kernel/git/stable/linux-stable.git/drivers/infiniband, branch linux-7.1.y

RDMA/siw: bound Read Response placement to the RREAD length

2026-07-18T14:55:50+00:00

commit 7d29f7e9dbd844cae4d3e559cf78324b9642fd6b upstream. In drivers/infiniband/sw/siw/siw_qp_rx.c, siw_proc_rresp() places each inbound Read Response DDP segment at sge->laddr + wqe->processed and then accumulates wqe->processed, but it never checks the running total against the sink buffer length on continuation segments. siw_check_sge() resolves and validates the sink memory only on the first fragment (the if (!*mem) branch), and siw_rresp_check_ntoh() compares the cumulative length against wqe->bytes only on the final segment (the !frx->more_ddp_segs guard). A connected siw peer that answers an outstanding RREAD with Read Response segments that keep the DDP Last flag clear, carrying more total payload than the RREAD requested, drives wqe->processed past the validated sink buffer; the next siw_rx_data() call writes out of bounds at sge->laddr + wqe->processed. siw runs iWARP over ordinary routable TCP, so the peer is the remote end of an established RDMA connection and needs no local privilege. Bound every segment before placement, exactly as siw_proc_send() and siw_proc_write() already do for their tagged and untagged paths, and terminate the connection with a base-or-bounds DDP error when the Read Response would overrun the sink buffer. This is the second receive-path length fix for this file. A separate change rejects an MPA FPDU length that underflows the per-fragment remainder in the header decode; that guard does not cover this case, because here each individual segment length is self-consistent and only the accumulated placement offset overruns the buffer. Fixes: 8b6a361b8c48 ("rdma/siw: receive path") Link: https://patch.msgid.link/r/20260602194700.2273758-1-michael.bommarito@gmail.com Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman

RDMA/core: Fix broadcast address falsely detected as local

2026-07-18T14:55:50+00:00

commit 942cd47faa2047b46dfd85745603eba9006973e6 upstream. When rdma_resolve_addr() is invoked with a broadcast destination on an IPoIB interface, is_dst_local() inspects the resolved route and incorrectly concludes that the address is local. As a result, the resolution fails with -ENODEV. The issue stems from using '&' to compare rt_type with RTN_LOCAL. The RTN_* values form a sequential enum, not a bitmask (RTN_LOCAL=2, RTN_BROADCAST=3). Thus, "rt_type & RTN_LOCAL" yields a non-zero result for a broadcast route as well. Replace '&' with '==' when comparing rt_type against RTN_LOCAL. Link: https://patch.msgid.link/r/20260609-fix-rdma-resolve-addr-v1-1-449b8b4e6c09@nvidia.com Cc: stable@vger.kernel.org Fixes: c31e4038c97f ("RDMA/core: Use route entry flag to decide on loopback traffic") Signed-off-by: Maher Sanalla Reviewed-by: Vlad Dumitrescu Signed-off-by: Edward Srouji Reviewed-by: Parav Pandit Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman

RDMA/rtrs-srv: Bound RDMA-Write length to chunk size in rdma_write_sg

2026-07-18T14:55:50+00:00

commit 963af8d97a8c6a117134a8d0db1415e0489200b1 upstream. When the server answers an RTRS READ, rdma_write_sg() builds the source scatter/gather entry for the IB_WR_RDMA_WRITE that returns data to the peer. Its length is taken directly from the wire descriptor: plist->length = le32_to_cpu(id->rd_msg->desc[0].len); rd_msg points into the chunk buffer that the remote peer filled via RDMA-WRITE-WITH-IMM (rtrs_srv_rdma_done() -> process_io_req() -> process_read()), so desc[0].len is attacker-controlled and, before this change, was only rejected when zero. The source address is the fixed chunk start (dma_addr[msg_id]) and the source lkey is the PD-wide local_dma_lkey, which is not tied to the chunk's MR mapping, so the verbs layer does not constrain the transfer length to max_chunk_size. msg_id and off are bounded against queue_depth and max_chunk_size in rtrs_srv_rdma_done(), but desc[0].len is a separate field that was not checked against the chunk size. A peer that advertises desc[0].len larger than max_chunk_size can make the posted RDMA write read past the chunk's mapped region. The resulting behaviour depends on the IOMMU configuration: with no IOMMU or in passthrough mode the read may extend into memory adjacent to the chunk and be returned to the peer, which can disclose host memory; with a translating IOMMU the out-of-range access is expected to fault and abort the connection. In either case the transfer exceeds what the protocol permits and is driven by a remote peer. Reject a descriptor length above max_chunk_size, mirroring the existing off >= max_chunk_size bound in rtrs_srv_rdma_done(). Legitimate clients do not exceed it: the client sets desc[0].len to its MR length, which is capped at the negotiated max_io_size (max_chunk_size - MAX_HDR_SIZE). Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://patch.msgid.link/r/20260612-master-v1-1-70cde5c6fdc9@gmail.com Reported-by: Yuhao Jiang Cc: stable@vger.kernel.org Signed-off-by: Zhenhao Wan Reviewed-by: Md Haris Iqbal Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman

RDMA/srp: bound SRP_RSP sense copy by the received length

2026-06-08T16:43:23+00:00

srp_process_rsp() copies sense data from rsp->data + resp_data_len, where resp_data_len is the full 32-bit value supplied by the SRP target and is never checked against the number of bytes actually received (wc->byte_len). The copy length is bounded to SCSI_SENSE_BUFFERSIZE, so at most 96 bytes are copied, but the source offset is not bounded. A malicious or compromised SRP target on the InfiniBand/RoCE fabric that the initiator has logged into can return an SRP_RSP with SRP_RSP_FLAG_SNSVALID set and a large resp_data_len. The receive buffer is allocated at the target-chosen max_ti_iu_len, so the source of the sense copy lands past the bytes actually received; with resp_data_len near 0xFFFFFFFF it is gigabytes past the buffer and the read faults. Copy the sense data only if it has not been truncated, that is, only if the response header, the response data, and the sense region fit within the bytes actually received; otherwise drop the sense and log. The in-tree iSER and NVMe-RDMA receive paths already bound their parse by wc->byte_len; this brings ib_srp into line with them. Fixes: aef9ec39c47f ("IB: Add SCSI RDMA Protocol (SRP) initiator") Link: https://patch.msgid.link/r/20260602220457.2542840-1-michael.bommarito@gmail.com Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito Reviewed-by: Bart Van Assche Signed-off-by: Jason Gunthorpe

IB/isert: Reject login PDUs shorter than ISER_HEADERS_LEN

2026-06-08T16:43:23+00:00

In drivers/infiniband/ulp/isert/ib_isert.c, isert_login_recv_done() computes the login request payload length as wc->byte_len minus ISER_HEADERS_LEN with no lower bound, and login_req_len is a signed int. A remote iSER initiator can post a login Send work request carrying fewer than ISER_HEADERS_LEN (76) bytes, so the subtraction underflows and login_req_len becomes negative. isert_rx_login_req() then reads that negative length back into a signed int, takes size = min(rx_buflen, MAX_KEY_VALUE_PAIRS), and because the min() is signed it keeps the negative value; the value is then passed as the memcpy() length and sign-extended to a multi-gigabyte size_t. The copy into the 8192-byte login->req_buf runs far out of bounds and faults, crashing the target node. The login phase precedes iSCSI authentication, so no credentials are required to reach this path. Reject any login PDU shorter than ISER_HEADERS_LEN before the subtraction, mirroring the existing early return on a failed work completion, so login_req_len can never go negative. The upper bound was already safe: a posted login buffer cannot deliver more than ISER_RX_PAYLOAD_SIZE, so the difference stays at or below MAX_KEY_VALUE_PAIRS and the existing min() clamps it; only the missing lower bound needs to be added. Fixes: b8d26b3be8b3 ("iser-target: Add iSCSI Extensions for RDMA (iSER) target driver") Link: https://patch.msgid.link/r/20260602194642.2273217-1-michael.bommarito@gmail.com Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito Signed-off-by: Jason Gunthorpe

