct amdgpu_device *adev); @@ -146,7 +148,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev); void amdgpu_ttm_fini(struct amdgpu_device *adev); void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable); - int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, uint64_t dst_offset, uint32_t byte_count, struct dma_resv *resv, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 75ece8a2f96b..3588f1c5a007 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -28,6 +28,8 @@ #include "amdgpu.h" #include "amdgpu_ucode.h" +#define AMDGPU_UCODE_NAME_MAX (128) + static void amdgpu_ucode_print_common_hdr(const struct common_firmware_header *hdr) { DRM_DEBUG("size_bytes: %u\n", le32_to_cpu(hdr->size_bytes)); @@ -1427,28 +1429,40 @@ void amdgpu_ucode_ip_version_decode(struct amdgpu_device *adev, int block_type, * * @adev: amdgpu device * @fw: pointer to load firmware to - * @fw_name: firmware to load + * @fmt: firmware name format string + * @...: variable arguments * * This is a helper that will use request_firmware and amdgpu_ucode_validate * to load and run basic validation on firmware. If the load fails, remap * the error code to -ENODEV, so that early_init functions will fail to load. */ int amdgpu_ucode_request(struct amdgpu_device *adev, const struct firmware **fw, - const char *fw_name) + const char *fmt, ...) { - int err = request_firmware(fw, fw_name, adev->dev); + char fname[AMDGPU_UCODE_NAME_MAX]; + va_list ap; + int r; + + va_start(ap, fmt); + r = vsnprintf(fname, sizeof(fname), fmt, ap); + va_end(ap); + if (r == sizeof(fname)) { + dev_warn(adev->dev, "amdgpu firmware name buffer overflow\n"); + return -EOVERFLOW; + } - if (err) + r = request_firmware(fw, fname, adev->dev); + if (r) return -ENODEV; - err = amdgpu_ucode_validate(*fw); - if (err) { - dev_dbg(adev->dev, "\"%s\" failed to validate\n", fw_name); + r = amdgpu_ucode_validate(*fw); + if (r) { + dev_dbg(adev->dev, "\"%s\" failed to validate\n", fname); release_firmware(*fw); *fw = NULL; } - return err; + return r; } /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index a3c04f711099..f4e5285c4dd6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -593,8 +593,9 @@ void amdgpu_ucode_print_rlc_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_sdma_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_psp_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_gpu_info_hdr(const struct common_firmware_header *hdr); +__printf(3, 4) int amdgpu_ucode_request(struct amdgpu_device *adev, const struct firmware **fw, - const char *fw_name); + const char *fmt, ...); void amdgpu_ucode_release(const struct firmware **fw); bool amdgpu_ucode_hdr_version(union amdgpu_firmware_header *hdr, uint16_t hdr_major, uint16_t hdr_minor); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 540e0f066b26..20e0e522fb51 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -195,7 +195,8 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); amdgpu_umc_handle_bad_pages(adev, ras_error_status); - if (err_data->ue_count && reset) { + if ((err_data->ue_count || err_data->de_count) && + (reset || (con && con->is_rma))) { con->gpu_reset_flags |= reset; amdgpu_ras_reset_gpu(adev); } @@ -211,6 +212,7 @@ int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev, .block = AMDGPU_RAS_BLOCK__UMC, }; struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); uint32_t timeout = timeout_ms; memset(&err_data, 0, sizeof(err_data)); @@ -243,9 +245,7 @@ int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev, kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); - if (reset) { - struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - + if (reset || (err_data.err_addr_cnt && con && con->is_rma)) { con->gpu_reset_flags |= reset; amdgpu_ras_reset_gpu(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index 5e2b7c340724..8d65b096db90 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -94,18 +94,14 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work); int amdgpu_vcn_early_init(struct amdgpu_device *adev) { char ucode_prefix[25]; - char fw_name[40]; int r, i; + amdgpu_ucode_ip_version_decode(adev, UVD_HWIP, ucode_prefix, sizeof(ucode_prefix)); for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - amdgpu_ucode_ip_version_decode(adev, UVD_HWIP, ucode_prefix, sizeof(ucode_prefix)); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s.bin", ucode_prefix); - if (amdgpu_ip_version(adev, UVD_HWIP, 0) == IP_VERSION(4, 0, 6) && - i == 1) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_%d.bin", ucode_prefix, i); - } - - r = amdgpu_ucode_request(adev, &adev->vcn.fw[i], fw_name); + if (i == 1 && amdgpu_ip_version(adev, UVD_HWIP, 0) == IP_VERSION(4, 0, 6)) + r = amdgpu_ucode_request(adev, &adev->vcn.fw[i], "amdgpu/%s_%d.bin", ucode_prefix, i); + else + r = amdgpu_ucode_request(adev, &adev->vcn.fw[i], "amdgpu/%s.bin", ucode_prefix); if (r) { amdgpu_ucode_release(&adev->vcn.fw[i]); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index b56d243c53ab..63f2286858c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -153,6 +153,20 @@ void amdgpu_virt_request_init_data(struct amdgpu_device *adev) } /** + * amdgpu_virt_ready_to_reset() - send ready to reset to host + * @adev: amdgpu device. + * Send ready to reset message to GPU hypervisor to signal we have stopped GPU + * activity and is ready for host FLR + */ +void amdgpu_virt_ready_to_reset(struct amdgpu_device *adev) +{ + struct amdgpu_virt *virt = &adev->virt; + + if (virt->ops && virt->ops->reset_gpu) + virt->ops->ready_to_reset(adev); +} + +/** * amdgpu_virt_wait_reset() - wait for reset gpu completed * @adev: amdgpu device. * Wait for GPU reset completed. @@ -979,6 +993,9 @@ u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 f return 0; } + if (amdgpu_device_skip_hw_access(adev)) + return 0; + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[xcc_id]; scratch_reg0 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg0; scratch_reg1 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg1; @@ -1055,6 +1072,9 @@ void amdgpu_sriov_wreg(struct amdgpu_device *adev, { u32 rlcg_flag; + if (amdgpu_device_skip_hw_access(adev)) + return; + if (!amdgpu_sriov_runtime(adev) && amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, true, &rlcg_flag)) { amdgpu_virt_rlcg_reg_rw(adev, offset, value, rlcg_flag, xcc_id); @@ -1072,6 +1092,9 @@ u32 amdgpu_sriov_rreg(struct amdgpu_device *adev, { u32 rlcg_flag; + if (amdgpu_device_skip_hw_access(adev)) + return 0; + if (!amdgpu_sriov_runtime(adev) && amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, false, &rlcg_flag)) return amdgpu_virt_rlcg_reg_rw(adev, offset, 0, rlcg_flag, xcc_id); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 0ec246c74570..f04cd1586c72 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -88,6 +88,7 @@ struct amdgpu_virt_ops { int (*rel_full_gpu)(struct amdgpu_device *adev, bool init); int (*req_init_data)(struct amdgpu_device *adev); int (*reset_gpu)(struct amdgpu_device *adev); + void (*ready_to_reset)(struct amdgpu_device *adev); int (*wait_reset)(struct amdgpu_device *adev); void (*trans_msg)(struct amdgpu_device *adev, enum idh_request req, u32 data1, u32 data2, u32 data3); @@ -347,6 +348,7 @@ int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_release_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_reset_gpu(struct amdgpu_device *adev); void amdgpu_virt_request_init_data(struct amdgpu_device *adev); +void amdgpu_virt_ready_to_reset(struct amdgpu_device *adev); int amdgpu_virt_wait_reset(struct amdgpu_device *adev); int amdgpu_virt_alloc_mm_table(struct amdgpu_device *adev); void amdgpu_virt_free_mm_table(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 8e3501e3765b..046949c4b695 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -128,6 +128,7 @@ struct amdgpu_mem_stats; (((uint64_t)(flags) & (~AMDGPU_PTE_MTYPE_GFX12_MASK)) | \ AMDGPU_PTE_MTYPE_GFX12_SHIFT(mtype)) +#define AMDGPU_PTE_DCC (1ULL << 58) #define AMDGPU_PTE_IS_PTE (1ULL << 63) /* PDE Block Fragment Size for gfx v12 */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c index 49881073ff58..bad232859972 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c @@ -232,13 +232,11 @@ int amdgpu_vpe_init_microcode(struct amdgpu_vpe *vpe) { struct amdgpu_device *adev = vpe->ring.adev; const struct vpe_firmware_header_v1_0 *vpe_hdr; - char fw_prefix[32], fw_name[64]; + char fw_prefix[32]; int ret; amdgpu_ucode_ip_version_decode(adev, VPE_HWIP, fw_prefix, sizeof(fw_prefix)); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s.bin", fw_prefix); - - ret = amdgpu_ucode_request(adev, &adev->vpe.fw, fw_name); + ret = amdgpu_ucode_request(adev, &adev->vpe.fw, "amdgpu/%s.bin", fw_prefix); if (ret) goto out; diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index d552e013354c..09715b506468 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -301,7 +301,7 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, (*ptr) += 4; if (print) DEBUG("IMM 0x%08X\n", val); - return val; + break; case ATOM_SRC_WORD0: case ATOM_SRC_WORD8: case ATOM_SRC_WORD16: @@ -309,7 +309,7 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, (*ptr) += 2; if (print) DEBUG("IMM 0x%04X\n", val); - return val; + break; case ATOM_SRC_BYTE0: case ATOM_SRC_BYTE8: case ATOM_SRC_BYTE16: @@ -318,9 +318,9 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, (*ptr)++; if (print) DEBUG("IMM 0x%02X\n", val); - return val; + break; } - break; + return val; case ATOM_ARG_PLL: idx = U8(*ptr); (*ptr)++; diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c index 6948ebda0fa2..952737de9411 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c @@ -107,7 +107,6 @@ static void cik_sdma_free_microcode(struct amdgpu_device *adev) static int cik_sdma_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err = 0, i; DRM_DEBUG("\n"); @@ -133,16 +132,18 @@ static int cik_sdma_init_microcode(struct amdgpu_device *adev) for (i = 0; i < adev->sdma.num_instances; i++) { if (i == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + "amdgpu/%s_sdma.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + "amdgpu/%s_sdma1.bin", chip_name); if (err) goto out; } out: if (err) { - pr_err("cik_sdma: Failed to load firmware \"%s\"\n", fw_name); + pr_err("cik_sdma: Failed to load firmware \"%s_sdma%s.bin\"\n", + chip_name, i == 0 ? "" : "1"); for (i = 0; i < adev->sdma.num_instances; i++) amdgpu_ucode_release(&adev->sdma.instance[i].fw); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 3b0d8d3af58a..18488c02d1cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -4116,7 +4116,6 @@ static void gfx_v10_0_check_gfxoff_flag(struct amdgpu_device *adev) static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) { - char fw_name[53]; char ucode_prefix[30]; const char *wks = ""; int err; @@ -4131,27 +4130,27 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) wks = "_wks"; amdgpu_ucode_ip_version_decode(adev, GC_HWIP, ucode_prefix, sizeof(ucode_prefix)); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp%s.bin", ucode_prefix, wks); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp%s.bin", ucode_prefix, wks); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_PFP); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me%s.bin", ucode_prefix, wks); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me%s.bin", ucode_prefix, wks); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_ME); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce%s.bin", ucode_prefix, wks); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce%s.bin", ucode_prefix, wks); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_CE); if (!amdgpu_sriov_vf(adev)) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", ucode_prefix); - err = request_firmware(&adev->gfx.rlc_fw, fw_name, adev->dev); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc.bin", ucode_prefix); if (err) goto out; @@ -4166,15 +4165,15 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) goto out; } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec%s.bin", ucode_prefix, wks); - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec%s.bin", ucode_prefix, wks); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1_JT); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2%s.bin", ucode_prefix, wks); - err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_mec2%s.bin", ucode_prefix, wks); if (!err) { amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC2); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC2_JT); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 19d95455780f..2a510351dfce 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -4415,7 +4415,9 @@ static int gfx_v11_0_gfxhub_enable(struct amdgpu_device *adev) false : true; adev->gfxhub.funcs->set_fault_enable_default(adev, value); - amdgpu_gmc_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); + /* TODO investigate why this and the hdp flush above is needed, + * are we missing a flush somewhere else? */ + adev->gmc.gmc_funcs->flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c index 9e7ce1e6bc06..9cd221ed240c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c @@ -85,6 +85,7 @@ static int gfx_v11_0_3_poison_consumption_handler(struct amdgpu_device *adev, if (entry && (entry->client_id == SOC21_IH_CLIENTID_GFX) && (entry->src_id == GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT) && !entry->vmid && !entry->pasid) { + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); uint32_t rlc_status0 = 0; rlc_status0 = RREG32_SOC15(GC, 0, regRLC_RLCS_FED_STATUS_0); @@ -96,7 +97,8 @@ static int gfx_v11_0_3_poison_consumption_handler(struct amdgpu_device *adev, ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET; } - amdgpu_ras_reset_gpu(adev); + if (con && !con->is_rma) + amdgpu_ras_reset_gpu(adev); } return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index 6419f98e32b6..460bf33a22b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -3192,7 +3192,9 @@ static int gfx_v12_0_gfxhub_enable(struct amdgpu_device *adev) false : true; adev->gfxhub.funcs->set_fault_enable_default(adev, value); - amdgpu_gmc_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); + /* TODO investigate why this and the hdp flush above is needed, + * are we missing a flush somewhere else? */ + adev->gmc.gmc_funcs->flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index d0992ce9fb47..564f0b9336b6 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -311,7 +311,6 @@ static const u32 verde_rlc_save_restore_register_list[] = static int gfx_v6_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err; const struct gfx_firmware_header_v1_0 *cp_hdr; const struct rlc_firmware_header_v1_0 *rlc_hdr; @@ -337,32 +336,32 @@ static int gfx_v6_0_init_microcode(struct amdgpu_device *adev) default: BUG(); } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp.bin", chip_name); if (err) goto out; cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data; adev->gfx.pfp_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); adev->gfx.pfp_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me.bin", chip_name); if (err) goto out; cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data; adev->gfx.me_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); adev->gfx.me_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce.bin", chip_name); if (err) goto out; cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data; adev->gfx.ce_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); adev->gfx.ce_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc.bin", chip_name); if (err) goto out; rlc_hdr = (const struct rlc_firmware_header_v1_0 *)adev->gfx.rlc_fw->data; @@ -371,7 +370,7 @@ static int gfx_v6_0_init_microcode(struct amdgpu_device *adev) out: if (err) { - pr_err("gfx6: Failed to load firmware \"%s\"\n", fw_name); + pr_err("gfx6: Failed to load firmware %s gfx firmware\n", chip_name); amdgpu_ucode_release(&adev->gfx.pfp_fw); amdgpu_ucode_release(&adev->gfx.me_fw); amdgpu_ucode_release(&adev->gfx.ce_fw); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 8f19b6ae8d5b..d84589137df9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -909,7 +909,6 @@ static void gfx_v7_0_free_microcode(struct amdgpu_device *adev) static int gfx_v7_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err; DRM_DEBUG("\n"); @@ -934,38 +933,38 @@ static int gfx_v7_0_init_microcode(struct amdgpu_device *adev) BUG(); } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp.bin", chip_name); if (err) goto out; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me.bin", chip_name); if (err) goto out; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce.bin", chip_name); if (err) goto out; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec.bin", chip_name); if (err) goto out; if (adev->asic_type == CHIP_KAVERI) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_mec2.bin", chip_name); if (err) goto out; } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc.bin", chip_name); out: if (err) { - pr_err("gfx7: Failed to load firmware \"%s\"\n", fw_name); + pr_err("gfx7: Failed to load firmware %s gfx firmware\n", chip_name); gfx_v7_0_free_microcode(adev); } return err; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 2f0e72caee1a..b4658c7db0e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -939,7 +939,6 @@ static void gfx_v8_0_free_microcode(struct amdgpu_device *adev) static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; @@ -982,15 +981,15 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) } if (adev->asic_type >= CHIP_POLARIS10 && adev->asic_type <= CHIP_POLARIS12) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp_2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp_2.bin", chip_name); if (err == -ENODEV) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp.bin", chip_name); } } else { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp.bin", chip_name); } if (err) goto out; @@ -999,15 +998,15 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) adev->gfx.pfp_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); if (adev->asic_type >= CHIP_POLARIS10 && adev->asic_type <= CHIP_POLARIS12) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me_2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me_2.bin", chip_name); if (err == -ENODEV) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me.bin", chip_name); } } else { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me.bin", chip_name); } if (err) goto out; @@ -1017,15 +1016,15 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) adev->gfx.me_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); if (adev->asic_type >= CHIP_POLARIS10 && adev->asic_type <= CHIP_POLARIS12) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce_2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce_2.bin", chip_name); if (err == -ENODEV) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce.bin", chip_name); } } else { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce.bin", chip_name); } if (err) goto out; @@ -1044,8 +1043,8 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) } else adev->virt.chained_ib_support = false; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc.bin", chip_name); if (err) goto out; rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; @@ -1093,15 +1092,15 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) adev->gfx.rlc.register_restore[i] = le32_to_cpu(tmp[i]); if (adev->asic_type >= CHIP_POLARIS10 && adev->asic_type <= CHIP_POLARIS12) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec_2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec_2.bin", chip_name); if (err == -ENODEV) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec.bin", chip_name); } } else { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec.bin", chip_name); } if (err) goto out; @@ -1112,15 +1111,15 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) if ((adev->asic_type != CHIP_STONEY) && (adev->asic_type != CHIP_TOPAZ)) { if (adev->asic_type >= CHIP_POLARIS10 && adev->asic_type <= CHIP_POLARIS12) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2_2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_mec2_2.bin", chip_name); if (err == -ENODEV) { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_mec2.bin", chip_name); } } else { - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_mec2.bin", chip_name); } if (!err) { cp_hdr = (const struct gfx_firmware_header_v1_0 *) @@ -1194,9 +1193,7 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) out: if (err) { - dev_err(adev->dev, - "gfx8: Failed to load firmware \"%s\"\n", - fw_name); + dev_err(adev->dev, "gfx8: Failed to load firmware %s gfx firmware\n", chip_name); amdgpu_ucode_release(&adev->gfx.pfp_fw); amdgpu_ucode_release(&adev->gfx.me_fw); amdgpu_ucode_release(&adev->gfx.ce_fw); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index ad99ee10fd4a..2929c8972ea7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -1378,23 +1378,22 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev) static int gfx_v9_0_init_cp_gfx_microcode(struct amdgpu_device *adev, char *chip_name) { - char fw_name[50]; int err; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.pfp_fw, + "amdgpu/%s_pfp.bin", chip_name); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_PFP); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.me_fw, + "amdgpu/%s_me.bin", chip_name); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_ME); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.ce_fw, + "amdgpu/%s_ce.bin", chip_name); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_CE); @@ -1411,7 +1410,6 @@ out: static int gfx_v9_0_init_rlc_microcode(struct amdgpu_device *adev, char *chip_name) { - char fw_name[53]; int err; const struct rlc_firmware_header_v2_0 *rlc_hdr; uint16_t version_major; @@ -1429,20 +1427,22 @@ static int gfx_v9_0_init_rlc_microcode(struct amdgpu_device *adev, if (!strcmp(chip_name, "picasso") && (((adev->pdev->revision >= 0xC8) && (adev->pdev->revision <= 0xCF)) || ((adev->pdev->revision >= 0xD8) && (adev->pdev->revision <= 0xDF)))) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc_am4.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc_am4.bin", chip_name); else if (!strcmp(chip_name, "raven") && (amdgpu_pm_load_smu_firmware(adev, &smu_version) == 0) && (smu_version >= 0x41e2b)) /** *SMC is loaded by SBIOS on APU and it's able to get the SMU version directly. */ - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_kicker_rlc.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_kicker_rlc.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc.bin", chip_name); if (err) goto out; - rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; + rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; version_major = le16_to_cpu(rlc_hdr->header.header_version_major); version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor); err = amdgpu_gfx_rlc_init_microcode(adev, version_major, version_minor); @@ -1466,28 +1466,27 @@ static bool gfx_v9_0_load_mec2_fw_bin_support(struct amdgpu_device *adev) static int gfx_v9_0_init_cp_compute_microcode(struct amdgpu_device *adev, char *chip_name) { - char fw_name[50]; int err; if (amdgpu_sriov_vf(adev) && (adev->asic_type == CHIP_ALDEBARAN)) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sjt_mec.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_sjt_mec.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); - - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec.bin", chip_name); if (err) goto out; + amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1_JT); if (gfx_v9_0_load_mec2_fw_bin_support(adev)) { if (amdgpu_sriov_vf(adev) && (adev->asic_type == CHIP_ALDEBARAN)) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sjt_mec2.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_sjt_mec2.