Linaro-mm-sig

linaro-mm-sig@lists.linaro.org

43 participants
6398 discussions

[PATCH v4 00/12] drm/amdgpu: use all SDMA instances for TTM clears and moves

by Pierre-Eric Pelloux-Prayer

The drm/ttm patch modifies TTM to support multiple contexts for the pipelined moves. Then amdgpu/ttm is updated to express dependencies between jobs explicitely, instead of relying on the ordering of execution guaranteed by the use of a single instance. With all of this in place, we can use multiple entities, with each having access to the available SDMA instances. This rework also gives the opportunity to merge the clear functions into a single one and to optimize a bit GART usage. Since v3 some patches have been already reviewed and merged separately: - https://lists.freedesktop.org/archives/amd-gfx/2026-January/137747.html - https://gitlab.freedesktop.org/drm/kernel/-/commit/ddf055b80a544d6f36f77be5… This version depend on them. v3: https://lists.freedesktop.org/archives/dri-devel/2025-November/537830.html Pierre-Eric Pelloux-Prayer (12): drm/amdgpu: allocate clear entities dynamically drm/amdgpu: allocate move entities dynamically drm/amdgpu: round robin through clear_entities in amdgpu_fill_buffer drm/amdgpu: use TTM_NUM_MOVE_FENCES when reserving fences drm/amdgpu: use multiple entities in amdgpu_move_blit drm/amdgpu: pass all the sdma scheds to amdgpu_mman drm/amdgpu: only use working sdma schedulers for ttm drm/amdgpu: create multiple clear/move ttm entities drm/amdgpu: give ttm entities access to all the sdma scheds drm/amdgpu: get rid of amdgpu_ttm_clear_buffer drm/amdgpu: rename amdgpu_fill_buffer as amdgpu_ttm_clear_buffer drm/amdgpu: split amdgpu_ttm_set_buffer_funcs_status in 2 funcs drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 + drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 5 +- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 16 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 4 +- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 17 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 329 ++++++++++-------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 29 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 6 +- drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 13 +- drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 8 +- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 8 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 15 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 12 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 11 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 14 +- drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 5 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 5 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c | 12 +- drivers/gpu/drm/amd/amdgpu/si_dma.c | 12 +- drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 5 +- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 3 +- .../amd/display/amdgpu_dm/amdgpu_dm_plane.c | 6 +- .../drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c | 6 +- 23 files changed, 300 insertions(+), 243 deletions(-) -- 2.43.0