RDMA: During rereg_mr ensure that REREG_ACCESS is compatible

2026-06-08T16:39:20+00:00

If IB_MR_REREG_ACCESS changes from RO to RW then the umem has to be re-evaluated to ensure it is properly pinned as RW. Since the umem is hidden inside each driver's mr struct add a ib_umem_check_rereg() function that each driver has to call before processing IB_MR_REREG_ACCESS. mlx4 has to retain its duplicate ib_access_writable check because it implements IB_MR_REREG_ACCESS | IB_MR_REREG_TRANS by changing both items in place sequentially while the MR is live, so it will continue to not support this combination. Cc: stable@vger.kernel.org Fixes: b40656aa7d55 ("RDMA/umem: remove FOLL_FORCE usage") Link: https://patch.msgid.link/r/0-v1-06fb1a2d6cf5+107-rereg_access_jgg@nvidia.com Reported-by: Philip Tsukerman Signed-off-by: Jason Gunthorpe

RDMA/core: Validate cpu_id against nr_cpu_ids in DMAH alloc

2026-06-03T17:55:48+00:00

The cpu_id attribute supplied by user space through UVERBS_ATTR_ALLOC_DMAH_CPU_ID is passed directly to cpumask_test_cpu() without first verifying that the value is within the valid CPU range. Passing such untrusted data to cpumask_test_cpu() may lead to an out-of-bounds read of the underlying cpumask bitmap: the helper expands to a test_bit() that indexes the bitmap by cpu_id / BITS_PER_LONG with no bound check. In addition, on kernels built with CONFIG_DEBUG_PER_CPU_MAPS it trips the WARN_ON_ONCE() in cpumask_check(); combined with panic_on_warn this turns a bad user input into a machine reboot. Reject any cpu_id that is not smaller than nr_cpu_ids with -EINVAL before it is used. Reported by Smatch. Fixes: d83edab562a4 ("RDMA/core: Introduce a DMAH object and its alloc/free APIs") Link: https://patch.msgid.link/r/20260525142136.28165-1-yishaih@nvidia.com Cc: stable@vger.kernel.org Reported-by: Dan Carpenter Closes: https://lore.kernel.org/r/ag68qoAW3P04J7pT@stanley.mountain/ Signed-off-by: Yishai Hadas Signed-off-by: Jason Gunthorpe

RDMA/umem: Fix truncation for block sizes >= 4G

2026-06-03T12:37:34+00:00

When the iommu is used the linearization of the mapping can give a single block that is very large split across multiple SG entries. When __rdma_block_iter_next() reassembles the split SG entries it is overflowing the 32 bit stack values and computed the wrong DMA addresses for blocks after the truncation. Use the right types to hold DMA addresses. Link: https://patch.msgid.link/r/1-v1-88303e9e509f+f7-ib_umem_types_jgg@nvidia.com Cc: stable@vger.kernel.org Fixes: a808273a495c ("RDMA/verbs: Add a DMA iterator to return aligned contiguous memory blocks") Signed-off-by: Jason Gunthorpe

RDMA/efa: Validate SQ ring size against max LLQ size

2026-06-02T17:48:59+00:00

Validate the SQ ring size against the device's max LLQ size. This ensures that when using 128-byte WQEs, userspace cannot exceed the queue limits. On create QP, userspace provides the SQ ring size (depth x WQE size) which is validated against the max LLQ size. Fixes: 40909f664d27 ("RDMA/efa: Add EFA verbs implementation") Link: https://patch.msgid.link/r/20260526081536.1203553-1-ynachum@amazon.com Reviewed-by: Michael Margolin Signed-off-by: Yonatan Nachum Signed-off-by: Jason Gunthorpe

RDMA/core: Validate the passed in fops for ib_get_ucaps()

2026-06-02T00:19:11+00:00

Sashiko pointed out it is not safe to rely only on the devt because char/block alias so if the user finds a block device with the same dev_t it can masquerade as a ucap cdev fd. Test the f_ops to only accept authentic cdevs. Link: https://patch.msgid.link/r/0-v1-fd9482545e37+1e25-ib_ucaps_fd_ops_jgg@nvidia.com Cc: stable@vger.kernel.org Fixes: 61e51682816d ("RDMA/uverbs: Introduce UCAP (User CAPabilities) API") Signed-off-by: Jason Gunthorpe