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name); - - /* ignore failures to load */ - err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec2_fw, + "amdgpu/%s_mec2.bin", chip_name); if (!err) { amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC2); amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC2_JT); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index aecc2bcea145..8d8763ebe027 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -372,15 +372,14 @@ static void gfx_v9_4_3_free_microcode(struct amdgpu_device *adev) static int gfx_v9_4_3_init_rlc_microcode(struct amdgpu_device *adev, const char *chip_name) { - char fw_name[30]; int err; const struct rlc_firmware_header_v2_0 *rlc_hdr; uint16_t version_major; uint16_t version_minor; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + "amdgpu/%s_rlc.bin", chip_name); if (err) goto out; rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; @@ -409,12 +408,10 @@ static void gfx_v9_4_3_check_if_need_gfxoff(struct amdgpu_device *adev) static int gfx_v9_4_3_init_cp_compute_microcode(struct amdgpu_device *adev, const char *chip_name) { - char fw_name[30]; int err; - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); - - err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, + "amdgpu/%s_mec.bin", chip_name); if (err) goto out; amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1); @@ -4203,9 +4200,10 @@ static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev, int xcc_i static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info) { - int i, j, k, counter, xcc_id, active_cu_number = 0; - u32 mask, bitmap, ao_bitmap, ao_cu_mask = 0; + int i, j, k, prev_counter, counter, xcc_id, active_cu_number = 0; + u32 mask, bitmap, ao_bitmap, ao_cu_mask = 0, tmp; unsigned disable_masks[4 * 4]; + bool is_symmetric_cus; if (!adev || !cu_info) return -EINVAL; @@ -4223,6 +4221,7 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, mutex_lock(&adev->grbm_idx_mutex); for (xcc_id = 0; xcc_id < NUM_XCC(adev->gfx.xcc_mask); xcc_id++) { + is_symmetric_cus = true; for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { mask = 1; @@ -4250,6 +4249,15 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, ao_cu_mask |= (ao_bitmap << (i * 16 + j * 8)); cu_info->ao_cu_bitmap[i][j] = ao_bitmap; } + if (i && is_symmetric_cus && prev_counter != counter) + is_symmetric_cus = false; + prev_counter = counter; + } + if (is_symmetric_cus) { + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_DEBUG); + tmp = REG_SET_FIELD(tmp, CP_CPC_DEBUG, CPC_HARVESTING_RELAUNCH_DISABLE, 1); + tmp = REG_SET_FIELD(tmp, CP_CPC_DEBUG, CPC_HARVESTING_DISPATCH_DISABLE, 1); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_DEBUG, tmp); } gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, xcc_id); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c index 16d3443f8a82..be78507ec0d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c @@ -532,6 +532,9 @@ static void gmc_v12_0_get_vm_pte(struct amdgpu_device *adev, is_system = (bo->tbo.resource->mem_type == TTM_PL_TT) || (bo->tbo.resource->mem_type == AMDGPU_PL_PREEMPT); + if (bo && bo->flags & AMDGPU_GEM_CREATE_GFX12_DCC) + *flags |= AMDGPU_PTE_DCC; + /* WA for HW bug */ if (is_system || ((bo_adev != adev) && coherent)) *flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_NC); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index 3e38d8bfcb69..d36725666b54 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -98,9 +98,7 @@ static void gmc_v6_0_mc_resume(struct amdgpu_device *adev) static int gmc_v6_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err; - bool is_58_fw = false; DRM_DEBUG("\n"); @@ -126,17 +124,13 @@ static int gmc_v6_0_init_microcode(struct amdgpu_device *adev) /* this memory configuration requires special firmware */ if (((RREG32(mmMC_SEQ_MISC0) & 0xff000000) >> 24) == 0x58) - is_58_fw = true; + chip_name = "si58"; - if (is_58_fw) - snprintf(fw_name, sizeof(fw_name), "amdgpu/si58_mc.bin"); - else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gmc.fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gmc.fw, "amdgpu/%s_mc.bin", chip_name); if (err) { dev_err(adev->dev, - "si_mc: Failed to load firmware \"%s\"\n", - fw_name); + "si_mc: Failed to load firmware \"%s_mc.bin\"\n", + chip_name); amdgpu_ucode_release(&adev->gmc.fw); } return err; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 85df8fc81065..994432fb57ea 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -130,7 +130,6 @@ static void gmc_v7_0_mc_resume(struct amdgpu_device *adev) static int gmc_v7_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err; DRM_DEBUG("\n"); @@ -153,11 +152,9 @@ static int gmc_v7_0_init_microcode(struct amdgpu_device *adev) return -EINVAL; } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mc.bin", chip_name); - - err = amdgpu_ucode_request(adev, &adev->gmc.fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gmc.fw, "amdgpu/%s_mc.bin", chip_name); if (err) { - pr_err("cik_mc: Failed to load firmware \"%s\"\n", fw_name); + pr_err("cik_mc: Failed to load firmware \"%s_mc.bin\"\n", chip_name); amdgpu_ucode_release(&adev->gmc.fw); } return err; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index fc97757e33d9..86488c052f82 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -212,7 +212,6 @@ static void gmc_v8_0_mc_resume(struct amdgpu_device *adev) static int gmc_v8_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err; DRM_DEBUG("\n"); @@ -255,10 +254,9 @@ static int gmc_v8_0_init_microcode(struct amdgpu_device *adev) return -EINVAL; } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mc.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->gmc.fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gmc.fw, "amdgpu/%s_mc.bin", chip_name); if (err) { - pr_err("mc: Failed to load firmware \"%s\"\n", fw_name); + pr_err("mc: Failed to load firmware \"%s_mc.bin\"\n", chip_name); amdgpu_ucode_release(&adev->gmc.fw); } return err; diff --git a/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c b/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c index 3e91a8e42c21..a9f5d9e4610d 100644 --- a/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c @@ -41,7 +41,6 @@ MODULE_FIRMWARE("amdgpu/gc_11_5_1_imu.bin"); static int imu_v11_0_init_microcode(struct amdgpu_device *adev) { - char fw_name[45]; char ucode_prefix[30]; int err; const struct imu_firmware_header_v1_0 *imu_hdr; @@ -50,11 +49,10 @@ static int imu_v11_0_init_microcode(struct amdgpu_device *adev) DRM_DEBUG("\n"); amdgpu_ucode_ip_version_decode(adev, GC_HWIP, ucode_prefix, sizeof(ucode_prefix)); - - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_imu.bin", ucode_prefix); - err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, "amdgpu/%s_imu.bin", ucode_prefix); if (err) goto out; + imu_hdr = (const struct imu_firmware_header_v1_0 *)adev->gfx.imu_fw->data; //adev->gfx.imu_feature_version = le32_to_cpu(imu_hdr->ucode_feature_version); @@ -75,8 +73,8 @@ static int imu_v11_0_init_microcode(struct amdgpu_device *adev) out: if (err) { dev_err(adev->dev, - "gfx11: Failed to load firmware \"%s\"\n", - fw_name); + "gfx11: Failed to load firmware \"%s_imu.bin\"\n", + ucode_prefix); amdgpu_ucode_release(&adev->gfx.imu_fw); } diff --git a/drivers/gpu/drm/amd/amdgpu/imu_v12_0.c b/drivers/gpu/drm/amd/amdgpu/imu_v12_0.c index 0c8ef908d112..1341f0292031 100644 --- a/drivers/gpu/drm/amd/amdgpu/imu_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/imu_v12_0.c @@ -39,7 +39,6 @@ MODULE_FIRMWARE("amdgpu/gc_12_0_1_imu.bin"); static int imu_v12_0_init_microcode(struct amdgpu_device *adev) { - char fw_name[40]; char ucode_prefix[15]; int err; const struct imu_firmware_header_v1_0 *imu_hdr; @@ -48,11 +47,10 @@ static int imu_v12_0_init_microcode(struct amdgpu_device *adev) DRM_DEBUG("\n"); amdgpu_ucode_ip_version_decode(adev, GC_HWIP, ucode_prefix, sizeof(ucode_prefix)); - - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_imu.bin", ucode_prefix); - err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, "amdgpu/%s_imu.bin", ucode_prefix); if (err) goto out; + imu_hdr = (const struct imu_firmware_header_v1_0 *)adev->gfx.imu_fw->data; adev->gfx.imu_fw_version = le32_to_cpu(imu_hdr->header.ucode_version); @@ -72,8 +70,8 @@ static int imu_v12_0_init_microcode(struct amdgpu_device *adev) out: if (err) { dev_err(adev->dev, - "gfx12: Failed to load firmware \"%s\"\n", - fw_name); + "gfx12: Failed to load firmware \"%s_imu.bin\"\n", + ucode_prefix); amdgpu_ucode_release(&adev->gfx.imu_fw); } @@ -119,7 +117,8 @@ static int imu_v12_0_load_microcode(struct amdgpu_device *adev) static int imu_v12_0_wait_for_reset_status(struct amdgpu_device *adev) { - int i, imu_reg_val = 0; + u32 imu_reg_val = 0; + int i; for (i = 0; i < adev->usec_timeout; i++) { imu_reg_val = RREG32_SOC15(GC, 0, regGFX_IMU_GFX_RESET_CTRL); @@ -138,7 +137,7 @@ static int imu_v12_0_wait_for_reset_status(struct amdgpu_device *adev) static void imu_v12_0_setup(struct amdgpu_device *adev) { - int imu_reg_val; + u32 imu_reg_val; WREG32_SOC15(GC, 0, regGFX_IMU_C2PMSG_ACCESS_CTRL0, 0xffffff); WREG32_SOC15(GC, 0, regGFX_IMU_C2PMSG_ACCESS_CTRL1, 0xffff); @@ -157,7 +156,7 @@ static void imu_v12_0_setup(struct amdgpu_device *adev) static int imu_v12_0_start(struct amdgpu_device *adev) { - int imu_reg_val; + u32 imu_reg_val; imu_reg_val = RREG32_SOC15(GC, 0, regGFX_IMU_CORE_CTRL); imu_reg_val &= 0xfffffffe; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c index 77595e9622da..71f43a5c7f72 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c @@ -581,7 +581,6 @@ static const struct amdgpu_ring_funcs jpeg_v1_0_decode_ring_vm_funcs = { static void jpeg_v1_0_set_dec_ring_funcs(struct amdgpu_device *adev) { adev->jpeg.inst->ring_dec->funcs = &jpeg_v1_0_decode_ring_vm_funcs; - DRM_INFO("JPEG decode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs jpeg_v1_0_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index ef3e42f6b841..99adf3625657 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -131,16 +131,11 @@ static int jpeg_v2_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; - int r; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, (adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0); - r = amdgpu_ring_test_helper(ring); - if (!r) - DRM_INFO("JPEG decode initialized successfully.