3 months, 1 week

[PATCH] dma-buf: heaps: cma: register a dmem region for each cma heap

by Eric Chanudet

The cma dma-buf heaps let userspace allocate buffers in CMA regions without enforcing limits. Register a dmem region per cma heap and charge against it when allocating a buffer in a cma heap. For the default cma region, two heaps may be created for the same cma range: commit 854acbe75ff4 ("dma-buf: heaps: Give default CMA heap a fixed name") Introduced /dev/dma_heap/default_cma_region commit 4f5f8baf7341 ("dma-buf: heaps: cma: Create CMA heap for each CMA reserved region") Created a CMA heap for each CMA region, which might create a duplicate heap to the default one, e.g: /dev/dma_heap/default_cma_region /dev/dma_heap/reserved Removing the legacy heap would break user API. So handle the special case by using one dmem between the two heaps to account charges correctly. Signed-off-by: Eric Chanudet <echanude(a)redhat.com> --- In continuation with introducing cgroup for the system heap[1], this behavior is enabled based on dma_heap.mem_accounting, disabled by default. dmem is chosen for CMA heaps as it allows limits to be set for each region backing each heap. There is one caveat for the default cma range that may accessible through two different cma heaps, which is treated as a special case. [1] https://lore.kernel.org/all/20260116-dmabuf-heap-system-memcg-v3-0-ecc6b62c… --- drivers/dma-buf/heaps/cma_heap.c | 51 ++++++++++++++++++++++++++++++++++++---- 1 file changed, 46 insertions(+), 5 deletions(-) diff --git a/drivers/dma-buf/heaps/cma_heap.c b/drivers/dma-buf/heaps/cma_heap.c index 49cc45fb42dd7200c3c14384bcfdbe85323454b1..608af8ad6bce7fe0321da6d8f1b65a69f5d8d950 100644 --- a/drivers/dma-buf/heaps/cma_heap.c +++ b/drivers/dma-buf/heaps/cma_heap.c @@ -27,6 +27,7 @@ #include <linux/scatterlist.h> #include <linux/slab.h> #include <linux/vmalloc.h> +#include <linux/cgroup_dmem.h> #define DEFAULT_CMA_NAME "default_cma_region" @@ -46,7 +47,9 @@ int __init dma_heap_cma_register_heap(struct cma *cma) struct cma_heap { struct dma_heap *heap; struct cma *cma; + struct dmem_cgroup_region *cg; }; +static struct dmem_cgroup_region *default_cma_cg; struct cma_heap_buffer { struct cma_heap *heap; @@ -58,6 +61,7 @@ struct cma_heap_buffer { pgoff_t pagecount; int vmap_cnt; void *vaddr; + struct dmem_cgroup_pool_state *pool; }; struct dma_heap_attachment { @@ -276,6 +280,7 @@ static void cma_heap_dma_buf_release(struct dma_buf *dmabuf) kfree(buffer->pages); /* release memory */ cma_release(cma_heap->cma, buffer->cma_pages, buffer->pagecount); + dmem_cgroup_uncharge(buffer->pool, buffer->len); kfree(buffer); } @@ -319,9 +324,16 @@ static struct dma_buf *cma_heap_allocate(struct dma_heap *heap, if (align > CONFIG_CMA_ALIGNMENT) align = CONFIG_CMA_ALIGNMENT; + if (mem_accounting) { + ret = dmem_cgroup_try_charge(cma_heap->cg, size, + &buffer->pool, NULL); + if (ret) + goto free_buffer; + } + cma_pages = cma_alloc(cma_heap->cma, pagecount, align, false); if (!cma_pages) - goto free_buffer; + goto uncharge_cgroup; /* Clear the cma pages */ if (PageHighMem(cma_pages)) { @@ -376,6 +388,8 @@ static struct dma_buf *cma_heap_allocate(struct dma_heap *heap, kfree(buffer->pages); free_cma: cma_release(cma_heap->cma, cma_pages, pagecount); +uncharge_cgroup: + dmem_cgroup_uncharge(buffer->pool, size); free_buffer: kfree(buffer); @@ -390,25 +404,52 @@ static int __init __add_cma_heap(struct cma *cma, const char *name) { struct dma_heap_export_info exp_info; struct cma_heap *cma_heap; + struct dmem_cgroup_region *region; + int ret; cma_heap = kzalloc(sizeof(*cma_heap), GFP_KERNEL); if (!cma_heap) return -ENOMEM; cma_heap->cma = cma; + /* + * If two heaps are created for the default cma region, use the same + * dmem for them. They both use the same memory pool. + */ + if (dev_get_cma_area(NULL) == cma && default_cma_cg) + region = default_cma_cg; + else { + region = dmem_cgroup_register_region(cma_get_size(cma), "cma/%s", name); + if (IS_ERR(region)) { + ret = PTR_ERR(region); + goto free_cma_heap; + } + } + cma_heap->cg = region; + exp_info.name = name; exp_info.ops = &cma_heap_ops; exp_info.priv = cma_heap; cma_heap->heap = dma_heap_add(&exp_info); if (IS_ERR(cma_heap->heap)) { - int ret = PTR_ERR(cma_heap->heap); - - kfree(cma_heap); - return ret; + ret = PTR_ERR(cma_heap->heap); + goto cg_unregister; } + if (dev_get_cma_area(NULL) == cma && !default_cma_cg) + default_cma_cg = region; + return 0; + +cg_unregister: + /* default_cma_cg == cma_heap->cg only for the duplicate heap. */ + if (default_cma_cg != cma_heap->cg) + dmem_cgroup_unregister_region(cma_heap->cg); +free_cma_heap: + kfree(cma_heap); + + return ret; } static int __init add_cma_heaps(void) --- base-commit: 3d65e4c276b32c03450261d114e495fda03c8e97 change-id: 20260128-dmabuf-heap-cma-dmem-f4120a2df4a8 Best regards, -- Eric Chanudet <echanude(a)redhat.com>

3 months, 1 week

Hire SAFEGUARD RECOVERY, a Bitcoin recovery specialist, to retrieve stolen or lost coins.

by sophia fourest

I invested $320,000 in Tether (USDT) on a fraudulent website after falling for a romantic scam. I felt completely helpless and in need of assistance after realizing I had been duped. I started looking for a hacker online and found SAFEGUARD RECOVERY. I had optimism because of his professionalism and knowledge. I'm happy to report that SAFEGUARD RECOVERY successfully recovered my stolen money after working relentlessly to do so! I am immensely appreciative of their help and heartily urge anyone in a comparable circumstance to use their services. I'm grateful Email: safeguardbitcoin(a)consultant.com WhatsApp: +44 7426 168300 Website: https://safeguardbitcoin.wixsite.com/safeguard-bitcoin--1

