diff options
75 files changed, 743 insertions, 576 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9c11535c44c6..447e734c362b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1239,6 +1239,14 @@ struct amdgpu_device { struct amdgpu_kfd_dev kfd; }; +/* + * MES FW uses address(mqd_addr + sizeof(struct mqd) + 3*sizeof(uint32_t)) + * as fence address and writes a 32 bit fence value to this address. + * Driver needs to allocate at least 4 DWs extra memory in addition to + * sizeof(struct mqd). Add 8 DWs and align to AMDGPU_GPU_PAGE_SIZE for safety. + */ +#define AMDGPU_MQD_SIZE_ALIGN(mqd_size) AMDGPU_GPU_PAGE_ALIGN(((mqd_size) + 32)) + static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev, uint8_t ip, uint8_t inst) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c index 66fb37b64388..87ec46c56a6e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c @@ -60,11 +60,9 @@ static int amdgpu_bo_list_entry_cmp(const void *_a, const void *_b) { const struct amdgpu_bo_list_entry *a = _a, *b = _b; - if (a->priority > b->priority) - return 1; - if (a->priority < b->priority) - return -1; - return 0; + BUILD_BUG_ON(AMDGPU_BO_LIST_MAX_PRIORITY >= INT_MAX); + + return (int)a->priority - (int)b->priority; } int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 8b5801d7109c..77578ecc6782 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -385,6 +385,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; struct amdgpu_ring *ring = &kiq->ring; u32 domain = AMDGPU_GEM_DOMAIN_GTT; + u32 gfx_mqd_size = max(adev->mqds[AMDGPU_HW_IP_GFX].mqd_size, mqd_size); + u32 compute_mqd_size = max(adev->mqds[AMDGPU_HW_IP_COMPUTE].mqd_size, mqd_size); #if !defined(CONFIG_ARM) && !defined(CONFIG_ARM64) /* Only enable on gfx10 and 11 for now to avoid changing behavior on older chips */ @@ -424,17 +426,17 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, for (i = 0; i < adev->gfx.num_gfx_rings; i++) { ring = &adev->gfx.gfx_ring[i]; if (!ring->mqd_obj) { - r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE, - domain, &ring->mqd_obj, + r = amdgpu_bo_create_kernel(adev, AMDGPU_MQD_SIZE_ALIGN(gfx_mqd_size), + PAGE_SIZE, domain, &ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); if (r) { dev_warn(adev->dev, "failed to create ring mqd bo (%d)", r); return r; } - ring->mqd_size = mqd_size; + ring->mqd_size = gfx_mqd_size; /* prepare MQD backup */ - adev->gfx.me.mqd_backup[i] = kzalloc(mqd_size, GFP_KERNEL); + adev->gfx.me.mqd_backup[i] = kzalloc(gfx_mqd_size, GFP_KERNEL); if (!adev->gfx.me.mqd_backup[i]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); return -ENOMEM; @@ -448,17 +450,17 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, j = i + xcc_id * adev->gfx.num_compute_rings; ring = &adev->gfx.compute_ring[j]; if (!ring->mqd_obj) { - r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE, - domain, &ring->mqd_obj, + r = amdgpu_bo_create_kernel(adev, AMDGPU_MQD_SIZE_ALIGN(compute_mqd_size), + PAGE_SIZE, domain, &ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); if (r) { dev_warn(adev->dev, "failed to create ring mqd bo (%d)", r); return r; } - ring->mqd_size = mqd_size; + ring->mqd_size = compute_mqd_size; /* prepare MQD backup */ - adev->gfx.mec.mqd_backup[j] = kzalloc(mqd_size, GFP_KERNEL); + adev->gfx.mec.mqd_backup[j] = kzalloc(compute_mqd_size, GFP_KERNEL); if (!adev->gfx.mec.mqd_backup[j]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index d9ff68a43178..b793ce17140c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -498,8 +498,13 @@ void amdgpu_gmc_filter_faults_remove(struct amdgpu_device *adev, uint64_t addr, if (adev->irq.retry_cam_enabled) return; + else if (adev->irq.ih1.ring_size) + ih = &adev->irq.ih1; + else if (adev->irq.ih_soft.enabled) + ih = &adev->irq.ih_soft; + else + return; - ih = &adev->irq.ih1; /* Get the WPTR of the last entry in IH ring */ last_wptr = amdgpu_ih_get_wptr(adev, ih); /* Order wptr with ring data. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 72ec455fa932..44f230d67da2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -235,7 +235,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, amdgpu_ring_ib_begin(ring); - if (ring->funcs->emit_gfx_shadow) + if (ring->funcs->emit_gfx_shadow && adev->gfx.cp_gfx_shadow) amdgpu_ring_emit_gfx_shadow(ring, shadow_va, csa_va, gds_va, init_shadow, vmid); @@ -291,7 +291,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, fence_flags | AMDGPU_FENCE_FLAG_64BIT); } - if (ring->funcs->emit_gfx_shadow && ring->funcs->init_cond_exec) { + if (ring->funcs->emit_gfx_shadow && ring->funcs->init_cond_exec && + adev->gfx.cp_gfx_shadow) { amdgpu_ring_emit_gfx_shadow(ring, 0, 0, 0, false, 0); amdgpu_ring_init_cond_exec(ring, ring->cond_exe_gpu_addr); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index 6e19836c5ff6..0d4c77c1b4b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -115,9 +115,6 @@ int amdgpu_mes_init(struct amdgpu_device *adev) adev->mes.adev = adev; - idr_init(&adev->mes.pasid_idr); - idr_init(&adev->mes.gang_id_idr); - idr_init(&adev->mes.queue_id_idr); ida_init(&adev->mes.doorbell_ida); spin_lock_init(&adev->mes.queue_id_lock); mutex_init(&adev->mes.mutex_hidden); @@ -252,9 +249,6 @@ error: &adev->mes.hung_queue_db_array_cpu_addr[i]); } - idr_destroy(&adev->mes.pasid_idr); - idr_destroy(&adev->mes.gang_id_idr); - idr_destroy(&adev->mes.queue_id_idr); ida_destroy(&adev->mes.doorbell_ida); mutex_destroy(&adev->mes.mutex_hidden); return r; @@ -283,9 +277,6 @@ void amdgpu_mes_fini(struct amdgpu_device *adev) amdgpu_mes_doorbell_free(adev); - idr_destroy(&adev->mes.pasid_idr); - idr_destroy(&adev->mes.gang_id_idr); - idr_destroy(&adev->mes.queue_id_idr); ida_destroy(&adev->mes.doorbell_ida); mutex_destroy(&adev->mes.mutex_hidden); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index 88685c58798e..bcf2a067dc41 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -77,9 +77,6 @@ struct amdgpu_mes { struct mutex mutex_hidden; - struct idr pasid_idr; - struct idr gang_id_idr; - struct idr queue_id_idr; struct ida doorbell_ida; spinlock_t queue_id_lock; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 64dd7a81bff5..469d04a39d7d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -1712,6 +1712,10 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control) dev_warn(adev->dev, "RAS records:%u exceeds 90%% of threshold:%d", control->ras_num_bad_pages, ras->bad_page_cnt_threshold); + if (amdgpu_bad_page_threshold != 0 && + control->ras_num_bad_pages >= ras->bad_page_cnt_threshold) + amdgpu_dpm_send_rma_reason(adev); + } else if (hdr->header == RAS_TABLE_HDR_BAD && amdgpu_bad_page_threshold != 0) { if (hdr->version >= RAS_TABLE_VER_V2_1) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index b0ebf19db10b..b700c2b91465 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -932,6 +932,17 @@ static int amdgpu_userq_input_args_validate(struct drm_device *dev, drm_file_err(filp, "invalidate userq queue va or size\n"); return -EINVAL; } + + if (!is_power_of_2(args->in.queue_size)) { + drm_file_err(filp, "Queue size must be a power of 2\n"); + return -EINVAL; + } + + if (args->in.queue_size < AMDGPU_GPU_PAGE_SIZE) { + drm_file_err(filp, "Queue size smaller than AMDGPU_GPU_PAGE_SIZE\n"); + return -EINVAL; + } + if (!args->in.wptr_va || !args->in.rptr_va) { drm_file_err(filp, "invalidate userq queue rptr or wptr\n"); return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 41bbedb8e157..1893ceeeb26c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -6880,7 +6880,7 @@ static int gfx_v10_0_kgq_init_queue(struct amdgpu_ring *ring, bool reset) memcpy_toio(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd)); /* reset the ring */ ring->wptr = 0; - *ring->wptr_cpu_addr = 0; + atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 3a4ca104b161..427975b5a1d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -4214,7 +4214,7 @@ static int gfx_v11_0_kgq_init_queue(struct amdgpu_ring *ring, bool reset) memcpy_toio(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd)); /* reset the ring */ ring->wptr = 0; - *ring->wptr_cpu_addr = 0; + atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); } @@ -6828,11 +6828,12 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; + bool use_mmio = false; int r; amdgpu_ring_reset_helper_begin(ring, timedout_fence); - r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false, 0); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, use_mmio, 0); if (r) { dev_warn(adev->dev, "reset via MES failed and try pipe reset %d\n", r); @@ -6841,16 +6842,18 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, return r; } - r = gfx_v11_0_kgq_init_queue(ring, true); - if (r) { - dev_err(adev->dev, "failed to init kgq\n"); - return r; - } + if (use_mmio) { + r = gfx_v11_0_kgq_init_queue(ring, true); + if (r) { + dev_err(adev->dev, "failed to init kgq\n"); + return r; + } - r = amdgpu_mes_map_legacy_queue(adev, ring, 0); - if (r) { - dev_err(adev->dev, "failed to remap kgq\n"); - return r; + r = amdgpu_mes_map_legacy_queue(adev, ring, 0); + if (r) { + dev_err(adev->dev, "failed to remap kgq\n"); + return r; + } } return amdgpu_ring_reset_helper_end(ring, timedout_fence); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index 40660b05f979..79ea1af363a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -3085,7 +3085,7 @@ static int gfx_v12_0_kgq_init_queue(struct amdgpu_ring *ring, bool reset) memcpy_toio(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd)); /* reset the ring */ ring->wptr = 0; - *ring->wptr_cpu_addr = 0; + atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); } @@ -5292,11 +5292,12 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; + bool use_mmio = false; int r; amdgpu_ring_reset_helper_begin(ring, timedout_fence); - r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false, 0); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, use_mmio, 0); if (r) { dev_warn(adev->dev, "reset via MES failed and try pipe reset %d\n", r); r = gfx_v12_reset_gfx_pipe(ring); @@ -5304,16 +5305,18 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, return r; } - r = gfx_v12_0_kgq_init_queue(ring, true); - if (r) { - dev_err(adev->dev, "failed to init kgq\n"); - return r; - } + if (use_mmio) { + r = gfx_v12_0_kgq_init_queue(ring, true); + if (r) { + dev_err(adev->dev, "failed to init kgq\n"); + return r; + } - r = amdgpu_mes_map_legacy_queue(adev, ring, 0); - if (r) { - dev_err(adev->dev, "failed to remap kgq\n"); - return r; + r = amdgpu_mes_map_legacy_queue(adev, ring, 0); + if (r) { + dev_err(adev->dev, "failed to remap kgq\n"); + return r; + } } return amdgpu_ring_reset_helper_end(ring, timedout_fence); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_1.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_1.c index 86cc90a66296..eb9725ae1607 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_1.