diff options
Diffstat (limited to 'drivers')
49 files changed, 466 insertions, 246 deletions
diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 582564f8dde6..d9d339b8b571 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -2021,6 +2021,7 @@ free: static void genpd_free_data(struct generic_pm_domain *genpd) { + put_device(&genpd->dev); if (genpd_is_cpu_domain(genpd)) free_cpumask_var(genpd->cpus); if (genpd->free_states) diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 1db04886def6..3011f7f9381b 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1062,13 +1062,13 @@ skip_format_initialization: /* Sanity check */ if (range_cfg->range_max < range_cfg->range_min) { - dev_err(map->dev, "Invalid range %d: %d < %d\n", i, + dev_err(map->dev, "Invalid range %d: %u < %u\n", i, range_cfg->range_max, range_cfg->range_min); goto err_range; } if (range_cfg->range_max > map->max_register) { - dev_err(map->dev, "Invalid range %d: %d > %d\n", i, + dev_err(map->dev, "Invalid range %d: %u > %u\n", i, range_cfg->range_max, map->max_register); goto err_range; } diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 997106fe73e4..65a1f1576e55 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -1624,9 +1624,12 @@ static void virtblk_remove(struct virtio_device *vdev) static int virtblk_freeze(struct virtio_device *vdev) { struct virtio_blk *vblk = vdev->priv; + struct request_queue *q = vblk->disk->queue; /* Ensure no requests in virtqueues before deleting vqs. */ - blk_mq_freeze_queue(vblk->disk->queue); + blk_mq_freeze_queue(q); + blk_mq_quiesce_queue_nowait(q); + blk_mq_unfreeze_queue(q); /* Ensure we don't receive any more interrupts */ virtio_reset_device(vdev); @@ -1650,8 +1653,8 @@ static int virtblk_restore(struct virtio_device *vdev) return ret; virtio_device_ready(vdev); + blk_mq_unquiesce_queue(vblk->disk->queue); - blk_mq_unfreeze_queue(vblk->disk->queue); return 0; } #endif diff --git a/drivers/dma/apple-admac.c b/drivers/dma/apple-admac.c index 356298e4dd22..5b1457f6e3bf 100644 --- a/drivers/dma/apple-admac.c +++ b/drivers/dma/apple-admac.c @@ -153,6 +153,8 @@ static int admac_alloc_sram_carveout(struct admac_data *ad, { struct admac_sram *sram; int i, ret = 0, nblocks; + ad->txcache.size = readl_relaxed(ad->base + REG_TX_SRAM_SIZE); + ad->rxcache.size = readl_relaxed(ad->base + REG_RX_SRAM_SIZE); if (dir == DMA_MEM_TO_DEV) sram = &ad->txcache; @@ -912,12 +914,7 @@ static int admac_probe(struct platform_device *pdev) goto free_irq; } - ad->txcache.size = readl_relaxed(ad->base + REG_TX_SRAM_SIZE); - ad->rxcache.size = readl_relaxed(ad->base + REG_RX_SRAM_SIZE); - dev_info(&pdev->dev, "Audio DMA Controller\n"); - dev_info(&pdev->dev, "imprint %x TX cache %u RX cache %u\n", - readl_relaxed(ad->base + REG_IMPRINT), ad->txcache.size, ad->rxcache.size); return 0; diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index c3b37168b21f..2d1ee284998e 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -1363,6 +1363,8 @@ at_xdmac_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value, return NULL; desc = at_xdmac_memset_create_desc(chan, atchan, dest, len, value); + if (!desc) + return NULL; list_add_tail(&desc->desc_node, &desc->descs_list); desc->tx_dma_desc.cookie = -EBUSY; diff --git a/drivers/dma/dw/acpi.c b/drivers/dma/dw/acpi.c index c510c109d2c3..b6452fffa657 100644 --- a/drivers/dma/dw/acpi.c +++ b/drivers/dma/dw/acpi.c @@ -8,13 +8,15 @@ static bool dw_dma_acpi_filter(struct dma_chan *chan, void *param) { + struct dw_dma *dw = to_dw_dma(chan->device); + struct dw_dma_chip_pdata *data = dev_get_drvdata(dw->dma.dev); struct acpi_dma_spec *dma_spec = param; struct dw_dma_slave slave = { .dma_dev = dma_spec->dev, .src_id = dma_spec->slave_id, .dst_id = dma_spec->slave_id, - .m_master = 0, - .p_master = 1, + .m_master = data->m_master, + .p_master = data->p_master, }; return dw_dma_filter(chan, &slave); diff --git a/drivers/dma/dw/internal.h b/drivers/dma/dw/internal.h index 563ce73488db..f1bd06a20cd6 100644 --- a/drivers/dma/dw/internal.h +++ b/drivers/dma/dw/internal.h @@ -51,11 +51,15 @@ struct dw_dma_chip_pdata { int (*probe)(struct dw_dma_chip *chip); int (*remove)(struct dw_dma_chip *chip); struct dw_dma_chip *chip; + u8 m_master; + u8 p_master; }; static __maybe_unused const struct dw_dma_chip_pdata dw_dma_chip_pdata = { .probe = dw_dma_probe, .remove = dw_dma_remove, + .m_master = 0, + .p_master = 1, }; static const struct dw_dma_platform_data idma32_pdata = { @@ -72,6 +76,8 @@ static __maybe_unused const struct dw_dma_chip_pdata idma32_chip_pdata = { .pdata = &idma32_pdata, .probe = idma32_dma_probe, .remove = idma32_dma_remove, + .m_master = 0, + .p_master = 0, }; static const struct dw_dma_platform_data xbar_pdata = { @@ -88,6 +94,8 @@ static __maybe_unused const struct dw_dma_chip_pdata xbar_chip_pdata = { .pdata = &xbar_pdata, .probe = idma32_dma_probe, .remove = idma32_dma_remove, + .m_master = 0, + .p_master = 0, }; #endif /* _DMA_DW_INTERNAL_H */ diff --git a/drivers/dma/dw/pci.c b/drivers/dma/dw/pci.c index ad2d4d012cf7..e8a0eb81726a 100644 --- a/drivers/dma/dw/pci.c +++ b/drivers/dma/dw/pci.c @@ -56,10 +56,10 @@ static int dw_pci_probe(struct pci_dev *pdev, const struct pci_device_id *pid) if (ret) return ret; - dw_dma_acpi_controller_register(chip->dw); - pci_set_drvdata(pdev, data); + dw_dma_acpi_controller_register(chip->dw); + return 0; } diff --git a/drivers/dma/fsl-edma-common.h b/drivers/dma/fsl-edma-common.h index 6028389de408..8d4ef6eed3ad 100644 --- a/drivers/dma/fsl-edma-common.h +++ b/drivers/dma/fsl-edma-common.h @@ -151,6 +151,7 @@ struct fsl_edma_chan { struct work_struct issue_worker; struct platform_device *pdev; struct device *pd_dev; + struct device_link *pd_dev_link; u32 srcid; struct clk *clk; int priority; diff --git a/drivers/dma/fsl-edma-main.c b/drivers/dma/fsl-edma-main.c index 8a0ae9054899..cd394eae47d1 100644 --- a/drivers/dma/fsl-edma-main.c +++ b/drivers/dma/fsl-edma-main.c @@ -384,10 +384,33 @@ static const struct of_device_id fsl_edma_dt_ids[] = { }; MODULE_DEVICE_TABLE(of, fsl_edma_dt_ids); +static void fsl_edma3_detach_pd(struct fsl_edma_engine *fsl_edma) +{ + struct fsl_edma_chan *fsl_chan; + int i; + + for (i = 0; i < fsl_edma->n_chans; i++) { + if (fsl_edma->chan_masked & BIT(i)) + continue; + fsl_chan = &fsl_edma->chans[i]; + if (fsl_chan->pd_dev_link) + device_link_del(fsl_chan->pd_dev_link); + if (fsl_chan->pd_dev) { + dev_pm_domain_detach(fsl_chan->pd_dev, false); + pm_runtime_dont_use_autosuspend(fsl_chan->pd_dev); + pm_runtime_set_suspended(fsl_chan->pd_dev); + } + } +} + +static void devm_fsl_edma3_detach_pd(void *data) +{ + fsl_edma3_detach_pd(data); +} + static int fsl_edma3_attach_pd(struct platform_device *pdev, struct fsl_edma_engine *fsl_edma) { struct fsl_edma_chan *fsl_chan; - struct device_link *link; struct device *pd_chan; struct device *dev; int i; @@ -403,15 +426,16 @@ static int fsl_edma3_attach_pd(struct platform_device *pdev, struct fsl_edma_eng pd_chan = dev_pm_domain_attach_by_id(dev, i); if (IS_ERR_OR_NULL(pd_chan)) { dev_err(dev, "Failed attach pd %d\n", i); - return -EINVAL; + goto detach; } - link = device_link_add(dev, pd_chan, DL_FLAG_STATELESS | + fsl_chan->pd_dev_link = device_link_add(dev, pd_chan, DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE); - if (!link) { + if (!fsl_chan->pd_dev_link) { dev_err(dev, "Failed to add device_link to %d\n", i); - return -EINVAL; + dev_pm_domain_detach(pd_chan, false); + goto detach; } fsl_chan->pd_dev = pd_chan; @@ -422,6 +446,10 @@ static int fsl_edma3_attach_pd(struct platform_device *pdev, struct fsl_edma_eng } return 0; + +detach: + fsl_edma3_detach_pd(fsl_edma); + return -EINVAL; } static int fsl_edma_probe(struct platform_device *pdev) @@ -522,6 +550,9 @@ static int fsl_edma_probe(struct platform_device *pdev) ret = fsl_edma3_attach_pd(pdev, fsl_edma); if (ret) return ret; + ret = devm_add_action_or_reset(&pdev->dev, devm_fsl_edma3_detach_pd, fsl_edma); + if (ret) + return ret; } INIT_LIST_HEAD(&fsl_edma->dma_dev.channels); diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c index 23b232b57518..ea48661e87ea 100644 --- a/drivers/dma/mv_xor.c +++ b/drivers/dma/mv_xor.c @@ -1393,6 +1393,7 @@ static int mv_xor_probe(struct platform_device *pdev) irq = irq_of_parse_and_map(np, 0); if (!irq) { ret = -ENODEV; + of_node_put(np); goto err_channel_add; } @@ -1401,6 +1402,7 @@ static int mv_xor_probe(struct