Linaro-mm-sig

linaro-mm-sig@lists.linaro.org

14 participants
3157 discussions

Re: [PATCH v1 17/20] drm/amdgpu: get rid of amdgpu_ttm_clear_buffer

by Christian König

On 11/4/25 09:35, Pierre-Eric Pelloux-Prayer wrote: > It's doing the same thing as amdgpu_fill_buffer(src_data=0), so drop it. > > The only caveat is that amdgpu_res_cleared() return value is only valid > right after allocation. > > Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer(a)amd.com> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 9 +-- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 86 ++++------------------ > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 3 - > 3 files changed, 18 insertions(+), 80 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c > index 4a69324bb730..410e9b68ff81 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c > @@ -723,15 +723,10 @@ int amdgpu_bo_create(struct amdgpu_device *adev, > > if (bp->flags & AMDGPU_GEM_CREATE_VRAM_CLEARED && > bo->tbo.resource->mem_type == TTM_PL_VRAM) { > - struct dma_fence *fence; > - > - r = amdgpu_ttm_clear_buffer(bo, bo->tbo.base.resv, &fence); > + r = amdgpu_fill_buffer(NULL, bo, 0, NULL, > + NULL, AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER); > if (unlikely(r)) > goto fail_unreserve; > - > - dma_resv_add_fence(bo->tbo.base.resv, fence, > - DMA_RESV_USAGE_KERNEL); > - dma_fence_put(fence); > } > if (!bp->resv) > amdgpu_bo_unreserve(bo); > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > index d88bdb2ac083..1f553c56f31d 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > @@ -2412,75 +2412,6 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, > return 0; > } > > -/** > - * amdgpu_ttm_clear_buffer - clear memory buffers > - * @bo: amdgpu buffer object > - * @resv: reservation object > - * @fence: dma_fence associated with the operation > - * > - * Clear the memory buffer resource. > - * > - * Returns: > - * 0 for success or a negative error code on failure. > - */ > -int amdgpu_ttm_clear_buffer(struct amdgpu_bo *bo, > - struct dma_resv *resv, > - struct dma_fence **fence) > -{ > - struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); > - struct amdgpu_ring *ring = adev->mman.buffer_funcs_rings[0]; > - struct amdgpu_ttm_entity *entity; > - struct amdgpu_res_cursor cursor; > - u64 addr; > - int r = 0; > - > - if (!adev->mman.buffer_funcs_enabled) > - return -EINVAL; > - > - if (!fence) > - return -EINVAL; > - entity = &adev->mman.clear_entities[0]; > - *fence = dma_fence_get_stub(); > - > - amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &cursor); > - > - mutex_lock(&entity->gart_window_lock); > - while (cursor.remaining) { > - struct dma_fence *next = NULL; > - u64 size; > - > - if (amdgpu_res_cleared(&cursor)) { > - amdgpu_res_next(&cursor, cursor.size); > - continue; > - } > - > - /* Never clear more than 256MiB at once to avoid timeouts */ > - size = min(cursor.size, 256ULL << 20); > - > - r = amdgpu_ttm_map_buffer(&entity->base, > - &bo->tbo, bo->tbo.resource, &cursor, > - entity->gart_window_id1, ring, false, &size, &addr, > - NULL, NULL); > - if (r) > - goto err; > - > - r = amdgpu_ttm_fill_mem(ring, &entity->base, 0, addr, size, resv, > - &next, true, > - AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER); > - if (r) > - goto err; > - > - dma_fence_put(*fence); > - *fence = next; > - > - amdgpu_res_next(&cursor, size); > - } > -err: > - mutex_unlock(&entity->gart_window_lock); > - > - return r; > -} > - > /** > * amdgpu_fill_buffer - fill a buffer with a given value > * @entity: optional entity to use. If NULL, the clearing entities will be > @@ -2508,6 +2439,9 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > struct amdgpu_res_cursor dst; > uint64_t cur_size, to; > int r, e, n_fences; > + /* The clear flag is only valid directly after allocation. */ > + bool consider_clear_flag = > + src_data == 0 && k_job_id == AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER; Absolutely clear NAK to that. Christian. > > /* The fences will be either added to the resv object or the last fence > * will be returned to the caller. In the latter case, all fill jobs will > @@ -2531,6 +2465,11 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > while (dst.remaining) { > cur_size = min(dst.size, 256ULL << 20); > > + if (consider_clear_flag && amdgpu_res_cleared(&dst)) { > + amdgpu_res_next(&dst, dst.size); > + continue; > + } > + > n_fences += 1; > amdgpu_res_next(&dst, cur_size); > } > @@ -2550,6 +2489,11 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &dst); > > while (dst.remaining) { > + if (consider_clear_flag && amdgpu_res_cleared(&dst)) { > + amdgpu_res_next(&dst, dst.size); > + continue; > + } > + > /* Never fill more than 256MiB at once to avoid timeouts */ > cur_size = min(dst.size, 256ULL << 20); > > @@ -2574,8 +2518,10 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > goto error; > } > > + > r = amdgpu_ttm_fill_mem(ring, &entity->base, > - src_data, to, cur_size, resv, > + src_data, to, cur_size, > + resv, > &fence, true, k_job_id); > if (r) { > mutex_unlock(&entity->gart_window_lock); > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > index c059a3d52b57..97e73919cb0c 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > @@ -182,9 +182,6 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, > struct dma_resv *resv, > struct dma_fence **fence, > bool vm_needs_flush, uint32_t copy_flags); > -int amdgpu_ttm_clear_buffer(struct amdgpu_bo *bo, > - struct dma_resv *resv, > - struct dma_fence **fence); > int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > struct amdgpu_bo *bo, > uint32_t src_data,

2 weeks

Re: [PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported through dma-buf

by Leon Romanovsky

On Tue, Nov 04, 2025 at 11:19:43AM -0800, Nicolin Chen wrote: > On Sun, Nov 02, 2025 at 10:00:48AM +0200, Leon Romanovsky wrote: > > Changelog: > > v6: > > * Fixed wrong error check from pcim_p2pdma_init(). > > * Documented pcim_p2pdma_provider() function. > > * Improved commit messages. > > * Added VFIO DMA-BUF selftest. > > * Added __counted_by(nr_ranges) annotation to struct vfio_device_feature_dma_buf. > > * Fixed error unwind when dma_buf_fd() fails. > > * Document latest changes to p2pmem. > > * Removed EXPORT_SYMBOL_GPL from pci_p2pdma_map_type. > > * Moved DMA mapping logic to DMA-BUF. > > * Removed types patch to avoid dependencies between subsystems. > > * Moved vfio_pci_dma_buf_move() in err_undo block. > > * Added nvgrace patch. > > I have verified this v6 using Jason's iommufd dmabuf branch: > https://github.com/jgunthorpe/linux/commits/iommufd_dmabuf/ > > by drafting a QEMU patch on top of Shameer's vSMMU v5 series: > https://github.com/nicolinc/qemu/commits/wip/iommufd_dmabuf/ > > with that, I see GPU BAR memory be correctly fetched in the QEMU: > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 0", offset: 0x0, size: 0x1000000 > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 2", offset: 0x0, size: 0x44f00000 > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 4", offset: 0x0, size: 0x17a0000000 > > Tested-by: Nicolin Chen <nicolinc(a)nvidia.com> Thanks a lot.