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_1.c @@ -30,6 +30,7 @@ #include "amdgpu_psp.h" #include "amdgpu_smu.h" #include "amdgpu_atomfirmware.h" +#include "amdgpu_userq_fence.h" #include "imu_v12_1.h" #include "soc_v1_0.h" #include "gfx_v12_1_pkt.h" @@ -2235,7 +2236,7 @@ static int gfx_v12_1_compute_mqd_init(struct amdgpu_device *adev, void *m, mqd->cp_hqd_queue_priority = prop->hqd_queue_priority; mqd->cp_mqd_stride_size = prop->mqd_stride_size ? prop->mqd_stride_size : - sizeof(struct v12_1_compute_mqd); + AMDGPU_MQD_SIZE_ALIGN(adev->mqds[AMDGPU_HW_IP_COMPUTE].mqd_size); mqd->cp_hqd_active = prop->hqd_active; @@ -3602,25 +3603,23 @@ static int gfx_v12_1_eop_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - int i, xcc_id; + u32 doorbell_offset = entry->src_data[0]; u8 me_id, pipe_id, queue_id; struct amdgpu_ring *ring; - uint32_t mes_queue_id = entry->src_data[0]; + int i, xcc_id; DRM_DEBUG("IH: CP EOP\n"); - if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) { - struct amdgpu_mes_queue *queue; + if (adev->enable_mes && doorbell_offset) { + struct amdgpu_userq_fence_driver *fence_drv = NULL; + struct xarray *xa = &adev->userq_xa; + unsigned long flags; - mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK; - - spin_lock(&adev->mes.queue_id_lock); - queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id); - if (queue) { - DRM_DEBUG("process mes queue id = %d\n", mes_queue_id); - amdgpu_fence_process(queue->ring); - } - spin_unlock(&adev->mes.queue_id_lock); + xa_lock_irqsave(xa, flags); + fence_drv = xa_load(xa, doorbell_offset); + if (fence_drv) + amdgpu_userq_fence_driver_process(fence_drv); + xa_unlock_irqrestore(xa, flags); } else { me_id = (entry->ring_id & 0x0c) >> 2; pipe_id = (entry->ring_id & 0x03) >> 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c index f2309d72bbe6..9508709abd49 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c @@ -289,7 +289,8 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, return -ENOMEM; } - r = amdgpu_userq_create_object(uq_mgr, &queue->mqd, mqd_hw_default->mqd_size); + r = amdgpu_userq_create_object(uq_mgr, &queue->mqd, + AMDGPU_MQD_SIZE_ALIGN(mqd_hw_default->mqd_size)); if (r) { DRM_ERROR("Failed to create MQD object for userqueue\n"); goto free_props; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 7811cbb1f7ba..e77e079fe833 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -1704,24 +1704,12 @@ static int sdma_v5_0_process_trap_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - uint32_t mes_queue_id = entry->src_data[0]; - DRM_DEBUG("IH: SDMA trap\n"); - if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) { - struct amdgpu_mes_queue *queue; - - mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK; - - spin_lock(&adev->mes.queue_id_lock); - queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id); - if (queue) { - DRM_DEBUG("process smda queue id = %d\n", mes_queue_id); - amdgpu_fence_process(queue->ring); - } - spin_unlock(&adev->mes.queue_id_lock); + if (drm_WARN_ON_ONCE(&adev->ddev, + adev->enable_mes && + (entry->src_data[0] & AMDGPU_FENCE_MES_QUEUE_FLAG))) return 0; - } switch (entry->client_id) { case SOC15_IH_CLIENTID_SDMA0: diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index dbe5b8f109f6..50b51965c211 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1617,24 +1617,12 @@ static int sdma_v5_2_process_trap_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - uint32_t mes_queue_id = entry->src_data[0]; - DRM_DEBUG("IH: SDMA trap\n"); - if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) { - struct amdgpu_mes_queue *queue; - - mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK; - - spin_lock(&adev->mes.queue_id_lock); - queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id); - if (queue) { - DRM_DEBUG("process smda queue id = %d\n", mes_queue_id); - amdgpu_fence_process(queue->ring); - } - spin_unlock(&adev->mes.queue_id_lock); + if (drm_WARN_ON_ONCE(&adev->ddev, + adev->enable_mes && + (entry->src_data[0] & AMDGPU_FENCE_MES_QUEUE_FLAG))) return 0; - } switch (entry->client_id) { case SOC15_IH_CLIENTID_SDMA0: diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c index 5bc45c3e00d1..0824cba48f2e 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c @@ -1494,24 +1494,13 @@ static int sdma_v7_1_process_trap_irq(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { int inst, instances, queue, xcc_id = 0; - uint32_t mes_queue_id = entry->src_data[0]; DRM_DEBUG("IH: SDMA trap\n"); - if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) { - struct amdgpu_mes_queue *queue; - - mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK; - - spin_lock(&adev->mes.queue_id_lock); - queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id); - if (queue) { - DRM_DEBUG("process smda queue id = %d\n", mes_queue_id); - amdgpu_fence_process(queue->ring); - } - spin_unlock(&adev->mes.queue_id_lock); + if (drm_WARN_ON_ONCE(&adev->ddev, + adev->enable_mes && + (entry->src_data[0] & AMDGPU_FENCE_MES_QUEUE_FLAG))) return 0; - } queue = entry->ring_id & 0xf; if (adev->gfx.funcs && adev->gfx.funcs->ih_node_to_logical_xcc) diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index 2da733b45c21..d9cc649d81ad 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -256,7 +256,13 @@ static u32 soc21_get_config_memsize(struct amdgpu_device *adev) static u32 soc21_get_xclk(struct amdgpu_device *adev) { - return adev->clock.spll.reference_freq; + u32 reference_clock = adev->clock.spll.reference_freq; + + /* reference clock is actually 99.81 Mhz rather than 100 Mhz */ + if ((adev->flags & AMD_IS_APU) && reference_clock == 10000) + return 9981; + + return reference_clock; } 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 dd247abce1ab..e78526a4e521 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -1742,11 +1742,11 @@ static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, goto unlock; } - r = vcn_v4_0_3_reset_jpeg_post_helper(adev, ring->me); if (pg_state) amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_JPEG, AMD_PG_STATE_GATE); mutex_unlock(&adev->jpeg.jpeg_pg_lock); + r = vcn_v4_0_3_reset_jpeg_post_helper(adev, ring->me); unlock: mutex_unlock(&vinst->engine_reset_mutex); diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h index 39bdc98b8b6d..54fa76f374c9 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h @@ -4587,14 +4587,14 @@ static const uint32_t cwsr_trap_gfx9_5_0_hex[] = { }; static const uint32_t cwsr_trap_gfx12_1_0_hex[] = { - 0xbfa00001, 0xbfa003ac, + 0xbfa00001, 0xbfa003be, 0xb0804009, 0xb8f8f804, 0x9178ff78, 0x00008c00, 0xb8fbf811, 0x8b6eff78, 0x00004000, 0xbfa10008, 0x8b6eff7b, 0x00000080, 0xbfa20018, 0x8b6ea07b, - 0xbfa200d1, 0xbf830010, + 0xbfa200da, 0xbf830010, 0xb8fbf811, 0xbfa0fffb, 0x8b6eff7b, 0x00000bd0, 0xbfa20010, 0xb8eef812, @@ -4605,7 +4605,7 @@ static const uint32_t cwsr_trap_gfx12_1_0_hex[] = { 0xf0000000, 0xbfa20005, 0x8b6fff6f, 0x00000200, 0xbfa20002, 0x8b6ea07b, - 0xbfa200bb, 0x9177ff77, + 0xbfa200c4, 0x9177ff77, 0x007fc000, 0xb8fa04a1, 0x847a967a, 0x8c777a77, 0xb8fa0421, 0x847a957a, @@ -4632,43 +4632,46 @@ static const uint32_t cwsr_trap_gfx12_1_0_hex[] = { 0xbfa00002, 0x806c846c, 0x826d806d, 0x8b6dff6d, 0x01ffffff, 0xb8fbf811, - 0xbf0d847b, 0xbfa20078, + 0xbf0d847b, 0xbfa20081, 0xf4003eb6, 0xf8000000, 0xbfc70000, 0xf4003bb6, 0xf8000008, 0x8b76ff7a, 0x80000000, 0xbfa20027, 0x9376ff7a, 0x00060019, 0x81f9a376, 0xbf0b8179, - 0xbfa20068, 0x81f9ac76, - 0xbf0b8179, 0xbfa20062, + 0xbfa2006e, 0x81f9ac76, + 0xbf0b8179, 0xbfa20068, 0x81f9b776, 0xbf0b8179, - 0xbfa2005f, 0x8b76ff7a, + 0xbfa20065, 0x8b76ff7a, 0x000001ff, 0xbf06ff76, - 0x000000fe, 0xbfa2005d, + 0x000000fe, 0xbfa20063, 0xbf06ff76, 0x000000ff, - 0xbfa20057, 0xbf06ff76, - 0x000000fa, 0xbfa20054, + 0xbfa2005d, 0xbf06ff76, + 0x000000fa, 0xbfa2005a, 0x81f9ff76, 0x000000e9, - 0xbf0b8179, 0xbfa20050, + 0xbf0b8179, 0xbfa20056, 0x8b76ff7b, 0xffff0000, 0xbf06ff76, 0xbf860000, - 0xbfa10051, 0x9376ff7b, + 0xbfa1005a, 0x9376ff7b, 0x0002000e, 0x8b79ff7b, 0x00003f00, 0x85798679, 0x8c767976, 0xb9763b01, - 0xbfa00049, 0x8b76ff7a, + 0xbfa00052, 0x8b76ff7a, 0xfc000000, 0xbf06ff76, - 0xd4000000, 0xbfa20013, + 0xd4000000, 0xbfa20019, 0xbf06ff76, 0xc8000000, - 0xbfa20027, 0x8b76ff7a, + 0xbfa2002d, 0x8b76ff7a, 0xff000000, 0xbf06ff76, - 0xcf000000, 0xbfa20039, + 0xcf000000, 0xbfa2003f, 0x8b79ff7a, 0xffff0000, + 0xbf06ff79, 0xcc330000, + 0xbfa2003d, 0xbf06ff79, + 0xcc880000, 0xbfa2003a, 0xbf06ff79, 0xcc350000, - 0xbfa20037, 0xbf06ff79, - 0xcc3a0000, 0xbfa20034, + 0xbfa2003a, 0xbf06ff79, + 0xcc3a0000, 0xbfa20037, 0xbf06ff76, 0xcc000000, - 0xbfa10031, 0x8b76ff7b, + 0xbfa10034, 0x8b76ff7b, 0x000001ff, 0xbf06ff76, 0x000000ff, 0xbfa20029, 0xbf06ff76, 0x000000fa, @@ -4691,86 +4694,92 @@ static const uint32_t cwsr_trap_gfx12_1_0_hex[] = { 0x000001ff, 0xbf06ff76, 0x000000ff, 0xbfa20003, 0xbfc70000, 0xbefb006e, - 0xbfa0ffad, 0xbfc70000, - 0xbefb006f, 0xbfa0ffaa, - 0xbfc70000, 0x857a9677, - 0xb97a04a1, 0x857a9577, - 0xb97a0421, 0x857a8e77, - 0xb97a3021, 0x8bfe7e7e, - 0x8bea6a6a, 0x85788978, - 0xb9783244, 0xbe804a6c, - 0xb8faf802, 0xbf0d987a, - 0xbfa10001, 0xbfb00000, - 0x8b6dff6d, 0x01ffffff, - 0xbefa0080, 0xb97a0151, - 0x9177ff77, 0x007fc000, - 0xb8fa04a1, 0x847a967a, - 0x8c777a77, 0xb8fa0421, - 0x847a957a, 0x8c777a77, - 0xb8fa3021, 0x847a8e7a, - 0x8c777a77, 0xb980f821, - 0x00000000, 0xbf0d847b, - 0xbfa20078, 0xf4003eb6, - 0xf8000000, 0xbfc70000, - 0xf4003bb6, 0xf8000008, - 0x8b76ff7a, 0x80000000, - 0xbfa20027, 0x9376ff7a, - 0x00060019, 0x81f9a376, + 0xbfa0ffa7, 0xbfc70000, + 0xbefb006f, 0xbfa0ffa4, + 0x80ec886c, 0x82ed806d, + 0xbfa0fff7, 0xbfc70000, + 0x857a9677, 0xb97a04a1, + 0x857a9577, 0xb97a0421, + 0x857a8e77, 0xb97a3021, + 0x8bfe7e7e, 0x8bea6a6a, + 0x85788978, 0xb9783244, + 0xbe804a6c, 0xb8faf802, + 0xbf0d987a, 0xbfa10001, + 0xbfb00000, 0x8b6dff6d, + 0x01ffffff, 0xbefa0080, + 0xb97a0151, 0x9177ff77, + 0x007fc000, 0xb8fa04a1, + 0x847a967a, 0x8c777a77, + 0xb8fa0421, 0x847a957a, + 0x8c777a77, 0xb8fa3021, + 0x847a8e7a, 0x8c777a77, + 0xb980f821, 0x00000000, + 0xbf0d847b, 0xbfa20081, + 0xf4003eb6, 0xf8000000, + 0xbfc70000, 0xf4003bb6, + 0xf8000008, 0x8b76ff7a, + 0x80000000, 0xbfa20027, + 0x9376ff7a, 0x00060019, + 0x81f9a376, 0xbf0b8179, + 0xbfa2006e, 0x81f9ac76, 0xbf0b8179, 0xbfa20068, - 0x81f9ac76, 0xbf0b8179, - 0xbfa20062, 0x81f9b776, - 0xbf0b8179, 0xbfa2005f, - 0x8b76ff7a, 0x000001ff, - 0xbf06ff76, 0x000000fe, + 0x81f9b776, 0xbf0b8179, + 0xbfa20065, 0x8b76ff7a, + 0x000001ff, 0xbf06ff76, + 0x000000fe, 0xbfa20063, + 0xbf06ff76, 0x000000ff, 0xbfa2005d, 0xbf06ff76, - 0x000000ff, 0xbfa20057, + 0x000000fa, 0xbfa2005a, + 0x81f9ff76, 0x000000e9, + 0xbf0b8179, 0xbfa20056, + 0x8b76ff7b, 0xffff0000, + 0xbf06ff76, 0xbf860000, + 0xbfa1005a, 0x9376ff7b, + 0x0002000e, 0x8b79ff7b, + 0x00003f00, 0x85798679, + 0x8c767976, 0xb9763b01, + 0xbfa00052, 0x8b76ff7a, + 0xfc000000, 0xbf06ff76, + 0xd4000000, 0xbfa20019, + 0xbf06ff76, 0xc8000000, + 0xbfa2002d, 0x8b76ff7a, + 0xff000000, 0xbf06ff76, + 0xcf000000, 0xbfa2003f, + 0x8b79ff7a, 0xffff0000, + 0xbf06ff79, 0xcc330000, + 0xbfa2003d, 0xbf06ff79, + 0xcc880000, 0xbfa2003a, + 0xbf06ff79, 0xcc350000, + 0xbfa2003a, 0xbf06ff79, + 0xcc3a0000, 0xbfa20037, + 0xbf06ff76, 0xcc000000, + 0xbfa10034, 0x8b76ff7b, + 0x000001ff, 0xbf06ff76, + 0x000000ff, 0xbfa20029, 0xbf06ff76, 0x000000fa, - 0xbfa20054, 0x81f9ff76, - 0x000000e9, 0xbf0b8179, - 0xbfa20050, 0x8b76ff7b, - 0xffff0000, 0xbf06ff76, - 0xbf860000, 0xbfa10051, - 0x9376ff7b, 0x0002000e, - 0x8b79ff7b, 0x00003f00, - 0x85798679, 0x8c767976, - 0xb9763b01, 0xbfa00049, - 0x8b76ff7a, 0xfc000000, - 0xbf06ff76, 0xd4000000, - 0xbfa20013, 0xbf06ff76, - 0xc8000000, 0xbfa20027, - 0x8b76ff7a, 0xff000000, - 0xbf06ff76, 0xcf000000, - 0xbfa20039, 0x8b79ff7a, - 0xffff0000, 0xbf06ff79, - 0xcc350000, 0xbfa20037, - 0xbf06ff79, 0xcc3a0000, - 0xbfa20034, 0xbf06ff76, - 0xcc000000, 0xbfa10031, - 0x8b76ff7b, 0x000001ff, - 0xbf06ff76, 0x000000ff, - 0xbfa20029, 0xbf06ff76, - 0x000000fa, 0xbfa20026, - 0x81f6ff76, 0x000000e9, - 0xbf0b8176, 0xbfa20022, - 0x8b76ff7b, 0x0003fe00, - 0xbf06ff76, 0x0001fe00, - 0xbfa2001d, 0x8b76ff7b, - 0x07fc0000, 0xbf06ff76, - 0x03fc0000, 0xbfa20018, - 0xbfa00014, 0x9376ff7a, - 0x00040016, 0x81f68176, - 0xbf0b8176, 0xbfa20012, - 0x9376ff7a, 0x00050011, + 0xbfa20026, 0x81f6ff76, + 0x000000e9, 0xbf0b8176, + 0xbfa20022, 