kernel/net/sunrpc/xprtrdma, branch linux-rolling-lts

xprtrdma: Decrement re_receiving on the early exit paths

2026-03-19T15:08:12Z

[ 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

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

2026-02-26T22:59:27Z

[ 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

svcrdma: use rc_pageoff for memcpy byte offset

2026-01-02T11:57:24Z

commit a8ee9099f30654917aa68f55d707b5627e1dbf77 upstream. svc_rdma_copy_inline_range added rc_curpage (page index) to the page base instead of the byte offset rc_pageoff. Use rc_pageoff so copies land within the current page. Found by ZeroPath (https://zeropath.com) Fixes: 8e122582680c ("svcrdma: Move svc_rdma_read_info::ri_pageno to struct svc_rdma_recv_ctxt") Cc: stable@vger.kernel.org Signed-off-by: Joshua Rogers Signed-off-by: Chuck Lever Signed-off-by: Greg Kroah-Hartman

svcrdma: return 0 on success from svc_rdma_copy_inline_range

2026-01-02T11:57:24Z

commit 94972027ab55b200e031059fd6c7a649f8248020 upstream. The function comment specifies 0 on success and -EINVAL on invalid parameters. Make the tail return 0 after a successful copy loop. Fixes: d7cc73972661 ("svcrdma: support multiple Read chunks per RPC") Cc: stable@vger.kernel.org Signed-off-by: Joshua Rogers Signed-off-by: Chuck Lever Signed-off-by: Greg Kroah-Hartman

svcrdma: bound check rq_pages index in inline path

2026-01-02T11:57:24Z

commit d1bea0ce35b6095544ee82bb54156fc62c067e58 upstream. svc_rdma_copy_inline_range indexed rqstp->rq_pages[rc_curpage] without verifying rc_curpage stays within the allocated page array. Add guards before the first use and after advancing to a new page. Fixes: d7cc73972661 ("svcrdma: support multiple Read chunks per RPC") Cc: stable@vger.kernel.org Signed-off-by: Joshua Rogers Signed-off-by: Chuck Lever Signed-off-by: Greg Kroah-Hartman

SUNRPC: Remove redundant __GFP_NOWARN

2025-09-23T17:29:50Z

GFP_NOWAIT already includes __GFP_NOWARN, so let's remove the redundant __GFP_NOWARN. Signed-off-by: Qianfeng Rong Acked-by: Chuck Lever Signed-off-by: Anna Schumaker

svcrdma: Adjust the number of entries in svc_rdma_send_ctxt::sc_pages

2025-05-15T20:16:26Z

Allow allocation of more entries in the sc_pages[] array when the maximum size of an RPC message is increased. Reviewed-by: Jeff Layton Reviewed-by: NeilBrown Signed-off-by: Chuck Lever

svcrdma: Adjust the number of entries in svc_rdma_recv_ctxt::rc_pages

2025-05-15T20:16:26Z

Allow allocation of more entries in the rc_pages[] array when the maximum size of an RPC message is increased. Reviewed-by: Jeff Layton Reviewed-by: NeilBrown Signed-off-by: Chuck Lever

sunrpc: Replace the rq_pages array with dynamically-allocated memory

2025-05-15T20:16:22Z

As a step towards making NFSD's maximum rsize and wsize variable at run-time, replace the fixed-size rq_vec[] array in struct svc_rqst with a chunk of dynamically-allocated memory. On a system with 8-byte pointers and 4KB pages, pahole reports that the rq_pages[] array is 2080 bytes. This patch replaces that with a single 8-byte pointer field. Reviewed-by: Jeff Layton Reviewed-by: NeilBrown Signed-off-by: Chuck Lever

svcrdma: Reduce the number of rdma_rw contexts per-QP

2025-05-15T20:16:21Z

There is an upper bound on the number of rdma_rw contexts that can be created per QP. This invisible upper bound is because rdma_create_qp() adds one or more additional SQEs for each ctxt that the ULP requests via qp_attr.cap.max_rdma_ctxs. The QP's actual Send Queue length is on the order of the sum of qp_attr.cap.max_send_wr and a factor times qp_attr.cap.max_rdma_ctxs. The factor can be up to three, depending on whether MR operations are required before RDMA Reads. This limit is not visible to RDMA consumers via dev->attrs. When the limit is surpassed, QP creation fails with -ENOMEM. For example: svcrdma's estimate of the number of rdma_rw contexts it needs is three times the number of pages in RPCSVC_MAXPAGES. When MAXPAGES is about 260, the internally-computed SQ length should be: 64 credits + 10 backlog + 3 * (3 * 260) = 2414 Which is well below the advertised qp_max_wr of 32768. If RPCSVC_MAXPAGES is increased to 4MB, that's 1040 pages: 64 credits + 10 backlog + 3 * (3 * 1040) = 9434 However, QP creation fails. Dynamic printk for mlx5 shows: calc_sq_size:618:(pid 1514): send queue size (9326 * 256 / 64 -> 65536) exceeds limits(32768) Although 9326 is still far below qp_max_wr, QP creation still fails. Because the total SQ length calculation is opaque to RDMA consumers, there doesn't seem to be much that can be done about this except for consumers to try to keep the requested rdma_rw ctxt count low. Fixes: 2da0f610e733 ("svcrdma: Increase the per-transport rw_ctx count") Reviewed-by: NeilBrown Reviewed-by: Christoph Hellwig Signed-off-by: Chuck Lever