2 weeks

Re: [PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported through dma-buf

by Leon Romanovsky

2 weeks

Re: [EXTERNAL] Re: [PATCH net-next v4 2/6] net: ti: icssg-prueth: Add XSK pool helpers

by Jakub Kicinski

On Tue, 4 Nov 2025 14:23:24 +0530 Meghana Malladi wrote: > > I tried honoring Jakub's comment to avoid freeing the rx memory wherever > > necessary. > > > > "In case of icssg driver, freeing the rx memory is necessary as the > > rx descriptor memory is owned by the cppi dma controller and can be > > mapped to a single memory model (pages/xdp buffers) at a given time. > > In order to remap it, the memory needs to be freed and reallocated." > > Just to make sure we are on the same page, does the above explanation > make sense to you or do you want me to make any changes in this series > for v5 ? No. Based on your reply below you seem to understand what is being asked, so you're expected to do it. > >> I think you should: > >> - stop the H/W from processing incoming packets, > >> - spool all the pending packets > >> - attach/detach the xsk_pool > >> - refill the ring > >> - re-enable the H/W > > > > Current implementation follows the same sequence: > > 1. Does a channel teardown -> stop incoming traffic > > 2. free the rx descriptors from free queue and completion queue -> spool > > all pending packets/descriptors > > 3. attach/detach the xsk pool > > 4. allocate rx descriptors and fill the freeq after mapping them to the > > correct memory buffers -> refill the ring > > 5. restart the NAPI - re-enable the H/W to recv the traffic > > > > I am still working on skipping 2 and 4 steps but this will be a long > > shot. Need to make sure all corner cases are getting covered. If this > > approach looks doable without causing any regressions I might post it as > > a followup patch later in the future.

2 weeks

Re: [PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported through dma-buf

by Jason Gunthorpe

On Tue, Nov 04, 2025 at 11:19:43AM -0800, Nicolin Chen wrote: > On Sun, Nov 02, 2025 at 10:00:48AM +0200, Leon Romanovsky wrote: > > Changelog: > > v6: > > * Fixed wrong error check from pcim_p2pdma_init(). > > * Documented pcim_p2pdma_provider() function. > > * Improved commit messages. > > * Added VFIO DMA-BUF selftest. > > * Added __counted_by(nr_ranges) annotation to struct vfio_device_feature_dma_buf. > > * Fixed error unwind when dma_buf_fd() fails. > > * Document latest changes to p2pmem. > > * Removed EXPORT_SYMBOL_GPL from pci_p2pdma_map_type. > > * Moved DMA mapping logic to DMA-BUF. > > * Removed types patch to avoid dependencies between subsystems. > > * Moved vfio_pci_dma_buf_move() in err_undo block. > > * Added nvgrace patch. > > I have verified this v6 using Jason's iommufd dmabuf branch: > https://github.com/jgunthorpe/linux/commits/iommufd_dmabuf/ > > by drafting a QEMU patch on top of Shameer's vSMMU v5 series: > https://github.com/nicolinc/qemu/commits/wip/iommufd_dmabuf/ > > with that, I see GPU BAR memory be correctly fetched in the QEMU: > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 0", offset: 0x0, size: 0x1000000 > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 2", offset: 0x0, size: 0x44f00000 > vfio_region_dmabuf Device 0009:01:00.0, region "0009:01:00.0 BAR 4", offset: 0x0, size: 0x17a0000000 Great thanks! This means we finally have a solution to that follow_pfn lifetime problem in type 1! What a long journey :) For those following along this same flow will be used with KVM to allow it to map VFIO as well. Confidential Compute will require this because some arches can't put confidential MMIO (or RAM) into a VMA. Jason