3 months, 1 week

[PATCH v6 0/8] dma-buf: Use revoke mechanism to invalidate shared buffers

by Leon Romanovsky

Changelog: v6: * Added Reviewed-by tags. * Changed for blocking wait_for_completion() in VFIO * Fixed race between ->attach and move_notify, where priv->revoked is flipped and lock is released. v5: https://patch.msgid.link/20260124-dmabuf-revoke-v5-0-f98fca917e96@nvidia.com * Documented the DMA-BUF expectations around DMA unmap. * Added wait support in VFIO for DMA unmap. * Reordered patches. * Improved commit messages to document even more. v4: https://lore.kernel.org/all/20260121-dmabuf-revoke-v4-0-d311cbc8633d@nvidia… * Changed DMA_RESV_USAGE_KERNEL to DMA_RESV_USAGE_BOOKKEEP. * Made .invalidate_mapping() truly optional. * Added patch which renames dma_buf_move_notify() to be dma_buf_invalidate_mappings(). * Restored dma_buf_attachment_is_dynamic() function. v3: https://lore.kernel.org/all/20260120-dmabuf-revoke-v3-0-b7e0b07b8214@nvidia… * Used Jason's wordings for commits and cover letter. * Removed IOMMUFD patch. * Renamed dma_buf_attachment_is_revoke() to be dma_buf_attach_revocable(). * Added patch to remove CONFIG_DMABUF_MOVE_NOTIFY. * Added Reviewed-by tags. * Called to dma_resv_wait_timeout() after dma_buf_move_notify() in VFIO. * Added dma_buf_attach_revocable() check to VFIO DMABUF attach function. * Slightly changed commit messages. v2: https://patch.msgid.link/20260118-dmabuf-revoke-v2-0-a03bb27c0875@nvidia.com * Changed series to document the revoke semantics instead of implementing it. v1: https://patch.msgid.link/20260111-dmabuf-revoke-v1-0-fb4bcc8c259b@nvidia.com ------------------------------------------------------------------------- This series is based on latest VFIO fix, which will be sent to Linus very soon. https://lore.kernel.org/all/20260121-vfio-add-pin-v1-1-4e04916b17f1@nvidia.… Thanks ------------------------------------------------------------------------- This series documents a dma-buf “revoke” mechanism: to allow a dma-buf exporter to explicitly invalidate (“kill”) a shared buffer after it has been distributed to importers, so that further CPU and device access is prevented and importers reliably observe failure. The change in this series is to properly document and use existing core “revoked” state on the dma-buf object and a corresponding exporter-triggered revoke operation. dma-buf has quietly allowed calling move_notify on pinned dma-bufs, even though legacy importers using dma_buf_attach() would simply ignore these calls. The intention was that move_notify() would tell the importer to expedite it's unmapping process and once the importer is fully finished with DMA it would unmap the dma-buf which finally signals that the importer is no longer ever going to touch the memory again. Importers that touch past their unmap() call can trigger IOMMU errors, AER and beyond, however read-and-discard access between move_notify() and unmap is allowed. Thus, we can define the exporter's revoke sequence for pinned dma-buf as: dma_resv_lock(dmabuf->resv, NULL); // Prevent new mappings from being established priv->revoked = true; // Tell all importers to eventually unmap dma_buf_invalidate_mappings(dmabuf); // Wait for any inprogress fences on the old mapping dma_resv_wait_timeout(dmabuf->resv, DMA_RESV_USAGE_BOOKKEEP, false, MAX_SCHEDULE_TIMEOUT); dma_resv_unlock(dmabuf->resv, NULL); // Wait for all importers to complete unmap wait_for_completion(&priv->unmapp_comp); However, dma-buf also supports importers that don't do anything on move_notify(), and will not unmap the buffer in bounded time. Since such importers would cause the above sequence to hang, a new mechanism is needed to detect incompatible importers. Introduce dma_buf_attach_revocable() which if true indicates the above sequence is safe to use and will complete in kernel-only bounded time for this attachment. Unfortunately dma_buf_attach_revocable() is going to fail for the popular RDMA pinned importer, which means we cannot introduce it to existing places using pinned move_notify() without potentially breaking existing userspace flows. Existing exporters that only trigger this flow for RAS errors should not call dma_buf_attach_revocable() and will suffer an unbounded block on the final completion, hoping that the userspace will notice the RAS and clean things up. Without revoke support on the RDMA pinned importers it doesn't seem like any other non-breaking option is currently possible. For new exporters, like VFIO and RDMA, that have userspace triggered revoke events, the unbouned sleep would not be acceptable. They can call dma_buf_attach_revocable() and will not work with the RDMA pinned importer from day 0, preventing regressions. In the process add documentation explaining the above details. Thanks Signed-off-by: Leon Romanovsky <leonro(a)nvidia.com> --- Leon Romanovsky (8): dma-buf: Rename .move_notify() callback to a clearer identifier dma-buf: Rename dma_buf_move_notify() to dma_buf_invalidate_mappings() dma-buf: Always build with DMABUF_MOVE_NOTIFY vfio: Wait for dma-buf invalidation to complete dma-buf: Make .invalidate_mapping() truly optional dma-buf: Add dma_buf_attach_revocable() vfio: Permit VFIO to work with pinned importers iommufd: Add dma_buf_pin() drivers/dma-buf/Kconfig | 12 ----- drivers/dma-buf/dma-buf.c | 69 +++++++++++++++++++----- drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 14 ++--- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +- drivers/gpu/drm/amd/amdkfd/Kconfig | 2 +- drivers/gpu/drm/virtio/virtgpu_prime.c | 2 +- drivers/gpu/drm/xe/tests/xe_dma_buf.c | 7 ++- drivers/gpu/drm/xe/xe_bo.c | 2 +- drivers/gpu/drm/xe/xe_dma_buf.c | 14 ++--- drivers/infiniband/core/umem_dmabuf.c | 13 ----- drivers/infiniband/hw/mlx5/mr.c | 2 +- drivers/iommu/iommufd/pages.c | 11 +++- drivers/iommu/iommufd/selftest.c | 2 +- drivers/vfio/pci/vfio_pci_dmabuf.c | 84 ++++++++++++++++++++++------- include/linux/dma-buf.h | 17 +++--- 15 files changed, 157 insertions(+), 96 deletions(-) --- base-commit: 61ceaf236115f20f4fdd7cf60f883ada1063349a change-id: 20251221-dmabuf-revoke-b90ef16e4236 Best regards, -- Leon Romanovsky <leonro(a)nvidia.com>