platform_device *pdev) if (IS_ERR(chan)) { ret = PTR_ERR(chan); irq_dispose_mapping(irq); + of_node_put(np); goto err_channel_add; } diff --git a/drivers/dma/tegra186-gpc-dma.c b/drivers/dma/tegra186-gpc-dma.c index 674cf6305283..029f45f7e37f 100644 --- a/drivers/dma/tegra186-gpc-dma.c +++ b/drivers/dma/tegra186-gpc-dma.c @@ -231,6 +231,7 @@ struct tegra_dma_channel { bool config_init; char name[30]; enum dma_transfer_direction sid_dir; + enum dma_status status; int id; int irq; int slave_id; @@ -393,6 +394,8 @@ static int tegra_dma_pause(struct tegra_dma_channel *tdc) tegra_dma_dump_chan_regs(tdc); } + tdc->status = DMA_PAUSED; + return ret; } @@ -419,6 +422,8 @@ static void tegra_dma_resume(struct tegra_dma_channel *tdc) val = tdc_read(tdc, TEGRA_GPCDMA_CHAN_CSRE); val &= ~TEGRA_GPCDMA_CHAN_CSRE_PAUSE; tdc_write(tdc, TEGRA_GPCDMA_CHAN_CSRE, val); + + tdc->status = DMA_IN_PROGRESS; } static int tegra_dma_device_resume(struct dma_chan *dc) @@ -544,6 +549,7 @@ static void tegra_dma_xfer_complete(struct tegra_dma_channel *tdc) tegra_dma_sid_free(tdc); tdc->dma_desc = NULL; + tdc->status = DMA_COMPLETE; } static void tegra_dma_chan_decode_error(struct tegra_dma_channel *tdc, @@ -716,6 +722,7 @@ static int tegra_dma_terminate_all(struct dma_chan *dc) tdc->dma_desc = NULL; } + tdc->status = DMA_COMPLETE; tegra_dma_sid_free(tdc); vchan_get_all_descriptors(&tdc->vc, &head); spin_unlock_irqrestore(&tdc->vc.lock, flags); @@ -769,6 +776,9 @@ static enum dma_status tegra_dma_tx_status(struct dma_chan *dc, if (ret == DMA_COMPLETE) return ret; + if (tdc->status == DMA_PAUSED) + ret = DMA_PAUSED; + spin_lock_irqsave(&tdc->vc.lock, flags); vd = vchan_find_desc(&tdc->vc, cookie); if (vd) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index af6c6d89e63a..fbee10927bfb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -467,28 +467,6 @@ uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev) return 100; } -void amdgpu_amdkfd_get_cu_info(struct amdgpu_device *adev, struct kfd_cu_info *cu_info) -{ - struct amdgpu_cu_info acu_info = adev->gfx.cu_info; - - memset(cu_info, 0, sizeof(*cu_info)); - if (sizeof(cu_info->cu_bitmap) != sizeof(acu_info.bitmap)) - return; - - cu_info->cu_active_number = acu_info.number; - cu_info->cu_ao_mask = acu_info.ao_cu_mask; - memcpy(&cu_info->cu_bitmap[0], &acu_info.bitmap[0], - sizeof(cu_info->cu_bitmap)); - cu_info->num_shader_engines = adev->gfx.config.max_shader_engines; - cu_info->num_shader_arrays_per_engine = adev->gfx.config.max_sh_per_se; - cu_info->num_cu_per_sh = adev->gfx.config.max_cu_per_sh; - cu_info->simd_per_cu = acu_info.simd_per_cu; - cu_info->max_waves_per_simd = acu_info.max_waves_per_simd; - cu_info->wave_front_size = acu_info.wave_front_size; - cu_info->max_scratch_slots_per_cu = acu_info.max_scratch_slots_per_cu; - cu_info->lds_size = acu_info.lds_size; -} - int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd, struct amdgpu_device **dmabuf_adev, uint64_t *bo_size, void *metadata_buffer, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 3134e6ad81d1..ff2b8ace438b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -235,8 +235,6 @@ void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev, uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev); uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev); -void amdgpu_amdkfd_get_cu_info(struct amdgpu_device *adev, - struct kfd_cu_info *cu_info); int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd, struct amdgpu_device **dmabuf_adev, uint64_t *bo_size, void *metadata_buffer, diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c index 71d1a2e3bac9..30210613dc5c 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c @@ -40,10 +40,12 @@ static void hdp_v4_0_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring) { - if (!ring || !ring->funcs->emit_wreg) - WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - else + if (!ring || !ring->funcs->emit_wreg) { + WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); + RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); + } } static void hdp_v4_0_invalidate_hdp(struct amdgpu_device *adev, @@ -53,11 +55,13 @@ static void hdp_v4_0_invalidate_hdp(struct amdgpu_device *adev, adev->ip_versions[HDP_HWIP][0] == IP_VERSION(4, 4, 2)) return; - if (!ring || !ring->funcs->emit_wreg) + if (!ring || !ring->funcs->emit_wreg) { WREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE, 1); - else + RREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE); + } else { amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET( HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1); + } } static void hdp_v4_0_query_ras_error_count(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c index a9ea23fa0def..d3962d469088 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c @@ -31,10 +31,12 @@ static void hdp_v5_0_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring) { - if (!ring || !ring->funcs->emit_wreg) - WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - else + if (!ring || !ring->funcs->emit_wreg) { + WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); + RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); + } } static void hdp_v5_0_invalidate_hdp(struct amdgpu_device *adev, @@ -42,6 +44,7 @@ static void hdp_v5_0_invalidate_hdp(struct amdgpu_device *adev, { if (!ring || !ring->funcs->emit_wreg) { WREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE, 1); + RREG32_SOC15_NO_KIQ(HDP, 0, mmHDP_READ_CACHE_INVALIDATE); } else { amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET( HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1); diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c index 063eba619f2f..b6d71ec1debf 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c @@ -31,10 +31,12 @@ static void hdp_v6_0_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring) { - if (!ring || !ring->funcs->emit_wreg) - WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); - else + if (!ring || !ring->funcs->emit_wreg) { + WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); + RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2); + } else { amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0); + } } static void hdp_v6_0_update_clock_gating(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index f76b7aee5c0a..29a02c175228 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -2037,11 +2037,12 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, uint32_t proximity_domain) { struct crat_header *crat_table = (struct crat_header *)pcrat_image; + struct amdgpu_gfx_config *gfx_info = &kdev->adev->gfx.config; + struct amdgpu_cu_info *cu_info = &kdev->adev->gfx.cu_info; struct crat_subtype_generic *sub_type_hdr; struct kfd_local_mem_info local_mem_info; struct kfd_topology_device *peer_dev; struct crat_subtype_computeunit *cu; - struct kfd_cu_info cu_info; int avail_size = *size; uint32_t total_num_of_cu; uint32_t nid = 0; @@ -2085,21 +2086,20 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, cu->flags |= CRAT_CU_FLAGS_GPU_PRESENT; cu->proximity_domain = proximity_domain; - amdgpu_amdkfd_get_cu_info(kdev->adev, &cu_info); - cu->num_simd_per_cu = cu_info.simd_per_cu; - cu->num_simd_cores = cu_info.simd_per_cu * - (cu_info.cu_active_number / kdev->kfd->num_nodes); - cu->max_waves_simd = cu_info.max_waves_per_simd; + cu->num_simd_per_cu = cu_info->simd_per_cu; + cu->num_simd_cores = cu_info->simd_per_cu * + (cu_info->number / kdev->kfd->num_nodes); + cu->max_waves_simd = cu_info->max_waves_per_simd; - cu->wave_front_size = cu_info.wave_front_size; - cu->array_count = cu_info.num_shader_arrays_per_engine * - cu_info.num_shader_engines; - total_num_of_cu = (cu->array_count * cu_info.num_cu_per_sh); + cu->wave_front_size = cu_info->wave_front_size; + cu->array_count = gfx_info->max_sh_per_se * + gfx_info->max_shader_engines; + total_num_of_cu = (cu->array_count * gfx_info->max_cu_per_sh); cu->processor_id_low = get_and_inc_gpu_processor_id(total_num_of_cu); - cu->num_cu_per_array = cu_info.num_cu_per_sh; - cu->max_slots_scatch_cu = cu_info.max_scratch_slots_per_cu; - cu->num_banks = cu_info.num_shader_engines; - cu->lds_size_in_kb = cu_info.lds_size; + cu->num_cu_per_array = gfx_info->max_cu_per_sh; + cu->max_slots_scatch_cu = cu_info->max_scratch_slots_per_cu; + cu->num_banks = gfx_info->max_shader_engines; + cu->lds_size_in_kb = cu_info->lds_size; cu->hsa_capability = 0; 