0x8b76ff7b, + 0x0003fe00, 0xbf06ff76, + 0x0001fe00, 0xbfa2001d, + 0x8b76ff7b, 0x07fc0000, + 0xbf06ff76, 0x03fc0000, + 0xbfa20018, 0xbfa00014, + 0x9376ff7a, 0x00040016, 0x81f68176, 0xbf0b8176, - 0xbfa2000d, 0x8b76ff7a, - 0x000001ff, 0xbf06ff76, - 0x000000ff, 0xbfa20008, - 0x8b76ff7b, 0x000001ff, + 0xbfa20012, 0x9376ff7a, + 0x00050011, 0x81f68176, + 0xbf0b8176, 0xbfa2000d, + 0x8b76ff7a, 0x000001ff, 0xbf06ff76, 0x000000ff, - 0xbfa20003, 0xbfc70000, - 0xbefb006e, 0xbfa0ffad, - 0xbfc70000, 0xbefb006f, - 0xbfa0ffaa, 0xbfc70000, + 0xbfa20008, 0x8b76ff7b, + 0x000001ff, 0xbf06ff76, + 0x000000ff, 0xbfa20003, + 0xbfc70000, 0xbefb006e, + 0xbfa0ffa7, 0xbfc70000, + 0xbefb006f, 0xbfa0ffa4, + 0x80ec886c, 0x82ed806d, + 0xbfa0fff7, 0xbfc70000, 0xbeee007e, 0xbeef007f, 0xbefe0180, 0xbefe4d84, 0xbf8a0000, 0x8b7aff7f, diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm index b1b1d15bb1e7..456db8199899 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm @@ -37,6 +37,7 @@ #define HAVE_CLUSTER_BARRIER (ASIC_FAMILY == CHIP_GC_12_0_3) #define CLUSTER_BARRIER_SERIALIZE_WORKAROUND (ASIC_FAMILY == CHIP_GC_12_0_3) #define RELAXED_SCHEDULING_IN_TRAP (ASIC_FAMILY == CHIP_GFX12) +#define HAVE_INSTRUCTION_FIXUP (ASIC_FAMILY == CHIP_GC_12_0_3) #define SINGLE_STEP_MISSED_WORKAROUND 1 //workaround for lost TRAP_AFTER_INST exception when SAVECTX raised #define HAVE_VALU_SGPR_HAZARD (ASIC_FAMILY == CHIP_GFX12) @@ -375,9 +376,9 @@ L_TRAP_CASE: L_EXIT_TRAP: s_and_b32 ttmp1, ttmp1, ADDRESS_HI32_MASK -#if HAVE_BANKED_VGPRS +#if HAVE_INSTRUCTION_FIXUP s_getreg_b32 s_save_excp_flag_priv, hwreg(HW_REG_WAVE_EXCP_FLAG_PRIV) - fixup_vgpr_bank_selection() + fixup_instruction() #endif #if HAVE_XNACK @@ -418,8 +419,8 @@ L_HAVE_VGPRS: save_and_clear_xnack_state_priv(s_save_tmp) #endif -#if HAVE_BANKED_VGPRS - fixup_vgpr_bank_selection() +#if HAVE_INSTRUCTION_FIXUP + fixup_instruction() #endif /* inform SPI the readiness and wait for SPI's go signal */ @@ -1400,8 +1401,8 @@ L_BARRIER_RESTORE_LOOP: L_BARRIER_RESTORE_DONE: end -#if HAVE_BANKED_VGPRS -function fixup_vgpr_bank_selection +#if HAVE_INSTRUCTION_FIXUP +function fixup_instruction // PC read may fault if memory violation has been asserted. // In this case no further progress is expected so fixup is not needed. s_bitcmp1_b32 s_save_excp_flag_priv, SQ_WAVE_EXCP_FLAG_PRIV_MEM_VIOL_SHIFT @@ -1480,8 +1481,13 @@ L_FIXUP_NOT_VOP12C: s_cmp_eq_u32 ttmp10, 0xcf000000 // If 31:24 = 0xcf, this is VOPD3 s_cbranch_scc1 L_FIXUP_THREE_DWORD // If VOPD3, 3 DWORD inst // Not VOP1, VOP2, VOPC, VOP3, VOP3SD, VOPD, or VOPD3. - // Might be in VOP3P, but we must ensure we are not VOP3PX2 + // Check if we are in the middle of VOP3PX. s_and_b32 ttmp13, ttmp14, 0xffff0000 // Bits 31:16 + s_cmp_eq_u32 ttmp13, 0xcc330000 // If 31:16 = 0xcc33, this is 8 bytes past VOP3PX + s_cbranch_scc1 L_FIXUP_VOP3PX_MIDDLE + s_cmp_eq_u32 ttmp13, 0xcc880000 // If 31:16 = 0xcc88, this is 8 bytes past VOP3PX + s_cbranch_scc1 L_FIXUP_VOP3PX_MIDDLE + // Might be in VOP3P, but we must ensure we are not VOP3PX2 s_cmp_eq_u32 ttmp13, 0xcc350000 // If 31:16 = 0xcc35, this is VOP3PX2 s_cbranch_scc1 L_FIXUP_DONE // If VOP3PX2, no fixup needed s_cmp_eq_u32 ttmp13, 0xcc3a0000 // If 31:16 = 0xcc3a, this is VOP3PX2 @@ -1542,6 +1548,11 @@ L_FIXUP_THREE_DWORD: s_mov_b32 ttmp15, ttmp3 // Move possible S_SET_VGPR_MSB into ttmp15 s_branch L_FIXUP_ONE_DWORD // Go to common logic that checks if it is S_SET_VGPR_MSB +L_FIXUP_VOP3PX_MIDDLE: + s_sub_co_u32 ttmp0, ttmp0, 8 // Rewind PC 8 bytes to beginning of instruction + s_sub_co_ci_u32 ttmp1, ttmp1, 0 + s_branch L_FIXUP_TWO_DWORD // 2 DWORD inst (2nd half of a 4 DWORD inst) + L_FIXUP_DONE: s_wait_kmcnt 0 // Ensure load of ttmp2 and ttmp3 is done end diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c index 27176b2dc714..8f8a0975f1a7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c @@ -1108,6 +1108,7 @@ int kfd_dbg_trap_device_snapshot(struct kfd_process *target, device_info.num_xcc = NUM_XCC(pdd->dev->xcc_mask); device_info.capability = topo_dev->node_props.capability; device_info.debug_prop = topo_dev->node_props.debug_prop; + device_info.capability2 = topo_dev->node_props.capability2; if (exception_clear_mask) pdd->exception_status &= ~exception_clear_mask; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index b542de9d50d1..804851632c4c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -676,7 +676,7 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm, /* Temporarily release dqm lock to avoid a circular lock dependency */ dqm_unlock(dqm); - q->mqd_mem_obj = mqd_mgr->allocate_mqd(mqd_mgr->dev, &q->properties); + q->mqd_mem_obj = mqd_mgr->allocate_mqd(mqd_mgr, &q->properties); dqm_lock(dqm); if (!q->mqd_mem_obj) { @@ -2002,7 +2002,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, dqm->asic_ops.init_sdma_vm(dqm, q, qpd); q->properties.tba_addr = qpd->tba_addr; q->properties.tma_addr = qpd->tma_addr; - q->mqd_mem_obj = mqd_mgr->allocate_mqd(mqd_mgr->dev, &q->properties); + q->mqd_mem_obj = mqd_mgr->allocate_mqd(mqd_mgr, &q->properties); if (!q->mqd_mem_obj) { retval = -ENOMEM; goto out_deallocate_doorbell; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 961a4b84e974..d987ff7ccfc9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -130,7 +130,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev, kq->queue->device = dev; - kq->queue->mqd_mem_obj = kq->mqd_mgr->allocate_mqd(kq->mqd_mgr->dev, + kq->queue->mqd_mem_obj = kq->mqd_mgr->allocate_mqd(kq->mqd_mgr, &kq->queue->properties); if (!kq->queue->mqd_mem_obj) goto err_allocate_mqd; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index f78b249e1a41..d88d0de58edd 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -46,9 +46,10 @@ int pipe_priority_map[] = { KFD_PIPE_PRIORITY_CS_HIGH }; -struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properties *q) +struct kfd_mem_obj *allocate_hiq_mqd(struct mqd_manager *mm, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj; + struct kfd_node *dev = mm->dev; mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); if (!mqd_mem_obj) @@ -61,10 +62,11 @@ struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properti return mqd_mem_obj; } -struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev, +struct kfd_mem_obj *allocate_sdma_mqd(struct mqd_manager *mm, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj; + struct kfd_node *dev = mm->dev; uint64_t offset; mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); @@ -290,6 +292,9 @@ void kfd_get_hiq_xcc_mqd(struct kfd_node *dev, struct kfd_mem_obj *mqd_mem_obj, uint64_t kfd_mqd_stride(struct mqd_manager *mm, struct queue_properties *q) { + if (KFD_GC_VERSION(mm->dev) >= IP_VERSION(11, 0, 0)) + return AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); + return mm->mqd_size; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h index 17cc1f25c8d0..2429d278ef0e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h @@ -68,7 +68,7 @@ */ extern int pipe_priority_map[]; struct mqd_manager { - struct kfd_mem_obj* (*allocate_mqd)(struct kfd_node *kfd, + struct kfd_mem_obj* (*allocate_mqd)(struct mqd_manager *mm, struct queue_properties *q); void (*init_mqd)(struct mqd_manager *mm, void **mqd, @@ -153,10 +153,10 @@ struct mqd_user_context_save_area_header { uint32_t wave_state_size; }; -struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, +struct kfd_mem_obj *allocate_hiq_mqd(struct mqd_manager *mm, struct queue_properties *q); -struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev, +struct kfd_mem_obj *allocate_sdma_mqd(struct mqd_manager *mm, struct queue_properties *q); void free_mqd_hiq_sdma(struct mqd_manager *mm, void *mqd, struct kfd_mem_obj *mqd_mem_obj); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c index 05f3ac2eaef9..90ac3a30e81d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c @@ -73,9 +73,10 @@ static void set_priority(struct cik_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { + struct kfd_node *kfd = mm->dev; struct kfd_mem_obj *mqd_mem_obj; if (kfd_gtt_sa_allocate(kfd, sizeof(struct cik_mqd), diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c index 1695dd78ede8..97055f808d4a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c @@ -73,9 +73,10 @@ static void set_priority(struct v10_compute_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { + struct kfd_node *kfd = mm->dev; struct kfd_mem_obj *mqd_mem_obj; if (kfd_gtt_sa_allocate(kfd, sizeof(struct v10_compute_mqd), diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c index 3c0ae28c5923..7e5a7ab6d0c0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c @@ -99,22 +99,14 @@ static void set_priority(struct v11_compute_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { + u32 mqd_size = AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); + struct kfd_node *node = mm->dev; struct kfd_mem_obj *mqd_mem_obj; - int size; - - /* - * MES write to areas beyond MQD size. So allocate - * 1 PAGE_SIZE memory for MQD is MES is enabled. - */ - if (node->kfd->shared_resources.enable_mes) - size = PAGE_SIZE; - else - size = sizeof(struct v11_compute_mqd); - if (kfd_gtt_sa_allocate(node, size, &mqd_mem_obj)) + if (kfd_gtt_sa_allocate(node, mqd_size, &mqd_mem_obj)) return NULL; return mqd_mem_obj; @@ -126,18 +118,13 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, { uint64_t addr; struct v11_compute_mqd *m; - int size; + u32 mqd_size = AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); uint32_t wa_mask = q->is_dbg_wa ? 0xffff : 0xffffffff; m = (struct v11_compute_mqd *) mqd_mem_obj->cpu_ptr; addr = mqd_mem_obj->gpu_addr; - if (mm->dev->kfd->shared_resources.enable_mes) - size = PAGE_SIZE; - else - size = sizeof(struct v11_compute_mqd); - - memset(m, 0, size); + memset(m, 0, mqd_size); m->header = 0xC0310800; m->compute_pipelinestat_enable = 1; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12.c index 729df66ebfb3..a51f217329db 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12.c @@ -80,16 +80,14 @@ static void set_priority(struct v12_compute_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { + u32 mqd_size = AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); + struct kfd_node *node = mm->dev; struct kfd_mem_obj *mqd_mem_obj; - /* - * Allocate one PAGE_SIZE memory for MQD as MES writes to areas beyond - * struct MQD size. - */ - if (kfd_gtt_sa_allocate(node, PAGE_SIZE, &mqd_mem_obj)) + if (kfd_gtt_sa_allocate(node, mqd_size, &mqd_mem_obj)) return NULL; return mqd_mem_obj; @@ -101,11 +99,12 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, { uint64_t addr; struct v12_compute_mqd *m; + u32 mqd_size = AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); m = (struct v12_compute_mqd *) mqd_mem_obj->cpu_ptr; addr = mqd_mem_obj->gpu_addr; - memset(m, 0, PAGE_SIZE); + memset(m, 0, mqd_size); m->header = 0xC0310800; m->compute_pipelinestat_enable = 1; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12_1.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12_1.c index 558216395a4d..d0776ba2cc99 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12_1.