\n"); - - return r; + return amdgpu_ring_test_helper(ring); } /** @@ -795,7 +790,6 @@ static const struct amdgpu_ring_funcs jpeg_v2_0_dec_ring_vm_funcs = { static void jpeg_v2_0_set_dec_ring_funcs(struct amdgpu_device *adev) { adev->jpeg.inst->ring_dec->funcs = &jpeg_v2_0_dec_ring_vm_funcs; - DRM_INFO("JPEG decode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs jpeg_v2_0_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index afeaf3c64e27..d8ef95c847c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -196,8 +196,6 @@ static int jpeg_v2_5_hw_init(void *handle) return r; } - DRM_INFO("JPEG decode initialized successfully.\n"); - return 0; } @@ -728,7 +726,6 @@ static void jpeg_v2_5_set_dec_ring_funcs(struct amdgpu_device *adev) else /* CHIP_ALDEBARAN */ adev->jpeg.inst[i].ring_dec->funcs = &jpeg_v2_6_dec_ring_vm_funcs; adev->jpeg.inst[i].ring_dec->me = i; - DRM_INFO("JPEG(%d) JPEG decode is enabled in VM mode\n", i); } } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index 1c7cf4800bf7..31cfa3ce6528 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -146,18 +146,11 @@ static int jpeg_v3_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; - int r; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, (adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0); - r = amdgpu_ring_test_helper(ring); - if (r) - return r; - - DRM_INFO("JPEG decode initialized successfully.\n"); - - return 0; + return amdgpu_ring_test_helper(ring); } /** @@ -593,7 +586,6 @@ static const struct amdgpu_ring_funcs jpeg_v3_0_dec_ring_vm_funcs = { static void jpeg_v3_0_set_dec_ring_funcs(struct amdgpu_device *adev) { adev->jpeg.inst->ring_dec->funcs = &jpeg_v3_0_dec_ring_vm_funcs; - DRM_INFO("JPEG decode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs jpeg_v3_0_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index 237fe5df5a8f..3dac8f259d7f 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -181,8 +181,6 @@ static int jpeg_v4_0_hw_init(void *handle) return r; } - DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n"); - return 0; } @@ -755,7 +753,6 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_dec_ring_vm_funcs = { static void jpeg_v4_0_set_dec_ring_funcs(struct amdgpu_device *adev) { adev->jpeg.inst->ring_dec->funcs = &jpeg_v4_0_dec_ring_vm_funcs; - DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs jpeg_v4_0_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c index d66af11aa66c..04d8966423de 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -341,7 +341,6 @@ static int jpeg_v4_0_3_hw_init(void *handle) } } } - DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n"); return 0; } @@ -1100,7 +1099,6 @@ static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev) adev->jpeg.inst[i].aid_id = jpeg_inst / adev->jpeg.num_inst_per_aid; } - DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs jpeg_v4_0_3_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c index 4c8f9772437b..f96ac6bce526 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c @@ -191,7 +191,6 @@ static int jpeg_v4_0_5_hw_init(void *handle) // TODO: Enable ring test with DPG support if (adev->pg_flags & AMD_PG_SUPPORT_JPEG_DPG) { - DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully under DPG Mode"); return 0; } @@ -205,9 +204,6 @@ static int jpeg_v4_0_5_hw_init(void *handle) return r; } - if (!r) - DRM_INFO("JPEG decode initialized successfully under SPG Mode\n"); - return 0; } @@ -805,7 +801,6 @@ static void jpeg_v4_0_5_set_dec_ring_funcs(struct amdgpu_device *adev) adev->jpeg.inst[i].ring_dec->funcs = &jpeg_v4_0_5_dec_ring_vm_funcs; adev->jpeg.inst[i].ring_dec->me = i; - DRM_DEV_INFO(adev->dev, "JPEG%d decode is enabled in VM mode\n", i); } } diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 8263b97c4466..3b1f6ad99100 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -154,18 +154,18 @@ static int mes_v11_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, void *pkt, int size, int api_status_off) { - int ndw = size / 4; - signed long r; - union MESAPI__MISC *x_pkt = pkt; - struct MES_API_STATUS *api_status; + union MESAPI__QUERY_MES_STATUS mes_status_pkt; + signed long timeout = 3000000; /* 3000 ms */ struct amdgpu_device *adev = mes->adev; struct amdgpu_ring *ring = &mes->ring; - unsigned long flags; - signed long timeout = 3000000; /* 3000 ms */ + struct MES_API_STATUS *api_status; + union MESAPI__MISC *x_pkt = pkt; const char *op_str, *misc_op_str; - u32 fence_offset; - u64 fence_gpu_addr; - u64 *fence_ptr; + unsigned long flags; + u64 status_gpu_addr; + u32 status_offset; + u64 *status_ptr; + signed long r; int ret; if (x_pkt->header.opcode >= MES_SCH_API_MAX) @@ -177,28 +177,38 @@ static int mes_v11_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, /* Worst case in sriov where all other 15 VF timeout, each VF needs about 600ms */ timeout = 15 * 600 * 1000; } - BUG_ON(size % 4 != 0); - ret = amdgpu_device_wb_get(adev, &fence_offset); + ret = amdgpu_device_wb_get(adev, &status_offset); if (ret) return ret; - fence_gpu_addr = - adev->wb.gpu_addr + (fence_offset * 4); - fence_ptr = (u64 *)&adev->wb.wb[fence_offset]; - *fence_ptr = 0; + + status_gpu_addr = adev->wb.gpu_addr + (status_offset * 4); + status_ptr = (u64 *)&adev->wb.wb[status_offset]; + *status_ptr = 0; spin_lock_irqsave(&mes->ring_lock, flags); - if (amdgpu_ring_alloc(ring, ndw)) { - spin_unlock_irqrestore(&mes->ring_lock, flags); - amdgpu_device_wb_free(adev, fence_offset); - return -ENOMEM; - } + r = amdgpu_ring_alloc(ring, (size + sizeof(mes_status_pkt)) / 4); + if (r) + goto error_unlock_free; api_status = (struct MES_API_STATUS *)((char *)pkt + api_status_off); - api_status->api_completion_fence_addr = fence_gpu_addr; + api_status->api_completion_fence_addr = status_gpu_addr; api_status->api_completion_fence_value = 1; - amdgpu_ring_write_multiple(ring, pkt, ndw); + amdgpu_ring_write_multiple(ring, pkt, size / 4); + + memset(&mes_status_pkt, 0, sizeof(mes_status_pkt)); + mes_status_pkt.header.type = MES_API_TYPE_SCHEDULER; + mes_status_pkt.header.opcode = MES_SCH_API_QUERY_SCHEDULER_STATUS; + mes_status_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS; + mes_status_pkt.api_status.api_completion_fence_addr = + ring->fence_drv.gpu_addr; + mes_status_pkt.api_status.api_completion_fence_value = + ++ring->fence_drv.sync_seq; + + amdgpu_ring_write_multiple(ring, &mes_status_pkt, + sizeof(mes_status_pkt) / 4); + amdgpu_ring_commit(ring); spin_unlock_irqrestore(&mes->ring_lock, flags); @@ -206,15 +216,16 @@ static int mes_v11_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, misc_op_str = mes_v11_0_get_misc_op_string(x_pkt); if (misc_op_str) - dev_dbg(adev->dev, "MES msg=%s (%s) was emitted\n", op_str, misc_op_str); + dev_dbg(adev->dev, "MES msg=%s (%s) was emitted\n", op_str, + misc_op_str); else if (op_str) dev_dbg(adev->dev, "MES msg=%s was emitted\n", op_str); else - dev_dbg(adev->dev, "MES msg=%d was emitted\n", x_pkt->header.opcode); + dev_dbg(adev->dev, "MES msg=%d was emitted\n", + x_pkt->header.opcode); - r = amdgpu_mes_fence_wait_polling(fence_ptr, (u64)1, timeout); - amdgpu_device_wb_free(adev, fence_offset); - if (r < 1) { + r = amdgpu_fence_wait_polling(ring, ring->fence_drv.sync_seq, timeout); + if (r < 1 || !