{"api_version":"1","generated_at":"2026-08-15T08:04:25+00:00","cve":"CVE-2026-72473","urls":{"html":"https://cve.report/CVE-2026-72473","api":"https://cve.report/api/cve/CVE-2026-72473.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-72473","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-72473"},"summary":{"title":"xprtrdma: Decouple req recycling from RPC completion","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nxprtrdma: Decouple req recycling from RPC completion\n\nrl_kref formerly served two distinct lifetimes through a single\nrefcount: it gated when a Reply could wake its RPC task, and it\ngated when an rpcrdma_req could return to its free pool. The\nmarshal path took the Send-side reference only when SGEs needed\nDMA-unmap (sc_unmap_count > 0), which made a Send carrying only\npre-registered buffers an exception: the Reply handler dropped\nrl_kref from 1 to 0 and freed the req while the HCA might still\nbe DMA-reading from its send buffer.\n\nGive rl_kref a narrower job. The RPC layer takes one reference\nwhen slot allocation hands a req out. rpcrdma_prepare_send_sges()\ntakes a Send-side reference unconditionally after WR preparation\nsucceeds. xprt_rdma_free_slot() and xprt_rdma_bc_free_rqst() drop\nthe RPC-layer reference; rpcrdma_sendctx_unmap() drops the\nSend-side reference. The req returns to its free pool only after\nboth owners have signed off.\n\nThe existing kref_init(&req->rl_kref) call in\nrpcrdma_prepare_send_sges() is removed. Initialization moves to\nthe slot-allocation paths (xprt_rdma_alloc_slot and\nrpcrdma_bc_rqst_get), and the release callback re-arms rl_kref\nbefore the req returns to a free pool. A re-init in the marshal\npath would discard the RPC-layer reference that already exists\non entry.\n\nThree invariants follow:\n\n  - Any rpcrdma_req held by an rpc_rqst has rl_kref >= 1.\n    xprt_rdma_alloc_slot(), rpcrdma_bc_rqst_get(), and the\n    backlog-wake branch in xprt_rdma_alloc_slot() each kref_init\n    rl_kref before publishing the req. Without this invariant,\n    an RPC task that aborts between slot allocation and marshal\n    (gss_refresh failure or signal during call_connect, for\n    example) would drive xprt_release() ->\n    xprt_rdma_free_slot() -> kref_put against a refcount of\n    zero, saturating refcount_t and stranding the slot.\n\n  - The Send-side reference is taken only after WR prep\n    succeeds. A mapping failure in rpcrdma_prepare_send_sges()\n    runs rpcrdma_sendctx_cancel(), which DMA-unmaps the sendctx\n    and clears sc_req without touching rl_kref. The sendctx\n    ring walks in rpcrdma_sendctx_put_locked() and\n    rpcrdma_sendctxs_destroy() skip entries with sc_req == NULL,\n    so a burst of -EIO marshal failures cannot hold reqs off\n    rb_send_bufs.\n\n  - The release callback re-arms rl_kref so the next consumer\n    enters with the invariant satisfied.\n\nReplies now complete the RPC directly. rpcrdma_reply_handler()\ncalls rpcrdma_complete_rqst() in place of kref_put on the\nnon-LocalInv branch. The LocalInv branch already completes the\nRPC from frwr_unmap_async() and is unaffected.\n\nBecause Send-side references can now outlive RPC completion,\nconnection teardown drains sendctx entries whose unsignaled\nSends never had a later signaled completion to walk the ring.\nrpcrdma_sendctxs_destroy() walks the active range and runs\nrpcrdma_sendctx_unmap() on each entry with a non-NULL sc_req\nbefore the request buffers are reset, and is moved ahead of\nrpcrdma_reqs_reset() in rpcrdma_xprt_disconnect() so the reqs\nare still in their pre-reset state when the Send-side refs are\nreleased.