2 weeks

Re: [PATCH v1 11/20] drm/amdgpu: use multiple entities in amdgpu_fill_buffer

by Christian König

On 11/4/25 09:35, Pierre-Eric Pelloux-Prayer wrote: > The benefits of using multiple entities is that multiple fill jobs > can run in parallel. Otherwise, even if the entity has access > to multiple engines, a burst of N independent jobs will all > run on the same engine because an entity guarantees the ordering > of execution matches the ordering of the submission. > > Callers can opt-out of this behavior by passing the entity they > want to use (see amdgpu_move_blit). That still sounds like a really bad idea to me. First of all we can't reserve so many fence slots in the release handler, previously we basically just relied on the fact that the BO will most likely be mostly idle. I think we should just use a single SDMA engine for each clear and distribute clearing different BOs over multiple engines. Regards, Christian. > > Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer(a)amd.com> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 84 ++++++++++++++++++------- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 1 + > 2 files changed, 64 insertions(+), 21 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > index c357a6d9763a..839ea8c7f6be 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > @@ -2224,6 +2224,7 @@ u32 amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) > adev->mman.clear_entities = kcalloc(num_clear_entities, > sizeof(struct amdgpu_ttm_entity), > GFP_KERNEL); > + atomic_set(&adev->mman.next_clear_entity, 0); > if (!adev->mman.clear_entities) > goto error_free_entity; > > @@ -2498,10 +2499,12 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > { > struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); > struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; > + struct dma_fence *fences[TTM_FENCES_MAX_SLOT_COUNT] = {}; > struct dma_fence *fence = NULL; > struct dma_resv *resv = NULL; > struct amdgpu_res_cursor dst; > - int r; > + uint64_t cur_size, to; > + int r, e, n_fences; > > /* The fences will be either added to the resv object or the last fence > * will be returned to the caller. In the latter case, all fill jobs will > @@ -2515,53 +2518,92 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_entity *entity, > } > > if (!entity) { > - entity = &adev->mman.clear_entities[0]; > resv = &bo->tbo.base._resv; > - r = dma_resv_reserve_fences(resv, 1); > + > + /* Determine how much fences we're going to add to the > + * resv object. > + */ > + n_fences = 0; > + amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &dst); > + while (dst.remaining) { > + cur_size = min(dst.size, 256ULL << 20); > + > + n_fences += 1; > + amdgpu_res_next(&dst, cur_size); > + } > + if (n_fences == 0) > + return 0; > + > + /* One slot per entity at most. */ > + n_fences = MIN(n_fences, adev->mman.num_clear_entities); > + > + r = dma_resv_reserve_fences(resv, n_fences); > if (r) > return r; > + } else { > + mutex_lock(&entity->gart_window_lock); > } > > amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &dst); > > - mutex_lock(&entity->gart_window_lock); > while (dst.remaining) { > - struct dma_fence *next; > - uint64_t cur_size, to; > - > /* Never fill more than 256MiB at once to avoid timeouts */ > cur_size = min(dst.size, 256ULL << 20); > > + if (resv) { > + /* Pick a new entity for each partial clear so they can > + * execute in parallel. > + */ > + e = atomic_inc_return(&adev->mman.next_clear_entity) % > + adev->mman.num_clear_entities; > + entity = &adev->mman.clear_entities[e]; > + mutex_lock(&entity->gart_window_lock); > + } > + > r = amdgpu_ttm_map_buffer(&entity->base, > &bo->tbo, bo->tbo.resource, &dst, > entity->gart_window_id1, ring, false, > &cur_size, &to, > dependency, > resv); > - if (r) > + if (r) { > + mutex_unlock(&entity->gart_window_lock); > goto error; > + } > > r = amdgpu_ttm_fill_mem(ring, &entity->base, > src_data, to, cur_size, resv, > - &next, true, k_job_id); > - if (r) > + &fence, true, k_job_id); > + if (r) { > + mutex_unlock(&entity->gart_window_lock); > goto error; > - > - if (resv) { > - dma_resv_add_fence(resv, next, DMA_RESV_USAGE_KERNEL); > - dma_fence_put(next); > - } else { > - dma_fence_put(fence); > - fence = next; > } > > amdgpu_res_next(&dst, cur_size); > + > + if (resv) { > + /* Delay the addition of the fences to resv, otherwise the next partial > + * clears will depend on this one. > + */ > + fences[e] = fence; > + mutex_unlock(&entity->gart_window_lock); > + } else { > + dma_fence_put(*f); > + *f = fence; > + } > } > error: > - mutex_unlock(&entity->gart_window_lock); > - if (f) > - *f = dma_fence_get(fence); > - dma_fence_put(fence); > + if (resv) { > + for (e = 0; e < adev->mman.num_clear_entities; e++) { > + if (fences[e]) { > + dma_resv_add_fence(resv, fences[e], DMA_RESV_USAGE_KERNEL); > + dma_fence_put(fences[e]); > + } > + } > + } else { > + mutex_unlock(&entity->gart_window_lock); > + } > + > return r; > } > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > index 38df2b5b4bc7..3fc31c7c6bfe 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > @@ -73,6 +73,7 @@ struct amdgpu_mman { > > struct amdgpu_ttm_entity default_entity; /* has no gart windows */ > struct amdgpu_ttm_entity *clear_entities; > + atomic_t next_clear_entity; > u32 num_clear_entities; > struct amdgpu_ttm_entity move_entities[TTM_FENCES_MAX_SLOT_COUNT]; > u32 num_move_entities;

2 weeks

Re: [PATCH v1 03/20] drm/amdgpu: remove direct_submit arg from amdgpu_copy_buffer

by Christian König

On 11/4/25 09:35, Pierre-Eric Pelloux-Prayer wrote: > It was always false. > > Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer(a)amd.com> Reviewed-by: Christian König <christian.koenig(a)amd.com> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c | 2 +- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 20 +++++++------------ > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 2 +- > drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 2 +- > 4 files changed, 10 insertions(+), 16 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c > index 199693369c7c..02c2479a8840 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c > @@ -39,7 +39,7 @@ static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size, > for (i = 0; i < n; i++) { > struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; > r = amdgpu_copy_buffer(ring, saddr, daddr, size, NULL, &fence, > - false, false, 0); > + false, 0); > if (r) > goto exit_do_move; > r = dma_fence_wait(fence, false); > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > index c66f00434991..fce22712396b 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > @@ -354,7 +354,7 @@ static int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, > } > > r = amdgpu_copy_buffer(ring, from, to, cur_size, resv, > - &next, false, true, copy_flags); > + &next, true, copy_flags); > if (r) > goto error; > > @@ -2211,16 +2211,13 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) > } > > static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev, > - bool direct_submit, > unsigned int num_dw, > struct dma_resv *resv, > bool vm_needs_flush, > struct amdgpu_job **job, > bool delayed, u64 k_job_id) > { > - enum amdgpu_ib_pool_type pool = direct_submit ? > - AMDGPU_IB_POOL_DIRECT : > - AMDGPU_IB_POOL_DELAYED; > + enum amdgpu_ib_pool_type pool = AMDGPU_IB_POOL_DELAYED; > int r; > struct drm_sched_entity *entity = delayed ? &adev->mman.low_pr : > &adev->mman.high_pr; > @@ -2246,7 +2243,7 @@ static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev, > int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, > uint64_t dst_offset, uint32_t byte_count, > struct dma_resv *resv, > - struct dma_fence **fence, bool direct_submit, > + struct dma_fence **fence, > bool vm_needs_flush, uint32_t copy_flags) > { > struct amdgpu_device *adev = ring->adev; > @@ -2256,7 +2253,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, > unsigned int i; > int r; > > - if (!direct_submit && !ring->sched.ready) { > + if (!ring->sched.ready) { > dev_err(adev->dev, > "Trying to move memory with ring turned off.\n"); > return -EINVAL; > @@ -2265,7 +2262,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, > max_bytes = adev->mman.buffer_funcs->copy_max_bytes; > num_loops = DIV_ROUND_UP(byte_count, max_bytes); > num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8); > - r = amdgpu_ttm_prepare_job(adev, direct_submit, num_dw, > + r = amdgpu_ttm_prepare_job(adev, num_dw, > resv, vm_needs_flush, &job, false, > AMDGPU_KERNEL_JOB_ID_TTM_COPY_BUFFER); > if (r) > @@ -2283,10 +2280,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, > > amdgpu_ring_pad_ib(ring, &job->ibs[0]); > WARN_ON(job->ibs[0].length_dw > num_dw); > - if (direct_submit) > - r = amdgpu_job_submit_direct(job, ring, fence); > - else > - *fence = amdgpu_job_submit(job); > + *fence = amdgpu_job_submit(job); > if (r) > goto error_free; > > @@ -2315,7 +2309,7 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data, > max_bytes = adev->mman.buffer_funcs->fill_max_bytes; > num_loops = DIV_ROUND_UP_ULL(byte_count, max_bytes); > num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->fill_num_dw, 8); > - r = amdgpu_ttm_prepare_job(adev, false, num_dw, resv, vm_needs_flush, > + r = amdgpu_ttm_prepare_job(adev, num_dw, resv, vm_needs_flush, > &job, delayed, k_job_id); > if (r) > return r; > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > index 577ee04ce0bf..50e40380fe95 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h > @@ -166,7 +166,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, > int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, > uint64_t dst_offset, uint32_t byte_count, > struct dma_resv *resv, > - struct dma_fence **fence, bool direct_submit, > + struct dma_fence **fence, > bool vm_needs_flush, uint32_t copy_flags); > int amdgpu_ttm_clear_buffer(struct amdgpu_bo *bo, > struct dma_resv *resv, > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c > index 46c84fc60af1..378af0b2aaa9 100644 > --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c > @@ -153,7 +153,7 @@ svm_migrate_copy_memory_gart(struct amdgpu_device *adev, dma_addr_t *sys, > } > > r = amdgpu_copy_buffer(ring, gart_s, gart_d, size * PAGE_SIZE, > - NULL, &next, false, true, 0); > + NULL, &next, true, 0); > if (r) { > dev_err(adev->dev, "fail %d to copy memory\n", r); > goto out_unlock;