*status_ptr) { if (misc_op_str) dev_err(adev->dev, "MES failed to respond to msg=%s (%s)\n", @@ -229,10 +240,19 @@ static int mes_v11_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, while (halt_if_hws_hang) schedule(); - return -ETIMEDOUT; + r = -ETIMEDOUT; + goto error_wb_free; } + amdgpu_device_wb_free(adev, status_offset); return 0; + +error_unlock_free: + spin_unlock_irqrestore(&mes->ring_lock, flags); + +error_wb_free: + amdgpu_device_wb_free(adev, status_offset); + return r; } static int convert_to_mes_queue_type(int queue_type) diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index f18fdda023c9..106eef1ff5cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -144,18 +144,18 @@ static int mes_v12_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, void *pkt, int size, int api_status_off) { - int ndw = size / 4; - signed long r; - union MESAPI__MISC *x_pkt = pkt; - struct MES_API_STATUS *api_status; + union MESAPI__QUERY_MES_STATUS mes_status_pkt; + signed long timeout = 3000000; /* 3000 ms */ struct amdgpu_device *adev = mes->adev; struct amdgpu_ring *ring = &mes->ring; - unsigned long flags; + struct MES_API_STATUS *api_status; + union MESAPI__MISC *x_pkt = pkt; const char *op_str, *misc_op_str; - signed long timeout = 3000000; /* 3000 ms */ - u32 fence_offset; - u64 fence_gpu_addr; - u64 *fence_ptr; + unsigned long flags; + u64 status_gpu_addr; + u32 status_offset; + u64 *status_ptr; + signed long r; int ret; if (x_pkt->header.opcode >= MES_SCH_API_MAX) @@ -167,28 +167,38 @@ static int mes_v12_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, /* Worst case in sriov where all other 15 VF timeout, each VF needs about 600ms */ timeout = 15 * 600 * 1000; } - BUG_ON(size % 4 != 0); - ret = amdgpu_device_wb_get(adev, &fence_offset); + ret = amdgpu_device_wb_get(adev, &status_offset); if (ret) return ret; - fence_gpu_addr = - adev->wb.gpu_addr + (fence_offset * 4); - fence_ptr = (u64 *)&adev->wb.wb[fence_offset]; - *fence_ptr = 0; + + status_gpu_addr = adev->wb.gpu_addr + (status_offset * 4); + status_ptr = (u64 *)&adev->wb.wb[status_offset]; + *status_ptr = 0; spin_lock_irqsave(&mes->ring_lock, flags); - if (amdgpu_ring_alloc(ring, ndw)) { - spin_unlock_irqrestore(&mes->ring_lock, flags); - amdgpu_device_wb_free(adev, fence_offset); - return -ENOMEM; - } + r = amdgpu_ring_alloc(ring, (size + sizeof(mes_status_pkt)) / 4); + if (r) + goto error_unlock_free; api_status = (struct MES_API_STATUS *)((char *)pkt + api_status_off); - api_status->api_completion_fence_addr = fence_gpu_addr; + api_status->api_completion_fence_addr = status_gpu_addr; api_status->api_completion_fence_value = 1; - amdgpu_ring_write_multiple(ring, pkt, ndw); + amdgpu_ring_write_multiple(ring, pkt, size / 4); + + memset(&mes_status_pkt, 0, sizeof(mes_status_pkt)); + mes_status_pkt.header.type = MES_API_TYPE_SCHEDULER; + mes_status_pkt.header.opcode = MES_SCH_API_QUERY_SCHEDULER_STATUS; + mes_status_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS; + mes_status_pkt.api_status.api_completion_fence_addr = + ring->fence_drv.gpu_addr; + mes_status_pkt.api_status.api_completion_fence_value = + ++ring->fence_drv.sync_seq; + + amdgpu_ring_write_multiple(ring, &mes_status_pkt, + sizeof(mes_status_pkt) / 4); + amdgpu_ring_commit(ring); spin_unlock_irqrestore(&mes->ring_lock, flags); @@ -196,16 +206,17 @@ static int mes_v12_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, misc_op_str = mes_v12_0_get_misc_op_string(x_pkt); if (misc_op_str) - dev_dbg(adev->dev, "MES msg=%s (%s) was emitted\n", op_str, misc_op_str); + dev_dbg(adev->dev, "MES msg=%s (%s) was emitted\n", op_str, + misc_op_str); else if (op_str) dev_dbg(adev->dev, "MES msg=%s was emitted\n", op_str); else - dev_dbg(adev->dev, "MES msg=%d was emitted\n", x_pkt->header.opcode); + dev_dbg(adev->dev, "MES msg=%d was emitted\n", + x_pkt->header.opcode); - r = amdgpu_mes_fence_wait_polling(fence_ptr, (u64)1, timeout); - amdgpu_device_wb_free(adev, fence_offset); + r = amdgpu_fence_wait_polling(ring, ring->fence_drv.sync_seq, timeout); + if (r < 1 || !*status_ptr) { - if (r < 1) { if (misc_op_str) dev_err(adev->dev, "MES failed to respond to msg=%s (%s)\n", op_str, misc_op_str); @@ -219,10 +230,19 @@ static int mes_v12_0_submit_pkt_and_poll_completion(struct amdgpu_mes *mes, while (halt_if_hws_hang) schedule(); - return -ETIMEDOUT; + r = -ETIMEDOUT; + goto error_wb_free; } + amdgpu_device_wb_free(adev, status_offset); return 0; + +error_unlock_free: + spin_unlock_irqrestore(&mes->ring_lock, flags); + +error_wb_free: + amdgpu_device_wb_free(adev, status_offset); + return r; } static int convert_to_mes_queue_type(int queue_type) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c index 92432cd2c0c7..9689e2b5d4e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c @@ -544,7 +544,7 @@ static int mmhub_v1_7_set_clockgating(struct amdgpu_device *adev, static void mmhub_v1_7_get_clockgating(struct amdgpu_device *adev, u64 *flags) { - int data, data1; + u32 data, data1; if (amdgpu_sriov_vf(adev)) *flags = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index 02fd45261399..a0cc8e218ca1 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -671,7 +671,7 @@ static int mmhub_v2_0_set_clockgating(struct amdgpu_device *adev, static void mmhub_v2_0_get_clockgating(struct amdgpu_device *adev, u64 *flags) { - int data, data1; + u32 data, data1; if (amdgpu_sriov_vf(adev)) *flags = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c index 8928f9160822..b4ce3375d3fd 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c @@ -613,7 +613,7 @@ static int mmhub_v3_3_set_clockgating(struct amdgpu_device *adev, static void mmhub_v3_3_get_clockgating(struct amdgpu_device *adev, u64 *flags) { - int data; + u32 data; if (amdgpu_sriov_vf(adev)) *flags = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index 1b7da4aff2b8..ff1b58e44689 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -657,7 +657,7 @@ static int mmhub_v9_4_set_clockgating(struct amdgpu_device *adev, static void mmhub_v9_4_get_clockgating(struct amdgpu_device *adev, u64 *flags) { - int data, data1; + u32 data, data1; if (amdgpu_sriov_vf(adev)) *flags = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index f4c47492e0cd..6b71ee85ee65 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -249,38 +249,30 @@ static int xgpu_ai_set_mailbox_ack_irq(struct amdgpu_device *adev, return 0; } -static void xgpu_ai_mailbox_flr_work(struct work_struct *work) +static void xgpu_ai_ready_to_reset(struct amdgpu_device *adev) { - struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); - struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); - int timeout = AI_MAILBOX_POLL_FLR_TIMEDOUT; - - /* block amdgpu_gpu_recover till msg FLR COMPLETE received, - * otherwise the mailbox msg will be ruined/reseted by - * the VF FLR. - */ - if (atomic_cmpxchg(&adev->reset_domain->in_gpu_reset, 0, 1) != 0) - return; - - down_write(&adev->reset_domain->sem); - - amdgpu_virt_fini_data_exchange(adev); - xgpu_ai_mailbox_trans_msg(adev, IDH_READY_TO_RESET, 0, 0, 0); +} +static int xgpu_ai_wait_reset(struct amdgpu_device *adev) +{ + int timeout = AI_MAILBOX_POLL_FLR_TIMEDOUT; do { if (xgpu_ai_mailbox_peek_msg(adev) == IDH_FLR_NOTIFICATION_CMPL) - goto flr_done; - + return 0; msleep(10); timeout -= 10; } while (timeout > 1); - dev_warn(adev->dev, "waiting IDH_FLR_NOTIFICATION_CMPL timeout\n"); + return -ETIME; +} -flr_done: - atomic_set(&adev->reset_domain->in_gpu_reset, 0); - up_write(&adev->reset_domain->sem); +static void xgpu_ai_mailbox_flr_work(struct work_struct *work) +{ + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); + struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); + + amdgpu_virt_fini_data_exchange(adev); /* Trigger recovery for world switch failure if no TDR */ if (amdgpu_device_should_recover_gpu(adev) @@ -417,7 +409,8 @@ const struct amdgpu_virt_ops xgpu_ai_virt_ops = { .req_full_gpu = xgpu_ai_request_full_gpu_access, .rel_full_gpu = xgpu_ai_release_full_gpu_access, .reset_gpu = xgpu_ai_request_reset, - .wait_reset = NULL, + .ready_to_reset = xgpu_ai_ready_to_reset, + .wait_reset = xgpu_ai_wait_reset, .trans_msg = xgpu_ai_mailbox_trans_msg, .req_init_data = xgpu_ai_request_init_data, .ras_poison_handler = xgpu_ai_ras_poison_handler, diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index 37b49a5ed2a1..22af30a15a5f 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -282,38 +282,30 @@ static int xgpu_nv_set_mailbox_ack_irq(struct amdgpu_device *adev, return 0; } -static void xgpu_nv_mailbox_flr_work(struct work_struct *work) +static void xgpu_nv_ready_to_reset(struct amdgpu_device *adev) { - struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); - struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); - int timeout = NV_MAILBOX_POLL_FLR_TIMEDOUT; - - /* block amdgpu_gpu_recover till msg FLR COMPLETE received, - * otherwise the mailbox msg will be ruined/reseted by - * the VF FLR. - */ - if (atomic_cmpxchg(&adev->reset_domain->in_gpu_reset, 0, 1) != 0) - return; - - down_write(&adev->reset_domain->sem); - - amdgpu_virt_fini_data_exchange(adev); - xgpu_nv_mailbox_trans_msg(adev, IDH_READY_TO_RESET, 0, 0, 0); +} +static int xgpu_nv_wait_reset(struct amdgpu_device *adev) +{ + int timeout = NV_MAILBOX_POLL_FLR_TIMEDOUT; do { if (xgpu_nv_mailbox_peek_msg(adev) == IDH_FLR_NOTIFICATION_CMPL) - goto flr_done; - + return 0; msleep(10); timeout -= 10; } while (timeout > 1); - dev_warn(adev->dev, "waiting IDH_FLR_NOTIFICATION_CMPL timeout\n"); + return -ETIME; +} -flr_done: - atomic_set(&adev->reset_domain->in_gpu_reset, 0); - up_write(&adev->reset_domain->sem); +static void xgpu_nv_mailbox_flr_work(struct work_struct *work) +{ + struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); + struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); + + amdgpu_virt_fini_data_exchange(adev); /* Trigger recovery for world switch failure if no TDR */ if (amdgpu_device_should_recover_gpu(adev) @@ -455,7 +447,8 @@ const struct amdgpu_virt_ops xgpu_nv_virt_ops = { .rel_full_gpu = xgpu_nv_release_full_gpu_access, .req_init_data = xgpu_nv_request_init_data, .reset_gpu = xgpu_nv_request_reset, - .wait_reset = NULL, + .ready_to_reset = xgpu_nv_ready_to_reset, + .wait_reset = xgpu_nv_wait_reset, .trans_msg = xgpu_nv_mailbox_trans_msg, .ras_poison_handler = xgpu_nv_ras_poison_handler, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c index 78cd07744ebe..e1d63bed84bf 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c @@ -515,12 +515,6 @@ static void xgpu_vi_mailbox_flr_work(struct work_struct *work) struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work); struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt); - /* wait until RCV_MSG become 3 */ - if (xgpu_vi_poll_msg(adev, IDH_FLR_NOTIFICATION_CMPL)) { - pr_err("failed to receive FLR_CMPL\n"); - return; - } - /* Trigger recovery due to world switch failure */ if (amdgpu_device_should_recover_gpu(adev)) { struct amdgpu_reset_context reset_context; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index 32cc60ce5521..8d80df94bd8b 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -414,6 +414,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device /* ras_controller_int is dedicated for nbif ras error, * not the global interrupt for sync flood */ + amdgpu_ras_set_fed(adev, true); amdgpu_ras_reset_gpu(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c index cc0248efa6b6..4d33c95a5116 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c @@ -32,7 +32,9 @@ #include "mp/mp_14_0_2_sh_mask.h" MODULE_FIRMWARE("amdgpu/psp_14_0_2_sos.bin"); +MODULE_FIRMWARE("amdgpu/psp_14_0_2_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_14_0_3_sos.bin"); +MODULE_FIRMWARE("amdgpu/psp_14_0_3_ta.bin"); /* For large FW files the time to complete can be very long */ #define USBC_PD_POLLING_LIMIT_S 240 @@ -66,6 +68,9 @@ static int psp_v14_0_init_microcode(struct psp_context *psp) err = psp_init_sos_microcode(psp, ucode_prefix); if (err) return err; + err = psp_init_ta_microcode(psp, ucode_prefix); + if (err) + return err; break; default: BUG(); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index ac8a9b9b3e52..725392522267 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -127,7 +127,6 @@ static void sdma_v2_4_free_microcode(struct amdgpu_device *adev) static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err = 0, i; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; @@ -145,10 +144,11 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) for (i = 0; i < adev->sdma.num_instances; i++) { if (i == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + "amdgpu/%s_sdma.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + "amdgpu/%s_sdma1.bin", chip_name); if (err) goto out; hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data; @@ -169,7 +169,8 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) out: if (err) { - pr_err("sdma_v2_4: Failed to load firmware \"%s\"\n", fw_name); + pr_err("sdma_v2_4: Failed to load firmware \"%s_sdma%s.bin\"\n", + chip_name, i == 0 ? "" : "1"); for (i = 0; i < adev->sdma.num_instances; i++) amdgpu_ucode_release(&adev->sdma.instance[i].fw); } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index b8ebdc4ae6f6..aa637541da58 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -267,7 +267,6 @@ static void sdma_v3_0_free_microcode(struct amdgpu_device *adev) static int sdma_v3_0_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err = 0, i; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; @@ -305,10 +304,11 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev) for (i = 0; i < adev->sdma.num_instances; i++) { if (i == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + "amdgpu/%s_sdma.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name); - err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, fw_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + "amdgpu/%s_sdma1.bin", chip_name); if (err) goto out; hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data; @@ -327,7 +327,8 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev) } out: if (err) { - pr_err("sdma_v3_0: Failed to load firmware \"%s\"\n", fw_name); + pr_err("sdma_v3_0: Failed to load firmware \"%s_sdma%s.bin\"\n", + chip_name, i == 0 ? "" : "1"); for (i = 0; i < adev->sdma.num_instances; i++) amdgpu_ucode_release(&adev->sdma.instance[i].fw); } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0_0_pkt_open.h b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0_0_pkt_open.h index 6af23e7888ca..d8cf830916b9 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0_0_pkt_open.h +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0_0_pkt_open.h @@ -91,6 +91,14 @@ #define SDMA_GCR_GLM_WB (1 << 4) #define SDMA_GCR_GL1_RANGE(x) (((x) & 0x3) << 2) #define SDMA_GCR_GLI_INV(x) (((x) & 0x3) << 0) + +#define SDMA_DCC_DATA_FORMAT(x) ((x) & 0x3f) +#define SDMA_DCC_NUM_TYPE(x) (((x) & 0x7) << 9) +#define SDMA_DCC_READ_CM(x) (((x) & 0x3) << 16) +#define SDMA_DCC_WRITE_CM(x) (((x) & 0x3) << 18) +#define SDMA_DCC_MAX_COM(x) (((x) & 0x3) << 24) +#define SDMA_DCC_MAX_UCOM(x) (((x) & 0x1) << 26) + /* ** Definitions for SDMA_PKT_COPY_LINEAR packet */ diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c index ab1dea77be6e..96514fd77e35 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c @@ -1568,13 +1568,22 @@ static void sdma_v7_0_emit_copy_buffer(struct amdgpu_ib *ib, { ib->ptr[ib->length_dw++] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_COPY) | SDMA_PKT_COPY_LINEAR_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR) | - SDMA_PKT_COPY_LINEAR_HEADER_TMZ((copy_flags & AMDGPU_COPY_FLAGS_TMZ) ? 1 : 0); + SDMA_PKT_COPY_LINEAR_HEADER_TMZ((copy_flags & AMDGPU_COPY_FLAGS_TMZ) ? 1 : 0) | + SDMA_PKT_COPY_LINEAR_HEADER_CPV((copy_flags & + (AMDGPU_COPY_FLAGS_READ_DECOMPRESSED | AMDGPU_COPY_FLAGS_WRITE_COMPRESSED)) ? 