3 months, 1 week

[PATCH v5 0/8] dma-buf: Use revoke mechanism to invalidate shared buffers

by Leon Romanovsky

Changelog: v5: * Documented the DMA-BUF expectations around DMA unmap. * Added wait support in VFIO for DMA unmap. * Reordered patches. * Improved commit messages to document even more. v4: https://lore.kernel.org/all/20260121-dmabuf-revoke-v4-0-d311cbc8633d@nvidia… * Changed DMA_RESV_USAGE_KERNEL to DMA_RESV_USAGE_BOOKKEEP. * Made .invalidate_mapping() truly optional. * Added patch which renames dma_buf_move_notify() to be dma_buf_invalidate_mappings(). * Restored dma_buf_attachment_is_dynamic() function. v3: https://lore.kernel.org/all/20260120-dmabuf-revoke-v3-0-b7e0b07b8214@nvidia… * Used Jason's wordings for commits and cover letter. * Removed IOMMUFD patch. * Renamed dma_buf_attachment_is_revoke() to be dma_buf_attach_revocable(). * Added patch to remove CONFIG_DMABUF_MOVE_NOTIFY. * Added Reviewed-by tags. * Called to dma_resv_wait_timeout() after dma_buf_move_notify() in VFIO. * Added dma_buf_attach_revocable() check to VFIO DMABUF attach function. * Slightly changed commit messages. v2: https://patch.msgid.link/20260118-dmabuf-revoke-v2-0-a03bb27c0875@nvidia.com * Changed series to document the revoke semantics instead of implementing it. v1: https://patch.msgid.link/20260111-dmabuf-revoke-v1-0-fb4bcc8c259b@nvidia.com ------------------------------------------------------------------------- This series is based on latest VFIO fix, which will be sent to Linus very soon. https://lore.kernel.org/all/20260121-vfio-add-pin-v1-1-4e04916b17f1@nvidia.… Thanks ------------------------------------------------------------------------- This series documents a dma-buf “revoke” mechanism: to allow a dma-buf exporter to explicitly invalidate (“kill”) a shared buffer after it has been distributed to importers, so that further CPU and device access is prevented and importers reliably observe failure. The change in this series is to properly document and use existing core “revoked” state on the dma-buf object and a corresponding exporter-triggered revoke operation. dma-buf has quietly allowed calling move_notify on pinned dma-bufs, even though legacy importers using dma_buf_attach() would simply ignore these calls. The intention was that move_notify() would tell the importer to expedite it's unmapping process and once the importer is fully finished with DMA it would unmap the dma-buf which finally signals that the importer is no longer ever going to touch the memory again. Importers that touch past their unmap() call can trigger IOMMU errors, AER and beyond, however read-and-discard access between move_notify() and unmap is allowed. Thus, we can define the exporter's revoke sequence for pinned dma-buf as: dma_resv_lock(dmabuf->resv, NULL); // Prevent new mappings from being established priv->revoked = true; // Tell all importers to eventually unmap dma_buf_invalidate_mappings(dmabuf); // Wait for any inprogress fences on the old mapping dma_resv_wait_timeout(dmabuf->resv, DMA_RESV_USAGE_BOOKKEEP, false, MAX_SCHEDULE_TIMEOUT); dma_resv_unlock(dmabuf->resv, NULL); // Wait for all importers to complete unmap wait_for_completion(&priv->unmapp_comp); However, dma-buf also supports importers that don't do anything on move_notify(), and will not unmap the buffer in bounded time. Since such importers would cause the above sequence to hang, a new mechanism is needed to detect incompatible importers. Introduce dma_buf_attach_revocable() which if true indicates the above sequence is safe to use and will complete in kernel-only bounded time for this attachment. Unfortunately dma_buf_attach_revocable() is going to fail for the popular RDMA pinned importer, which means we cannot introduce it to existing places using pinned move_notify() without potentially breaking existing userspace flows. Existing exporters that only trigger this flow for RAS errors should not call dma_buf_attach_revocable() and will suffer an unbounded block on the final completion, hoping that the userspace will notice the RAS and clean things up. Without revoke support on the RDMA pinned importers it doesn't seem like any other non-breaking option is currently possible. For new exporters, like VFIO and RDMA, that have userspace triggered revoke events, the unbouned sleep would not be acceptable. They can call dma_buf_attach_revocable() and will not work with the RDMA pinned importer from day 0, preventing regressions. In the process add documentation explaining the above details. Thanks Signed-off-by: Leon Romanovsky <leonro(a)nvidia.com> --- Leon Romanovsky (8): dma-buf: Rename .move_notify() callback to a clearer identifier dma-buf: Rename dma_buf_move_notify() to dma_buf_invalidate_mappings() dma-buf: Always build with DMABUF_MOVE_NOTIFY vfio: Wait for dma-buf invalidation to complete dma-buf: Make .invalidate_mapping() truly optional dma-buf: Add dma_buf_attach_revocable() vfio: Permit VFIO to work with pinned importers iommufd: Add dma_buf_pin() drivers/dma-buf/Kconfig | 12 ---- drivers/dma-buf/dma-buf.c | 69 +++++++++++++++++----- drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 14 ++--- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +- drivers/gpu/drm/amd/amdkfd/Kconfig | 2 +- drivers/gpu/drm/virtio/virtgpu_prime.c | 2 +- drivers/gpu/drm/xe/tests/xe_dma_buf.c | 7 +-- drivers/gpu/drm/xe/xe_bo.c | 2 +- drivers/gpu/drm/xe/xe_dma_buf.c | 14 ++--- drivers/infiniband/core/umem_dmabuf.c | 13 ----- drivers/infiniband/hw/mlx5/mr.c | 2 +- drivers/iommu/iommufd/pages.c | 11 +++- drivers/iommu/iommufd/selftest.c | 2 +- drivers/vfio/pci/vfio_pci_dmabuf.c | 90 +++++++++++++++++++++++------ include/linux/dma-buf.h | 17 +++--- 15 files changed, 164 insertions(+), 95 deletions(-) --- base-commit: 61ceaf236115f20f4fdd7cf60f883ada1063349a change-id: 20251221-dmabuf-revoke-b90ef16e4236 Best regards, -- Leon Romanovsky <leonro(a)nvidia.com>