2 weeks

Re: [PATCH v1 02/20] drm/ttm: rework pipelined eviction fence handling

by Christian König

On 11/4/25 09:35, Pierre-Eric Pelloux-Prayer wrote: > Until now ttm stored a single pipelined eviction fence which means > drivers had to use a single entity for these evictions. > > To lift this requirement, this commit allows up to 8 entities to > be used. > > Ideally a dma_resv object would have been used as a container of > the eviction fences, but the locking rules makes it complex. > dma_resv all have the same ww_class, which means "Attempting to > lock more mutexes after ww_acquire_done." is an error. > > One alternative considered was to introduced a 2nd ww_class for > specific resv to hold a single "transient" lock (= the resv lock > would only be held for a short period, without taking any other > locks). > > The other option, is to statically reserve a fence array, and > extend the existing code to deal with N fences, instead of 1. > > The driver is still responsible to reserve the correct number > of fence slots. > > Lastly ttm_resource_manager.pipelined_eviction.n_fences is > initialized to 1, so the new behavior is opt-in. > > Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer(a)amd.com> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 8 ++- > .../gpu/drm/ttm/tests/ttm_bo_validate_test.c | 13 +++-- > drivers/gpu/drm/ttm/tests/ttm_resource_test.c | 5 +- > drivers/gpu/drm/ttm/ttm_bo.c | 56 ++++++++++++------- > drivers/gpu/drm/ttm/ttm_bo_util.c | 36 ++++++++++-- > drivers/gpu/drm/ttm/ttm_resource.c | 45 ++++++++++----- > include/drm/ttm/ttm_resource.h | 34 ++++++++--- > 7 files changed, 139 insertions(+), 58 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > index 326476089db3..c66f00434991 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c > @@ -2156,7 +2156,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) > { > struct ttm_resource_manager *man = ttm_manager_type(&adev->mman.bdev, TTM_PL_VRAM); > uint64_t size; > - int r; > + int r, i; > > if (!adev->mman.initialized || amdgpu_in_reset(adev) || > adev->mman.buffer_funcs_enabled == enable || adev->gmc.is_app_apu) > @@ -2190,8 +2190,10 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) > } else { > drm_sched_entity_destroy(&adev->mman.high_pr); > drm_sched_entity_destroy(&adev->mman.low_pr); > - dma_fence_put(man->move); > - man->move = NULL; > + for (i = 0; i < TTM_FENCES_MAX_SLOT_COUNT; i++) { > + dma_fence_put(man->pipelined_eviction.fences[i]); > + man->pipelined_eviction.fences[i] = NULL; > + } > } > > /* this just adjusts TTM size idea, which sets lpfn to the correct value */ > diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > index 3148f5d3dbd6..1396674e1923 100644 > --- a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > +++ b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > @@ -651,7 +651,8 @@ static void ttm_bo_validate_move_fence_signaled(struct kunit *test) > int err; > > man = ttm_manager_type(priv->ttm_dev, mem_type); > - man->move = dma_fence_get_stub(); > + man->pipelined_eviction.n_fences = 1; > + man->pipelined_eviction.fences[0] = dma_fence_get_stub(); > > bo = ttm_bo_kunit_init(test, test->priv, size, NULL); > bo->type = bo_type; > @@ -668,7 +669,7 @@ static void ttm_bo_validate_move_fence_signaled(struct kunit *test) > KUNIT_EXPECT_EQ(test, ctx.bytes_moved, size); > > ttm_bo_put(bo); > - dma_fence_put(man->move); > + dma_fence_put(man->pipelined_eviction.fences[0]); > } > > static const struct ttm_bo_validate_test_case ttm_bo_validate_wait_cases[] = { > @@ -732,9 +733,10 @@ static void ttm_bo_validate_move_fence_not_signaled(struct kunit *test) > > spin_lock_init(&fence_lock); > man = ttm_manager_type(priv->ttm_dev, fst_mem); > - man->move = alloc_mock_fence(test); > + man->pipelined_eviction.n_fences = 1; > + man->pipelined_eviction.fences[0] = alloc_mock_fence(test); > > - task = kthread_create(threaded_fence_signal, man->move, "move-fence-signal"); > + task = kthread_create(threaded_fence_signal, man->pipelined_eviction.fences[0], "move-fence-signal"); > if (IS_ERR(task)) > KUNIT_FAIL(test, "Couldn't create move fence signal task\n"); > > @@ -742,7 +744,8 @@ static void ttm_bo_validate_move_fence_not_signaled(struct kunit *test) > err = ttm_bo_validate(bo, placement_val, &ctx_val); > dma_resv_unlock(bo->base.resv); > > - dma_fence_wait_timeout(man->move, false, MAX_SCHEDULE_TIMEOUT); > + dma_fence_wait_timeout(man->pipelined_eviction.fences[0], false, MAX_SCHEDULE_TIMEOUT); > + man->pipelined_eviction.fences[0] = NULL; > > KUNIT_EXPECT_EQ(test, err, 0); > KUNIT_EXPECT_EQ(test, ctx_val.bytes_moved, size); > diff --git a/drivers/gpu/drm/ttm/tests/ttm_resource_test.c b/drivers/gpu/drm/ttm/tests/ttm_resource_test.c > index e6ea2bd01f07..6dfdf759a491 100644 > --- a/drivers/gpu/drm/ttm/tests/ttm_resource_test.c > +++ b/drivers/gpu/drm/ttm/tests/ttm_resource_test.c > @@ -207,6 +207,7 @@ static void ttm_resource_manager_init_basic(struct kunit *test) > struct ttm_resource_test_priv *priv = test->priv; > struct ttm_resource_manager *man; > size_t size = SZ_16K; > + int i; > > man = kunit_kzalloc(test, sizeof(*man), GFP_KERNEL); > KUNIT_ASSERT_NOT_NULL(test, man); > @@ -216,8 +217,8 @@ static void ttm_resource_manager_init_basic(struct kunit *test) > KUNIT_ASSERT_PTR_EQ(test, man->bdev, priv->devs->ttm_dev); > KUNIT_ASSERT_EQ(test, man->size, size); > KUNIT_ASSERT_EQ(test, man->usage, 0); > - KUNIT_ASSERT_NULL(test, man->move); > - KUNIT_ASSERT_NOT_NULL(test, &man->move_lock); > + for (i = 0; i < TTM_FENCES_MAX_SLOT_COUNT; i++) > + KUNIT_ASSERT_NULL(test, man->pipelined_eviction.fences[i]); > > for (int i = 0; i < TTM_MAX_BO_PRIORITY; ++i) > KUNIT_ASSERT_TRUE(test, list_empty(&man->lru[i])); > diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c > index f4d9e68b21e7..bc6d4a6c6d70 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo.c > +++ b/drivers/gpu/drm/ttm/ttm_bo.c > @@ -658,34 +658,48 @@ void ttm_bo_unpin(struct ttm_buffer_object *bo) > EXPORT_SYMBOL(ttm_bo_unpin); > > /* > - * Add the last move fence to the BO as kernel dependency and reserve a new > - * fence slot. > + * Add the pipelined eviction fencesto the BO as kernel dependency and reserve new > + * fence slots. > */ > -static int ttm_bo_add_move_fence(struct ttm_buffer_object *bo, > - struct ttm_resource_manager *man, > - bool no_wait_gpu) > +static int ttm_bo_add_pipelined_eviction_fences(struct ttm_buffer_object *bo, > + struct ttm_resource_manager *man, > + bool no_wait_gpu) > { > + struct dma_fence *fences_to_add[TTM_FENCES_MAX_SLOT_COUNT] = {}; > struct dma_fence *fence; > - int ret; > + bool all_signaled = true, signaled; > + int i, n = 0; > > - spin_lock(&man->move_lock); > - fence = dma_fence_get(man->move); > - spin_unlock(&man->move_lock); > + spin_lock(&man->pipelined_eviction.lock); > + for (i = 0; i < man->pipelined_eviction.n_fences; i++) { > + fence = man->pipelined_eviction.fences[i]; > + if (!fence) > + continue; > + signaled = dma_fence_is_signaled(fence); > > - if (!fence) > + if (signaled) { > + dma_fence_put(man->pipelined_eviction.fences[i]); > + man->pipelined_eviction.fences[i] = NULL; Please completely drop that, only check if the fences are signaled when the no_wait_gpu flag is set. > + } else { > + all_signaled = false; > + if (no_wait_gpu) { > + spin_unlock(&man->pipelined_eviction.lock); > + return -EBUSY; > + } > + fences_to_add[n++] = dma_fence_get(fence); > + } > + } > + spin_unlock(&man->pipelined_eviction.lock); > + > + if (all_signaled) > return 0; > > - if (no_wait_gpu) { > - ret = dma_fence_is_signaled(fence) ? 0 : -EBUSY; > - dma_fence_put(fence); > - return ret; > + for (i = 0; i < n; i++) { > + dma_resv_add_fence(bo->base.resv, fences_to_add[i], DMA_RESV_USAGE_KERNEL); > + dma_fence_put(fences_to_add[i]); > } > > - dma_resv_add_fence(bo->base.resv, fence, DMA_RESV_USAGE_KERNEL); > - > - ret = dma_resv_reserve_fences(bo->base.resv, 1); > - dma_fence_put(fence); > - return ret; > + return dma_resv_reserve_fences(bo->base.resv, TTM_FENCES_MAX_SLOT_COUNT); Please separate out a patch where the call to dma_resv_reserve_fences() is removed here. > } > > /** > @@ -718,7 +732,7 @@ static int ttm_bo_alloc_resource(struct ttm_buffer_object *bo, > int i, ret; > > ticket = dma_resv_locking_ctx(bo->base.resv); > - ret = dma_resv_reserve_fences(bo->base.resv, 1); > + ret = dma_resv_reserve_fences(bo->base.resv, TTM_FENCES_MAX_SLOT_COUNT); > if (unlikely(ret)) > return ret; > > @@ -757,7 +771,7 @@ static int ttm_bo_alloc_resource(struct ttm_buffer_object *bo, > return ret; > } > > - ret = ttm_bo_add_move_fence(bo, man, ctx->no_wait_gpu); > + ret = ttm_bo_add_pipelined_eviction_fences(bo, man, ctx->no_wait_gpu); > if (unlikely(ret)) { > ttm_resource_free(bo, res); > if (ret == -EBUSY) > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > index acbbca9d5c92..ada8af965acf 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > @@ -258,7 +258,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, > ret = dma_resv_trylock(&fbo->base.base._resv); > WARN_ON(!ret); > > - ret = dma_resv_reserve_fences(&fbo->base.base._resv, 1); > + ret = dma_resv_reserve_fences(&fbo->base.base._resv, TTM_FENCES_MAX_SLOT_COUNT); > if (ret) { > dma_resv_unlock(&fbo->base.base._resv); > kfree(fbo); > @@ -646,6 +646,8 @@ static void ttm_bo_move_pipeline_evict(struct ttm_buffer_object *bo, > { > struct ttm_device *bdev = bo->bdev; > struct ttm_resource_manager *from; > + struct dma_fence *tmp; > + int i, free_slot = -1; > > from = ttm_manager_type(bdev, bo->resource->mem_type); > > @@ -653,13 +655,35 @@ static void ttm_bo_move_pipeline_evict(struct ttm_buffer_object *bo, > * BO doesn't have a TTM we need to bind/unbind. Just remember > * this eviction and free up the allocation > */ > - spin_lock(&from->move_lock); > - if (!from->move || dma_fence_is_later(fence, from->move)) { > - dma_fence_put(from->move); > - from->move = dma_fence_get(fence); > + spin_lock(&from->pipelined_eviction.lock); > + for (i = 0; i < from->pipelined_eviction.n_fences; i++) { > + tmp = from->pipelined_eviction.fences[i]; > + if (!tmp) { > + if (free_slot < 0) > + free_slot = i; > + continue; Just break here. > + } > + if (fence->context != tmp->context) > + continue; > + if (dma_fence_is_later(fence, tmp)) { > + dma_fence_put(tmp); > + free_slot = i; > + break; > + } > + goto unlock; > + } > + if (free_slot >= 