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 4d9a406925e1..43fa260ddbce 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -197,6 +197,21 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q, if (dqm->is_hws_hang) return -EIO; + if (!pdd->proc_ctx_cpu_ptr) { + r = amdgpu_amdkfd_alloc_gtt_mem(adev, + AMDGPU_MES_PROC_CTX_SIZE, + &pdd->proc_ctx_bo, + &pdd->proc_ctx_gpu_addr, + &pdd->proc_ctx_cpu_ptr, + false); + if (r) { + dev_err(adev->dev, + "failed to allocate process context bo\n"); + return r; + } + memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE); + } + memset(&queue_input, 0x0, sizeof(struct mes_add_queue_input)); queue_input.process_id = qpd->pqm->process->pasid; queue_input.page_table_base_addr = qpd->page_table_base; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index 6604a3f99c5e..b22a036523b7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -373,7 +373,8 @@ int kfd_init_apertures(struct kfd_process *process) pdd = kfd_create_process_device_data(dev, process); if (!pdd) { - pr_err("Failed to create process device data\n"); + dev_err(dev->adev->dev, + "Failed to create process device data\n"); return -ENOMEM; } /* diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 68d13c4fac8f..2c529339ff65 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -68,7 +68,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev, kq->mqd_mgr = dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]; break; default: - pr_err("Invalid queue type %d\n", type); + dev_err(dev->adev->dev, "Invalid queue type %d\n", type); return false; } @@ -78,13 +78,14 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev, prop.doorbell_ptr = kfd_get_kernel_doorbell(dev->kfd, &prop.doorbell_off); if (!prop.doorbell_ptr) { - pr_err("Failed to initialize doorbell"); + dev_err(dev->adev->dev, "Failed to initialize doorbell"); goto err_get_kernel_doorbell; } retval = kfd_gtt_sa_allocate(dev, queue_size, &kq->pq); if (retval != 0) { - pr_err("Failed to init pq queues size %d\n", queue_size); + dev_err(dev->adev->dev, "Failed to init pq queues size %d\n", + queue_size); goto err_pq_allocate_vidmem; } @@ -332,7 +333,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_node *dev, if (kq_initialize(kq, dev, type, KFD_KERNEL_QUEUE_SIZE)) return kq; - pr_err("Failed to init kernel queue\n"); + dev_err(dev->adev->dev, "Failed to init kernel queue\n"); kfree(kq); return NULL; @@ -351,26 +352,26 @@ static __attribute__((unused)) void test_kq(struct kfd_node *dev) uint32_t *buffer, i; int retval; - pr_err("Starting kernel queue test\n"); + dev_err(dev->adev->dev, "Starting kernel queue test\n"); kq = kernel_queue_init(dev, KFD_QUEUE_TYPE_HIQ); if (unlikely(!kq)) { - pr_err(" Failed to initialize HIQ\n"); - pr_err("Kernel queue test failed\n"); + dev_err(dev->adev->dev, " Failed to initialize HIQ\n"); + dev_err(dev->adev->dev, "Kernel queue test failed\n"); return; } retval = kq_acquire_packet_buffer(kq, 5, &buffer); if (unlikely(retval != 0)) { - pr_err(" Failed to acquire packet buffer\n"); - pr_err("Kernel queue test failed\n"); + dev_err(dev->adev->dev, " Failed to acquire packet buffer\n"); + dev_err(dev->adev->dev, "Kernel queue test failed\n"); return; } for (i = 0; i < 5; i++) buffer[i] = kq->nop_packet; kq_submit_packet(kq); - pr_err("Ending kernel queue test\n"); + dev_err(dev->adev->dev, "Ending kernel queue test\n"); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index 4c3f37980311..0edae9ded68a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -99,7 +99,8 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, const uint32_t *cu_mask, uint32_t cu_mask_count, uint32_t *se_mask, uint32_t inst) { - struct kfd_cu_info cu_info; + struct amdgpu_cu_info *cu_info = &mm->dev->adev->gfx.cu_info; + struct amdgpu_gfx_config *gfx_info = &mm->dev->adev->gfx.config; uint32_t cu_per_sh[KFD_MAX_NUM_SE][KFD_MAX_NUM_SH_PER_SE] = {0}; bool wgp_mode_req = KFD_GC_VERSION(mm->dev) >= IP_VERSION(10, 0, 0); uint32_t en_mask = wgp_mode_req ? 0x3 : 0x1; @@ -108,9 +109,7 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, int inc = cu_inc * NUM_XCC(mm->dev->xcc_mask); int xcc_inst = inst + ffs(mm->dev->xcc_mask) - 1; - amdgpu_amdkfd_get_cu_info(mm->dev->adev, &cu_info); - - cu_active_per_node = cu_info.cu_active_number / mm->dev->kfd->num_nodes; + cu_active_per_node = cu_info->number / mm->dev->kfd->num_nodes; if (cu_mask_count > cu_active_per_node) cu_mask_count = cu_active_per_node; @@ -118,13 +117,16 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, * Returning with no CU's enabled will hang the queue, which should be * attention grabbing. */ - if (cu_info.num_shader_engines > KFD_MAX_NUM_SE) { - pr_err("Exceeded KFD_MAX_NUM_SE, chip reports %d\n", cu_info.num_shader_engines); + if (gfx_info->max_shader_engines > KFD_MAX_NUM_SE) { + dev_err(mm->dev->adev->dev, + "Exceeded KFD_MAX_NUM_SE, chip reports %d\n", + gfx_info->max_shader_engines); return; } - if (cu_info.num_shader_arrays_per_engine > KFD_MAX_NUM_SH_PER_SE) { - pr_err("Exceeded KFD_MAX_NUM_SH, chip reports %d\n", - cu_info.num_shader_arrays_per_engine * cu_info.num_shader_engines); + if (gfx_info->max_sh_per_se > KFD_MAX_NUM_SH_PER_SE) { + dev_err(mm->dev->adev->dev, + "Exceeded KFD_MAX_NUM_SH, chip reports %d\n", + gfx_info->max_sh_per_se * gfx_info->max_shader_engines); return; } @@ -142,10 +144,10 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, * See note on Arcturus cu_bitmap layout in gfx_v9_0_get_cu_info. * See note on GFX11 cu_bitmap layout in gfx_v11_0_get_cu_info. */ - for (se = 0; se < cu_info.num_shader_engines; se++) - for (sh = 0; sh < cu_info.num_shader_arrays_per_engine; sh++) + for (se = 0; se < gfx_info->max_shader_engines; se++) + for (sh = 0; sh < gfx_info->max_sh_per_se; sh++) cu_per_sh[se][sh] = hweight32( - cu_info.cu_bitmap[xcc_inst][se % 4][sh + (se / 4) * + cu_info->bitmap[xcc_inst][se % 4][sh + (se / 4) * cu_bitmap_sh_mul]); /* Symmetrically map cu_mask to all SEs & SHs: @@ -184,13 +186,13 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, * * First ensure all CUs are disabled, then enable user specified CUs. */ - for (i = 0; i < cu_info.num_shader_engines; i++) + for (i = 0; i < gfx_info->max_shader_engines; i++) se_mask[i] = 0; i = inst; for (cu = 0; cu < 16; cu += cu_inc) { - for (sh = 0; sh < cu_info.num_shader_arrays_per_engine; sh++) { - for (se = 0; se < cu_info.num_shader_engines; se++) { + for (sh = 0; sh < gfx_info->max_sh_per_se; sh++) { + for (se = 0; se < gfx_info->max_shader_engines; se++) { if (cu_per_sh[se][sh] > cu) { if (cu_mask[i / 32] & (en_mask << (i % 32))) se_mask[se] |= en_mask << (cu + sh * 16); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c index 401096c103b2..ecb38a6e8013 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c @@ -45,7 +45,8 @@ static void pm_calc_rlib_size(struct packet_manager *pm, unsigned int process_count, queue_count, compute_queue_count, gws_queue_count; unsigned int map_queue_size; unsigned int max_proc_per_quantum = 1; - struct kfd_node *dev = pm->dqm->dev; + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; process_count = pm->dqm->processes_count; queue_count = pm->dqm->active_queue_count; @@ -59,14 +60,14 @@ static void pm_calc_rlib_size(struct packet_manager *pm, */ *over_subscription = false; - if (dev->max_proc_per_quantum > 1) - max_proc_per_quantum = dev->max_proc_per_quantum; + if (node->max_proc_per_quantum > 1) + max_proc_per_quantum = node->max_proc_per_quantum; if ((process_count > max_proc_per_quantum) || compute_queue_count > get_cp_queues_num(pm->dqm) || gws_queue_count > 1) { *over_subscription = true; - pr_debug("Over subscribed runlist\n"); + dev_dbg(dev, "Over subscribed runlist\n"); } map_queue_size = pm->pmf->map_queues_size; @@ -81,7 +82,7 @@ static void pm_calc_rlib_size(struct packet_manager *pm, if (*over_subscription) *rlib_size += pm->pmf->runlist_size; - pr_debug("runlist ib size %d\n", *rlib_size); + dev_dbg(dev, "runlist ib size %d\n", *rlib_size); } static int pm_allocate_runlist_ib(struct packet_manager *pm, @@ -90,6 +91,8 @@ static int pm_allocate_runlist_ib(struct packet_manager *pm, unsigned int *rl_buffer_size, bool *is_over_subscription) { + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; int retval; if (WARN_ON(pm->allocated)) @@ -99,11 +102,10 @@ static int pm_allocate_runlist_ib(struct packet_manager *pm, mutex_lock(&pm->lock); - retval = kfd_gtt_sa_allocate(pm->dqm->dev, *rl_buffer_size, - &pm->ib_buffer_obj); + retval = kfd_gtt_sa_allocate(node, *rl_buffer_size, &pm->ib_buffer_obj); if (retval) { - pr_err("Failed to allocate runlist IB\n"); + dev_err(dev, "Failed to allocate runlist IB\n"); goto out; } @@ -125,6 +127,8 @@ static int pm_create_runlist_ib(struct packet_manager *pm, { unsigned int alloc_size_bytes; unsigned int *rl_buffer, rl_wptr, i; + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; int retval, processes_mapped; struct device_process_node *cur; struct qcm_process_device *qpd; @@ -142,7 +146,7 @@ static int pm_create_runlist_ib(struct packet_manager *pm, *rl_size_bytes = alloc_size_bytes; pm->ib_size_bytes = alloc_size_bytes; - pr_debug("Building runlist ib process count: %d queues count %d\n", + dev_dbg(dev, "Building runlist ib process count: %d queues count %d\n", pm->dqm->processes_count, pm->dqm->active_queue_count); /* build the run list ib packet */ @@ -150,7 +154,7 @@ static int pm_create_runlist_ib(struct packet_manager *pm, qpd = cur->qpd; /* build map process packet */ if (processes_mapped >= pm->dqm->processes_count) { - pr_debug("Not enough space left in runlist IB\n"); + dev_dbg(dev, "Not enough space left in runlist IB\n"); pm_release_ib(pm); return -ENOMEM; } @@ -167,7 +171,8 @@ static int pm_create_runlist_ib(struct packet_manager *pm, if (!kq->queue->properties.is_active) continue; - pr_debug("static_queue, mapping kernel q %d, is debug status %d\n", + dev_dbg(dev, + "static_queue, mapping kernel q %d, is debug status %d\n", kq->queue->queue, qpd->is_debug); retval = pm->pmf->map_queues(pm, @@ -186,7 +191,8 @@ static int pm_create_runlist_ib(struct packet_manager *pm, if (!q->properties.is_active) continue; - pr_debug("static_queue, mapping user queue %d, is debug status %d\n", + dev_dbg(dev, + "static_queue, mapping user queue %d, is debug status %d\n", q->queue, qpd->is_debug); retval = pm->pmf->map_queues(pm, @@ -203,11 +209,13 @@ static int pm_create_runlist_ib(struct packet_manager *pm, } } - pr_debug("Finished map process and queues to runlist\n"); + dev_dbg(dev, "Finished map process and queues to runlist\n"); if (is_over_subscription) { if (!pm->is_over_subscription) - pr_warn("Runlist is getting oversubscribed. Expect reduced ROCm performance.\n"); + dev_warn( + dev, + "Runlist is getting oversubscribed. Expect reduced ROCm performance.\n"); retval = pm->pmf->runlist(pm, &rl_buffer[rl_wptr], *rl_gpu_addr, alloc_size_bytes / sizeof(uint32_t), @@ -272,6 +280,8 @@ void pm_uninit(struct packet_manager *pm, bool hanging) int pm_send_set_resources(struct packet_manager *pm, struct scheduling_resources *res) { + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; uint32_t *buffer, size; int retval = 0; @@ -281,7 +291,7 @@ int pm_send_set_resources(struct packet_manager *pm, size / sizeof(uint32_t), (unsigned int **)&buffer); if (!buffer) { - pr_err("Failed to allocate buffer on kernel queue\n"); + dev_err(dev, "Failed to allocate buffer on kernel queue\n"); retval = -ENOMEM; goto out; } @@ -343,6 +353,8 @@ fail_create_runlist_ib: int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address, uint64_t fence_value) { + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; uint32_t *buffer, size; int retval = 0; @@ -354,7 +366,7 @@ int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address, kq_acquire_packet_buffer(pm->priv_queue, size / sizeof(uint32_t), (unsigned int **)&buffer); if (!buffer) { - pr_err("Failed to allocate buffer on kernel queue\n"); + dev_err(dev, "Failed to allocate buffer on kernel queue\n"); retval = -ENOMEM; goto out; } @@ -372,6 +384,8 @@ out: int pm_update_grace_period(struct packet_manager *pm, uint32_t grace_period) { + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; int retval = 0; uint32_t *buffer, size; @@ -385,7 +399,8 @@ int pm_update_grace_period(struct packet_manager *pm, uint32_t grace_period) (unsigned int **)&buffer); if (!buffer) { - pr_err("Failed to allocate buffer on kernel queue\n"); + dev_err(dev, + "Failed to allocate buffer on kernel queue\n"); retval = -ENOMEM; goto out; } @@ -406,6 +421,8 @@ int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_unmap_queues_filter filter, uint32_t filter_param, bool reset) { + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; uint32_t *buffer, size; int retval = 0; @@ -414,7 +431,7 @@ int pm_send_unmap_queue(struct packet_manager *pm, kq_acquire_packet_buffer(pm->priv_queue, size / sizeof(uint32_t), (unsigned int **)&buffer); if (!buffer) { - pr_err("Failed to allocate buffer on kernel queue\n"); + dev_err(dev, "Failed to allocate buffer on kernel queue\n"); retval = -ENOMEM; goto out; } @@ -463,6 +480,8 @@ out: int pm_debugfs_hang_hws(struct packet_manager *pm) { + struct kfd_node *node = pm->dqm->dev; + struct device *dev = node->adev->dev; uint32_t *buffer, size; int r = 0; @@ -474,16 +493,16 @@ int pm_debugfs_hang_hws(struct packet_manager *pm) kq_acquire_packet_buffer(pm->priv_queue, size / sizeof(uint32_t), (unsigned int **)&buffer); if (!buffer) { - pr_err("Failed to allocate buffer on kernel queue\n"); + dev_err(dev, "Failed to allocate buffer on kernel queue\n"); r = -ENOMEM; goto out; } memset(buffer, 0x55, size); kq_submit_packet(pm->priv_queue); - pr_info("Submitting %x %x %x %x %x %x %x to HIQ to hang the HWS.", - buffer[0], buffer[1], buffer[2], buffer[3], - buffer[4], buffer[5], buffer[6]); + dev_info(dev, "Submitting %x %x %x %x %x %x %x to HIQ to hang the HWS.", + buffer[0], buffer[1], buffer[2], buffer[3], buffer[4], + buffer[5], buffer[6]); out: mutex_unlock(&pm->lock); return r; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index fd640a061c96..64346c71c62a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -1046,7 +1046,8 @@ static void kfd_process_destroy_pdds(struct kfd_process *p) kfd_free_process_doorbells(pdd->dev->kfd, pdd); - if (pdd->dev->kfd->shared_resources.enable_mes) + if (pdd->dev->kfd->shared_resources.enable_mes && + pdd->proc_ctx_cpu_ptr) amdgpu_amdkfd_free_gtt_mem(pdd->dev->adev, &pdd->proc_ctx_bo); /* @@ -1308,7 +1309,8 @@ int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep) if (IS_ERR_VALUE(qpd->tba_addr)) { int err = qpd->tba_addr; - pr_err("Failure to set tba address. error %d.\n", err); + dev_err(dev->adev->dev, + "Failure to set tba address. error %d.\n", err); qpd->tba_addr = 0; qpd->cwsr_kaddr = NULL; return err; @@ -1571,7 +1573,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, struct kfd_process *p) { struct kfd_process_device *pdd = NULL; - int retval = 0; if (WARN_ON_ONCE(p->n_pdds >= MAX_GPU_INSTANCE)) return NULL; @@ -1595,20 +1596,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, pdd->user_gpu_id = dev->id; atomic64_set(&pdd->evict_duration_counter, 0); - if (dev->kfd->shared_resources.enable_mes) { - retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, - AMDGPU_MES_PROC_CTX_SIZE, - &pdd->proc_ctx_bo, - &pdd->proc_ctx_gpu_addr, - &pdd->proc_ctx_cpu_ptr, - false); - if (retval) { - pr_err("failed to allocate process context bo\n"); - goto err_free_pdd; - } - memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE); - } - p->pdds[p->n_pdds++] = pdd; if (kfd_dbg_is_per_vmid_supported(pdd->dev)) pdd->spi_dbg_override = pdd->dev->kfd2kgd->disable_debug_trap( @@ -1620,10 +1607,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, idr_init(&pdd->alloc_idr); return pdd; - -err_free_pdd: - kfree(pdd); - return NULL; } /** @@ -1667,7 +1650,7 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, &p->kgd_process_info, &p->ef); if (ret) { - pr_err("Failed to create process VM object\n"); + dev_err(dev->adev->dev, "Failed to create process VM object\n"); return ret; } pdd->drm_priv = drm_file->private_data; @@ -1714,7 +1697,7 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev, pdd = kfd_get_process_device_data(dev, p); if (!pdd) { - pr_err("Process device data doesn't exist\n"); + dev_err(dev->adev->dev, "Process device data doesn't exist\n"); return ERR_PTR(-ENOMEM); } @@ -1824,6 +1807,7 @@ int kfd_process_evict_queues(struct kfd_process *p, uint32_t trigger) for (i = 0; i < p->n_pdds; i++) { struct kfd_process_device *pdd = p->pdds[i]; + struct device *dev = pdd->dev->adev->dev; kfd_smi_event_queue_eviction(pdd->dev, p->lead_thread->pid, trigger); @@ -1835,7 +1819,7 @@ int kfd_process_evict_queues(struct kfd_process *p, uint32_t trigger) * them been add back since they actually not be saved right now. */ if (r && r != -EIO) { - pr_err("Failed to evict process queues\n"); + dev_err(dev, "Failed to evict process queues\n"); goto fail; } n_evicted++; @@ -1857,7 +1841,8 @@ fail: if (pdd->dev->dqm->ops.restore_process_queues(pdd->dev->dqm, &pdd->qpd)) - pr_err("Failed to restore queues\n"); + dev_err(pdd->dev->adev->dev, + "Failed to restore queues\n"); n_evicted--; } @@ -1873,13 +1858,14 @@ int kfd_process_restore_queues(struct kfd_process *p) for (i = 0; i < p->n_pdds; i++) { struct kfd_process_device *pdd = p->pdds[i]; + struct device *dev = pdd->dev->adev->dev; kfd_smi_event_queue_restore(pdd->dev, p->lead_thread->pid); r = pdd->dev->dqm->ops.restore_process_queues(pdd->dev->dqm, &pdd->qpd); if (r) { - pr_err("Failed to restore process queues\n"); + dev_err(dev, "Failed to restore process queues\n"); if (!ret) ret = r; } @@ -2039,7 +2025,7 @@ int kfd_reserved_mem_mmap(struct kfd_node *dev, struct kfd_process *process, struct qcm_process_device *qpd; if ((vma->vm_end - vma->vm_start) != KFD_CWSR_TBA_TMA_SIZE) { - pr_err("Incorrect CWSR mapping size.