kernel/linux.git/net/sunrpc, branch v5.15.212

xprtrdma: Fix bcall rep leak and unbounded peek

2026-07-24T13:51:38+00:00

[ Upstream commit c7653d5cebc8492c77ec0415b5e9c0fb3e644bc6 ] rpcrdma_is_bcall() decodes a reply's first words to decide whether the frame is a backchannel call. Two issues in that decode path let a short or malformed reply leak the receive buffer and drain the Receive queue. First, the speculative peek p = xdr_inline_decode(xdr, 0); /* five p++ reads follow */ asks xdr_inline_decode() for zero bytes, which returns xdr->p without consulting xdr->end. The five subsequent __be32 reads can then walk up to 20 bytes past the wire payload into stale regbuf contents and misclassify the reply as a backchannel call. Second, after the post-peek p = xdr_inline_decode(xdr, 3 * sizeof(*p)); if (unlikely(!p)) return true; the short-header arm returns true without calling rpcrdma_bc_receive_call(). The contract with the caller is that a true return transfers ownership of rep to the backchannel path: rpcrdma_reply_handler() if (rpcrdma_is_bcall(r_xprt, rep)) return; /* bare return, skips out_post */ ... out_post: rpcrdma_post_recvs(r_xprt, credits + ...); Because rpcrdma_bc_receive_call() never ran, no one took rep, but rpcrdma_reply_handler still bare-returns past rpcrdma_rep_put() and rpcrdma_post_recvs(). The rep, with its persistently DMA-mapped receive buffer, is orphaned on rb_all_reps and freed only at transport teardown. This completion reposts nothing, so its slot is reclaimed only when a later forward-channel reply reaches out_post and rpcrdma_post_recvs() allocates a fresh rep to backfill; absent that traffic the Receive queue drains and the peer's Sends draw RNR NAKs. Fix by consulting xdr->end after the zero-length peek so the five __be32 reads cannot run unless 20 bytes of wire payload remain. A byte-precise comparison against xdr->end is required because a non-4-aligned receive rounds the stream's word count up past the true payload. Also return false from the short-header arm so the reply falls through the normal out_norqst cleanup chain (rpcrdma_rep_put() plus rpcrdma_post_recvs()). Fixes: 41c8f70f5a3d ("xprtrdma: Harden backchannel call decoding") Assisted-by: kres:claude-opus-4-7 Signed-off-by: Chris Mason Signed-off-by: Chuck Lever Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin

use less confusing names for iov_iter direction initializers

2026-06-19T11:33:46+00:00

[ Upstream commit de4eda9de2d957ef2d6a8365a01e26a435e958cb ] READ/WRITE proved to be actively confusing - the meanings are "data destination, as used with read(2)" and "data source, as used with write(2)", but people keep interpreting those as "we read data from it" and "we write data to it", i.e. exactly the wrong way. Call them ITER_DEST and ITER_SOURCE - at least that is harder to misinterpret... Signed-off-by: Al Viro Stable-dep-of: a4f0b001782b ("vsock/virtio: reset connection on receiving queue overflow") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

SUNRPC: Do not dereference non-socket transports in sysfs

2026-06-01T15:35:44+00:00

[ Upstream commit 421ab1be43bd015ffe744f4ea25df4f19d1ce6fe ] Do not cast the struct xprt to a sock_xprt unless we know it is a UDP or TCP transport. Otherwise the call to lock the mutex will scribble over whatever structure is actually there. This has been seen to cause hard system lockups when the underlying transport was RDMA. Fixes: b49ea673e119 ("SUNRPC: lock against ->sock changing during sysfs read") Cc: stable@vger.kernel.org Signed-off-by: Trond Myklebust Signed-off-by: Sasha Levin

SUNRPC: Check if the xprt is connected before handling sysfs reads

2026-06-01T15:35:44+00:00

[ Upstream commit 17f09d3f619a7ad2d2b021b4e5246f08225b1b0f ] xprts don't immediately reconnect when changing the "dstaddr" property, instead this gets handled the next time an operation uses the transport. This could lead to NULL pointer dereferences when trying to read sysfs files between the disconnect and reconnect operations. Fix this by returning an error if the xprt is not connected. Signed-off-by: Anna Schumaker Signed-off-by: Trond Myklebust Stable-dep-of: 421ab1be43bd ("SUNRPC: Do not dereference non-socket transports in sysfs") Signed-off-by: Sasha Levin