0) { Drop free_slot and check i here. > + from->pipelined_eviction.fences[free_slot] = dma_fence_get(fence); > + } else { > + WARN(1, "not enough fence slots for all fence contexts"); > + spin_unlock(&from->pipelined_eviction.lock); > + dma_fence_wait(fence, false); > + goto end; > } > - spin_unlock(&from->move_lock); > > +unlock: > + spin_unlock(&from->pipelined_eviction.lock); > +end: > ttm_resource_free(bo, &bo->resource); > } > > diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c > index e2c82ad07eb4..ae0d4621cc55 100644 > --- a/drivers/gpu/drm/ttm/ttm_resource.c > +++ b/drivers/gpu/drm/ttm/ttm_resource.c > @@ -523,14 +523,19 @@ void ttm_resource_manager_init(struct ttm_resource_manager *man, > { > unsigned i; > > - spin_lock_init(&man->move_lock); > man->bdev = bdev; > man->size = size; > man->usage = 0; > > for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) > INIT_LIST_HEAD(&man->lru[i]); > - man->move = NULL; > + spin_lock_init(&man->pipelined_eviction.lock); > + for (i = 0; i < TTM_FENCES_MAX_SLOT_COUNT; i++) > + man->pipelined_eviction.fences[i] = NULL; > + /* Can be overridden by drivers that wants to use more than 1 entity > + * for moves and evictions (limited to TTM_FENCES_MAX_SLOT_COUNT). > + */ > + man->pipelined_eviction.n_fences = 1; > } > EXPORT_SYMBOL(ttm_resource_manager_init); > > @@ -551,7 +556,7 @@ int ttm_resource_manager_evict_all(struct ttm_device *bdev, > .no_wait_gpu = false, > }; > struct dma_fence *fence; > - int ret; > + int ret, i; > > do { > ret = ttm_bo_evict_first(bdev, man, &ctx); > @@ -561,18 +566,32 @@ int ttm_resource_manager_evict_all(struct ttm_device *bdev, > if (ret && ret != -ENOENT) > return ret; > > - spin_lock(&man->move_lock); > - fence = dma_fence_get(man->move); > - spin_unlock(&man->move_lock); > + ret = 0; > > - if (fence) { > - ret = dma_fence_wait(fence, false); > - dma_fence_put(fence); > - if (ret) > - return ret; > - } > + do { > + fence = NULL; > > - return 0; > + spin_lock(&man->pipelined_eviction.lock); > + for (i = 0; i < man->pipelined_eviction.n_fences; i++) { > + fence = man->pipelined_eviction.fences[i]; > + man->pipelined_eviction.fences[i] = NULL; Drop that. We should never set man->pipelined_eviction.fences to NULL. Potentially even initialize all move fences with a stub fence. > + if (fence) > + break; > + } > + spin_unlock(&man->pipelined_eviction.lock); > + > + if (fence) { > + ret = dma_fence_wait(fence, false); > + dma_fence_put(fence); > + > + if (ret) > + break; > + } else { > + break; > + } > + } while (1); > + > + return ret; > } > EXPORT_SYMBOL(ttm_resource_manager_evict_all); > > diff --git a/include/drm/ttm/ttm_resource.h b/include/drm/ttm/ttm_resource.h > index f49daa504c36..898c429b37ad 100644 > --- a/include/drm/ttm/ttm_resource.h > +++ b/include/drm/ttm/ttm_resource.h > @@ -50,6 +50,15 @@ struct io_mapping; > struct sg_table; > struct scatterlist; > > +/** > + * define TTM_FENCES_MAX_SLOT_COUNT - How many entities can be used for evictions > + * > + * Pipelined evictions can be spread on multiple entities. This > + * is the max number of entities that can be used by the driver > + * for that purpose. > + */ > +#define TTM_FENCES_MAX_SLOT_COUNT 8 Make that TTM_NUM_MOVE_FENCES. > + > /** > * enum ttm_lru_item_type - enumerate ttm_lru_item subclasses > */ > @@ -180,8 +189,10 @@ struct ttm_resource_manager_func { > * @size: Size of the managed region. > * @bdev: ttm device this manager belongs to > * @func: structure pointer implementing the range manager. See above > - * @move_lock: lock for move fence > - * @move: The fence of the last pipelined move operation. > + * @pipelined_eviction.lock: lock for eviction fences > + * @pipelined_eviction.n_fences: The number of fences allowed in the array. If > + * 0, pipelined evictions aren't used. > + * @pipelined_eviction.fences: The fences of the last pipelined move operation. > * @lru: The lru list for this memory type. > * > * This structure is used to identify and manage memory types for a device. > @@ -195,12 +206,15 @@ struct ttm_resource_manager { > struct ttm_device *bdev; > uint64_t size; > const struct ttm_resource_manager_func *func; > - spinlock_t move_lock; > > - /* > - * Protected by @move_lock. > + /* This is very similar to a dma_resv object, but locking rules make > + * it difficult to use a it in this context. > */ > - struct dma_fence *move; > + struct { > + spinlock_t lock; > + int n_fences; > + struct dma_fence *fences[TTM_FENCES_MAX_SLOT_COUNT]; > + } pipelined_eviction; Drop the separate structure, just make move an array instead. And also drop n_fences. Just always take a look at all fences. Regards, Christian. > > /* > * Protected by the bdev->lru_lock. > @@ -421,8 +435,12 @@ static inline bool ttm_resource_manager_used(struct ttm_resource_manager *man) > static inline void > ttm_resource_manager_cleanup(struct ttm_resource_manager *man) > { > - dma_fence_put(man->move); > - man->move = NULL; > + int i; > + > + for (i = 0; i < TTM_FENCES_MAX_SLOT_COUNT; i++) { > + dma_fence_put(man->pipelined_eviction.fences[i]); > + man->pipelined_eviction.fences[i] = NULL; > + } > } > > void ttm_lru_bulk_move_init(struct ttm_lru_bulk_move *bulk);

2 weeks

[PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported through dma-buf

by Leon Romanovsky

Changelog: v6: * Fixed wrong error check from pcim_p2pdma_init(). * Documented pcim_p2pdma_provider() function. * Improved commit messages. * Added VFIO DMA-BUF selftest. * Added __counted_by(nr_ranges) annotation to struct vfio_device_feature_dma_buf. * Fixed error unwind when dma_buf_fd() fails. * Document latest changes to p2pmem. * Removed EXPORT_SYMBOL_GPL from pci_p2pdma_map_type. * Moved DMA mapping logic to DMA-BUF. * Removed types patch to avoid dependencies between subsystems. * Moved vfio_pci_dma_buf_move() in err_undo block. * Added nvgrace patch. v5: https://lore.kernel.org/all/cover.1760368250.git.leon@kernel.org * Rebased on top of v6.18-rc1. * Added more validation logic to make sure that DMA-BUF length doesn't overflow in various scenarios. * Hide kernel config from the users. * Fixed type conversion issue. DMA ranges are exposed with u64 length, but DMA-BUF uses "unsigned int" as a length for SG entries. * Added check to prevent from VFIO drivers which reports BAR size different from PCI, do not use DMA-BUF functionality. v4: https://lore.kernel.org/all/cover.1759070796.git.leon@kernel.org * Split pcim_p2pdma_provider() to two functions, one that initializes array of providers and another to return right provider pointer. v3: https://lore.kernel.org/all/cover.1758804980.git.leon@kernel.org * Changed pcim_p2pdma_enable() to be pcim_p2pdma_provider(). * Cache provider in vfio_pci_dma_buf struct instead of BAR index. * Removed misleading comment from pcim_p2pdma_provider(). * Moved MMIO check to be in pcim_p2pdma_provider(). v2: https://lore.kernel.org/all/cover.1757589589.git.leon@kernel.org/ * Added extra patch which adds new CONFIG, so next patches can reuse * it. * Squashed "PCI/P2PDMA: Remove redundant bus_offset from map state" into the other patch. * Fixed revoke calls to be aligned with true->false semantics. * Extended p2pdma_providers to be per-BAR and not global to whole * device. * Fixed possible race between dmabuf states and revoke. * Moved revoke to PCI BAR zap block. v1: https://lore.kernel.org/all/cover.1754311439.git.leon@kernel.org * Changed commit messages. * Reused DMA_ATTR_MMIO attribute. * Returned support for multiple DMA ranges per-dMABUF. v0: https://lore.kernel.org/all/cover.1753274085.git.leonro@nvidia.com --------------------------------------------------------------------------- Based on "[PATCH v6 00/16] dma-mapping: migrate to physical address-based API" https://lore.kernel.org/all/cover.1757423202.git.leonro@nvidia.com/ series. --------------------------------------------------------------------------- This series extends the VFIO PCI subsystem to support exporting MMIO regions from PCI device BARs as dma-buf objects, enabling safe sharing of non-struct page memory with controlled lifetime management. This allows RDMA and other subsystems to import dma-buf FDs and build them into memory regions for PCI P2P operations. The series supports a use case for SPDK where a NVMe device will be owned by SPDK through VFIO but interacting with a RDMA device. The RDMA device may directly access the NVMe CMB or directly manipulate the NVMe device's doorbell using PCI P2P. However, as a general mechanism, it can support many other scenarios with VFIO. This dmabuf approach can be usable by iommufd as well for generic and safe P2P mappings. In addition to the SPDK use-case mentioned above, the capability added in this patch series can also be useful when a buffer (located in device memory such as VRAM) needs to be shared between any two dGPU devices or instances (assuming one of them is bound to VFIO PCI) as long as they are P2P DMA compatible. The implementation provides a revocable attachment mechanism using dma-buf move operations. MMIO regions are normally pinned as BARs don't change physical addresses, but access is revoked when the VFIO device is closed or a PCI reset is issued. This ensures kernel self-defense against potentially hostile userspace. The series includes significant refactoring of the PCI P2PDMA subsystem to separate core P2P functionality from memory allocation features, making it more modular and suitable for VFIO use cases that don't need struct page support. ----------------------------------------------------------------------- The series is based originally on https://lore.kernel.org/all/20250307052248.405803-1-vivek.kasireddy@intel.c… but heavily rewritten to be based on DMA physical API. ----------------------------------------------------------------------- The WIP branch can be found here: https://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma.git/log/?h=… Thanks --- Jason Gunthorpe (2): PCI/P2PDMA: Document DMABUF model vfio/nvgrace: Support get_dmabuf_phys Leon Romanovsky (7): PCI/P2PDMA: Separate the mmap() support from the core logic PCI/P2PDMA: Simplify bus address mapping API PCI/P2PDMA: Refactor to separate core P2P functionality from memory allocation PCI/P2PDMA: Provide an access to pci_p2pdma_map_type() function dma-buf: provide phys_vec to scatter-gather mapping routine vfio/pci: Enable peer-to-peer DMA transactions by default vfio/pci: Add dma-buf export support for MMIO regions Vivek Kasireddy (2): vfio: Export vfio device get and put registration helpers vfio/pci: Share the core device pointer while invoking feature functions Documentation/driver-api/pci/p2pdma.rst | 95 +++++++--- block/blk-mq-dma.c | 2 +- drivers/dma-buf/dma-buf.c | 235 ++++++++++++++++++++++++ drivers/iommu/dma-iommu.c | 4 +- drivers/pci/p2pdma.c | 182 +++++++++++++----- drivers/vfio/pci/Kconfig | 3 + drivers/vfio/pci/Makefile | 1 + drivers/vfio/pci/nvgrace-gpu/main.c | 56 ++++++ drivers/vfio/pci/vfio_pci.c | 5 + drivers/vfio/pci/vfio_pci_config.c | 22 ++- drivers/vfio/pci/vfio_pci_core.c | 56 ++++-- drivers/vfio/pci/vfio_pci_dmabuf.c | 315 ++++++++++++++++++++++++++++++++ drivers/vfio/pci/vfio_pci_priv.h | 23 +++ drivers/vfio/vfio_main.c | 2 + include/linux/dma-buf.h | 18 ++ include/linux/pci-p2pdma.h | 120 +++++++----- include/linux/vfio.h | 2 + include/linux/vfio_pci_core.h | 42 +++++ include/uapi/linux/vfio.h | 27 +++ kernel/dma/direct.c | 4 +- mm/hmm.c | 2 +- 21 files changed, 1077 insertions(+), 139 deletions(-) --- base-commit: 3a8660878839faadb4f1a6dd72c3179c1df56787 change-id: 20251016-dmabuf-vfio-6cef732adf5a Best regards, -- Leon Romanovsky <leonro(a)nvidia.com>