3 months, 1 week

[RFC v2 0/8] dma-buf: Add support for mapping dmabufs via interconnects

by Vivek Kasireddy

In a typical dma-buf use case, a dmabuf exporter makes its buffer buffer available to an importer by mapping it using DMA APIs such as dma_map_sgtable() or dma_map_resource(). However, this is not desirable in some cases where the exporter and importer are directly connected via a physical or virtual link (or interconnect) and the importer can access the buffer without having it DMA mapped. So, to address this scenario, this patch series adds APIs to map/ unmap dmabufs via interconnects and also provides a helper to identify the first common interconnect between the exporter and importer. Furthermore, this patch series also adds support for IOV interconnect in the vfio-pci driver and Intel Xe driver. The IOV interconnect is a virtual interconnect between an SRIOV physical function (PF) and its virtual functions (VFs). And, for the IOV interconnect, the addresses associated with a buffer are shared using an xarray (instead of an sg_table) that is populated with entries of type struct range. The dma-buf patches in this series are based on ideas/suggestions provided by Jason Gunthorpe, Christian Koenig and Thomas Hellström. Changelog: RFC -> RFCv2: - Add documentation for the new dma-buf APIs and types (Thomas) - Change the interconnect type from enum to unique pointer (Thomas) - Moved the new dma-buf APIs to a separate file - Store a copy of the interconnect matching data in the attachment - Simplified the macros to create and match interconnects - Use struct device instead of struct pci_dev in match data - Replace DRM_INTERCONNECT_DRIVER with XE_INTERCONNECT_VRAM during address encoding (Matt, Thomas) - Drop is_devmem_external and instead rely on bo->dma_data.dma_addr to check for imported VRAM BOs (Matt) - Pass XE_PAGE_SIZE as the last parameter to xe_bo_addr (Matt) - Add a check to prevent malicious VF from accessing other VF's addresses (Thomas) - Fallback to legacy (map_dma_buf) mapping method if mapping via interconnect fails Patchset overview: Patch 1-3: Add dma-buf APIs to map/unmap and match Patch 4: Add support for IOV interconnect in vfio-pci driver Patch 5: Add support for IOV interconnect in Xe driver Patch 6-8: Create and use a new dma_addr array for LMEM based dmabuf BOs to store translated addresses (DPAs) This series is rebased on top of the following repo: https://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma.git/log/?h=… Associated Qemu patch series: https://lore.kernel.org/qemu-devel/20251003234138.85820-1-vivek.kasireddy@i… Associated vfio-pci patch series: https://lore.kernel.org/dri-devel/cover.1760368250.git.leon@kernel.org/ This series is tested using the following method: - Run Qemu with the following relevant options: qemu-system-x86_64 -m 4096m .... -device ioh3420,id=root_port1,bus=pcie.0 -device x3130-upstream,id=upstream1,bus=root_port1 -device xio3130-downstream,id=downstream1,bus=upstream1,chassis=9 -device xio3130-downstream,id=downstream2,bus=upstream1,chassis=10 -device vfio-pci,host=0000:03:00.1,bus=downstream1 -device virtio-gpu,max_outputs=1,blob=true,xres=1920,yres=1080,bus=downstream2 -display gtk,gl=on -object memory-backend-memfd,id=mem1,size=4096M -machine q35,accel=kvm,memory-backend=mem1 ... - Run Gnome Wayland with the following options in the Guest VM: # cat /usr/lib/udev/rules.d/61-mutter-primary-gpu.rules ENV{DEVNAME}=="/dev/dri/card1", TAG+="mutter-device-preferred-primary", TAG+="mutter-device-disable-kms-modifiers" # XDG_SESSION_TYPE=wayland dbus-run-session -- /usr/bin/gnome-shell --wayland --no-x11 & Cc: Jason Gunthorpe <jgg(a)nvidia.com> Cc: Leon Romanovsky <leonro(a)nvidia.com> Cc: Christian Koenig <christian.koenig(a)amd.com> Cc: Sumit Semwal <sumit.semwal(a)linaro.org> Cc: Thomas Hellström <thomas.hellstrom(a)linux.intel.com> Cc: Simona Vetter <simona.vetter(a)ffwll.ch> Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: Dongwon Kim <dongwon.kim(a)intel.com> Vivek Kasireddy (8): dma-buf: Add support for map/unmap APIs for interconnects dma-buf: Add a helper to match interconnects between exporter/importer dma-buf: Create and expose IOV interconnect to all exporters/importers vfio/pci/dmabuf: Add support for IOV interconnect drm/xe/dma_buf: Add support for IOV interconnect drm/xe/pf: Add a helper function to get a VF's backing object in LMEM drm/xe/bo: Create new dma_addr array for dmabuf BOs associated with VFs drm/xe/pt: Add an additional check for dmabuf BOs while doing bind drivers/dma-buf/Makefile | 2 +- drivers/dma-buf/dma-buf-interconnect.c | 164 +++++++++++++++++++++ drivers/dma-buf/dma-buf.c | 12 +- drivers/gpu/drm/xe/xe_bo.c | 162 ++++++++++++++++++-- drivers/gpu/drm/xe/xe_bo_types.h | 6 + drivers/gpu/drm/xe/xe_dma_buf.c | 17 ++- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 24 +++ drivers/gpu/drm/xe/xe_gt_sriov_pf_config.h | 1 + drivers/gpu/drm/xe/xe_pt.c | 8 + drivers/gpu/drm/xe/xe_sriov_pf_types.h | 19 +++ drivers/vfio/pci/vfio_pci_dmabuf.c | 135 ++++++++++++++++- include/linux/dma-buf-interconnect.h | 122 +++++++++++++++ include/linux/dma-buf.h | 41 ++++++ 13 files changed, 691 insertions(+), 22 deletions(-) create mode 100644 drivers/dma-buf/dma-buf-interconnect.c create mode 100644 include/linux/dma-buf-interconnect.h -- 2.50.1