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v12_1.c @@ -32,17 +32,6 @@ #include "amdgpu_amdkfd.h" #include "kfd_device_queue_manager.h" -#define MQD_SIZE (2 * PAGE_SIZE) - -static uint64_t mqd_stride_v12_1(struct mqd_manager *mm, - struct queue_properties *q) -{ - if (q->type == KFD_QUEUE_TYPE_COMPUTE) - return MQD_SIZE; - else - return PAGE_SIZE; -} - static inline struct v12_1_compute_mqd *get_mqd(void *mqd) { return (struct v12_1_compute_mqd *)mqd; @@ -145,23 +134,17 @@ static void set_priority(struct v12_1_compute_mqd *m, struct queue_properties *q m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { + u32 mqd_size = AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); + struct kfd_node *node = mm->dev; struct kfd_mem_obj *mqd_mem_obj; - unsigned int size; - /* - * Allocate two PAGE_SIZE memory for Compute MQD as MES writes to areas beyond - * struct MQD size. Size of the Compute MQD is 1 PAGE_SIZE. - * For SDMA MQD, we allocate 1 Page_size. - */ if (q->type == KFD_QUEUE_TYPE_COMPUTE) - size = MQD_SIZE * NUM_XCC(node->xcc_mask); - else - size = PAGE_SIZE; + mqd_size *= NUM_XCC(node->xcc_mask); - if (kfd_gtt_sa_allocate(node, size, &mqd_mem_obj)) + if (kfd_gtt_sa_allocate(node, mqd_size, &mqd_mem_obj)) return NULL; return mqd_mem_obj; @@ -173,11 +156,12 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, { uint64_t addr; struct v12_1_compute_mqd *m; + u32 mqd_size = AMDGPU_MQD_SIZE_ALIGN(mm->mqd_size); m = (struct v12_1_compute_mqd *) mqd_mem_obj->cpu_ptr; addr = mqd_mem_obj->gpu_addr; - memset(m, 0, MQD_SIZE); + memset(m, 0, mqd_size); m->header = 0xC0310800; m->compute_pipelinestat_enable = 1; @@ -680,7 +664,7 @@ struct mqd_manager *mqd_manager_init_v12_1(enum KFD_MQD_TYPE type, mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v12_1_compute_mqd); mqd->get_wave_state = get_wave_state_v12_1; - mqd->mqd_stride = mqd_stride_v12_1; + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c index d867dccae675..dcf4bbfa641b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c @@ -120,10 +120,11 @@ static bool mqd_on_vram(struct amdgpu_device *adev) } } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { int retval; + struct kfd_node *node = mm->dev; struct kfd_mem_obj *mqd_mem_obj = NULL; /* For V9 only, due to a HW bug, the control stack of a user mode diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c index c1fafc502515..09483f0862d4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c @@ -76,9 +76,10 @@ static void set_priority(struct vi_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd, +static struct kfd_mem_obj *allocate_mqd(struct mqd_manager *mm, struct queue_properties *q) { + struct kfd_node *kfd = mm->dev; struct kfd_mem_obj *mqd_mem_obj; if (kfd_gtt_sa_allocate(kfd, sizeof(struct vi_mqd), diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index b4ba60e90c47..0b4fc654e76f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -11867,7 +11867,7 @@ static int dm_check_cursor_fb(struct amdgpu_crtc *new_acrtc, * check tiling flags when the FB doesn't have a modifier. */ if (!(fb->flags & DRM_MODE_FB_MODIFIERS)) { - if (adev->family >= AMDGPU_FAMILY_GC_12_0_0) { + if (adev->family == AMDGPU_FAMILY_GC_12_0_0) { linear = AMDGPU_TILING_GET(afb->tiling_flags, GFX12_SWIZZLE_MODE) == 0; } else if (adev->family >= AMDGPU_FAMILY_AI) { linear = AMDGPU_TILING_GET(afb->tiling_flags, SWIZZLE_MODE) == 0; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c index d95608abb7c4..d3e62f511c8f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c @@ -278,7 +278,7 @@ static int amdgpu_dm_plane_validate_dcc(struct amdgpu_device *adev, if (!dcc->enable) return 0; - if (adev->family < AMDGPU_FAMILY_GC_12_0_0 && + if (adev->family != AMDGPU_FAMILY_GC_12_0_0 && format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) return -EINVAL; @@ -901,7 +901,7 @@ int amdgpu_dm_plane_fill_plane_buffer_attributes(struct amdgpu_device *adev, upper_32_bits(chroma_addr); } - if (adev->family >= AMDGPU_FAMILY_GC_12_0_0) { + if (adev->family == AMDGPU_FAMILY_GC_12_0_0) { ret = amdgpu_dm_plane_fill_gfx12_plane_attributes_from_modifiers(adev, afb, format, rotation, plane_size, tiling_info, dcc, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index ad8ce46ce310..cb85b7ac2697 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -7506,7 +7506,7 @@ struct dc_update_scratch_space *dc_update_planes_and_stream_init( .stream = stream, .stream_update = stream_update, .update_v3 = version >= DCN_VERSION_4_01 || version == DCN_VERSION_3_2 || version == DCN_VERSION_3_21, - .do_clear_update_flags = version >= DCN_VERSION_3_2 || version == DCN_VERSION_3_01, + .do_clear_update_flags = version >= DCN_VERSION_1_0, }; return scratch; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 9349cccc8438..f59020f1a722 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -515,6 +515,19 @@ bool dc_stream_program_cursor_position( } } + /* apply manual trigger */ + int i; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; + + /* trigger event on first pipe with current stream */ + if (stream == pipe_ctx->stream) { + pipe_ctx->stream_res.tg->funcs->program_manual_trigger(pipe_ctx->stream_res.tg); + break; + } + } + return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index d377be76360c..ab19b6230945 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -63,7 +63,7 @@ struct dcn_dsc_reg_state; struct dcn_optc_reg_state; struct dcn_dccg_reg_state; -#define DC_VER "3.2.366" +#define DC_VER "3.2.367" /** * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC @@ -1210,6 +1210,7 @@ struct dc_debug_options { bool disable_deferred_minimal_transitions; unsigned int num_fast_flips_to_steady_state_override; bool enable_dmu_recovery; + unsigned int force_vmin_threshold; }; diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.c index 7847c1c4927b..97ef8281a476 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.c @@ -944,3 +944,21 @@ void hubbub1_construct(struct hubbub *hubbub, hubbub1->debug_test_index_pstate = 0xB; } +void dcn10_hubbub_global_timer_enable(struct hubbub *hubbub, bool enable, uint32_t refdiv) +{ + struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); + + if (refdiv > 0) + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, refdiv); + + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, enable ? 1 : 0); +} + +void dcn10_hubbub_read_fb_aperture(struct hubbub *hubbub, uint32_t *fb_base_value, uint32_t *fb_offset_value) +{ + struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); + + REG_GET(DCHUBBUB_SDPIF_FB_BASE, SDPIF_FB_BASE, fb_base_value); + REG_GET(DCHUBBUB_SDPIF_FB_OFFSET, SDPIF_FB_OFFSET, fb_offset_value); +} + diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.h index 0a29a758d013..990d3cd8e050 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn10/dcn10_hubbub.h @@ -519,4 +519,8 @@ bool hubbub1_program_pstate_watermarks( unsigned int refclk_mhz, bool safe_to_lower); +void dcn10_hubbub_global_timer_enable(struct hubbub *hubbub, bool enable, uint32_t refdiv); + +void dcn10_hubbub_read_fb_aperture(struct hubbub *hubbub, uint32_t *fb_base_value, uint32_t *fb_offset_value); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index 8a23763ca98e..c1586364ecd4 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -2678,8 +2678,7 @@ static void mmhub_read_vm_context0_settings(struct dcn10_hubp *hubp1, uint32_t fb_base_value; uint32_t fb_offset_value; - REG_GET(DCHUBBUB_SDPIF_FB_BASE, SDPIF_FB_BASE, &fb_base_value); - REG_GET(DCHUBBUB_SDPIF_FB_OFFSET, SDPIF_FB_OFFSET, &fb_offset_value); + dcn10_hubbub_read_fb_aperture(hws->ctx->dc->res_pool->hubbub, &fb_base_value, &fb_offset_value); REG_GET(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, PAGE_DIRECTORY_ENTRY_HI32, &vm0->pte_base.high_part); @@ -3474,7 +3473,7 @@ void dcn10_set_static_screen_control(struct pipe_ctx **pipe_ctx, triggers, params->num_frames); } -static void dcn10_config_stereo_parameters( +void dcn10_config_stereo_parameters( struct dc_stream_state *stream, struct crtc_stereo_flags *flags) { enum view_3d_format view_format = stream->view_format; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h index 57d30ea225f2..476095c5dd0c 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h @@ -217,5 +217,7 @@ void dcn10_update_visual_confirm_color( void dcn10_reset_surface_dcc_and_tiling(struct pipe_ctx *pipe_ctx, struct dc_plane_state *plane_state, bool clear_tiling); +void dcn10_config_stereo_parameters( + struct dc_stream_state *stream, struct crtc_stereo_flags *flags); #endif /* __DC_HWSS_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index 77ac7b22b8dc..a76436dcbe40 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -46,6 +46,7 @@ #include "dchubbub.h" #include "reg_helper.h" #include "dcn10/dcn10_cm_common.h" +#include "dcn10/dcn10_hubbub.h" #include "vm_helper.h" #include "dccg.h" #include "dc_dmub_srv.h" @@ -3058,9 +3059,17 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) dccg->funcs->enable_symclk32_se(dccg, dp_hpo_inst, phyd32clk); } } else { - if (dccg->funcs->enable_symclk_se) - dccg->funcs->enable_symclk_se(dccg, stream_enc->stream_enc_inst, + if (dccg->funcs->enable_symclk_se && link_enc) { + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA + && link->cur_link_settings.link_rate == LINK_RATE_UNKNOWN + && !link->link_status.link_active) { + if (dccg->funcs->disable_symclk_se) + dccg->funcs->disable_symclk_se(dccg, stream_enc->stream_enc_inst, link_enc->transmitter - TRANSMITTER_UNIPHY_A); + } else + dccg->funcs->enable_symclk_se(dccg, stream_enc->stream_enc_inst, + link_enc->transmitter - TRANSMITTER_UNIPHY_A); + } } if (dc->res_pool->dccg->funcs->set_pixel_rate_div) @@ -3145,8 +3154,7 @@ void dcn20_fpga_init_hw(struct dc *dc) REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); + dcn10_hubbub_global_timer_enable(dc->res_pool->hubbub, true, 2); if (REG(REFCLK_CNTL)) REG_WRITE(REFCLK_CNTL, 0); // diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c index 1635e5a552ad..482053c4ad22 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c @@ -39,6 +39,7 @@ #include "dccg.h" #include "clk_mgr.h" #include "reg_helper.h" +#include "dcn10/dcn10_hubbub.h" #define CTX \ hws->ctx diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c index 81bcadf5e57e..c02ddada723f 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c @@ -36,6 +36,7 @@ #include "dcn10/dcn10_cm_common.h" #include "dcn30/dcn30_cm_common.h" #include "reg_helper.h" +#include "dcn10/dcn10_hubbub.h" #include "abm.h" #include "clk_mgr.h" #include "hubp.h" diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c index 20f700b59847..2adbcc105aa6 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c @@ -32,6 +32,7 @@ #include "dce/dce_hwseq.h" #include "clk_mgr.h" #include "reg_helper.h" +#include "dcn10/dcn10_hubbub.h" #include "abm.h" #include "hubp.h" #include "dchubbub.h" diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c index 2675d7dca586..f7e16fee7594 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c @@ -32,6 +32,7 @@ #include "dce/dce_hwseq.h" #include "clk_mgr.h" #include "reg_helper.h" +#include "dcn10/dcn10_hubbub.h" #include "abm.h" #include "hubp.h" #include "dchubbub.h" diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c index 8d10aac9c510..46985eb2a623 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c @@ -2022,9 +2022,7 @@ bool dcn20_fast_validate_bw( dcn20_merge_pipes_for_validate(dc, context); - DC_FP_START(); pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); - DC_FP_END(); *pipe_cnt_out = pipe_cnt; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c index 2060acd5ae09..967e813a45e5 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c @@ -785,9 +785,7 @@ bool dcn21_fast_validate_bw(struct dc *dc, dcn20_merge_pipes_for_validate(dc, context); - DC_FP_START(); pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); - DC_FP_END(); *pipe_cnt_out = pipe_cnt; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c index 875ae97489d3..1cdbb65da4a3 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c @@ -1772,7 +1772,7 @@ static int dcn401_get_power_profile(const