\n"); + dev_err(dev->adev->dev, "Incorrect CWSR mapping size.\n"); return -EINVAL; } @@ -2051,7 +2037,8 @@ int kfd_reserved_mem_mmap(struct kfd_node *dev, struct kfd_process *process, qpd->cwsr_kaddr = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(KFD_CWSR_TBA_TMA_SIZE)); if (!qpd->cwsr_kaddr) { - pr_err("Error allocating per process CWSR buffer.\n"); + dev_err(dev->adev->dev, + "Error allocating per process CWSR buffer.\n"); return -ENOMEM; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 8362a71ab707..3885bb53f019 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -1537,7 +1537,6 @@ out: /* Helper function. See kfd_fill_gpu_cache_info for parameter description */ static int fill_in_l1_pcache(struct kfd_cache_properties **props_ext, struct kfd_gpu_cache_info *pcache_info, - struct kfd_cu_info *cu_info, int cu_bitmask, int cache_type, unsigned int cu_processor_id, int cu_block) @@ -1599,7 +1598,8 @@ static int fill_in_l1_pcache(struct kfd_cache_properties **props_ext, /* Helper function. See kfd_fill_gpu_cache_info for parameter description */ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, struct kfd_gpu_cache_info *pcache_info, - struct kfd_cu_info *cu_info, + struct amdgpu_cu_info *cu_info, + struct amdgpu_gfx_config *gfx_info, int cache_type, unsigned int cu_processor_id, struct kfd_node *knode) { @@ -1610,7 +1610,7 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, start = ffs(knode->xcc_mask) - 1; end = start + NUM_XCC(knode->xcc_mask); - cu_sibling_map_mask = cu_info->cu_bitmap[start][0][0]; + cu_sibling_map_mask = cu_info->bitmap[start][0][0]; cu_sibling_map_mask &= ((1 << pcache_info[cache_type].num_cu_shared) - 1); first_active_cu = ffs(cu_sibling_map_mask); @@ -1646,15 +1646,15 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, k = 0; for (xcc = start; xcc < end; xcc++) { - for (i = 0; i < cu_info->num_shader_engines; i++) { - for (j = 0; j < cu_info->num_shader_arrays_per_engine; j++) { + for (i = 0; i < gfx_info->max_shader_engines; i++) { + for (j = 0; j < gfx_info->max_sh_per_se; j++) { pcache->sibling_map[k] = (uint8_t)(cu_sibling_map_mask & 0xFF); pcache->sibling_map[k+1] = (uint8_t)((cu_sibling_map_mask >> 8) & 0xFF); pcache->sibling_map[k+2] = (uint8_t)((cu_sibling_map_mask >> 16) & 0xFF); pcache->sibling_map[k+3] = (uint8_t)((cu_sibling_map_mask >> 24) & 0xFF); k += 4; - cu_sibling_map_mask = cu_info->cu_bitmap[xcc][i % 4][j + i / 4]; + cu_sibling_map_mask = cu_info->bitmap[xcc][i % 4][j + i / 4]; cu_sibling_map_mask &= ((1 << pcache_info[cache_type].num_cu_shared) - 1); } } @@ -1679,16 +1679,14 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct unsigned int cu_processor_id; int ret; unsigned int num_cu_shared; - struct kfd_cu_info cu_info; - struct kfd_cu_info *pcu_info; + struct amdgpu_cu_info *cu_info = &kdev->adev->gfx.cu_info; + struct amdgpu_gfx_config *gfx_info = &kdev->adev->gfx.config; int gpu_processor_id; struct kfd_cache_properties *props_ext; int num_of_entries = 0; int num_of_cache_types = 0; struct kfd_gpu_cache_info cache_info[KFD_MAX_CACHE_TYPES]; - amdgpu_amdkfd_get_cu_info(kdev->adev, &cu_info); - pcu_info = &cu_info; gpu_processor_id = dev->node_props.simd_id_base; @@ -1715,12 +1713,12 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct cu_processor_id = gpu_processor_id; if (pcache_info[ct].cache_level == 1) { for (xcc = start; xcc < end; xcc++) { - for (i = 0; i < pcu_info->num_shader_engines; i++) { - for (j = 0; j < pcu_info->num_shader_arrays_per_engine; j++) { - for (k = 0; k < pcu_info->num_cu_per_sh; k += pcache_info[ct].num_cu_shared) { + for (i = 0; i < gfx_info->max_shader_engines; i++) { + for (j = 0; j < gfx_info->max_sh_per_se; j++) { + for (k = 0; k < gfx_info->max_cu_per_sh; k += pcache_info[ct].num_cu_shared) { - ret = fill_in_l1_pcache(&props_ext, pcache_info, pcu_info, - pcu_info->cu_bitmap[xcc][i % 4][j + i / 4], ct, + ret = fill_in_l1_pcache(&props_ext, pcache_info, + cu_info->bitmap[xcc][i % 4][j + i / 4], ct, cu_processor_id, k); if (ret < 0) @@ -1733,9 +1731,9 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct /* Move to next CU block */ num_cu_shared = ((k + pcache_info[ct].num_cu_shared) <= - pcu_info->num_cu_per_sh) ? + gfx_info->max_cu_per_sh) ? pcache_info[ct].num_cu_shared : - (pcu_info->num_cu_per_sh - k); + (gfx_info->max_cu_per_sh - k); cu_processor_id += num_cu_shared; } } @@ -1743,7 +1741,7 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct } } else { ret = fill_in_l2_l3_pcache(&props_ext, pcache_info, - pcu_info, ct, cu_processor_id, kdev); + cu_info, gfx_info, ct, cu_processor_id, kdev); if (ret < 0) break; @@ -1922,10 +1920,11 @@ int kfd_topology_add_device(struct kfd_node *gpu) { uint32_t gpu_id; struct kfd_topology_device *dev; - struct kfd_cu_info cu_info; int res = 0; int i; const char *asic_name = amdgpu_asic_name[gpu->adev->asic_type]; + struct amdgpu_gfx_config *gfx_info = &gpu->adev->gfx.config; + struct amdgpu_cu_info *cu_info = &gpu->adev->gfx.cu_info; gpu_id = kfd_generate_gpu_id(gpu); if (gpu->xcp && !gpu->xcp->ddev) { @@ -1963,9 +1962,6 @@ int kfd_topology_add_device(struct kfd_node *gpu) /* Fill-in additional information that is not available in CRAT but * needed for the topology */ - - amdgpu_amdkfd_get_cu_info(dev->gpu->adev, &cu_info); - for (i = 0; i < KFD_TOPOLOGY_PUBLIC_NAME_SIZE-1; i++) { dev->node_props.name[i] = __tolower(asic_name[i]); if (asic_name[i] == '\0') @@ -1974,7 +1970,7 @@ int kfd_topology_add_device(struct kfd_node *gpu) dev->node_props.name[i] = '\0'; dev->node_props.simd_arrays_per_engine = - cu_info.num_shader_arrays_per_engine; + gfx_info->max_sh_per_se; dev->node_props.gfx_target_version = gpu->kfd->device_info.gfx_target_version; @@ -2055,7 +2051,7 @@ int kfd_topology_add_device(struct kfd_node *gpu) */ if (dev->gpu->adev->asic_type == CHIP_CARRIZO) { dev->node_props.simd_count = - cu_info.simd_per_cu * cu_info.cu_active_number; + cu_info->simd_per_cu * cu_info->number; dev->node_props.max_waves_per_simd = 10; } diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index 3b5a56585c4b..c653a7f4d5e5 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -57,20 +57,6 @@ struct kfd_vm_fault_info { bool prot_exec; }; -struct kfd_cu_info { - uint32_t num_shader_engines; - uint32_t num_shader_arrays_per_engine; - uint32_t num_cu_per_sh; - uint32_t cu_active_number; - uint32_t cu_ao_mask; - uint32_t simd_per_cu; - uint32_t max_waves_per_simd; - uint32_t wave_front_size; - uint32_t max_scratch_slots_per_cu; - uint32_t lds_size; - uint32_t cu_bitmap[AMDGPU_MAX_GC_INSTANCES][4][4]; -}; - /* For getting GPU local memory information from KGD */ struct kfd_local_mem_info { uint64_t local_mem_size_private; diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c index 90bfb1e988fb..d6c5de190c27 100644 --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c @@ -4033,9 +4033,10 @@ static void drm_dp_mst_up_req_work(struct work_struct *work) static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr) { struct drm_dp_pending_up_req *up_req; + struct drm_dp_mst_branch *mst_primary; if (!drm_dp_get_one_sb_msg(mgr, true, NULL)) - goto out; + goto out_clear_reply; if (!mgr->up_req_recv.have_eomt) return 0; @@ -4053,10 +4054,19 @@ static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr) drm_dbg_kms(mgr->dev, "Received unknown up req type, ignoring: %x\n", up_req->msg.req_type); kfree(up_req); - goto out; + goto out_clear_reply; + } + + mutex_lock(&mgr->lock); + mst_primary = mgr->mst_primary; + if (!mst_primary || !drm_dp_mst_topology_try_get_mstb(mst_primary)) { + mutex_unlock(&mgr->lock); + kfree(up_req); + goto out_clear_reply; } + mutex_unlock(&mgr->lock); - drm_dp_send_up_ack_reply(mgr, mgr->mst_primary, up_req->msg.req_type, + drm_dp_send_up_ack_reply(mgr, mst_primary, up_req->msg.req_type, false); if (up_req->msg.req_type == DP_CONNECTION_STATUS_NOTIFY) { @@ -4073,13 +4083,13 @@ static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr) conn_stat->peer_device_type); mutex_lock(&mgr->probe_lock); - handle_csn = mgr->mst_primary->link_address_sent; + handle_csn = mst_primary->link_address_sent; mutex_unlock(&mgr->probe_lock); if (!handle_csn) { drm_dbg_kms(mgr->dev, "Got CSN before finish topology probing. Skip it."); kfree(up_req); - goto out; + goto out_put_primary; } } else if (up_req->msg.req_type == DP_RESOURCE_STATUS_NOTIFY) { const struct drm_dp_resource_status_notify *res_stat = @@ -4096,7 +4106,9 @@ static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr) mutex_unlock(&mgr->up_req_lock); queue_work(system_long_wq, &mgr->up_req_work); -out: +out_put_primary: + drm_dp_mst_topology_put_mstb(mst_primary); +out_clear_reply: memset(&mgr->up_req_recv, 0, sizeof(struct drm_dp_sideband_msg_rx)); return 0; } diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c index 0951bfdc89cf..913823a063c9 100644 --- a/drivers/i2c/busses/i2c-imx.c +++ b/drivers/i2c/busses/i2c-imx.c @@ -286,6 +286,7 @@ static const struct of_device_id i2c_imx_dt_ids[] = { { .compatible = "fsl,imx6sll-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx6sx-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx6ul-i2c", .data = &imx6_i2c_hwdata, }, + { .compatible = "fsl,imx7d-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx7s-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx8mm-i2c", .data = &imx6_i2c_hwdata, }, { .compatible = "fsl,imx8mn-i2c", .data = &imx6_i2c_hwdata, }, diff --git a/drivers/i2c/busses/i2c-microchip-corei2c.c b/drivers/i2c/busses/i2c-microchip-corei2c.c index 0b0a1c4d17ca..b0a51695138a 100644 --- a/drivers/i2c/busses/i2c-microchip-corei2c.c +++ b/drivers/i2c/busses/i2c-microchip-corei2c.c @@ -93,27 +93,35 @@ * @base: pointer to register struct * @dev: device reference * @i2c_clk: clock reference for i2c input clock + * @msg_queue: pointer to the messages requiring sending * @buf: pointer to msg buffer for easier use * @msg_complete: xfer completion object * @adapter: core i2c abstraction * @msg_err: error code for completed message * @bus_clk_rate: current i2c bus clock rate * @isr_status: cached copy of local ISR status + * @total_num: total number of messages to be sent/received + * @current_num: index of the current message being sent/received * @msg_len: number of bytes transferred in msg * @addr: address of the current slave + * @restart_needed: whether or not a repeated start is required after current message */ struct mchp_corei2c_dev { void __iomem *base; struct device *dev; struct clk *i2c_clk; + struct i2c_msg *msg_queue; u8 *buf; struct completion msg_complete; struct i2c_adapter adapter; int msg_err; + int total_num; + int current_num; u32 bus_clk_rate; u32 isr_status; u16 msg_len; u8 addr; + bool restart_needed; }; static void mchp_corei2c_core_disable(struct mchp_corei2c_dev *idev) @@ -222,6 +230,47 @@ static int mchp_corei2c_fill_tx(struct mchp_corei2c_dev *idev) return 0; } +static void mchp_corei2c_next_msg(struct mchp_corei2c_dev *idev) +{ + struct i2c_msg *this_msg; + u8 ctrl; + + if (idev->current_num >= idev->total_num) { + complete(&idev->msg_complete); + return; + } + + /* + * If there's been an error, the isr needs to return control + * to the "main" part of the driver, so as not to keep sending + * messages once it completes and clears the SI bit. + */ + if (idev->msg_err) { + complete(&idev->msg_complete); + return; + } + + this_msg = idev->msg_queue++; + + if (idev->current_num < (idev->total_num - 1)) { + struct i2c_msg *next_msg = idev->msg_queue; + + idev->restart_needed = next_msg->flags & I2C_M_RD; + } else { + idev->restart_needed = false; + } + + idev->addr = i2c_8bit_addr_from_msg(this_msg); + idev->msg_len = this_msg->len; + idev->buf = this_msg->buf; + + ctrl = readb(idev->base + CORE_I2C_CTRL); + ctrl |= CTRL_STA; + writeb(ctrl, idev->base + CORE_I2C_CTRL); + + idev->current_num++; +} + static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) { u32 status = idev->isr_status; @@ -238,8 +287,6 @@ static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) ctrl &= ~CTRL_STA; writeb(idev->addr, idev->base + CORE_I2C_DATA); writeb(ctrl, idev->base + CORE_I2C_CTRL); - if (idev->msg_len == 0) - finished = true; break; case STATUS_M_ARB_LOST: idev->msg_err = -EAGAIN; @@ -247,10 +294,14 @@ static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) break; case STATUS_M_SLAW_ACK: case STATUS_M_TX_DATA_ACK: - if (idev->msg_len > 0) + if (idev->msg_len > 0) { mchp_corei2c_fill_tx(idev); - else - last_byte = true; + } else { + if (idev->restart_needed) + finished = true; + else + last_byte = true; + } break; case STATUS_M_TX_DATA_NACK: case STATUS_M_SLAR_NACK: @@ -287,7 +338,7 @@ static irqreturn_t mchp_corei2c_handle_isr(struct mchp_corei2c_dev *idev) mchp_corei2c_stop(idev); if (last_byte || finished) - complete(&idev->msg_complete); + mchp_corei2c_next_msg(idev); return IRQ_HANDLED; } @@ -311,21 +362,48 @@ static irqreturn_t mchp_corei2c_isr(int irq, void *_dev) return ret; } -static int mchp_corei2c_xfer_msg(struct mchp_corei2c_dev *idev, - struct i2c_msg *msg) +static int mchp_corei2c_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, + int num) { - u8 ctrl; + struct mchp_corei2c_dev *idev = i2c_get_adapdata(adap); + struct i2c_msg *this_msg = msgs; unsigned long time_left; + u8 ctrl; + + mchp_corei2c_core_enable(idev); + + /* + * The isr controls the flow of a transfer, this info needs to be saved + * to a location that it can access the queue information from. + */ + idev->restart_needed = false; + idev->msg_queue = msgs; + idev->total_num = num; + idev->current_num = 0; - idev->addr = i2c_8bit_addr_from_msg(msg); - idev->msg_len = msg->len; - idev->buf = msg->buf; + /* + * But the first entry to the isr is triggered by the start in this + * function, so the first message needs to be "dequeued". + */ + idev->addr = i2c_8bit_addr_from_msg(this_msg); + idev->msg_len = this_msg->len; + idev->buf = this_msg->buf; idev->msg_err = 0; - reinit_completion(&idev->msg_complete); + if (idev->total_num > 1) { + struct i2c_msg *next_msg = msgs + 1; - mchp_corei2c_core_enable(idev); + idev->restart_needed = next_msg->flags & I2C_M_RD; + } + idev->current_num++; + idev->msg_queue++; + + reinit_completion(&idev->msg_complete); + + /* + * Send the first start to pass control to the isr + */ ctrl = readb(idev->base + CORE_I2C_CTRL); ctrl |= CTRL_STA; writeb(ctrl, idev->base + CORE_I2C_CTRL); @@ -335,20 +413,8 @@ static int mchp_corei2c_xfer_msg(struct mchp_corei2c_dev *idev, if (!time_left) return -ETIMEDOUT; - return idev->msg_err; -} - -static int mchp_corei2c_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, - int num) -{ - struct mchp_corei2c_dev *idev = i2c_get_adapdata(adap); - int i, ret; - - for (i = 0; i < num; i++) { - ret = mchp_corei2c_xfer_msg(idev, msgs++); - if (ret) - return ret; - } + if (idev->msg_err) + return idev->msg_err; return num; } diff --git a/drivers/media/dvb-frontends/dib3000mb.c b/drivers/media/dvb-frontends/dib3000mb.c index c598b2a63325..7c452ddd9e40 100644 --- a/drivers/media/dvb-frontends/dib3000mb.c +++ b/drivers/media/dvb-frontends/dib3000mb.c @@ -51,7 +51,7 @@ MODULE_PARM_DESC(debug, "set debugging level (1=info,2=xfer,4=setfe,8=getfe (|-a static int dib3000_read_reg(struct dib3000_state *state, u16 reg) { u8 wb[] = { ((reg >> 8) | 0x80) & 0xff, reg & 0xff }; - u8 rb[2]; + u8 rb[2] = {}; struct i2c_msg msg[] = { { .addr = state->config.demod_address, .flags = 0, .buf = wb, .len = 2 }, { .addr = state->config.demod_address, .flags = I2C_M_RD, .buf = rb, .len = 2 }, diff --git a/drivers/mtd/nand/raw/arasan-nand-controller.c b/drivers/mtd/nand/raw/arasan-nand-controller.c index a492051c46f5..bde396b359c3 100644 --- a/drivers/mtd/nand/raw/arasan-nand-controller.c +++ b/drivers/mtd/nand/raw/arasan-nand-controller.c @@ -1410,8 +1410,8 @@ static int anfc_parse_cs(struct arasan_nfc *nfc) * case, the "not" chosen CS is assigned to nfc->spare_cs and selected * whenever a GPIO CS must be asserted. */ - if (nfc->cs_array && nfc->ncs > 2) { - if (!nfc->cs_array[0] && !nfc->cs_array[1]) { + if (nfc->cs_array) { + if (nfc->ncs > 2 && !nfc->cs_array[0] && !nfc->cs_array[1]) { dev_err(nfc->dev, "Assign a single native CS when using GPIOs\n"); return -EINVAL; @@ -1479,8 +1479,15 @@ static int anfc_probe(struct platform_device *pdev) static void anfc_remove(struct platform_device *pdev) { + int i; struct arasan_nfc *nfc = platform_get_drvdata(pdev); + for (i = 0; i < nfc->ncs; i++) { + if (nfc->cs_array[i]) { + gpiod_put(nfc->cs_array[i]); + } + } + anfc_chips_cleanup(nfc); } diff --git a/drivers/mtd/nand/raw/atmel/pmecc.c b/drivers/mtd/nand/raw/atmel/pmecc.c index a22aab4ed4e8..3c7dee1be21d 100644 --- a/drivers/mtd/nand/raw/atmel/pmecc.c +++ b/drivers/mtd/nand/raw/atmel/pmecc.c @@ -380,10 +380,8 @@ atmel_pmecc_create_user(struct atmel_pmecc *pmecc, user->delta = user->dmu + req->ecc.strength + 1; gf_tables = atmel_pmecc_get_gf_tables(req); - if (IS_ERR(gf_tables)) { - kfree(user); + if (IS_ERR(gf_tables)) return ERR_CAST(gf_tables); - } user->gf_tables = gf_tables; diff --git a/drivers/mtd/nand/raw/diskonchip.c b/drivers/mtd/nand/raw/diskonchip.c index 2068025d5639..594e13a852c4 100644 --- a/drivers/mtd/nand/raw/diskonchip.c +++ b/drivers/mtd/nand/raw/diskonchip.c @@ -1098,7 +1098,7 @@ static inline int __init inftl_partscan(struct mtd_info *mtd, struct mtd_partiti (i == 0) && (ip->firstUnit > 0)) { parts[0].name = " DiskOnChip IPL / Media Header partition"; parts[0].offset = 0; - parts[0].size = mtd->erasesize * ip->firstUnit; + parts[0].size = (uint64_t)mtd->erasesize * ip->firstUnit; numparts = 1; } diff --git a/drivers/pci/msi/irqdomain.c b/drivers/pci/msi/irqdomain.c index cfd84a899c82..dfc15e7507f9 100644 --- a/drivers/pci/msi/irqdomain.c +++ b/drivers/pci/msi/irqdomain.c @@ -330,8 +330,11 @@ bool pci_msi_domain_supports(struct pci_dev *pdev, unsigned int feature_mask, domain = dev_get_msi_domain(&pdev->dev); - if (!domain || !irq_domain_is_hierarchy(domain)) - return mode == ALLOW_LEGACY; + if (!domain || !irq_domain_is_hierarchy(domain)) { + if (IS_ENABLED(CONFIG_PCI_MSI_ARCH_FALLBACKS)) + return mode == ALLOW_LEGACY; + return false; + } if (!irq_domain_is_msi_parent(domain)) { /* diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c index 2d117cb74832..053bb9fac6e3 100644 --- a/drivers/pci/msi/msi.c +++ b/drivers/pci/msi/msi.c @@ -429,6 +429,10 @@ int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec, if (WARN_ON_ONCE(dev->msi_enabled)) return -EINVAL; + /* Test for the availability of MSI support */ + if (!pci_msi_domain_supports(dev, 0, ALLOW_LEGACY)) + return -ENOTSUPP; + nvec = pci_msi_vec_count(dev); if (nvec < 0) return nvec; diff --git a/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c b/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c index 4c10cafded4e..530b571607c0 100644 --- a/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c +++ b/drivers/phy/broadcom/phy-brcm-usb-init-synopsys.c @@ -323,6 +323,12 @@ static void usb_init_common_7216(struct brcm_usb_init_params *params) void __iomem *ctrl = params->regs[BRCM_REGS_CTRL]; USB_CTRL_UNSET(ctrl, USB_PM, XHC_S2_CLK_SWITCH_EN); + + /* + * The PHY might be in a bad state if it is already powered + * up. Toggle the power just in case. + */ + USB_CTRL_SET(ctrl, USB_PM, USB_PWRDN); USB_CTRL_UNSET(ctrl, USB_PM, USB_PWRDN); /* 1 millisecond - for USB clocks to settle down */ diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index 96a0b1e111f3..a892e1d7e2d0 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -140,8 +140,10 @@ static struct phy_provider *of_phy_provider_lookup(struct device_node *node) return phy_provider; for_each_child_of_node(phy_provider->children, child) - if (child == node) + if (child == node) { + of_node_put(child); return phy_provider; + } } return ERR_PTR(-EPROBE_DEFER); @@ -577,8 +579,10 @@ static struct phy *_of_phy_get(struct device_node *np, int index) return ERR_PTR(-ENODEV); /* This phy type handled by the usb-phy subsystem for now */ - if (of_device_is_compatible(args.np, "usb-nop-xceiv")) - return ERR_PTR(-ENODEV); + if (of_device_is_compatible(args.np, "usb-nop-xceiv")) { + phy = ERR_PTR(-ENODEV); + goto out_put_node; + } mutex_lock(&phy_provider_mutex); phy_provider = of_phy_provider_lookup(args.np); @@ -600,6 +604,7 @@ out_put_module: out_unlock: mutex_unlock(&phy_provider_mutex); +out_put_node: of_node_put(args.np); return phy; @@ -685,7 +690,7 @@ void devm_phy_put(struct device *dev, struct phy *phy) if (!phy) return; - r = devres_destroy(dev, devm_phy_release, devm_phy_match, phy); + r = devres_release(dev, devm_phy_release, devm_phy_match, phy); dev_WARN_ONCE(dev, r, "couldn't find PHY resource\n"); } EXPORT_SYMBOL_GPL(devm_phy_put); @@ -1069,7 +1074,7 @@ void devm_phy_destroy(struct device *dev, struct phy *phy) { int r; - r = devres_destroy(dev, devm_phy_consume, devm_phy_match, phy); + r = devres_release(dev, devm_phy_consume, devm_phy_match, phy); dev_WARN_ONCE(dev, r, "couldn't find PHY resource\n"); } EXPORT_SYMBOL_GPL(devm_phy_destroy); @@ -1207,12 +1212,12 @@ EXPORT_SYMBOL_GPL(of_phy_provider_unregister); * of_phy_provider_unregister to unregister the phy provider. */ void devm_of_phy_provider_unregister(struct device *dev, - struct phy_provider *phy_provider) + struct phy_provider *phy_provider) { int r; - r = devres_destroy(dev, devm_phy_provider_release, devm_phy_match, - phy_provider); + r = devres_release(dev, devm_phy_provider_release, devm_phy_match, + phy_provider); dev_WARN_ONCE(dev, r, "couldn't find PHY provider device resource\n"); } EXPORT_SYMBOL_GPL(devm_of_phy_provider_unregister); diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c index d5a726c13e39..c697d01b2a2a 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-usb.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-usb.c @@ -1088,7 +1088,7 @@ static const struct qmp_phy_init_tbl sc8280xp_usb3_uniphy_rx_tbl[] = { QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FASTLOCK_FO_GAIN, 0x2f), QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FASTLOCK_COUNT_LOW, 0xff), QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FASTLOCK_COUNT_HIGH, 0x0f), - QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_SO_GAIN, 0x0a), + QMP_PHY_INIT_CFG(QSERDES_V5_RX_UCDR_FO_GAIN, 0x0a), QMP_PHY_INIT_CFG(QSERDES_V5_RX_VGA_CAL_CNTRL1, 0x54), QMP_PHY_INIT_CFG(QSERDES_V5_RX_VGA_CAL_CNTRL2, 0x0f), QMP_PHY_INIT_CFG(QSERDES_V5_RX_RX_EQU_ADAPTOR_CNTRL2, 0x0f), diff --git a/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c b/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c index 26b157f53f3d..9c231094ba35 100644 --- a/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c +++ b/drivers/phy/rockchip/phy-rockchip-naneng-combphy.c @@ -309,7 +309,7 @@ static int rockchip_combphy_parse_dt(struct device *dev, struct rockchip_combphy priv->ext_refclk = device_property_present(dev, "rockchip,ext-refclk"); - priv->phy_rst = devm_reset_control_array_get_exclusive(dev); + priv->phy_rst = devm_reset_control_get(dev, "phy"); if (IS_ERR(priv->phy_rst)) return dev_err_probe(dev, PTR_ERR(priv->phy_rst), "failed to get phy reset\n"); diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c index af3da303e2b1..cba515ce3444 100644 --- a/drivers/platform/x86/asus-nb-wmi.c +++ b/drivers/platform/x86/asus-nb-wmi.c @@ -590,6 +590,7 @@ static const struct key_entry asus_nb_wmi_keymap[] = { { KE_KEY, 0xC4, { KEY_KBDILLUMUP } }, { KE_KEY, 0xC5, { KEY_KBDILLUMDOWN } }, { KE_IGNORE, 0xC6, }, /* Ambient Light Sensor notification */ + { KE_IGNORE, 0xCF, }, /* AC mode */ { KE_KEY, 0xFA, { KEY_PROG2 } }, /* Lid flip action */ { KE_KEY, 0xBD, { KEY_PROG2 } }, /* Lid flip action on ROG xflow laptops */ { KE_END, 0}, diff --git a/drivers/power/supply/gpio-charger.c b/drivers/power/supply/gpio-charger.c index 68212b39785b..6139f736ecbe 100644 --- a/drivers/power/supply/gpio-charger.c +++ b/drivers/power/supply/gpio-charger.c @@ -67,6 +67,14 @@ static int set_charge_current_limit(struct gpio_charger *gpio_charger, int val) if (gpio_charger->current_limit_map[i].limit_ua <= val) break; } + + /* + * If a valid charge current limit isn't found, default to smallest + * current limitation for safety reasons. + */ + if (i >= gpio_charger->current_limit_map_size) + i = gpio_charger->current_limit_map_size - 1; + mapping = gpio_charger->current_limit_map[i]; for (i = 0; i < ndescs; i++) { diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c index 3d4f13da1ae8..4cc93cb79b8b 100644 --- a/drivers/scsi/megaraid/megaraid_sas_base.c +++ b/drivers/scsi/megaraid/megaraid_sas_base.c @@ -8904,8 +8904,11 @@ megasas_aen_polling(struct work_struct *work) (ld_target_id / MEGASAS_MAX_DEV_PER_CHANNEL), (ld_target_id % MEGASAS_MAX_DEV_PER_CHANNEL), 0); - if (sdev1) + if (sdev1) { + mutex_unlock(&instance->reset_mutex); megasas_remove_scsi_device(sdev1); + mutex_lock(&instance->reset_mutex); + } event_type = SCAN_VD_CHANNEL; break; diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c b/drivers/scsi/mpt3sas/mpt3sas_base.c index 8acf586dc8b2..a5d12b95fbd0 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.c +++ b/drivers/scsi/mpt3sas/mpt3sas_base.c @@ -7050,11 +7050,12 @@ _base_handshake_req_reply_wait(struct MPT3SAS_ADAPTER *ioc, int request_bytes, int i; u8 failed; __le32 *mfp; + int ret_val; /* make sure doorbell is not in use */ if ((ioc->base_readl_ext_retry(&ioc->chip->Doorbell) & MPI2_DOORBELL_USED)) { ioc_err(ioc, "doorbell is in use (line=%d)\n", __LINE__); - return -EFAULT; + goto doorbell_diag_reset; } /* clear pending doorbell interrupts from previous state changes */ @@ -7144,6 +7145,10 @@ _base_handshake_req_reply_wait(struct MPT3SAS_ADAPTER *ioc, int request_bytes, le32_to_cpu(mfp[i])); } return 0; + +doorbell_diag_reset: + ret_val = _base_diag_reset(ioc); + return ret_val; } /** diff --git a/drivers/scsi/qla1280.h b/drivers/scsi/qla1280.h index d309e2ca14de..dea2290b37d4 100644 --- a/drivers/scsi/qla1280.h +++ b/drivers/scsi/qla1280.h @@ -116,12 +116,12 @@ struct device_reg { uint16_t id_h; /* ID high */ uint16_t cfg_0; /* Configuration 0 */ #define ISP_CFG0_HWMSK 0x000f /* Hardware revision mask */ -#define ISP_CFG0_1020 BIT_0 /* ISP1020 */ -#define ISP_CFG0_1020A BIT_1 /* ISP1020A */ -#define ISP_CFG0_1040 BIT_2 /* ISP1040 */ -#define ISP_CFG0_1040A BIT_3 /* ISP1040A */ -#define ISP_CFG0_1040B BIT_4 /* ISP1040B */ -#define ISP_CFG0_1040C BIT_5 /* ISP1040C */ +#define ISP_CFG0_1020 1 /* ISP1020 */ +#define ISP_CFG0_1020A 2 /* ISP1020A */ +#define ISP_CFG0_1040 3 /* ISP1040 */ +#define ISP_CFG0_1040A 4 /* ISP1040A */ +#define ISP_CFG0_1040B 5 /* ISP1040B */ +#define ISP_CFG0_1040C 6 /* ISP1040C */ uint16_t cfg_1; /* Configuration 1 */ #define ISP_CFG1_F128 BIT_6 /* 128-byte FIFO threshold */ #define ISP_CFG1_F64 BIT_4|BIT_5 /* 128-byte FIFO threshold */ diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c index 7ceb982040a5..d0b55c1fa908 100644 --- a/drivers/scsi/storvsc_drv.c +++ b/drivers/scsi/storvsc_drv.c @@ -149,6 +149,8 @@ struct hv_fc_wwn_packet { */ static int vmstor_proto_version; +static bool hv_dev_is_fc(struct hv_device *hv_dev); + #define STORVSC_LOGGING_NONE 0 #define STORVSC_LOGGING_ERROR 1 #define STORVSC_LOGGING_WARN 2 @@ -1138,6 +1140,7 @@ static void storvsc_on_io_completion(struct storvsc_device *stor_device, * not correctly handle: * INQUIRY command with page code parameter set to 0x80 * MODE_SENSE command with cmd[2] == 0x1c + * MAINTENANCE_IN is not supported by HyperV FC passthrough * * Setup srb and scsi status so this won't be fatal. * We do this so we can distinguish truly fatal failues @@ -1145,7 +1148,9 @@ static void storvsc_on_io_completion(struct storvsc_device *stor_device, */ if ((stor_pkt->vm_srb.cdb[0] == INQUIRY) || - (stor_pkt->vm_srb.cdb[0] == MODE_SENSE)) { + (stor_pkt->vm_srb.cdb[0] == MODE_SENSE) || + (stor_pkt->vm_srb.cdb[0] == MAINTENANCE_IN && + hv_dev_is_fc(device))) { vstor_packet->vm_srb.scsi_status = 0; vstor_packet->vm_srb.srb_status = SRB_STATUS_SUCCESS; } diff --git a/drivers/spi/spi-intel-pci.c b/drivers/spi/spi-intel-pci.c index 4337ca51d7aa..5c0dec90eec1 100644 --- a/drivers/spi/spi-intel-pci.c +++ b/drivers/spi/spi-intel-pci.c @@ -86,6 +86,8 @@ static const struct pci_device_id intel_spi_pci_ids[] = { { PCI_VDEVICE(INTEL, 0xa324), (unsigned long)&cnl_info }, { PCI_VDEVICE(INTEL, 0xa3a4), (unsigned long)&cnl_info }, { PCI_VDEVICE(INTEL, 0xa823), (unsigned long)&cnl_info }, + { PCI_VDEVICE(INTEL, 0xe323), (unsigned long)&cnl_info }, + { PCI_VDEVICE(INTEL, 0xe423), (unsigned long)&cnl_info }, { }, }; MODULE_DEVICE_TABLE(pci, intel_spi_pci_ids); diff --git a/drivers/spi/spi-omap2-mcspi.c b/drivers/spi/spi-omap2-mcspi.c index ddf1c684bcc7..3cfd262c1abc 100644 --- a/drivers/spi/spi-omap2-mcspi.c +++ b/drivers/spi/spi-omap2-mcspi.c @@ -1521,10 +1521,10 @@ static int omap2_mcspi_probe(struct platform_device *pdev) } mcspi->ref_clk = devm_clk_get_optional_enabled(&pdev->dev, NULL); - if (mcspi->ref_clk) - mcspi->ref_clk_hz = clk_get_rate(mcspi->ref_clk); - else + if (IS_ERR(mcspi->ref_clk)) mcspi->ref_clk_hz = OMAP2_MCSPI_MAX_FREQ; + else + mcspi->ref_clk_hz = clk_get_rate(mcspi->ref_clk); ctlr->max_speed_hz = mcspi->ref_clk_hz; ctlr->min_speed_hz = mcspi->ref_clk_hz >> 15; diff --git a/drivers/watchdog/it87_wdt.c b/drivers/watchdog/it87_wdt.c index 843f9f8e3917..239947df613d 100644 --- a/drivers/watchdog/it87_wdt.c +++ b/drivers/watchdog/it87_wdt.c @@ -20,6 +20,8 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include <linux/bits.h> +#include <linux/dmi.h> #include <linux/init.h> #include <linux/io.h> #include <linux/kernel.h> @@ -40,6 +42,7 @@ #define VAL 0x2f /* Logical device Numbers LDN */ +#define EC 0x04 #define GPIO 0x07 /* Configuration Registers and Functions */ @@ -71,6 +74,12 @@ #define IT8784_ID 0x8784 #define IT8786_ID 0x8786 +/* Environment Controller Configuration Registers LDN=0x04 */ +#define SCR1 0xfa + +/* Environment Controller Bits SCR1 */ +#define WDT_PWRGD 0x20 + /* GPIO Configuration Registers LDN=0x07 */ #define WDTCTRL 0x71 #define WDTCFG 0x72 @@ -233,6 +242,21 @@ static int wdt_set_timeout(struct watchdog_device *wdd, unsigned int t) return ret; } +enum { + IT87_WDT_OUTPUT_THROUGH_PWRGD = BIT(0), +}; + +static const struct dmi_system_id it87_quirks[] = { + { + /* Qotom Q30900P (IT8786) */ + .matches = { + DMI_EXACT_MATCH(DMI_BOARD_NAME, "QCML04"), + }, + .driver_data = (void *)IT87_WDT_OUTPUT_THROUGH_PWRGD, + }, + {} +}; + static const struct watchdog_info ident = { .options = WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE | WDIOF_KEEPALIVEPING, .firmware_version = 1, @@ -254,8 +278,10 @@ static struct watchdog_device wdt_dev = { static int __init it87_wdt_init(void) { + const struct dmi_system_id *dmi_id; u8 chip_rev; u8 ctrl; + int quirks = 0; int rc; rc = superio_enter(); @@ -266,6 +292,10 @@ static int __init it87_wdt_init(void) chip_rev = superio_inb(CHIPREV) & 0x0f; superio_exit(); + dmi_id = dmi_first_match(it87_quirks); + if (dmi_id) + quirks = (long)dmi_id->driver_data; + switch (chip_type) { case IT8702_ID: max_units = 255; @@ -326,6 +356,15 @@ static int __init it87_wdt_init(void) superio_outb(0x00, WDTCTRL); } + if (quirks & IT87_WDT_OUTPUT_THROUGH_PWRGD) { + superio_select(EC); + ctrl = superio_inb(SCR1); + if (!(ctrl & WDT_PWRGD)) { + ctrl |= WDT_PWRGD; + superio_outb(ctrl, SCR1); + } + } + superio_exit(); if (timeout < 1 || timeout > max_units * 60) { diff --git a/drivers/watchdog/mtk_wdt.c b/drivers/watchdog/mtk_wdt.c index 0559d9f2d97b..66bb68ceb14c 100644 --- a/drivers/watchdog/mtk_wdt.c +++ b/drivers/watchdog/mtk_wdt.c @@ -10,6 +10,7 @@ */ #include <dt-bindings/reset/mt2712-resets.h> +#include <dt-bindings/reset/mediatek,mt6735-wdt.h> #include <dt-bindings/reset/mediatek,mt6795-resets.h> #include <dt-bindings/reset/mt7986-resets.h> #include <dt-bindings/reset/mt8183-resets.h> @@ -81,6 +82,10 @@ static const struct mtk_wdt_data mt2712_data = { .toprgu_sw_rst_num = MT2712_TOPRGU_SW_RST_NUM, }; +static const struct mtk_wdt_data mt6735_data = { + .toprgu_sw_rst_num = MT6735_TOPRGU_RST_NUM, +}; + static const struct mtk_wdt_data mt6795_data = { .toprgu_sw_rst_num = MT6795_TOPRGU_SW_RST_NUM, }; @@ -448,6 +453,7 @@ static int mtk_wdt_resume(struct device *dev) static const struct of_device_id mtk_wdt_dt_ids[] = { { .compatible = "mediatek,mt2712-wdt", .data = &mt2712_data }, { .compatible = "mediatek,mt6589-wdt" }, + { .compatible = "mediatek,mt6735-wdt", .data = &mt6735_data }, { .compatible = "mediatek,mt6795-wdt", .data = &mt6795_data }, { .compatible = "mediatek,mt7986-wdt", .data = &mt7986_data }, { .compatible = "mediatek,mt8183-wdt", .data = &mt8183_data }, |