3 months, 2 weeks

[PATCH rdma-next v2 0/3] RDMA: Add support for exporting dma-buf file descriptors

by Edward Srouji

This patch series introduces dma-buf export support for RDMA/InfiniBand devices, enabling userspace applications to export RDMA PCI-backed memory regions (such as device memory or mlx5 UAR pages) as dma-buf file descriptors. This allows PCI device memory to be shared with other kernel subsystems (e.g., graphics or media) or between userspace processes, via the standard dma-buf interface, avoiding unnecessary copies and enabling efficient peer-to-peer (P2P) DMA transfers. See [1] for background on dma-buf. As part of this series, we introduce a new uverbs object of type FD for dma-buf export, along with the corresponding APIs for allocation and teardown. This object encapsulates all attributes required to export a dma-buf. The implementation enforces P2P-only mappings and properly manages resource lifecycle, including: - Cleanup during driver removal or RDMA context destruction. - Revocation via dma_buf_move_notify() when the underlying mmap entries are removed. - Refactors common cleanup logic for reuse across FD uobject types. The infrastructure is generic within uverbs, allowing individual drivers to easily integrate and supply their vendor-specific implementation. The mlx5 driver is the first consumer of this new API, providing: - Initialization of PCI peer-to-peer DMA support. - mlx5-specific implementations of the mmap_get_pfns and pgoff_to_mmap_entry device operations required for dma-buf export. [1] https://docs.kernel.org/driver-api/dma-buf.html Signed-off-by: Yishai Hadas <yishaih(a)nvidia.com> Signed-off-by: Edward Srouji <edwards(a)nvidia.com> --- Changes in v2: - Split the FD uobject refactoring into a separate patch ("RDMA: Add support for exporting dma-buf file descriptors") - Remove redundant revoked check from attach callback. It is checked during map - Add pin callback that returns -EOPNOTSUPP to explicitly refuse pinned importers - Wait for pending fences after dma_buf_move_notify() using dma_resv_wait_timeout() to ensure hardware has completed all in-flight operations before proceeding - Link to v1: https://lore.kernel.org/r/20260108-dmabuf-export-v1-0-6d47d46580d3@nvidia.c… --- Yishai Hadas (3): RDMA/uverbs: Support external FD uobjects RDMA/uverbs: Add DMABUF object type and operations RDMA/mlx5: Implement DMABUF export ops drivers/infiniband/core/Makefile | 1 + drivers/infiniband/core/device.c | 2 + drivers/infiniband/core/ib_core_uverbs.c | 22 +++ drivers/infiniband/core/rdma_core.c | 63 ++++---- drivers/infiniband/core/rdma_core.h | 1 + drivers/infiniband/core/uverbs.h | 10 ++ drivers/infiniband/core/uverbs_std_types_dmabuf.c | 176 ++++++++++++++++++++++ drivers/infiniband/core/uverbs_uapi.c | 1 + drivers/infiniband/hw/mlx5/main.c | 72 +++++++++ include/rdma/ib_verbs.h | 9 ++ include/rdma/uverbs_types.h | 1 + include/uapi/rdma/ib_user_ioctl_cmds.h | 10 ++ 12 files changed, 342 insertions(+), 26 deletions(-) --- base-commit: 325e3b5431ddd27c5f93156b36838a351e3b2f72 change-id: 20260108-dmabuf-export-0d598058dd1e Best regards, -- Edward Srouji <edwards(a)nvidia.com>