1 : 0); + ib->ptr[ib->length_dw++] = byte_count - 1; ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */ ib->ptr[ib->length_dw++] = lower_32_bits(src_offset); ib->ptr[ib->length_dw++] = upper_32_bits(src_offset); ib->ptr[ib->length_dw++] = lower_32_bits(dst_offset); ib->ptr[ib->length_dw++] = upper_32_bits(dst_offset); + + if ((copy_flags & (AMDGPU_COPY_FLAGS_READ_DECOMPRESSED | AMDGPU_COPY_FLAGS_WRITE_COMPRESSED))) + ib->ptr[ib->length_dw++] = SDMA_DCC_DATA_FORMAT(4) | SDMA_DCC_NUM_TYPE(4) | + ((copy_flags & AMDGPU_COPY_FLAGS_READ_DECOMPRESSED) ? SDMA_DCC_READ_CM(2) : 0) | + ((copy_flags & AMDGPU_COPY_FLAGS_WRITE_COMPRESSED) ? SDMA_DCC_WRITE_CM(1) : 0) | + SDMA_DCC_MAX_COM(1) | SDMA_DCC_MAX_UCOM(1); } /** @@ -1603,7 +1612,6 @@ static const struct amdgpu_buffer_funcs sdma_v7_0_buffer_funcs = { .copy_max_bytes = 0x400000, .copy_num_dw = 7, .emit_copy_buffer = sdma_v7_0_emit_copy_buffer, - .fill_max_bytes = 0x400000, .fill_num_dw = 5, .emit_fill_buffer = sdma_v7_0_emit_fill_buffer, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index cb253bd3a2a2..a280b9fecb77 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -202,24 +202,17 @@ static int vcn_v1_0_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; for (i = 0; i < adev->vcn.num_enc_rings; ++i) { ring = &adev->vcn.inst->ring_enc[i]; r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; } ring = adev->jpeg.inst->ring_dec; r = amdgpu_ring_test_helper(ring); - if (r) - goto done; - -done: - if (!r) - DRM_INFO("VCN decode and encode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); return r; } @@ -2043,7 +2036,6 @@ static const struct amdgpu_ring_funcs vcn_v1_0_enc_ring_vm_funcs = { static void vcn_v1_0_set_dec_ring_funcs(struct amdgpu_device *adev) { adev->vcn.inst->ring_dec.funcs = &vcn_v1_0_dec_ring_vm_funcs; - DRM_INFO("VCN decode is enabled in VM mode\n"); } static void vcn_v1_0_set_enc_ring_funcs(struct amdgpu_device *adev) @@ -2052,8 +2044,6 @@ static void vcn_v1_0_set_enc_ring_funcs(struct amdgpu_device *adev) for (i = 0; i < adev->vcn.num_enc_rings; ++i) adev->vcn.inst->ring_enc[i].funcs = &vcn_v1_0_enc_ring_vm_funcs; - - DRM_INFO("VCN encode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs vcn_v1_0_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index f18fd61c435e..d3d096909a7f 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -237,7 +237,7 @@ static int vcn_v2_0_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; //Disable vcn decode for sriov if (amdgpu_sriov_vf(adev)) @@ -247,15 +247,10 @@ static int vcn_v2_0_hw_init(void *handle) ring = &adev->vcn.inst->ring_enc[i]; r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; } -done: - if (!r) - DRM_INFO("VCN decode and encode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); - - return r; + return 0; } /** @@ -2074,7 +2069,6 @@ static const struct amdgpu_ring_funcs vcn_v2_0_enc_ring_vm_funcs = { static void vcn_v2_0_set_dec_ring_funcs(struct amdgpu_device *adev) { adev->vcn.inst->ring_dec.funcs = &vcn_v2_0_dec_ring_vm_funcs; - DRM_INFO("VCN decode is enabled in VM mode\n"); } static void vcn_v2_0_set_enc_ring_funcs(struct amdgpu_device *adev) @@ -2083,8 +2077,6 @@ static void vcn_v2_0_set_enc_ring_funcs(struct amdgpu_device *adev) for (i = 0; i < adev->vcn.num_enc_rings; ++i) adev->vcn.inst->ring_enc[i].funcs = &vcn_v2_0_enc_ring_vm_funcs; - - DRM_INFO("VCN encode is enabled in VM mode\n"); } static const struct amdgpu_irq_src_funcs vcn_v2_0_irq_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index baec14bde2a2..96f60c303161 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -314,22 +314,17 @@ static int vcn_v2_5_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; for (i = 0; i < adev->vcn.num_enc_rings; ++i) { ring = &adev->vcn.inst[j].ring_enc[i]; r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; } } } -done: - if (!r) - DRM_INFO("VCN decode and encode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); - return r; } @@ -1710,7 +1705,6 @@ static void vcn_v2_5_set_dec_ring_funcs(struct amdgpu_device *adev) continue; adev->vcn.inst[i].ring_dec.funcs = &vcn_v2_5_dec_ring_vm_funcs; adev->vcn.inst[i].ring_dec.me = i; - DRM_INFO("VCN(%d) decode is enabled in VM mode\n", i); } } @@ -1725,7 +1719,6 @@ static void vcn_v2_5_set_enc_ring_funcs(struct amdgpu_device *adev) adev->vcn.inst[j].ring_enc[i].funcs = &vcn_v2_5_enc_ring_vm_funcs; adev->vcn.inst[j].ring_enc[i].me = j; } - DRM_INFO("VCN(%d) encode is enabled in VM mode\n", j); } } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index 6b31cf4b8aac..24f947751c46 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -303,7 +303,7 @@ static int vcn_v3_0_hw_init(void *handle) if (amdgpu_sriov_vf(adev)) { r = vcn_v3_0_start_sriov(adev); if (r) - goto done; + return r; /* initialize VCN dec and enc ring buffers */ for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -348,24 +348,18 @@ static int vcn_v3_0_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; for (j = 0; j < adev->vcn.num_enc_rings; ++j) { ring = &adev->vcn.inst[i].ring_enc[j]; r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; } } } return 0; -done: - if (!r) - DRM_INFO("VCN decode and encode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); - - return r; } /** @@ -2041,8 +2035,6 @@ static void vcn_v3_0_set_dec_ring_funcs(struct amdgpu_device *adev) else adev->vcn.inst[i].ring_dec.funcs = &vcn_v3_0_dec_sw_ring_vm_funcs; adev->vcn.inst[i].ring_dec.me = i; - DRM_INFO("VCN(%d) decode%s is enabled in VM mode\n", i, - DEC_SW_RING_ENABLED?"(Software Ring)":""); } } @@ -2058,8 +2050,6 @@ static void vcn_v3_0_set_enc_ring_funcs(struct amdgpu_device *adev) adev->vcn.inst[i].ring_enc[j].funcs = &vcn_v3_0_enc_ring_vm_funcs; adev->vcn.inst[i].ring_enc[j].me = i; } - if (adev->vcn.num_enc_rings > 0) - DRM_INFO("VCN(%d) encode is enabled in VM mode\n", i); } } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index ac1b8ead03b3..f6d96a44d75f 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -258,7 +258,7 @@ static int vcn_v4_0_hw_init(void *handle) if (amdgpu_sriov_vf(adev)) { r = vcn_v4_0_start_sriov(adev); if (r) - goto done; + return r; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { if (adev->vcn.harvest_config & (1 << i)) @@ -269,7 +269,6 @@ static int vcn_v4_0_hw_init(void *handle) ring->wptr_old = 0; vcn_v4_0_unified_ring_set_wptr(ring); ring->sched.ready = true; - } } else { for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -283,18 +282,11 @@ static int vcn_v4_0_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; - + return r; } } return 0; -done: - if (!r) - DRM_INFO("VCN decode and encode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); - - return r; } /** @@ -1900,8 +1892,6 @@ static void vcn_v4_0_set_unified_ring_funcs(struct amdgpu_device *adev) adev->vcn.inst[i].ring_enc[0].funcs = (const struct amdgpu_ring_funcs *)&vcn_v4_0_unified_ring_vm_funcs; adev->vcn.inst[i].ring_enc[0].me = i; - - DRM_INFO("VCN(%d) encode/decode are enabled in VM mode\n", i); } } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 2279d8fce03d..f53054e39ebb 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -210,7 +210,7 @@ static int vcn_v4_0_3_hw_init(void *handle) if (amdgpu_sriov_vf(adev)) { r = vcn_v4_0_3_start_sriov(adev); if (r) - goto done; + return r; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { ring = &adev->vcn.inst[i].ring_enc[0]; @@ -246,15 +246,10 @@ static int vcn_v4_0_3_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; } } -done: - if (!r) - DRM_DEV_INFO(adev->dev, "VCN decode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); - return r; } @@ -1450,7 +1445,6 @@ static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev) adev->vcn.inst[i].aid_id = vcn_inst / adev->vcn.num_inst_per_aid; } - DRM_DEV_INFO(adev->dev, "VCN decode is enabled in VM mode\n"); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index 81fb99729f37..f45495de6875 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -234,16 +234,10 @@ static int vcn_v4_0_5_hw_init(void *handle) r = amdgpu_ring_test_helper(ring); if (r) - goto done; + return r; } return 0; -done: - if (!r) - DRM_INFO("VCN decode and encode initialized successfully(under %s).\n", - (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); - - return r; } /** @@ -1558,8 +1552,6 @@ static void vcn_v4_0_5_set_unified_ring_funcs(struct amdgpu_device *adev) adev->vcn.inst[i].ring_enc[0].funcs = &vcn_v4_0_5_unified_ring_vm_funcs; adev->vcn.inst[i].ring_enc[0].me = i; - - DRM_INFO("VCN(%d) encode/decode are enabled in VM mode\n", i); } } |