2 weeks, 1 day

Re: [PATCH v6 00/11] vfio/pci: Allow MMIO regions to be exported through dma-buf

by Jason Gunthorpe

On Mon, Nov 03, 2025 at 12:07:12PM -0800, Alex Mastro wrote: > On Sun, Nov 02, 2025 at 10:00:48AM +0200, Leon Romanovsky wrote: > > Changelog: > > v6: > > * Fixed wrong error check from pcim_p2pdma_init(). > > * Documented pcim_p2pdma_provider() function. > > * Improved commit messages. > > * Added VFIO DMA-BUF selftest. > > * Added __counted_by(nr_ranges) annotation to struct vfio_device_feature_dma_buf. > > * Fixed error unwind when dma_buf_fd() fails. > > * Document latest changes to p2pmem. > > * Removed EXPORT_SYMBOL_GPL from pci_p2pdma_map_type. > > * Moved DMA mapping logic to DMA-BUF. > > * Removed types patch to avoid dependencies between subsystems. > > * Moved vfio_pci_dma_buf_move() in err_undo block. > > * Added nvgrace patch. > > Thanks Leon. Attaching a toy program which sanity tests the dma-buf export UAPI > by feeding the allocated dma-buf into an dma-buf importer (libibverbs + CX-7). Oh! Here is my toy program to do the same with iommufd as the importer: #define _GNU_SOURCE #define __user #include <fcntl.h> #include <stdio.h> #include <stdlib.h> #include <sys/ioctl.h> #include <unistd.h> #include "include/uapi/linux/vfio.h" #include "include/uapi/linux/iommufd.h" #include <string.h> #include <sys/mman.h> #include <errno.h> int main(int argc, const char *argv[]) { int vfio_dev_fd, iommufd_fd, ret; // Open the per-device VFIO file (e.g., /dev/vfio/devices/vfio3) vfio_dev_fd = open("/dev/vfio/devices/vfio0", O_RDWR); if (vfio_dev_fd < 0) { perror("Failed to open VFIO per-device file"); return 1; } // Open /dev/iommu for iommufd iommufd_fd = open("/dev/iommu", O_RDWR); if (iommufd_fd < 0) { perror("Failed to open /dev/iommu"); close(vfio_dev_fd); return 1; } // Bind device FD to iommufd struct vfio_device_bind_iommufd bind = { .argsz = sizeof(bind), .flags = 0, .iommufd = iommufd_fd, }; ret = ioctl(vfio_dev_fd, VFIO_DEVICE_BIND_IOMMUFD, &bind); if (ret < 0) { perror("VFIO_DEVICE_BIND_IOMMUFD failed"); close(vfio_dev_fd); close(iommufd_fd); return 1; } // Allocate an IOAS (I/O address space) struct iommu_ioas_alloc alloc_data = { .size = sizeof(alloc_data), .flags = 0, }; ret = ioctl(iommufd_fd, IOMMU_IOAS_ALLOC, &alloc_data); if (ret < 0) { perror("IOMMU_IOAS_ALLOC failed"); close(vfio_dev_fd); close(iommufd_fd); return 1; } // Attach the device to the IOAS struct vfio_device_attach_iommufd_pt attach_data = { .argsz = sizeof(attach_data), .flags = 0, .pt_id = alloc_data.out_ioas_id, }; ret = ioctl(vfio_dev_fd, VFIO_DEVICE_ATTACH_IOMMUFD_PT, &attach_data); if (ret < 0) { perror("VFIO_DEVICE_ATTACH_IOMMUFD_PT failed"); close(vfio_dev_fd); close(iommufd_fd); return 1; } #if 0 int mapfd = memfd_create("test", MFD_CLOEXEC); if (mapfd == -1) { perror("memfd_create failed"); return 1; } ftruncate(mapfd, 4096); #else struct dmabuf_arg { struct vfio_device_feature hdr; struct vfio_device_feature_dma_buf dma_buf; struct vfio_region_dma_range range; } dma_buf_feature = { .hdr = { .argsz = sizeof(dma_buf_feature), .flags = VFIO_DEVICE_FEATURE_GET | VFIO_DEVICE_FEATURE_DMA_BUF }, .dma_buf = { .region_index = VFIO_PCI_BAR0_REGION_INDEX, .open_flags = O_CLOEXEC, .nr_ranges = 1 }, .range = { .length = 4096 }, }; ret = ioctl(vfio_dev_fd, VFIO_DEVICE_FEATURE, &dma_buf_feature); if (ret < 0) { perror("VFIO_DEVICE_FEATURE_GET failed"); return 1; } int mapfd = ret; #endif struct iommu_ioas_map_file map_file = { .size = sizeof(map_file), .flags = IOMMU_IOAS_MAP_WRITEABLE | IOMMU_IOAS_MAP_READABLE, .ioas_id = alloc_data.out_ioas_id, .fd = mapfd, .start = 0, .length = 4096, }; ret = ioctl(iommufd_fd, IOMMU_IOAS_MAP_FILE, &map_file); if (ret < 0) { perror("IOMMU_IOAS_MAP_FILE failed"); return 1; } printf("Successfully attached device to IOAS ID: %u\n", alloc_data.out_ioas_id); close(vfio_dev_fd); close(iommufd_fd); return 0; }

2 weeks, 1 day

← Newer
1
2
3
4
5
6
7
8
...
316
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-mm-sig