kernel/linux.git/drivers/infiniband/core, branch v6.6.155

RDMA/umem: Add helpers for umem dmabuf revoke lock

2026-08-03T09:15:12+00:00

[ Upstream commit 3a0b171302eea1732a168e26db3b8461f51cc1f9 ] Added helpers to acquire and release the umem dmabuf revoke lock. The intent is to avoid the need for drivers to peek into the ib_umem_dmabuf internals to get the dma_resv_lock and bring us one step closer to abstracting ib_umem_dmabuf away from drivers in general. Signed-off-by: Jacob Moroni Link: https://patch.msgid.link/20260305170826.3803155-5-jmoroni@google.com Signed-off-by: Leon Romanovsky Stable-dep-of: a846aecb931b ("RDMA/irdma: Prevent rereg_mr for non-mem regions") Signed-off-by: Sasha Levin

RDMA/umem: Add pinned revocable dmabuf import interface

2026-08-03T09:15:12+00:00

[ Upstream commit ff85a2ebacbdaec9f28c4660c991295ace93cd1c ] Added an interface for importing a pinned but revocable dmabuf. This interface can be used by drivers that are capable of revocation so that they can import dmabufs from exporters that may require it, such as VFIO. This interface implements a two step process, where drivers will first call ib_umem_dmabuf_get_pinned_revocable_and_lock() which will pin and map the dmabuf (and provide a functional move_notify/invalidate_mappings callback), but will return with the lock still held so that the driver can then populate the callback via ib_umem_dmabuf_set_revoke_locked() without races from concurrent revocations. This scheme also allows for easier integration with drivers that may not have actually allocated their internal MR objects at the time of the get_pinned_revocable* call. Signed-off-by: Jacob Moroni Link: https://patch.msgid.link/20260305170826.3803155-4-jmoroni@google.com Signed-off-by: Leon Romanovsky Stable-dep-of: a846aecb931b ("RDMA/irdma: Prevent rereg_mr for non-mem regions") Signed-off-by: Sasha Levin

RDMA/umem: Move umem dmabuf revoke logic into helper function

2026-08-03T09:15:12+00:00

[ Upstream commit 797291a66ce346c96114b72222fc290d402da005 ] This same logic will eventually be reused from within the invalidate_mappings callback which already has the dma_resv_lock held, so break it out into a separate function so it can be reused. Signed-off-by: Jacob Moroni Link: https://patch.msgid.link/20260305170826.3803155-3-jmoroni@google.com Signed-off-by: Leon Romanovsky Stable-dep-of: a846aecb931b ("RDMA/irdma: Prevent rereg_mr for non-mem regions") Signed-off-by: Sasha Levin

RDMA/umem: Add ib_umem_dmabuf_get_pinned_and_lock helper

2026-08-03T09:15:12+00:00

[ Upstream commit 553dfa8cbd0c6d36adae042d9738ddf8f8765ac7 ] Move the inner logic of ib_umem_dmabuf_get_pinned_with_dma_device() to a new static function that returns with the lock held upon success. The intent is to allow reuse for the future get_pinned_revocable_and_lock function. Signed-off-by: Jacob Moroni Link: https://patch.msgid.link/20260305170826.3803155-2-jmoroni@google.com Signed-off-by: Leon Romanovsky Stable-dep-of: a846aecb931b ("RDMA/irdma: Prevent rereg_mr for non-mem regions") Signed-off-by: Sasha Levin

RDMA/umem: Introduce an option to revoke DMABUF umem

2026-08-03T09:15:12+00:00

[ Upstream commit 253c61dc256b3e6be65657f78b4a8452163ce00f ] Introduce an option to revoke DMABUF umem. This option will retain the umem allocation while revoking its DMA mapping. Furthermore, any subsequent attempts to map the pages should fail once the umem has been revoked. This functionality will be utilized in the upcoming patches in the series, where we aim to delay umem deallocation until the mkey deregistration. However, we must unmap its pages immediately. Signed-off-by: Yishai Hadas Link: https://patch.msgid.link/a38270f2fe4a194868ca2312f4c1c760e51bcbff.1722512548.git.leon@kernel.org Signed-off-by: Leon Romanovsky Stable-dep-of: a846aecb931b ("RDMA/irdma: Prevent rereg_mr for non-mem regions") Signed-off-by: Sasha Levin