3 months, 3 weeks

[PATCH v3 0/7] dma-buf: Use revoke mechanism to invalidate shared buffers

by Leon Romanovsky

Changelog: v3: * Used Jason's wordings for commits and cover letter. * Removed IOMMUFD patch. * Renamed dma_buf_attachment_is_revoke() to be dma_buf_attach_revocable(). * Added patch to remove CONFIG_DMABUF_MOVE_NOTIFY. * Added Reviewed-by tags. * Called to dma_resv_wait_timeout() after dma_buf_move_notify() in VFIO. * Added dma_buf_attach_revocable() check to VFIO DMABUF attach function. * Slightly changed commit messages. v2: https://patch.msgid.link/20260118-dmabuf-revoke-v2-0-a03bb27c0875@nvidia.com * Changed series to document the revoke semantics instead of implementing it. v1: https://patch.msgid.link/20260111-dmabuf-revoke-v1-0-fb4bcc8c259b@nvidia.com ------------------------------------------------------------------------- This series documents a dma-buf “revoke” mechanism: to allow a dma-buf exporter to explicitly invalidate (“kill”) a shared buffer after it has been distributed to importers, so that further CPU and device access is prevented and importers reliably observe failure. The change in this series is to properly document and use existing core “revoked” state on the dma-buf object and a corresponding exporter-triggered revoke operation. dma-buf has quietly allowed calling move_notify on pinned dma-bufs, even though legacy importers using dma_buf_attach() would simply ignore these calls. RDMA saw this and needed to use allow_peer2peer=true, so implemented a new-style pinned importer with an explicitly non-working move_notify() callback. This has been tolerable because the existing exporters are thought to only call move_notify() on a pinned DMABUF under RAS events and we have been willing to tolerate the UAF that results by allowing the importer to continue to use the mapping in this rare case. VFIO wants to implement a pin supporting exporter that will issue a revoking move_notify() around FLRs and a few other user triggerable operations. Since this is much more common we are not willing to tolerate the security UAF caused by interworking with non-move_notify() supporting drivers. Thus till now VFIO has required dynamic importers, even though it never actually moves the buffer location. To allow VFIO to work with pinned importers, according to how dma-buf was intended, we need to allow VFIO to detect if an importer is legacy or RDMA and does not actually implement move_notify(). Introduce a new function that exporters can call to detect these less capable importers. VFIO can then refuse to accept them during attach. In theory all exporters that call move_notify() on pinned dma-buf's should call this function, however that would break a number of widely used NIC/GPU flows. Thus for now do not spread this further than VFIO until we can understand how much of RDMA can implement the full semantic. In the process clarify how move_notify is intended to be used with pinned dma-bufs. Thanks Signed-off-by: Leon Romanovsky <leonro(a)nvidia.com> --- Leon Romanovsky (7): dma-buf: Rename .move_notify() callback to a clearer identifier dma-buf: Always build with DMABUF_MOVE_NOTIFY dma-buf: Document RDMA non-ODP invalidate_mapping() special case dma-buf: Add check function for revoke semantics iommufd: Pin dma-buf importer for revoke semantics vfio: Wait for dma-buf invalidation to complete vfio: Validate dma-buf revocation semantics drivers/dma-buf/Kconfig | 12 ----- drivers/dma-buf/dma-buf.c | 69 +++++++++++++++++++++++------ drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 14 +++--- drivers/gpu/drm/amd/amdkfd/Kconfig | 2 +- drivers/gpu/drm/virtio/virtgpu_prime.c | 2 +- drivers/gpu/drm/xe/tests/xe_dma_buf.c | 7 ++- drivers/gpu/drm/xe/xe_dma_buf.c | 14 +++--- drivers/infiniband/core/umem_dmabuf.c | 13 +----- drivers/infiniband/hw/mlx5/mr.c | 2 +- drivers/iommu/iommufd/pages.c | 11 ++++- drivers/vfio/pci/vfio_pci_dmabuf.c | 8 ++++ include/linux/dma-buf.h | 9 ++-- 12 files changed, 96 insertions(+), 67 deletions(-) --- base-commit: 9ace4753a5202b02191d54e9fdf7f9e3d02b85eb change-id: 20251221-dmabuf-revoke-b90ef16e4236 Best regards, -- Leon Romanovsky <leonro(a)nvidia.com>

3 months, 3 weeks

[PATCH rdma-next 0/2] RDMA: Add support for exporting dma-buf file descriptors

by Edward Srouji

This patch series introduces dma-buf export support for RDMA/InfiniBand devices, enabling userspace applications to export RDMA PCI-backed memory regions (such as device memory or mlx5 UAR pages) as dma-buf file descriptors. This allows PCI device memory to be shared with other kernel subsystems (e.g., graphics or media) or between userspace processes, via the standard dma-buf interface, avoiding unnecessary copies and enabling efficient peer-to-peer (P2P) DMA transfers. See [1] for background on dma-buf. As part of this series, we introduce a new uverbs object of type FD for dma-buf export, along with the corresponding APIs for allocation and teardown. This object encapsulates all attributes required to export a dma-buf. The implementation enforces P2P-only mappings and properly manages resource lifecycle, including: - Cleanup during driver removal or RDMA context destruction. - Revocation via dma_buf_move_notify() when the underlying mmap entries are removed. - Refactors common cleanup logic for reuse across FD uobject types. The infrastructure is generic within uverbs, allowing individual drivers to easily integrate and supply their vendor-specific implementation. The mlx5 driver is the first consumer of this new API, providing: - Initialization of PCI peer-to-peer DMA support. - mlx5-specific implementations of the mmap_get_pfns and pgoff_to_mmap_entry device operations required for dma-buf export. [1] https://docs.kernel.org/driver-api/dma-buf.html Signed-off-by: Yishai Hadas <yishaih(a)nvidia.com> Signed-off-by: Edward Srouji <edwards(a)nvidia.com> --- Yishai Hadas (2): RDMA/uverbs: Add DMABUF object type and operations RDMA/mlx5: Implement DMABUF export ops drivers/infiniband/core/Makefile | 1 + drivers/infiniband/core/device.c | 2 + drivers/infiniband/core/ib_core_uverbs.c | 19 +++ drivers/infiniband/core/rdma_core.c | 63 ++++---- drivers/infiniband/core/rdma_core.h | 1 + drivers/infiniband/core/uverbs.h | 10 ++ drivers/infiniband/core/uverbs_std_types_dmabuf.c | 172 ++++++++++++++++++++++ drivers/infiniband/core/uverbs_uapi.c | 1 + drivers/infiniband/hw/mlx5/main.c | 72 +++++++++ include/rdma/ib_verbs.h | 9 ++ include/rdma/uverbs_types.h | 1 + include/uapi/rdma/ib_user_ioctl_cmds.h | 10 ++ 12 files changed, 335 insertions(+), 26 deletions(-) --- base-commit: 325e3b5431ddd27c5f93156b36838a351e3b2f72 change-id: 20260108-dmabuf-export-0d598058dd1e Best regards, -- Edward Srouji <edwards(a)nvidia.com>

3 months, 3 weeks

[PATCH v4 0/8] dma-buf: Use revoke mechanism to invalidate shared buffers

by Leon Romanovsky

Changelog: v4: * Changed DMA_RESV_USAGE_KERNEL to DMA_RESV_USAGE_BOOKKEEP. * Made .invalidate_mapping() truly optional. * Added patch which renames dma_buf_move_notify() to be dma_buf_invalidate_mappings(). * Restored dma_buf_attachment_is_dynamic() function. v3: https://lore.kernel.org/all/20260120-dmabuf-revoke-v3-0-b7e0b07b8214@nvidia… * Used Jason's wordings for commits and cover letter. * Removed IOMMUFD patch. * Renamed dma_buf_attachment_is_revoke() to be dma_buf_attach_revocable(). * Added patch to remove CONFIG_DMABUF_MOVE_NOTIFY. * Added Reviewed-by tags. * Called to dma_resv_wait_timeout() after dma_buf_move_notify() in VFIO. * Added dma_buf_attach_revocable() check to VFIO DMABUF attach function. * Slightly changed commit messages. v2: https://patch.msgid.link/20260118-dmabuf-revoke-v2-0-a03bb27c0875@nvidia.com * Changed series to document the revoke semantics instead of implementing it. v1: https://patch.msgid.link/20260111-dmabuf-revoke-v1-0-fb4bcc8c259b@nvidia.com ------------------------------------------------------------------------- This series documents a dma-buf “revoke” mechanism: to allow a dma-buf exporter to explicitly invalidate (“kill”) a shared buffer after it has been distributed to importers, so that further CPU and device access is prevented and importers reliably observe failure. The change in this series is to properly document and use existing core “revoked” state on the dma-buf object and a corresponding exporter-triggered revoke operation. dma-buf has quietly allowed calling move_notify on pinned dma-bufs, even though legacy importers using dma_buf_attach() would simply ignore these calls. RDMA saw this and needed to use allow_peer2peer=true, so implemented a new-style pinned importer with an explicitly non-working move_notify() callback. This has been tolerable because the existing exporters are thought to only call move_notify() on a pinned DMABUF under RAS events and we have been willing to tolerate the UAF that results by allowing the importer to continue to use the mapping in this rare case. VFIO wants to implement a pin supporting exporter that will issue a revoking move_notify() around FLRs and a few other user triggerable operations. Since this is much more common we are not willing to tolerate the security UAF caused by interworking with non-move_notify() supporting drivers. Thus till now VFIO has required dynamic importers, even though it never actually moves the buffer location. To allow VFIO to work with pinned importers, according to how dma-buf was intended, we need to allow VFIO to detect if an importer is legacy or RDMA and does not actually implement move_notify(). In theory all exporters that call move_notify() on pinned dma-buf's should call this function, however that would break a number of widely used NIC/GPU flows. Thus for now do not spread this further than VFIO until we can understand how much of RDMA can implement the full semantic. In the process clarify how move_notify is intended to be used with pinned dma-bufs. Thanks Signed-off-by: Leon Romanovsky <leonro(a)nvidia.com> --- Leon Romanovsky (8): dma-buf: Rename .move_notify() callback to a clearer identifier dma-buf: Rename dma_buf_move_notify() to dma_buf_invalidate_mappings() dma-buf: Always build with DMABUF_MOVE_NOTIFY dma-buf: Make .invalidate_mapping() truly optional dma-buf: Add check function for revoke semantics iommufd: Pin dma-buf importer for revoke semantics vfio: Wait for dma-buf invalidation to complete vfio: Validate dma-buf revocation semantics drivers/dma-buf/Kconfig | 12 ------- drivers/dma-buf/dma-buf.c | 53 ++++++++++++++++++++++------- drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 14 +++----- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +- drivers/gpu/drm/amd/amdkfd/Kconfig | 2 +- drivers/gpu/drm/virtio/virtgpu_prime.c | 2 +- drivers/gpu/drm/xe/tests/xe_dma_buf.c | 7 ++-- drivers/gpu/drm/xe/xe_bo.c | 2 +- drivers/gpu/drm/xe/xe_dma_buf.c | 14 +++----- drivers/infiniband/core/umem_dmabuf.c | 13 ------- drivers/infiniband/hw/mlx5/mr.c | 2 +- drivers/iommu/iommufd/pages.c | 11 ++++-- drivers/iommu/iommufd/selftest.c | 2 +- drivers/vfio/pci/vfio_pci_dmabuf.c | 13 +++++-- include/linux/dma-buf.h | 9 ++--- 15 files changed, 84 insertions(+), 74 deletions(-) --- base-commit: 9ace4753a5202b02191d54e9fdf7f9e3d02b85eb change-id: 20251221-dmabuf-revoke-b90ef16e4236 Best regards, -- Leon Romanovsky <leonro(a)nvidia.com>

3 months, 3 weeks

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-mm-sig