struct dc_state *context) return dpm_level; } -static unsigned int dcn401_get_vstartup_for_pipe(struct pipe_ctx *pipe_ctx) +unsigned int dcn401_get_vstartup_for_pipe(struct pipe_ctx *pipe_ctx) { return pipe_ctx->global_sync.dcn4x.vstartup_lines; } diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h index 406246a9867e..08bec1755617 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h @@ -28,6 +28,8 @@ enum dc_status dcn401_validate_bandwidth(struct dc *dc, void dcn401_prepare_mcache_programming(struct dc *dc, struct dc_state *context); +unsigned int dcn401_get_vstartup_for_pipe(struct pipe_ctx *pipe_ctx); + /* Following are definitions for run time init of reg offsets */ /* HUBP */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 6d82973ccc18..18e0bdfd6ff4 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -139,6 +139,33 @@ */ #define DMUB_CMD_PSR_CONTROL_VERSION_1 0x1 +/** + * + * dirty rect cmd version legacy + */ +#define DMUB_CMD_DIRTY_RECTS_VERSION_UNKNOWN 0x0 +/** + * dirty rect cmd version with multi edp support + */ +#define DMUB_CMD_DIRTY_RECTS_VERSION_1 0x1 +/** + * dirty rect cmd version with external monitor support + */ +#define DMUB_CMD_DIRTY_RECTS_VERSION_2 0x2 + +/** + * + * Cursor update cmd version legacy + */ +#define DMUB_CMD_CURSOR_UPDATE_VERSION_UNKNOWN 0x0 +/** + * Cursor update cmd version with multi edp support + */ +#define DMUB_CMD_CURSOR_UPDATE_VERSION_1 0x1 +/** + * Cursor update cmd version with external monitor support + */ +#define DMUB_CMD_CURSOR_UPDATE_VERSION_2 0x2 /** * ABM control version legacy @@ -3929,7 +3956,7 @@ struct dmub_cmd_update_dirty_rect_data { */ union dmub_psr_su_debug_flags debug_flags; /** - * OTG HW instance. + * Pipe index. */ uint8_t pipe_idx; /** @@ -3937,7 +3964,7 @@ struct dmub_cmd_update_dirty_rect_data { */ uint8_t dirty_rect_count; /** - * PSR control version. + * dirty rects cmd version. */ uint8_t cmd_version; /** @@ -3946,6 +3973,14 @@ struct dmub_cmd_update_dirty_rect_data { * Currently the support is only for 0 or 1 */ uint8_t panel_inst; + /** + * OTG HW instance + */ + uint8_t otg_inst; + /** + * Padding for 4 byte alignment + */ + uint8_t padding[3]; }; /** @@ -4071,11 +4106,11 @@ struct dmub_cmd_update_cursor_payload0 { */ uint8_t enable; /** - * OTG HW instance. + * Pipe index. */ uint8_t pipe_idx; /** - * PSR control version. + * Cursor update cmd version. */ uint8_t cmd_version; /** @@ -4089,6 +4124,14 @@ struct dmub_cmd_update_cursor_payload0 { * Registers contains Hubp & Dpp modules */ struct dmub_cursor_position_cfg position_cfg; + /** + * OTG HW instance + */ + uint8_t otg_inst; + /** + * Padding for 4 byte alignment + */ + uint8_t padding[3]; }; struct dmub_cmd_update_cursor_payload1 { @@ -6629,9 +6672,9 @@ struct dmub_cmd_pr_copy_settings_data { */ uint8_t su_y_granularity; /** - * @pad: Align structure to 4 byte boundary. + * @main_link_activity_option: Indicates main link activity option selected */ - uint8_t pad; + uint8_t main_link_activity_option; }; /** diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c index 6a2d35756c8c..639f9835e5e9 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c @@ -419,6 +419,9 @@ void dmub_dcn35_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmu boot_options.bits.enable_non_transparent_setconfig = params->enable_non_transparent_setconfig; boot_options.bits.lower_hbr3_phy_ssc = params->lower_hbr3_phy_ssc; boot_options.bits.disable_dpia_bw_allocation = params->disable_dpia_bw_allocation; + boot_options.bits.bootcrc_en_at_preos = dmub_dcn35_get_fw_boot_option(dmub).bits.bootcrc_en_at_preos; + boot_options.bits.bootcrc_en_at_S0i3 = dmub_dcn35_get_fw_boot_option(dmub).bits.bootcrc_en_at_S0i3; + boot_options.bits.bootcrc_boot_mode = dmub_dcn35_get_fw_boot_option(dmub).bits.bootcrc_boot_mode; REG_WRITE(DMCUB_SCRATCH14, boot_options.all); } diff --git a/drivers/gpu/drm/amd/display/include/dpcd_defs.h b/drivers/gpu/drm/amd/display/include/dpcd_defs.h index 7d8359a7d99d..1afa10e85eb5 100644 --- a/drivers/gpu/drm/amd/display/include/dpcd_defs.h +++ b/drivers/gpu/drm/amd/display/include/dpcd_defs.h @@ -224,6 +224,7 @@ enum dpcd_psr_sink_states { #define DP_SINK_PR_PIXEL_DEVIATION_PER_LINE 0x379 #define DP_SINK_PR_MAX_NUMBER_OF_DEVIATION_LINE 0x37A #define DP_SINK_EMISSION_RATE 0x37E +#define DP_SINK_PR_FRAME_SKIP_COUNT 0x337 /* Remove once drm_dp_helper.h is updated upstream */ #ifndef DP_TOTAL_LTTPR_CNT diff --git a/drivers/gpu/drm/amd/include/mes_v12_api_def.h b/drivers/gpu/drm/amd/include/mes_v12_api_def.h index 83e4f0142324..e541a43714a1 100644 --- a/drivers/gpu/drm/amd/include/mes_v12_api_def.h +++ b/drivers/gpu/drm/amd/include/mes_v12_api_def.h @@ -492,8 +492,10 @@ union MESAPI__SUSPEND { union MES_API_HEADER header; /* false - suspend all gangs; true - specific gang */ struct { - uint32_t suspend_all_gangs : 1; - uint32_t reserved : 31; + uint32_t suspend_all_gangs : 1; // suspend all compute gangs (can be set together with suspend_all_sdma_gangs) + uint32_t query_status : 1; + uint32_t suspend_all_sdma_gangs : 1; // suspend all sdma gangs (can be set together with suspend_all_gangs) + uint32_t reserved : 29; }; /* gang_context_addr is valid only if suspend_all = false */ diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 4214f7314963..feadf604b474 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -80,15 +80,15 @@ int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, enum ip_power_state pwr_state = gate ? POWER_STATE_OFF : POWER_STATE_ON; bool is_vcn = block_type == AMD_IP_BLOCK_TYPE_VCN; + mutex_lock(&adev->pm.mutex); + if (atomic_read(&adev->pm.pwr_state[block_type]) == pwr_state && (!is_vcn || adev->vcn.num_vcn_inst == 1)) { dev_dbg(adev->dev, "IP block%d already in the target %s state!", block_type, gate ? "gate" : "ungate"); - return 0; + goto out_unlock; } - mutex_lock(&adev->pm.mutex); - switch (block_type) { case AMD_IP_BLOCK_TYPE_UVD: case AMD_IP_BLOCK_TYPE_VCE: @@ -115,6 +115,7 @@ int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, if (!ret) atomic_set(&adev->pm.pwr_state[block_type], pwr_state); +out_unlock: mutex_unlock(&adev->pm.mutex); return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index 6b6b05e8f736..75897ac203c3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -619,6 +619,9 @@ int amdgpu_smu_ras_send_msg(struct amdgpu_device *adev, enum smu_message_type ms struct smu_context *smu = adev->powerplay.pp_handle; int ret = -EOPNOTSUPP; + if (!smu) + return ret; + if (smu->ppt_funcs && smu->ppt_funcs->ras_send_msg) ret = smu->ppt_funcs->ras_send_msg(smu, msg, param, read_arg); @@ -688,12 +691,8 @@ static int smu_sys_set_pp_table(void *handle, return ret; } -static int smu_get_driver_allowed_feature_mask(struct smu_context *smu) +static int smu_init_driver_allowed_feature_mask(struct smu_context *smu) { - struct smu_feature *feature = &smu->smu_feature; - uint32_t allowed_feature_mask[SMU_FEATURE_MAX/32]; - int ret = 0; - /* * With SCPM enabled, the allowed featuremasks setting(via * PPSMC_MSG_SetAllowedFeaturesMaskLow/High) is not permitted. @@ -702,22 +701,13 @@ static int smu_get_driver_allowed_feature_mask(struct smu_context *smu) * such scenario. */ if (smu->adev->scpm_enabled) { - bitmap_fill(feature->allowed, SMU_FEATURE_MAX); + smu_feature_list_set_all(smu, SMU_FEATURE_LIST_ALLOWED); return 0; } - bitmap_zero(feature->allowed, SMU_FEATURE_MAX); + smu_feature_list_clear_all(smu, SMU_FEATURE_LIST_ALLOWED); - ret = smu_get_allowed_feature_mask(smu, allowed_feature_mask, - SMU_FEATURE_MAX/32); - if (ret) - return ret; - - bitmap_or(feature->allowed, feature->allowed, - (unsigned long *)allowed_feature_mask, - feature->feature_num); - - return ret; + return smu_init_allowed_features(smu); } static int smu_set_funcs(struct amdgpu_device *adev) @@ -1365,9 +1355,7 @@ static int smu_sw_init(struct amdgpu_ip_block *ip_block) int i, ret; smu->pool_size = adev->pm.smu_prv_buffer_size; - smu->smu_feature.feature_num = SMU_FEATURE_MAX; - bitmap_zero(smu->smu_feature.supported, SMU_FEATURE_MAX); - bitmap_zero(smu->smu_feature.allowed, SMU_FEATURE_MAX); + smu_feature_init(smu, SMU_FEATURE_MAX); INIT_WORK(&smu->throttling_logging_work, smu_throttling_logging_work_fn); INIT_WORK(&smu->interrupt_work, smu_interrupt_work_fn); @@ -1656,7 +1644,6 @@ static void smu_wbrf_fini(struct smu_context *smu) static int smu_smc_hw_setup(struct smu_context *smu) { - struct smu_feature *feature = &smu->smu_feature; struct amdgpu_device *adev = smu->adev; uint8_t pcie_gen = 0, pcie_width = 0; uint64_t features_supported; @@ -1819,9 +1806,8 @@ static int smu_smc_hw_setup(struct smu_context *smu) dev_err(adev->dev, "Failed to retrieve supported dpm features!\n"); return ret; } - bitmap_copy(feature->supported, - (unsigned long *)&features_supported, - feature->feature_num); + smu_feature_list_set_bits(smu, SMU_FEATURE_LIST_SUPPORTED, + (unsigned long *)&features_supported); if (!smu_is_dpm_running(smu)) dev_info(adev->dev, "dpm has been disabled\n"); @@ -1952,7 +1938,7 @@ static int smu_hw_init(struct amdgpu_ip_block *ip_block) if (!smu->pm_enabled) return 0; - ret = smu_get_driver_allowed_feature_mask(smu); + ret = smu_init_driver_allowed_feature_mask(smu); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index 1def04826f10..7c63c631f6d4 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -472,10 +472,20 @@ struct smu_power_context { }; #define SMU_FEATURE_MAX (64) + +struct smu_feature_bits { + DECLARE_BITMAP(bits, SMU_FEATURE_MAX); +}; + +enum smu_feature_list { + SMU_FEATURE_LIST_SUPPORTED, + SMU_FEATURE_LIST_ALLOWED, + SMU_FEATURE_LIST_MAX, +}; + struct smu_feature { uint32_t feature_num; - DECLARE_BITMAP(supported, SMU_FEATURE_MAX); - DECLARE_BITMAP(allowed, SMU_FEATURE_MAX); + struct smu_feature_bits bits[SMU_FEATURE_LIST_MAX]; }; struct smu_clocks { @@ -802,11 +812,10 @@ struct pptable_funcs { int (*run_btc)(struct smu_context *smu); /** - * @get_allowed_feature_mask: Get allowed feature mask. - * &feature_mask: Array to store feature mask. - * &num: Elements in &feature_mask. + * @init_allowed_features: Initialize allowed features bitmap. + * Directly sets allowed features using smu_feature wrapper functions. */ - int (*get_allowed_feature_mask)(struct smu_context *smu, uint32_t *feature_mask, uint32_t num); + int (*init_allowed_features)(struct smu_context *smu); /** * @get_current_power_state: Get the current power state. @@ -1974,4 +1983,158 @@ int amdgpu_smu_ras_send_msg(struct amdgpu_device *adev, enum smu_message_type ms void smu_feature_cap_set(struct smu_context *smu, enum smu_feature_cap_id fea_id); bool smu_feature_cap_test(struct smu_context *smu, enum smu_feature_cap_id fea_id); + +static inline bool smu_feature_bits_is_set(const struct smu_feature_bits *bits, + unsigned int bit) +{ + if (bit >= SMU_FEATURE_MAX) + return false; + + return test_bit(bit, bits->bits); +} + +static inline void smu_feature_bits_set_bit(struct smu_feature_bits *bits, + unsigned int bit) +{ + if (bit < SMU_FEATURE_MAX) + __set_bit(bit, bits->bits); +} + +static inline void smu_feature_bits_clear_bit(struct smu_feature_bits *bits, + unsigned int bit) +{ + if (bit < SMU_FEATURE_MAX) + __clear_bit(bit, bits->bits); +} + +static inline void smu_feature_bits_clearall(struct smu_feature_bits *bits) +{ + bitmap_zero(bits->bits, SMU_FEATURE_MAX); +} + +static inline void smu_feature_bits_fill(struct smu_feature_bits *bits) +{ + bitmap_fill(bits->bits, SMU_FEATURE_MAX); +} + +static inline bool +smu_feature_bits_test_mask(const struct smu_feature_bits *bits, + const unsigned long *mask) +{ + return