RDMA/umem: Add support for creating pinned DMABUF umem with a given dma device

2026-08-03T09:15:12+00:00

[ Upstream commit 682358fd35dece838e6ae2d9d6a69fc0b9a9d411 ] Add support for creating pinned DMABUF umem with a specified DMA device instead of the DMA device of the given IB device. This API will be utilized in the upcoming patches of the series when multiple path DMAs are implemented. Signed-off-by: Yishai Hadas Link: https://patch.msgid.link/038aad36a43797e5591b20ba81051fc5758124f9.1722512548.git.leon@kernel.org Signed-off-by: Leon Romanovsky Stable-dep-of: a846aecb931b ("RDMA/irdma: Prevent rereg_mr for non-mem regions") Signed-off-by: Sasha Levin

RDMA/cma: Fix hardware address comparison length in netevent callback

2026-08-03T09:15:12+00:00

[ Upstream commit 18313833e2c6de222a4f6c072da759d0d5888528 ] The cited commit hardcoded the hardware address comparison len to ETH_ALEN. This breaks IPoIB, which uses 20-byte addresses. By truncating the memcmp, the CMA may incorrectly assume the target address is unchanged and fails to abort the stalled connection. Fix this by replacing ETH_ALEN with the dynamic neigh->dev->addr_len to correctly evaluate the full address regardless of the link layer. Fixes: 925d046e7e52 ("RDMA/core: Add a netevent notifier to cma") Signed-off-by: Or Gerlitz Link: https://patch.msgid.link/20260617-fix-cma-ipoib-v1-1-03f869344304@ddn.com Signed-off-by: Leon Romanovsky Signed-off-by: Sasha Levin

IB/mad: Drop unmatched RMPP responses before reassembly

2026-08-03T09:15:11+00:00

[ Upstream commit d2e52d610b9b09694261632340b801a421e0b0c5 ] Kernel-handled RMPP receive processing starts reassembly for active DATA responses before the response is matched to an outstanding send. The normal match happens later, after ib_process_rmpp_recv_wc() has either assembled a complete message or consumed the segment. That ordering lets an unsolicited response that routes to a kernel RMPP agent by the high TID bits allocate or extend RMPP receive state before the full TID and source address are checked against a real request. A reordered burst can therefore reach the receive-side insertion path even though the response would not match any send. For kernel-handled RMPP DATA responses, require the existing ib_find_send_mad() match before entering RMPP reassembly. The matcher already checks the full TID, management class and source address/GID against the agent wait, backlog and in-flight send lists. If there is no match, drop the response without creating RMPP state. This leaves the RMPP window behavior unchanged and only rejects responses that have no corresponding request. Fixes: fa619a77046b ("[PATCH] IB: Add RMPP implementation") Assisted-by: Codex:gpt-5-5-xhigh Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/3170ff3bc389a930bb1641f2caa394a0b2241579.1780774907.git.michael.bommarito@gmail.com Signed-off-by: Leon Romanovsky Signed-off-by: Sasha Levin

IB/cm: Fix av cm device leak on an error path in cm_init_av_by_path()

2026-07-24T14:03:07+00:00

[ Upstream commit 9b2207bc5cdb955bdae34b3eec80f04979e17081 ] Codex pointed out that cm_init_av_by_path() can call cm_set_av_port() which takes a reference on the cm device, but then can immediately return error if ib_init_ah_attr_from_path() fails. Since callers like ib_send_cm_req() put the av on the stack this leaks that cm device reference. Re-order cm_init_av_by_path() so it doesn't touch the av until it has done all its failable work, and then update the av in one shot so it is either left alone or fully init'd. Sashiko also pointed out that the cm_destroy_av() prior to cm_init_av_by_path() is harmful as it leaves the AV broken in the error case and thus the REJ won't send. Since cm_init_av_by_path() is now atomic it is safe to delete the cm_destroy_av(). On succees the av from cm_init_av_for_response() is cleaned up by cm_init_av_by_path(), on failure the 'goto rejected' guarentees the av is destroyed during ib_destroy_cm_id(). Fixes: 76039ac9095f ("IB/cm: Protect cm_dev, cm_ports and mad_agent with kref and lock") Link: https://patch.msgid.link/r/0-v1-38292501f539+14f-ib_cm_av_leak_jgg@nvidia.com Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

RDMA/counter: Fix incorrect port index in rdma_counter_init() error cleanup

2026-07-24T14:03:02+00:00

[ Upstream commit b86fd95805a7bd4c5b9465c9e7f75e45bbe7eb6f ] The error cleanup loop in rdma_counter_init() iterates with variable 'i' but accesses dev->port_data[port] instead of dev->port_data[i]. This causes the failed port's hstats to be freed multiple times while leaking hstats of previously initialized ports. Fixes: 56594ae1d250 ("RDMA/core: Annotate destroy of mutex to ensure that it is released as unlocked") Link: https://patch.msgid.link/r/20260520104546.1776253-3-cuitao@kylinos.cn Signed-off-by: Tao Cui Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin