diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu')
20 files changed, 265 insertions, 100 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 5afaf6016b4a..0b14b5373783 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -717,7 +717,7 @@ int amdgpu_queue_mgr_fini(struct amdgpu_device *adev, struct amdgpu_queue_mgr *mgr); int amdgpu_queue_mgr_map(struct amdgpu_device *adev, struct amdgpu_queue_mgr *mgr, - int hw_ip, int instance, int ring, + u32 hw_ip, u32 instance, u32 ring, struct amdgpu_ring **out_ring); /* @@ -1572,18 +1572,14 @@ struct amdgpu_device { /* sdma */ struct amdgpu_sdma sdma; - union { - struct { - /* uvd */ - struct amdgpu_uvd uvd; + /* uvd */ + struct amdgpu_uvd uvd; - /* vce */ - struct amdgpu_vce vce; - }; + /* vce */ + struct amdgpu_vce vce; - /* vcn */ - struct amdgpu_vcn vcn; - }; + /* vcn */ + struct amdgpu_vcn vcn; /* firmwares */ struct amdgpu_firmware firmware; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index 47d1c132ac40..1e3e9be7d77e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -379,29 +379,50 @@ static int kgd_hqd_sdma_load(struct kgd_dev *kgd, void *mqd) { struct amdgpu_device *adev = get_amdgpu_device(kgd); struct cik_sdma_rlc_registers *m; + unsigned long end_jiffies; uint32_t sdma_base_addr; + uint32_t data; m = get_sdma_mqd(mqd); sdma_base_addr = get_sdma_base_addr(m); - WREG32(sdma_base_addr + mmSDMA0_RLC0_VIRTUAL_ADDR, - m->sdma_rlc_virtual_addr); + WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, + m->sdma_rlc_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, - m->sdma_rlc_rb_base); + end_jiffies = msecs_to_jiffies(2000) + jiffies; + while (true) { + data = RREG32(sdma_base_addr + mmSDMA0_RLC0_CONTEXT_STATUS); + if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK) + break; + if (time_after(jiffies, end_jiffies)) + return -ETIME; + usleep_range(500, 1000); + } + if (m->sdma_engine_id) { + data = RREG32(mmSDMA1_GFX_CONTEXT_CNTL); + data = REG_SET_FIELD(data, SDMA1_GFX_CONTEXT_CNTL, + RESUME_CTX, 0); + WREG32(mmSDMA1_GFX_CONTEXT_CNTL, data); + } else { + data = RREG32(mmSDMA0_GFX_CONTEXT_CNTL); + data = REG_SET_FIELD(data, SDMA0_GFX_CONTEXT_CNTL, + RESUME_CTX, 0); + WREG32(mmSDMA0_GFX_CONTEXT_CNTL, data); + } + WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, + m->sdma_rlc_doorbell); + WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, 0); + WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR, 0); + WREG32(sdma_base_addr + mmSDMA0_RLC0_VIRTUAL_ADDR, + m->sdma_rlc_virtual_addr); + WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, m->sdma_rlc_rb_base); WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE_HI, m->sdma_rlc_rb_base_hi); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_LO, m->sdma_rlc_rb_rptr_addr_lo); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR_ADDR_HI, m->sdma_rlc_rb_rptr_addr_hi); - - WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, - m->sdma_rlc_doorbell); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, m->sdma_rlc_rb_cntl); @@ -574,9 +595,9 @@ static int kgd_hqd_sdma_destroy(struct kgd_dev *kgd, void *mqd, } WREG32(sdma_base_addr + mmSDMA0_RLC0_DOORBELL, 0); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_RPTR, 0); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_WPTR, 0); - WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_BASE, 0); + WREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL, + RREG32(sdma_base_addr + mmSDMA0_RLC0_RB_CNTL) | + SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c index c21adf60a7f2..057e1ecd83ce 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c @@ -59,12 +59,6 @@ static bool check_atom_bios(uint8_t *bios, size_t size) return false; } - tmp = bios[0x18] | (bios[0x19] << 8); - if (bios[tmp + 0x14] != 0x0) { - DRM_INFO("Not an x86 BIOS ROM\n"); - return false; - } - bios_header_start = bios[0x48] | (bios[0x49] << 8); if (!bios_header_start) { DRM_INFO("Can't locate bios header\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 6c78623e1386..57abf7abd7a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -409,6 +409,10 @@ static bool amdgpu_cs_try_evict(struct amdgpu_cs_parser *p, if (candidate->robj == validated) break; + /* We can't move pinned BOs here */ + if (bo->pin_count) + continue; + other = amdgpu_mem_type_to_domain(bo->tbo.mem.mem_type); /* Check if this BO is in one of the domains we need space for */ @@ -1495,8 +1499,11 @@ out: memset(wait, 0, sizeof(*wait)); wait->out.status = (r > 0); wait->out.first_signaled = first; - /* set return value 0 to indicate success */ - r = array[first]->error; + + if (first < fence_count && array[first]) + r = array[first]->error; + else + r = 0; err_free_fence_array: for (i = 0; i < fence_count; i++) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 2d792cdc094c..3573ecdb06ee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3261,9 +3261,9 @@ static ssize_t amdgpu_debugfs_regs_read(struct file *f, char __user *buf, pm_pg_lock = (*pos >> 23) & 1; if (*pos & (1ULL << 62)) { - se_bank = (*pos >> 24) & 0x3FF; - sh_bank = (*pos >> 34) & 0x3FF; - instance_bank = (*pos >> 44) & 0x3FF; + se_bank = (*pos & GENMASK_ULL(33, 24)) >> 24; + sh_bank = (*pos & GENMASK_ULL(43, 34)) >> 34; + instance_bank = (*pos & GENMASK_ULL(53, 44)) >> 44; if (se_bank == 0x3FF) se_bank = 0xFFFFFFFF; @@ -3337,9 +3337,9 @@ static ssize_t amdgpu_debugfs_regs_write(struct file *f, const char __user *buf, pm_pg_lock = (*pos >> 23) & 1; if (*pos & (1ULL << 62)) { - se_bank = (*pos >> 24) & 0x3FF; - sh_bank = (*pos >> 34) & 0x3FF; - instance_bank = (*pos >> 44) & 0x3FF; + se_bank = (*pos & GENMASK_ULL(33, 24)) >> 24; + sh_bank = (*pos & GENMASK_ULL(43, 34)) >> 34; + instance_bank = (*pos & GENMASK_ULL(53, 44)) >> 44; if (se_bank == 0x3FF) se_bank = 0xFFFFFFFF; @@ -3687,12 +3687,12 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf, return -EINVAL; /* decode offset */ - offset = (*pos & 0x7F); - se = ((*pos >> 7) & 0xFF); - sh = ((*pos >> 15) & 0xFF); - cu = ((*pos >> 23) & 0xFF); - wave = ((*pos >> 31) & 0xFF); - simd = ((*pos >> 37) & 0xFF); + offset = (*pos & GENMASK_ULL(6, 0)); + se = (*pos & GENMASK_ULL(14, 7)) >> 7; + sh = (*pos & GENMASK_ULL(22, 15)) >> 15; + cu = (*pos & GENMASK_ULL(30, 23)) >> 23; + wave = (*pos & GENMASK_ULL(36, 31)) >> 31; + simd = (*pos & GENMASK_ULL(44, 37)) >> 37; /* switch to the specific se/sh/cu */ mutex_lock(&adev->grbm_idx_mutex); @@ -3737,14 +3737,14 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, return -EINVAL; /* decode offset */ - offset = (*pos & 0xFFF); /* in dwords */ - se = ((*pos >> 12) & 0xFF); - sh = ((*pos >> 20) & 0xFF); - cu = ((*pos >> 28) & 0xFF); - wave = ((*pos >> 36) & 0xFF); - simd = ((*pos >> 44) & 0xFF); - thread = ((*pos >> 52) & 0xFF); - bank = ((*pos >> 60) & 1); + offset = *pos & GENMASK_ULL(11, 0); + se = (*pos & GENMASK_ULL(19, 12)) >> 12; + sh = (*pos & GENMASK_ULL(27, 20)) >> 20; + cu = (*pos & GENMASK_ULL(35, 28)) >> 28; + wave = (*pos & GENMASK_ULL(43, 36)) >> 36; + simd = (*pos & GENMASK_ULL(51, 44)) >> 44; + thread = (*pos & GENMASK_ULL(59, 52)) >> 52; + bank = (*pos & GENMASK_ULL(61, 60)) >> 60; data = kmalloc_array(1024, sizeof(*data), GFP_KERNEL); if (!data) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index ec96bb1f9eaf..c2f414ffb2cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -536,7 +536,7 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x686c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x687f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, /* Raven */ - {0x1002, 0x15dd, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RAVEN|AMD_IS_APU|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x15dd, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RAVEN|AMD_IS_APU}, {0, 