bitmap_intersects(bits->bits, mask, SMU_FEATURE_MAX); +} + +static inline void smu_feature_bits_from_arr32(struct smu_feature_bits *bits, + const uint32_t *arr, + unsigned int nbits) +{ + bitmap_from_arr32(bits->bits, arr, nbits); +} + +static inline void +smu_feature_bits_to_arr32(const struct smu_feature_bits *bits, uint32_t *arr, + unsigned int nbits) +{ + bitmap_to_arr32(arr, bits->bits, nbits); +} + +static inline bool smu_feature_bits_empty(const struct smu_feature_bits *bits, + unsigned int nbits) +{ + return bitmap_empty(bits->bits, nbits); +} + +static inline void smu_feature_bits_copy(struct smu_feature_bits *dst, + const unsigned long *src, + unsigned int nbits) +{ + bitmap_copy(dst->bits, src, nbits); +} + +static inline struct smu_feature_bits * +__smu_feature_get_list(struct smu_context *smu, enum smu_feature_list list) +{ + if (unlikely(list >= SMU_FEATURE_LIST_MAX)) { + dev_warn(smu->adev->dev, "Invalid feature list: %d\n", list); + return &smu->smu_feature.bits[SMU_FEATURE_LIST_SUPPORTED]; + } + + return &smu->smu_feature.bits[list]; +} + +static inline bool smu_feature_list_is_set(struct smu_context *smu, + enum smu_feature_list list, + unsigned int bit) +{ + if (bit >= smu->smu_feature.feature_num) + return false; + + return smu_feature_bits_is_set(__smu_feature_get_list(smu, list), bit); +} + +static inline void smu_feature_list_set_bit(struct smu_context *smu, + enum smu_feature_list list, + unsigned int bit) +{ + if (bit >= smu->smu_feature.feature_num) + return; + + smu_feature_bits_set_bit(__smu_feature_get_list(smu, list), bit); +} + +static inline void smu_feature_list_clear_bit(struct smu_context *smu, + enum smu_feature_list list, + unsigned int bit) +{ + if (bit >= smu->smu_feature.feature_num) + return; + + smu_feature_bits_clear_bit(__smu_feature_get_list(smu, list), bit); +} + +static inline void smu_feature_list_set_all(struct smu_context *smu, + enum smu_feature_list list) +{ + smu_feature_bits_fill(__smu_feature_get_list(smu, list)); +} + +static inline void smu_feature_list_clear_all(struct smu_context *smu, + enum smu_feature_list list) +{ + smu_feature_bits_clearall(__smu_feature_get_list(smu, list)); +} + +static inline bool smu_feature_list_is_empty(struct smu_context *smu, + enum smu_feature_list list) +{ + return smu_feature_bits_empty(__smu_feature_get_list(smu, list), + smu->smu_feature.feature_num); +} + +static inline void smu_feature_list_set_bits(struct smu_context *smu, + enum smu_feature_list dst_list, + const unsigned long *src) +{ + smu_feature_bits_copy(__smu_feature_get_list(smu, dst_list), src, + smu->smu_feature.feature_num); +} + +static inline void smu_feature_list_to_arr32(struct smu_context *smu, + enum smu_feature_list list, + uint32_t *arr) +{ + smu_feature_bits_to_arr32(__smu_feature_get_list(smu, list), arr, + smu->smu_feature.feature_num); +} + +static inline void smu_feature_init(struct smu_context *smu, int feature_num) +{ + if (!feature_num || smu->smu_feature.feature_num != 0) + return; + + smu->smu_feature.feature_num = feature_num; + smu_feature_list_clear_all(smu, SMU_FEATURE_LIST_SUPPORTED); + smu_feature_list_clear_all(smu, SMU_FEATURE_LIST_ALLOWED); +} + #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index efeaa3d57712..b0d6b7b0946d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -55,6 +55,7 @@ #define SMUQ10_TO_UINT(x) ((x) >> 10) #define SMUQ10_FRAC(x) ((x) & 0x3ff) #define SMUQ10_ROUND(x) ((SMUQ10_TO_UINT(x)) + ((SMUQ10_FRAC(x)) >= 0x200)) +#define SMU_V13_SOFT_FREQ_ROUND(x) ((x) + 1) extern const int pmfw_decoded_link_speed[5]; extern const int pmfw_decoded_link_width[7]; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h index 613d4d36f32f..b453e6efc7c9 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h @@ -56,6 +56,7 @@ extern const int decoded_link_width[8]; #define DECODE_GEN_SPEED(gen_speed_idx) (decoded_link_speed[gen_speed_idx]) #define DECODE_LANE_WIDTH(lane_width_idx) (decoded_link_width[lane_width_idx]) +#define SMU_V14_SOFT_FREQ_ROUND(x) ((x) + 1) struct smu_14_0_max_sustainable_clocks { uint32_t display_clock; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 7c5ce6a6e2ca..b22a0e91826d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -345,14 +345,9 @@ static int arcturus_init_smc_tables(struct smu_context *smu) } static int -arcturus_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +arcturus_init_allowed_features(struct smu_context *smu) { - if (num > 2) - return -EINVAL; - - /* pptable will handle the features to enable */ - memset(feature_mask, 0xFF, sizeof(uint32_t) * num); + smu_feature_list_set_all(smu, SMU_FEATURE_LIST_ALLOWED); return 0; } @@ -1877,7 +1872,7 @@ static ssize_t arcturus_get_gpu_metrics(struct smu_context *smu, static const struct pptable_funcs arcturus_ppt_funcs = { /* init dpm */ - .get_allowed_feature_mask = arcturus_get_allowed_feature_mask, + .init_allowed_features = arcturus_init_allowed_features, /* btc */ .run_btc = arcturus_run_btc, /* dpm/clk tables */ diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 1f84654bbc85..f14eed052526 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -275,89 +275,83 @@ static bool is_asic_secure(struct smu_context *smu) } static int -navi10_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +navi10_init_allowed_features(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - if (num > 2) - return -EINVAL; - - memset(feature_mask, 0, sizeof(uint32_t) * num); - - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_PREFETCHER_BIT) - | FEATURE_MASK(FEATURE_DPM_MP0CLK_BIT) - | FEATURE_MASK(FEATURE_RSMU_SMN_CG_BIT) - | FEATURE_MASK(FEATURE_DS_SOCCLK_BIT) - | FEATURE_MASK(FEATURE_PPT_BIT) - | FEATURE_MASK(FEATURE_TDC_BIT) - | FEATURE_MASK(FEATURE_GFX_EDC_BIT) - | FEATURE_MASK(FEATURE_APCC_PLUS_BIT) - | FEATURE_MASK(FEATURE_VR0HOT_BIT) - | FEATURE_MASK(FEATURE_FAN_CONTROL_BIT) - | FEATURE_MASK(FEATURE_THERMAL_BIT) - | FEATURE_MASK(FEATURE_LED_DISPLAY_BIT) - | FEATURE_MASK(FEATURE_DS_LCLK_BIT) - | FEATURE_MASK(FEATURE_DS_DCEFCLK_BIT) - | FEATURE_MASK(FEATURE_FW_DSTATE_BIT) - | FEATURE_MASK(FEATURE_BACO_BIT) - | FEATURE_MASK(FEATURE_GFX_SS_BIT) - | FEATURE_MASK(FEATURE_APCC_DFLL_BIT) - | FEATURE_MASK(FEATURE_FW_CTF_BIT) - | FEATURE_MASK(FEATURE_OUT_OF_BAND_MONITOR_BIT) - | FEATURE_MASK(FEATURE_TEMP_DEPENDENT_VMIN_BIT); + smu_feature_list_clear_all(smu, SMU_FEATURE_LIST_ALLOWED); + + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_PREFETCHER_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_MP0CLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_RSMU_SMN_CG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_SOCCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_PPT_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_TDC_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_EDC_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_APCC_PLUS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VR0HOT_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FAN_CONTROL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_THERMAL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_LED_DISPLAY_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_LCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_DCEFCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_DSTATE_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_BACO_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_SS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_APCC_DFLL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_CTF_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_OUT_OF_BAND_MONITOR_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_TEMP_DEPENDENT_VMIN_BIT); if (adev->pm.pp_feature & PP_SCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_GFXCLK_BIT); if (adev->pm.pp_feature & PP_PCIE_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_LINK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_LINK_BIT); if (adev->pm.pp_feature & PP_DCEFCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_DCEFCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_DCEFCLK_BIT); if (adev->pm.pp_feature & PP_ULV_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_ULV_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_ULV_BIT); if (adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_GFXCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_GFXCLK_BIT); if (adev->pm.pp_feature & PP_GFXOFF_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFXOFF_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFXOFF_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_MMHUB) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MMHUB_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MMHUB_PG_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_ATHUB) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ATHUB_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_ATHUB_PG_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_VCN) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VCN_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VCN_PG_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_JPEG) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_JPEG_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_JPEG_PG_BIT); if (smu->dc_controlled_by_gpio) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ACDC_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_ACDC_BIT); if (adev->pm.pp_feature & PP_SOCCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_SOCCLK_BIT); - /* DPM UCLK enablement should be skipped for navi10 A0 secure board */ if (!(is_asic_secure(smu) && (amdgpu_ip_version(adev, MP1_HWIP, 0) == IP_VERSION(11, 0, 0)) && (adev->rev_id == 0)) && - (adev->pm.pp_feature & PP_MCLK_DPM_MASK)) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT) - | FEATURE_MASK(FEATURE_MEM_VDDCI_SCALING_BIT) - | FEATURE_MASK(FEATURE_MEM_MVDD_SCALING_BIT); + (adev->pm.pp_feature & PP_MCLK_DPM_MASK)) { + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_UCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MEM_VDDCI_SCALING_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MEM_MVDD_SCALING_BIT); + } - /* DS SOCCLK enablement should be skipped for navi10 A0 secure board */ if (is_asic_secure(smu) && (amdgpu_ip_version(adev, MP1_HWIP, 0) == IP_VERSION(11, 0, 0)) && (adev->rev_id == 0)) - *(uint64_t *)feature_mask &= - ~FEATURE_MASK(FEATURE_DS_SOCCLK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_SOCCLK_BIT); return 0; } @@ -3277,7 +3271,7 @@ static int navi10_set_config_table(struct smu_context *smu, } static const struct pptable_funcs navi10_ppt_funcs = { - .get_allowed_feature_mask = navi10_get_allowed_feature_mask, + .init_allowed_features = navi10_init_allowed_features, .set_default_dpm_table = navi10_set_default_dpm_table, .dpm_set_vcn_enable = navi10_dpm_set_vcn_enable, .dpm_set_jpeg_enable = navi10_dpm_set_jpeg_enable, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index f930ba2733e9..98a02fc08214 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -276,85 +276,82 @@ static const uint8_t sienna_cichlid_throttler_map[] = { }; static int -sienna_cichlid_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +sienna_cichlid_init_allowed_features(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - if (num > 2) - return -EINVAL; - - memset(feature_mask, 