SUNRPC: lock against ->sock changing during sysfs read

2026-06-01T15:35:11+00:00

[ Upstream commit b49ea673e119f59c71645e2f65b3ccad857c90ee ] ->sock can be set to NULL asynchronously unless ->recv_mutex is held. So it is important to hold that mutex. Otherwise a sysfs read can trigger an oops. Commit 17f09d3f619a ("SUNRPC: Check if the xprt is connected before handling sysfs reads") appears to attempt to fix this problem, but it only narrows the race window. Fixes: 17f09d3f619a ("SUNRPC: Check if the xprt is connected before handling sysfs reads") Fixes: a8482488a7d6 ("SUNRPC query transport's source port") Signed-off-by: NeilBrown Signed-off-by: Anna Schumaker Signed-off-by: Robert Garcia Signed-off-by: Greg Kroah-Hartman

sunrpc: fix cache_request leak in cache_release

2026-04-18T08:33:17+00:00

commit 17ad31b3a43b72aec3a3d83605891e1397d0d065 upstream. When a reader's file descriptor is closed while in the middle of reading a cache_request (rp->offset != 0), cache_release() decrements the request's readers count but never checks whether it should free the request. In cache_read(), when readers drops to 0 and CACHE_PENDING is clear, the cache_request is removed from the queue and freed along with its buffer and cache_head reference. cache_release() lacks this cleanup. The only other path that frees requests with readers == 0 is cache_dequeue(), but it runs only when CACHE_PENDING transitions from set to clear. If that transition already happened while readers was still non-zero, cache_dequeue() will have skipped the request, and no subsequent call will clean it up. Add the same cleanup logic from cache_read() to cache_release(): after decrementing readers, check if it reached 0 with CACHE_PENDING clear, and if so, dequeue and free the cache_request. Reported-by: NeilBrown Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@kernel.org Signed-off-by: Jeff Layton Signed-off-by: Chuck Lever Signed-off-by: Greg Kroah-Hartman

xprtrdma: Decrement re_receiving on the early exit paths

2026-04-18T08:33:11+00:00

[ Upstream commit 7b6275c80a0c81c5f8943272292dfe67730ce849 ] In the event that rpcrdma_post_recvs() fails to create a work request (due to memory allocation failure, say) or otherwise exits early, we should decrement ep->re_receiving before returning. Otherwise we will hang in rpcrdma_xprt_drain() as re_receiving will never reach zero and the completion will never be triggered. On a system with high memory pressure, this can appear as the following hung task: INFO: task kworker/u385:17:8393 blocked for more than 122 seconds. Tainted: G S E 6.19.0 #3 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:kworker/u385:17 state:D stack:0 pid:8393 tgid:8393 ppid:2 task_flags:0x4248060 flags:0x00080000 Workqueue: xprtiod xprt_autoclose [sunrpc] Call Trace: __schedule+0x48b/0x18b0 ? ib_post_send_mad+0x247/0xae0 [ib_core] schedule+0x27/0xf0 schedule_timeout+0x104/0x110 __wait_for_common+0x98/0x180 ? __pfx_schedule_timeout+0x10/0x10 wait_for_completion+0x24/0x40 rpcrdma_xprt_disconnect+0x444/0x460 [rpcrdma] xprt_rdma_close+0x12/0x40 [rpcrdma] xprt_autoclose+0x5f/0x120 [sunrpc] process_one_work+0x191/0x3e0 worker_thread+0x2e3/0x420 ? __pfx_worker_thread+0x10/0x10 kthread+0x10d/0x230 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x273/0x2b0 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 Fixes: 15788d1d1077 ("xprtrdma: Do not refresh Receive Queue while it is draining") Signed-off-by: Eric Badger Reviewed-by: Chuck Lever Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin

SUNRPC: fix gss_auth kref leak in gss_alloc_msg error path

2026-03-04T12:19:55+00:00

commit dd2fdc3504592d85e549c523b054898a036a6afe upstream. Commit 5940d1cf9f42 ("SUNRPC: Rebalance a kref in auth_gss.c") added a kref_get(&gss_auth->kref) call to balance the gss_put_auth() done in gss_release_msg(), but forgot to add a corresponding kref_put() on the error path when kstrdup_const() fails. If service_name is non-NULL and kstrdup_const() fails, the function jumps to err_put_pipe_version which calls put_pipe_version() and kfree(gss_msg), but never releases the gss_auth reference. This leads to a kref leak where the gss_auth structure is never freed. Add a forward declaration for gss_free_callback() and call kref_put() in the err_put_pipe_version error path to properly release the reference taken earlier. Fixes: 5940d1cf9f42 ("SUNRPC: Rebalance a kref in auth_gss.c") Cc: stable@vger.kernel.org Signed-off-by: Daniel Hodges Signed-off-by: Anna Schumaker Signed-off-by: Greg Kroah-Hartman

SUNRPC: auth_gss: fix memory leaks in XDR decoding error paths

2026-03-04T12:19:55+00:00

commit 3e6397b056335cc56ef0e9da36c95946a19f5118 upstream. The gssx_dec_ctx(), gssx_dec_status(), and gssx_dec_name() functions allocate memory via gssx_dec_buffer(), which calls kmemdup(). When a subsequent decode operation fails, these functions return immediately without freeing previously allocated buffers, causing memory leaks. The leak in gssx_dec_ctx() is particularly relevant because the caller (gssp_accept_sec_context_upcall) initializes several buffer length fields to non-zero values, resulting in memory allocation: struct gssx_ctx rctxh = { .exported_context_token.len = GSSX_max_output_handle_sz, .mech.len = GSS_OID_MAX_LEN, .src_name.display_name.len = GSSX_max_princ_sz, .targ_name.display_name.len = GSSX_max_princ_sz }; If, for example, gssx_dec_name() succeeds for src_name but fails for targ_name, the memory allocated for exported_context_token, mech, and src_name.display_name remains unreferenced and cannot be reclaimed. Add error handling with goto-based cleanup to free any previously allocated buffers before returning an error. Reported-by: Xingjing Deng Closes: https://lore.kernel.org/linux-nfs/CAK+ZN9qttsFDu6h1FoqGadXjMx1QXqPMoYQ=6O9RY4SxVTvKng@mail.gmail.com/ Fixes: 1d658336b05f ("SUNRPC: Add RPC based upcall mechanism for RPCGSS auth") Cc: stable@vger.kernel.org Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever Signed-off-by: Greg Kroah-Hartman

RDMA/core: add rdma_rw_max_sge() helper for SQ sizing

2026-03-04T12:19:43+00:00

[ Upstream commit afcae7d7b8a278a6c29e064f99e5bafd4ac1fb37 ] svc_rdma_accept() computes sc_sq_depth as the sum of rq_depth and the number of rdma_rw contexts (ctxts). This value is used to allocate the Send CQ and to initialize the sc_sq_avail credit pool. However, when the device uses memory registration for RDMA operations, rdma_rw_init_qp() inflates the QP's max_send_wr by a factor of three per context to account for REG and INV work requests. The Send CQ and credit pool remain sized for only one work request per context, causing Send Queue exhaustion under heavy NFS WRITE workloads. Introduce rdma_rw_max_sge() to compute the actual number of Send Queue entries required for a given number of rdma_rw contexts. Upper layer protocols call this helper before creating a Queue Pair so that their Send CQs and credit accounting match the QP's true capacity. Update svc_rdma_accept() to use rdma_rw_max_sge() when computing sc_sq_depth, ensuring the credit pool reflects the work requests that rdma_rw_init_qp() will reserve. Reviewed-by: Christoph Hellwig Fixes: 00bd1439f464 ("RDMA/rw: Support threshold for registration vs scattering to local pages") Signed-off-by: Chuck Lever Link: https://patch.msgid.link/20260128005400.25147-5-cel@kernel.org Signed-off-by: Leon Romanovsky Signed-off-by: Sasha Levin