0, 0} }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index bd5b8065c32e..2fa95aef74d5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -268,9 +268,10 @@ void amdgpu_fence_process(struct amdgpu_ring *ring) * * Checks for fence activity. */ -static void amdgpu_fence_fallback(unsigned long arg) +static void amdgpu_fence_fallback(struct timer_list *t) { - struct amdgpu_ring *ring = (void *)arg; + struct amdgpu_ring *ring = from_timer(ring, t, + fence_drv.fallback_timer); amdgpu_fence_process(ring); } @@ -422,8 +423,7 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring, atomic_set(&ring->fence_drv.last_seq, 0); ring->fence_drv.initialized = false; - setup_timer(&ring->fence_drv.fallback_timer, amdgpu_fence_fallback, - (unsigned long)ring); + timer_setup(&ring->fence_drv.fallback_timer, amdgpu_fence_fallback, 0); ring->fence_drv.num_fences_mask = num_hw_submission * 2 - 1; spin_lock_init(&ring->fence_drv.lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index a418df1b9422..e87eedcc0da9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -63,6 +63,11 @@ retry: flags, NULL, resv, 0, &bo); if (r) { if (r != -ERESTARTSYS) { + if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) { + flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED; + goto retry; + } + if (initial_domain == AMDGPU_GEM_DOMAIN_VRAM) { initial_domain |= AMDGPU_GEM_DOMAIN_GTT; goto retry; @@ -323,7 +328,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, r = amdgpu_ttm_tt_get_user_pages(bo->tbo.ttm, bo->tbo.ttm->pages); if (r) - goto unlock_mmap_sem; + goto release_object; r = amdgpu_bo_reserve(bo, true); if (r) @@ -348,9 +353,6 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, free_pages: release_pages(bo->tbo.ttm->pages, bo->tbo.ttm->num_pages); -unlock_mmap_sem: - up_read(¤t->mm->mmap_sem); - release_object: drm_gem_object_put_unlocked(gobj); @@ -556,9 +558,8 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, if (args->va_address < AMDGPU_VA_RESERVED_SIZE) { dev_err(&dev->pdev->dev, - "va_address 0x%lX is in reserved area 0x%X\n", - (unsigned long)args->va_address, - AMDGPU_VA_RESERVED_SIZE); + "va_address 0x%LX is in reserved area 0x%LX\n", + args->va_address, AMDGPU_VA_RESERVED_SIZE); return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c index 33535d347734..00e0ce10862f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c @@ -71,12 +71,6 @@ static int amdgpu_gtt_mgr_fini(struct ttm_mem_type_manager *man) { struct amdgpu_gtt_mgr *mgr = man->priv; - spin_lock(&mgr->lock); - if (!drm_mm_clean(&mgr->mm)) { - spin_unlock(&mgr->lock); - return -EBUSY; - } - drm_mm_takedown(&mgr->mm); spin_unlock(&mgr->lock); kfree(mgr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index d6df5728df7f..6c570d4e4516 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -946,6 +946,10 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, struct amdgpu_device *adev = dev_get_drvdata(dev); umode_t effective_mode = attr->mode; + /* no skipping for powerplay */ + if (adev->powerplay.cgs_device) + return effective_mode; + /* Skip limit attributes if DPM is not enabled */ if (!adev->pm.dpm_enabled && (attr == &sensor_dev_attr_temp1_crit.dev_attr.attr || diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c index 90af8e82b16a..ae9c106979d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c @@ -169,10 +169,14 @@ struct dma_buf *amdgpu_gem_prime_export(struct drm_device *dev, int flags) { struct amdgpu_bo *bo = gem_to_amdgpu_bo(gobj); + struct dma_buf *buf; if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm) || bo->flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID) return ERR_PTR(-EPERM); - return drm_gem_prime_export(dev, gobj, flags); + buf = drm_gem_prime_export(dev, gobj, flags); + if (!IS_ERR(buf)) + buf->file->f_mapping = dev->anon_inode->i_mapping; + return buf; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c index 190e28cb827e..93d86619e802 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c @@ -63,7 +63,7 @@ static int amdgpu_update_cached_map(struct amdgpu_queue_mapper *mapper, static int amdgpu_identity_map(struct amdgpu_device *adev, struct amdgpu_queue_mapper *mapper, - int ring, + u32 ring, struct amdgpu_ring **out_ring) { switch (mapper->hw_ip) { @@ -121,7 +121,7 @@ static enum amdgpu_ring_type amdgpu_hw_ip_to_ring_type(int hw_ip) static int amdgpu_lru_map(struct amdgpu_device *adev, struct amdgpu_queue_mapper *mapper, - int user_ring, bool lru_pipe_order, + u32 user_ring, bool lru_pipe_order, struct amdgpu_ring **out_ring) { int r, i, j; @@ -208,7 +208,7 @@ int amdgpu_queue_mgr_fini(struct amdgpu_device *adev, */ int amdgpu_queue_mgr_map(struct amdgpu_device *adev, struct amdgpu_queue_mgr *mgr, - int hw_ip, int instance, int ring, + u32 hw_ip, u32 instance, u32 ring, struct amdgpu_ring **out_ring) { int r, ip_num_rings; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index aa914256b4bc..bae77353447b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -94,7 +94,8 @@ struct amdgpu_bo_list_entry; #define AMDGPU_MMHUB 1 /* hardcode that limit for now */ -#define AMDGPU_VA_RESERVED_SIZE (8 << 20) +#define AMDGPU_VA_RESERVED_SIZE (8ULL << 20) + /* max vmids dedicated for process */ #define AMDGPU_VM_MAX_RESERVED_VMID 1 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 26e900627971..4acca92f6a52 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -68,11 +68,6 @@ static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man) struct amdgpu_vram_mgr *mgr = man->priv; spin_lock(&mgr->lock); - if (!drm_mm_clean(&mgr->mm)) { - spin_unlock(&mgr->lock); - return -EBUSY; - } - drm_mm_takedown(&mgr->mm); spin_unlock(&mgr->lock); kfree(mgr); diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index 793b1470284d..a296f7bbe57c 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1023,22 +1023,101 @@ static const struct amdgpu_allowed_register_entry cik_allowed_read_registers[] = {mmPA_SC_RASTER_CONFIG_1, true}, }; -static uint32_t cik_read_indexed_register(struct amdgpu_device *adev, - u32 se_num, u32 sh_num, - u32 reg_offset) + +static uint32_t cik_get_register_value(struct amdgpu_device *adev, + bool indexed, u32 se_num, + u32 sh_num, u32 reg_offset) { - uint32_t val; + if (indexed) { + uint32_t val; + unsigned se_idx = (se_num == 0xffffffff) ? 0 : se_num; + unsigned sh_idx = (sh_num == 0xffffffff) ? 0 : sh_num; + + switch (reg_offset) { + case mmCC_RB_BACKEND_DISABLE: + return adev->gfx.config.rb_config[se_idx][sh_idx].rb_backend_disable; + case mmGC_USER_RB_BACKEND_DISABLE: + return adev->gfx.config.rb_config[se_idx][sh_idx].user_rb_backend_disable; + case mmPA_SC_RASTER_CONFIG: + return adev->gfx.config.rb_config[se_idx][sh_idx].raster_config; + case mmPA_SC_RASTER_CONFIG_1: + return adev->gfx.config.rb_config[se_idx][sh_idx].raster_config_1; + } - mutex_lock(&adev->grbm_idx_mutex); - if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + mutex_lock(&adev->grbm_idx_mutex); + if (se_num != 0xffffffff || sh_num != 0xffffffff) + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); - val = RREG32(reg_offset); + val = RREG32(reg_offset); - if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); - mutex_unlock(&adev->grbm_idx_mutex); - return val; + if (se_num != 0xffffffff || sh_num != 0xffffffff) + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + mutex_unlock(&adev->grbm_idx_mutex); + return val; + } else { + unsigned idx; + + switch (reg_offset) { + case mmGB_ADDR_CONFIG: + return adev->gfx.config.gb_addr_config; + case mmMC_ARB_RAMCFG: + return adev->gfx.config.mc_arb_ramcfg; + case mmGB_TILE_MODE0: + case mmGB_TILE_MODE1: + case mmGB_TILE_MODE2: + case mmGB_TILE_MODE3: + case mmGB_TILE_MODE4: + case mmGB_TILE_MODE5: + case mmGB_TILE_MODE6: + case mmGB_TILE_MODE7: + case mmGB_TILE_MODE8: + case mmGB_TILE_MODE9: + case mmGB_TILE_MODE10: + case mmGB_TILE_MODE11: + case mmGB_TILE_MODE12: + case mmGB_TILE_MODE13: + case mmGB_TILE_MODE14: + case mmGB_TILE_MODE15: + case mmGB_TILE_MODE16: + case mmGB_TILE_MODE17: + case mmGB_TILE_MODE18: + case mmGB_TILE_MODE19: + case mmGB_TILE_MODE20: + case mmGB_TILE_MODE21: + case mmGB_TILE_MODE22: + case mmGB_TILE_MODE23: + case mmGB_TILE_MODE24: + case mmGB_TILE_MODE25: + case mmGB_TILE_MODE26: + case mmGB_TILE_MODE27: + case mmGB_TILE_MODE28: + case mmGB_TILE_MODE29: + case mmGB_TILE_MODE30: + case mmGB_TILE_MODE31: + idx = (reg_offset - mmGB_TILE_MODE0); + return adev->gfx.config.tile_mode_array[idx]; + case mmGB_MACROTILE_MODE0: + case mmGB_MACROTILE_MODE1: + case mmGB_MACROTILE_MODE2: + case mmGB_MACROTILE_MODE3: + case mmGB_MACROTILE_MODE4: + case mmGB_MACROTILE_MODE5: + case mmGB_MACROTILE_MODE6: + case mmGB_MACROTILE_MODE7: + case mmGB_MACROTILE_MODE8: + case mmGB_MACROTILE_MODE9: + case mmGB_MACROTILE_MODE10: + case mmGB_MACROTILE_MODE11: + case mmGB_MACROTILE_MODE12: + case mmGB_MACROTILE_MODE13: + case mmGB_MACROTILE_MODE14: + case mmGB_MACROTILE_MODE15: + idx = (reg_offset - mmGB_MACROTILE_MODE0); + return adev->gfx.config.macrotile_mode_array[idx]; + default: + return RREG32(reg_offset); + } + } } static int cik_read_register(struct amdgpu_device *adev, u32 se_num, @@ -1048,13 +1127,13 @@ static int cik_read_register(struct amdgpu_device *adev, u32 se_num, *value = 0; for (i = 0; i < ARRAY_SIZE(cik_allowed_read_registers); i++) { + bool indexed = cik_allowed_read_registers[i].grbm_indexed; + if (reg_offset != cik_allowed_read_registers[i].reg_offset) continue; - *value = cik_allowed_read_registers[i].grbm_indexed ? - cik_read_indexed_register(adev, se_num, - sh_num, reg_offset) : - RREG32(reg_offset); + *value = cik_get_register_value(adev, indexed, se_num, sh_num, + reg_offset); return 0; } return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 00868764a0dd..419ba0ce7ee5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -1819,6 +1819,22 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev) adev->gfx.config.backend_enable_mask, num_rb_pipes); } + + /* cache the values for userspace */ + for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { + for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { + gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff); + adev->gfx.config.rb_config[i][j].rb_backend_disable = + RREG32(mmCC_RB_BACKEND_DISABLE); + adev->gfx.config.rb_config[i][j].user_rb_backend_disable = + RREG32(mmGC_USER_RB_BACKEND_DISABLE); + adev->gfx.config.rb_config[i][j].raster_config = + RREG32(mmPA_SC_RASTER_CONFIG); + adev->gfx.config.rb_config[i][j].raster_config_1 = + RREG32(mmPA_SC_RASTER_CONFIG_1); + } + } + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); mutex_unlock(&adev->grbm_idx_mutex); } @@ -4670,6 +4686,14 @@ static int gfx_v7_0_sw_fini(void *handle) gfx_v7_0_cp_compute_fini(adev); gfx_v7_0_rlc_fini(adev); gfx_v7_0_mec_fini(adev); + amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, + &adev->gfx.rlc.clear_state_gpu_addr, + (void **)&adev->gfx.rlc.cs_ptr); + if (adev->gfx.rlc.cp_table_size) { + amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, + &adev->gfx.rlc.cp_table_gpu_addr, + (void **)&adev->gfx.rlc.cp_table_ptr); + } gfx_v7_0_free_microcode(adev); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index b8002ac3e536..9ecdf621a74a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -2118,6 +2118,15 @@ static int gfx_v8_0_sw_fini(void *handle) gfx_v8_0_mec_fini(adev); gfx_v8_0_rlc_fini(adev); + amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, + &adev->gfx.rlc.clear_state_gpu_addr, + (void **)&adev->gfx.rlc.cs_ptr); + if ((adev->asic_type == CHIP_CARRIZO) || + (adev->asic_type == CHIP_STONEY)) { + amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, + &adev->gfx.rlc.cp_table_gpu_addr, + (void **)&adev->gfx.rlc.cp_table_ptr); + } gfx_v8_0_free_microcode(adev); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 7f15bb2c5233..da43813d67a4 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -207,6 +207,12 @@ static const u32 golden_settings_gc_9_1_rv1[] = SOC15_REG_OFFSET(GC, 0, mmTD_CNTL), 0x01bd9f33, 0x00000800 }; +static const u32 golden_settings_gc_9_x_common[] = +{ + SOC15_REG_OFFSET(GC, 0, mmGRBM_CAM_INDEX), 0xffffffff, 0x00000000, + SOC15_REG_OFFSET(GC, 0, mmGRBM_CAM_DATA), 0xffffffff, 0x2544c382 +}; + #define VEGA10_GB_ADDR_CONFIG_GOLDEN 0x2a114042 #define RAVEN_GB_ADDR_CONFIG_GOLDEN 0x24000042 @@ -242,6 +248,9 @@ static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev) default: break; } + + amdgpu_program_register_sequence(adev, golden_settings_gc_9_x_common, + (const u32)ARRAY_SIZE(golden_settings_gc_9_x_common)); } static void gfx_v9_0_scratch_init(struct amdgpu_device *adev) @@ -988,12 +997,22 @@ static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, start + SQIND_WAVE_SGPRS_OFFSET, size, dst); } +static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, + uint32_t wave, uint32_t thread, + uint32_t start, uint32_t size, + uint32_t *dst) +{ + wave_read_regs( + adev, simd, wave, thread, + start + SQIND_WAVE_VGPRS_OFFSET, size, dst); +} static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { .get_gpu_clock_counter = &gfx_v9_0_get_gpu_clock_counter, .select_se_sh = &gfx_v9_0_select_se_sh, .read_wave_data = &gfx_v9_0_read_wave_data, .read_wave_sgprs = &gfx_v9_0_read_wave_sgprs, + .read_wave_vgprs = &gfx_v9_0_read_wave_vgprs, }; static void gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) @@ -1449,6 +1468,14 @@ static int gfx_v9_0_sw_fini(void *handle) gfx_v9_0_mec_fini(adev); gfx_v9_0_ngg_fini(adev); + amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, + &adev->gfx.rlc.clear_state_gpu_addr, + (void **)&adev->gfx.rlc.cs_ptr); + if (adev->asic_type == CHIP_RAVEN) { + amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, + &adev->gfx.rlc.cp_table_gpu_addr, + (void **)&adev->gfx.rlc.cp_table_ptr); + } gfx_v9_0_free_microcode(adev); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 621699331e09..c8f1aebeac7a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -392,7 +392,16 @@ static int gmc_v9_0_early_init(void *handle) static int gmc_v9_0_late_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - unsigned vm_inv_eng[AMDGPU_MAX_VMHUBS] = { 3, 3 }; + /* + * The latest engine allocation on gfx9 is: + * Engine 0, 1: idle + * Engine 2, 3: firmware + * Engine 4~13: amdgpu ring, subject to change when ring number changes + * Engine 14~15: idle + * Engine 16: kfd tlb invalidation + * Engine 17: Gart flushes + */ + unsigned vm_inv_eng[AMDGPU_MAX_VMHUBS] = { 4, 4 }; unsigned i; for(i = 0; i < adev->num_rings; ++i) { @@ -405,9 +414,9 @@ static int gmc_v9_0_late_init(void *handle) ring->funcs->vmhub); } - /* Engine 17 is used for GART flushes */ + /* Engine 16 is used for KFD and 17 for GART flushes */ for(i = 0; i < AMDGPU_MAX_VMHUBS; ++i) - BUG_ON(vm_inv_eng[i] > 17); + BUG_ON(vm_inv_eng[i] > 16); return amdgpu_irq_get(adev, &adev->mc.vm_fault, 0); } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 1eb4d79d6e30..0450ac5ba6b6 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -1175,7 +1175,7 @@ static const struct amdgpu_irq_src_funcs vcn_v1_0_irq_funcs = { static void vcn_v1_0_set_irq_funcs(struct amdgpu_device *adev) { - adev->uvd.irq.num_types = adev->vcn.num_enc_rings + 1; + adev->vcn.irq.num_types = adev->vcn.num_enc_rings + 1; adev->vcn.irq.funcs = &vcn_v1_0_irq_funcs; } |