0, sizeof(uint32_t) * num); - - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_PREFETCHER_BIT) - | FEATURE_MASK(FEATURE_DPM_FCLK_BIT) - | FEATURE_MASK(FEATURE_DPM_MP0CLK_BIT) - | FEATURE_MASK(FEATURE_DS_SOCCLK_BIT) - | FEATURE_MASK(FEATURE_DS_DCEFCLK_BIT) - | FEATURE_MASK(FEATURE_DS_FCLK_BIT) - | FEATURE_MASK(FEATURE_DS_UCLK_BIT) - | FEATURE_MASK(FEATURE_FW_DSTATE_BIT) - | FEATURE_MASK(FEATURE_DF_CSTATE_BIT) - | FEATURE_MASK(FEATURE_RSMU_SMN_CG_BIT) - | FEATURE_MASK(FEATURE_GFX_SS_BIT) - | FEATURE_MASK(FEATURE_VR0HOT_BIT) - | FEATURE_MASK(FEATURE_PPT_BIT) - | FEATURE_MASK(FEATURE_TDC_BIT) - | FEATURE_MASK(FEATURE_BACO_BIT) - | FEATURE_MASK(FEATURE_APCC_DFLL_BIT) - | FEATURE_MASK(FEATURE_FW_CTF_BIT) - | FEATURE_MASK(FEATURE_FAN_CONTROL_BIT) - | FEATURE_MASK(FEATURE_THERMAL_BIT) - | FEATURE_MASK(FEATURE_OUT_OF_BAND_MONITOR_BIT); + smu_feature_list_clear_all(smu, SMU_FEATURE_LIST_ALLOWED); + + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_PREFETCHER_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_FCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_MP0CLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_SOCCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_DCEFCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_FCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_UCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_DSTATE_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DF_CSTATE_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_RSMU_SMN_CG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_SS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VR0HOT_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_PPT_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_TDC_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_BACO_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_APCC_DFLL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_CTF_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FAN_CONTROL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_THERMAL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_OUT_OF_BAND_MONITOR_BIT); if (adev->pm.pp_feature & PP_SCLK_DPM_MASK) { - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFX_GPO_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_GFXCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_GFX_GPO_BIT); } if ((adev->pm.pp_feature & PP_GFX_DCS_MASK) && (amdgpu_ip_version(adev, MP1_HWIP, 0) > IP_VERSION(11, 0, 7)) && !(adev->flags & AMD_IS_APU)) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_DCS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_DCS_BIT); - if (adev->pm.pp_feature & PP_MCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT) - | FEATURE_MASK(FEATURE_MEM_VDDCI_SCALING_BIT) - | FEATURE_MASK(FEATURE_MEM_MVDD_SCALING_BIT); + if (adev->pm.pp_feature & PP_MCLK_DPM_MASK) { + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_UCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MEM_VDDCI_SCALING_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MEM_MVDD_SCALING_BIT); + } if (adev->pm.pp_feature & PP_PCIE_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_LINK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_LINK_BIT); if (adev->pm.pp_feature & PP_DCEFCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_DCEFCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_DCEFCLK_BIT); if (adev->pm.pp_feature & PP_SOCCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_SOCCLK_BIT); if (adev->pm.pp_feature & PP_ULV_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_ULV_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_ULV_BIT); if (adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_GFXCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_GFXCLK_BIT); if (adev->pm.pp_feature & PP_GFXOFF_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFXOFF_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFXOFF_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_ATHUB) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ATHUB_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_ATHUB_PG_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_MMHUB) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MMHUB_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MMHUB_PG_BIT); if (smu->adev->pg_flags & AMD_PG_SUPPORT_VCN || smu->adev->pg_flags & AMD_PG_SUPPORT_JPEG) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MM_DPM_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MM_DPM_PG_BIT); if (smu->dc_controlled_by_gpio) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ACDC_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_ACDC_BIT); if (amdgpu_device_should_use_aspm(adev)) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_LCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_LCLK_BIT); return 0; } @@ -3085,7 +3082,7 @@ out: } static const struct pptable_funcs sienna_cichlid_ppt_funcs = { - .get_allowed_feature_mask = sienna_cichlid_get_allowed_feature_mask, + .init_allowed_features = sienna_cichlid_init_allowed_features, .set_default_dpm_table = sienna_cichlid_set_default_dpm_table, .dpm_set_vcn_enable = sienna_cichlid_dpm_set_vcn_enable, .dpm_set_jpeg_enable = sienna_cichlid_dpm_set_jpeg_enable, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c index eb1b9faf8e5c..1d0f9f8ddf9b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c @@ -750,12 +750,13 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu) int ret = 0; uint32_t feature_mask[2]; - if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || feature->feature_num < 64) { + if (smu_feature_list_is_empty(smu, SMU_FEATURE_LIST_ALLOWED) || + feature->feature_num < 64) { ret = -EINVAL; goto failed; } - bitmap_to_arr32(feature_mask, feature->allowed, 64); + smu_feature_list_to_arr32(smu, SMU_FEATURE_LIST_ALLOWED, feature_mask); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, feature_mask[1], NULL); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index 9de0b676bb7b..3b6a34644a92 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -329,14 +329,9 @@ static int aldebaran_init_smc_tables(struct smu_context *smu) return smu_v13_0_init_smc_tables(smu); } -static int aldebaran_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +static int aldebaran_init_allowed_features(struct smu_context *smu) { - if (num > 2) - return -EINVAL; - - /* pptable will handle the features to enable */ - memset(feature_mask, 0xFF, sizeof(uint32_t) * num); + smu_feature_list_set_all(smu, SMU_FEATURE_LIST_ALLOWED); return 0; } @@ -1967,7 +1962,7 @@ static int aldebaran_send_hbm_bad_channel_flag(struct smu_context *smu, static const struct pptable_funcs aldebaran_ppt_funcs = { /* init dpm */ - .get_allowed_feature_mask = aldebaran_get_allowed_feature_mask, + .init_allowed_features = aldebaran_init_allowed_features, /* dpm/clk tables */ .set_default_dpm_table = aldebaran_set_default_dpm_table, .populate_umd_state_clk = aldebaran_populate_umd_state_clk, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 51f96fdcec24..3b0aa6a2e78e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -761,11 +761,11 @@ int smu_v13_0_set_allowed_mask(struct smu_context *smu) int ret = 0; uint32_t feature_mask[2]; - if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || + if (smu_feature_list_is_empty(smu, SMU_FEATURE_LIST_ALLOWED) || feature->feature_num < 64) return -EINVAL; - bitmap_to_arr32(feature_mask, feature->allowed, 64); + smu_feature_list_to_arr32(smu, SMU_FEATURE_LIST_ALLOWED, feature_mask); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, feature_mask[1], NULL); @@ -1554,6 +1554,7 @@ int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, return clk_id; if (max > 0) { + max = SMU_V13_SOFT_FREQ_ROUND(max); if (automatic) param = (uint32_t)((clk_id << 16) | 0xffff); else diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index ce52b616b935..9c4298736b28 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -287,49 +287,44 @@ static const uint8_t smu_v13_0_0_throttler_map[] = { }; static int -smu_v13_0_0_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +smu_v13_0_0_init_allowed_features(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - if (num > 2) - return -EINVAL; - - memset(feature_mask, 0xff, sizeof(uint32_t) * num); + smu_feature_list_set_all(smu, SMU_FEATURE_LIST_ALLOWED); if (!(adev->pm.pp_feature & PP_SCLK_DPM_MASK)) { - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFX_IMU_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_GFXCLK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_IMU_BIT); } if (!(adev->pg_flags & AMD_PG_SUPPORT_ATHUB) || !(adev->pg_flags & AMD_PG_SUPPORT_MMHUB)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_ATHUB_MMHUB_PG_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_ATHUB_MMHUB_PG_BIT); if (!(adev->pm.pp_feature & PP_SOCCLK_DPM_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_SOCCLK_BIT); - /* PMFW 78.58 contains a critical fix for gfxoff feature */ if ((smu->smc_fw_version < 0x004e3a00) || !(adev->pm.pp_feature & PP_GFXOFF_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFXOFF_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFXOFF_BIT); if (!(adev->pm.pp_feature & PP_MCLK_DPM_MASK)) { - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_UCLK_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_VMEMP_SCALING_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_VDDIO_MEM_SCALING_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_UCLK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VMEMP_SCALING_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VDDIO_MEM_SCALING_BIT); } if (!(adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DS_GFXCLK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_GFXCLK_BIT); if (!(adev->pm.pp_feature & PP_PCIE_DPM_MASK)) { - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_LINK_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DS_LCLK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_LINK_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_LCLK_BIT); } if (!(adev->pm.pp_feature & PP_ULV_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFX_ULV_BIT); + smu_feature_list_clear_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_ULV_BIT); return 0; } @@ -3130,7 +3125,7 @@ static int smu_v13_0_0_update_pcie_parameters(struct smu_context *smu, } static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { - .get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask, + .init_allowed_features = smu_v13_0_0_init_allowed_features, .set_default_dpm_table = smu_v13_0_0_set_default_dpm_table, .i2c_init = smu_v13_0_0_i2c_control_init, .i2c_fini = smu_v13_0_0_i2c_control_fini, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index 1e82c43c851a..bd893e95515f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -742,15 +742,9 @@ static int smu_v13_0_6_fini_smc_tables(struct smu_context *smu) return smu_v13_0_fini_smc_tables(smu); } -static int smu_v13_0_6_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, - uint32_t num) +static int smu_v13_0_6_init_allowed_features(struct smu_context *smu) { - if (num > 2) - return -EINVAL; - - /* pptable will handle the features to enable */ - memset(feature_mask, 0xFF, sizeof(uint32_t) * num); + smu_feature_list_set_all(smu, SMU_FEATURE_LIST_ALLOWED); return 0; } @@ -3836,7 +3830,7 @@ static int smu_v13_0_6_get_ras_smu_drv(struct smu_context *smu, const struct ras static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { /* init dpm */ - .get_allowed_feature_mask = smu_v13_0_6_get_allowed_feature_mask, + .init_allowed_features = smu_v13_0_6_init_allowed_features, /* dpm/clk tables */ .set_default_dpm_table = smu_v13_0_6_set_default_dpm_table, .populate_umd_state_clk = smu_v13_0_6_populate_umd_state_clk, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index 0375e8484b2a..415766dbfe6c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -265,71 +265,67 @@ static const uint8_t smu_v13_0_7_throttler_map[] = { }; static int -smu_v13_0_7_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +smu_v13_0_7_init_allowed_features(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - if (num > 2) - return -EINVAL; - - memset(feature_mask, 0, sizeof(uint32_t) * num); + smu_feature_list_clear_all(smu, SMU_FEATURE_LIST_ALLOWED); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FW_DATA_READ_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_DATA_READ_BIT); if (adev->pm.pp_feature & PP_SCLK_DPM_MASK) { - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_IMU_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFX_POWER_OPTIMIZER_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_GFXCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_IMU_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_GFX_POWER_OPTIMIZER_BIT); } if (adev->pm.pp_feature & PP_GFXOFF_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFXOFF_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFXOFF_BIT); if (adev->pm.pp_feature & PP_MCLK_DPM_MASK) { - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_FCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VMEMP_SCALING_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VDDIO_MEM_SCALING_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_UCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_FCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VMEMP_SCALING_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VDDIO_MEM_SCALING_BIT); } - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_SOCCLK_BIT); if (adev->pm.pp_feature & PP_PCIE_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_LINK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_LINK_BIT); if (adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_GFXCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_GFXCLK_BIT); if (adev->pm.pp_feature & PP_ULV_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_ULV_BIT); - - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_LCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_MP0CLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MM_DPM_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_VCN_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_FCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DF_CSTATE_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_THROTTLERS_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VR0HOT_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FW_CTF_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FAN_CONTROL_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_SOCCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFXCLK_SPREAD_SPECTRUM_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MEM_TEMP_READ_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FW_DSTATE_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_SOC_MPCLK_DS_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_BACO_MPCLK_DS_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_PCC_DFLL_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_SOC_CG_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_BACO_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_ULV_BIT); + + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_LCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_MP0CLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MM_DPM_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_VCN_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_FCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DF_CSTATE_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_THROTTLERS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_VR0HOT_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_CTF_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FAN_CONTROL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DS_SOCCLK_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFXCLK_SPREAD_SPECTRUM_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_MEM_TEMP_READ_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_FW_DSTATE_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_SOC_MPCLK_DS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_BACO_MPCLK_DS_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_GFX_PCC_DFLL_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_SOC_CG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_BACO_BIT); if (adev->pm.pp_feature & PP_DCEFCLK_DPM_MASK) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_DCN_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_DPM_DCN_BIT); if ((adev->pg_flags & AMD_PG_SUPPORT_ATHUB) && (adev->pg_flags & AMD_PG_SUPPORT_MMHUB)) - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ATHUB_MMHUB_PG_BIT); + smu_feature_list_set_bit(smu, SMU_FEATURE_LIST_ALLOWED, FEATURE_ATHUB_MMHUB_PG_BIT); return 0; } @@ -2736,7 +2732,7 @@ static int smu_v13_0_7_update_pcie_parameters(struct smu_context *smu, } static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { - .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, + .init_allowed_features = smu_v13_0_7_init_allowed_features, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, .is_dpm_running = smu_v13_0_7_is_dpm_running, .init_microcode = smu_v13_0_init_microcode, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c index f85ba23f9d99..cabbd234c6e2 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c @@ -746,11 +746,11 @@ int smu_v14_0_set_allowed_mask(struct smu_context *smu) int ret = 0; uint32_t feature_mask[2]; - if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || + if (smu_feature_list_is_empty(smu, SMU_FEATURE_LIST_ALLOWED) || feature->feature_num < 64) return -EINVAL; - bitmap_to_arr32(feature_mask, feature->allowed, 64); + smu_feature_list_to_arr32(smu, SMU_FEATURE_LIST_ALLOWED, feature_mask); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, feature_mask[1], NULL); @@ -1177,6 +1177,7 @@ int smu_v14_0_set_soft_freq_limited_range(struct smu_context *smu, return clk_id; if (max > 0) { + max = SMU_V14_SOFT_FREQ_ROUND(max); if (automatic) param = (uint32_t)((clk_id << 16) | 0xffff); else diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index faae1da81bd4..3c351ee41e68 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -264,53 +264,9 @@ static const uint8_t smu_v14_0_2_throttler_map[] = { [THROTTLER_FIT_BIT] = (SMU_THROTTLER_FIT_BIT), }; -static int -smu_v14_0_2_get_allowed_feature_mask(struct smu_context *smu, - uint32_t *feature_mask, uint32_t num) +static int smu_v14_0_2_init_allowed_features(struct smu_context *smu) { - struct amdgpu_device *adev = smu->adev; - /*u32 smu_version;*/ - - if (num > 2) - return -EINVAL; - - memset(feature_mask, 0xff, sizeof(uint32_t) * num); - - if (adev->pm.pp_feature & PP_SCLK_DPM_MASK) { - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT); - *(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_IMU_BIT); - } -#if 0 - if (!(adev->pg_flags & AMD_PG_SUPPORT_ATHUB) || - !(adev->pg_flags & AMD_PG_SUPPORT_MMHUB)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_ATHUB_MMHUB_PG_BIT); - - if (!(adev->pm.pp_feature & PP_SOCCLK_DPM_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT); - - /* PMFW 78.58 contains a critical fix for gfxoff feature */ - smu_cmn_get_smc_version(smu, NULL, &smu_version); - if ((smu_version < 0x004e3a00) || - !(adev->pm.pp_feature & PP_GFXOFF_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFXOFF_BIT); - - if (!(adev->pm.pp_feature & PP_MCLK_DPM_MASK)) { - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_UCLK_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_VMEMP_SCALING_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_VDDIO_MEM_SCALING_BIT); - } - - if (!(adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DS_GFXCLK_BIT); - - if (!(adev->pm.pp_feature & PP_PCIE_DPM_MASK)) { - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_LINK_BIT); - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DS_LCLK_BIT); - } - - if (!(adev->pm.pp_feature & PP_ULV_MASK)) - *(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFX_ULV_BIT); -#endif + smu_feature_list_set_all(smu, SMU_FEATURE_LIST_ALLOWED); return 0; } @@ -2796,7 +2752,7 @@ static int smu_v14_0_2_set_power_limit(struct smu_context *smu, } static const struct pptable_funcs smu_v14_0_2_ppt_funcs = { - .get_allowed_feature_mask = smu_v14_0_2_get_allowed_feature_mask, + .init_allowed_features = smu_v14_0_2_init_allowed_features, .set_default_dpm_table = smu_v14_0_2_set_default_dpm_table, .i2c_init = smu_v14_0_2_i2c_control_init, .i2c_fini = smu_v14_0_2_i2c_control_fini, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu15/smu_v15_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu15/smu_v15_0.c index 6557085a7c72..d77eaac556d9 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu15/smu_v15_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu15/smu_v15_0.c @@ -715,11 +715,11 @@ int smu_v15_0_set_allowed_mask(struct smu_context *smu) int ret = 0; uint32_t feature_mask[2]; - if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || + if (smu_feature_list_is_empty(smu, SMU_FEATURE_LIST_ALLOWED) || feature->feature_num < 64) return -EINVAL; - bitmap_to_arr32(feature_mask, feature->allowed, 64); + smu_feature_list_to_arr32(smu, SMU_FEATURE_LIST_ALLOWED, feature_mask); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, feature_mask[1], NULL); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 24835600c1cd..9bb7e3760c0f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -677,7 +677,6 @@ int smu_cmn_to_asic_specific_index(struct smu_context *smu, int smu_cmn_feature_is_supported(struct smu_context *smu, enum smu_feature_mask mask) { - struct smu_feature *feature = &smu->smu_feature; int feature_id; feature_id = smu_cmn_to_asic_specific_index(smu, @@ -686,9 +685,8 @@ int smu_cmn_feature_is_supported(struct smu_context *smu, if (feature_id < 0) return 0; - WARN_ON(feature_id > feature->feature_num); - - return test_bit(feature_id, feature->supported); + return smu_feature_list_is_set(smu, SMU_FEATURE_LIST_SUPPORTED, + feature_id); } static int __smu_get_enabled_features(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h index 0f7778410a3a..24848da90234 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h @@ -70,7 +70,7 @@ #define smu_apply_clocks_adjust_rules(smu) smu_ppt_funcs(apply_clocks_adjust_rules, 0, smu) #define smu_notify_smc_display_config(smu) smu_ppt_funcs(notify_smc_display_config, 0, smu) #define smu_run_btc(smu) smu_ppt_funcs(run_btc, 0, smu) -#define smu_get_allowed_feature_mask(smu, feature_mask, num) smu_ppt_funcs(get_allowed_feature_mask, 0, smu, feature_mask, num) +#define smu_init_allowed_features(smu) smu_ppt_funcs(init_allowed_features, 0, smu) #define smu_set_watermarks_table(smu, clock_ranges) smu_ppt_funcs(set_watermarks_table, 0, smu, clock_ranges) #define smu_thermal_temperature_range_update(smu, range, rw) smu_ppt_funcs(thermal_temperature_range_update, 0, smu, range, rw) #define smu_register_irq_handler(smu) smu_ppt_funcs(register_irq_handler, 0, smu) diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h index 047bcb1cc078..e72359370857 100644 --- a/include/uapi/linux/kfd_ioctl.h +++ b/include/uapi/linux/kfd_ioctl.h @@ -149,6 +149,8 @@ struct kfd_dbg_device_info_entry { __u32 num_xcc; __u32 capability; __u32 debug_prop; + __u32 capability2; + __u32 pad; }; /* For kfd_ioctl_set_memory_policy_args.default_policy and alternate_policy */ |