\n\nThe drain creates a teardown-ordering hazard on the backchannel\npath. With the new lifetime, releasing a bc_prealloc req from\nrpcrdma_req_release() re-adds it to bc_pa_list. The disconnect\nin xprt_rdma_destroy() runs after xprt_destroy_backchannel() has\nalready emptied bc_pa_list, so the drained reqs would otherwise\nleak. xprt_rdma_destroy() now runs xprt_rdma_bc_destroy(xprt, 0)\na second time after the disconnect to reclaim them.","state":"PUBLISHED","assigner":"Linux","published_at":"2026-08-15 06:22:21","updated_at":"2026-08-15 06:22:21"},"problem_types":[],"metrics":[],"references":[{"url":"https://git.kernel.org/stable/c/9f3d9b68c1c6c51746e5ecdb52b2e6a2901de37e","name":"https://git.kernel.org/stable/c/9f3d9b68c1c6c51746e5ecdb52b2e6a2901de37e","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/740975054a1970c0cf15f70ac39724a064f45847","name":"https://git.kernel.org/stable/c/740975054a1970c0cf15f70ac39724a064f45847","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/e7632089523acddcdd8f090ad19e96fb3107b04d","name":"https://git.kernel.org/stable/c/e7632089523acddcdd8f090ad19e96fb3107b04d","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/e786233d2e0bbff9a82e43f02ae3a46ab4b08ec3","name":"https://git.kernel.org/stable/c/e786233d2e0bbff9a82e43f02ae3a46ab4b08ec3","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/8203f760a72bd39a3b66bc4eff0aa272a99fe22b","name":"https://git.kernel.org/stable/c/8203f760a72bd39a3b66bc4eff0aa272a99fe22b","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/53442c7d0c888e51b8bc3da196970a669cc6b294","name":"https://git.kernel.org/stable/c/53442c7d0c888e51b8bc3da196970a669cc6b294","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-72473","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-72473","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 0ab115237025f5e379620bbcd56a02697d07b002 740975054a1970c0cf15f70ac39724a064f45847 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 0ab115237025f5e379620bbcd56a02697d07b002 9f3d9b68c1c6c51746e5ecdb52b2e6a2901de37e git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 0ab115237025f5e379620bbcd56a02697d07b002 e7632089523acddcdd8f090ad19e96fb3107b04d git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 0ab115237025f5e379620bbcd56a02697d07b002 53442c7d0c888e51b8bc3da196970a669cc6b294 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 0ab115237025f5e379620bbcd56a02697d07b002 8203f760a72bd39a3b66bc4eff0aa272a99fe22b git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 0ab115237025f5e379620bbcd56a02697d07b002 e786233d2e0bbff9a82e43f02ae3a46ab4b08ec3 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 5.3","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 5.3 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.1.178 6.1.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.6.145 6.6.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.12.97 6.12.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.18.40 6.18.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.1.5 7.1.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2-rc1 * original_commit_for_fix","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":null,"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"defaultStatus":"unaffected","product":"Linux","programFiles":["net/sunrpc/xprtrdma/backchannel.c","net/sunrpc/xprtrdma/rpc_rdma.c","net/sunrpc/xprtrdma/transport.c","net/sunrpc/xprtrdma/verbs.c","net/sunrpc/xprtrdma/xprt_rdma.h"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"740975054a1970c0cf15f70ac39724a064f45847","status":"affected","version":"0ab115237025f5e379620bbcd56a02697d07b002","versionType":"git"},{"lessThan":"9f3d9b68c1c6c51746e5ecdb52b2e6a2901de37e","status":"affected","version":"0ab115237025f5e379620bbcd56a02697d07b002","versionType":"git"},{"lessThan":"e7632089523acddcdd8f090ad19e96fb3107b04d","status":"affected","version":"0ab115237025f5e379620bbcd56a02697d07b002","versionType":"git"},{"lessThan":"53442c7d0c888e51b8bc3da196970a669cc6b294","status":"affected","version":"0ab115237025f5e379620bbcd56a02697d07b002","versionType":"git"},{"lessThan":"8203f760a72bd39a3b66bc4eff0aa272a99fe22b","status":"affected","version":"0ab115237025f5e379620bbcd56a02697d07b002","versionType":"git"},{"lessThan":"e786233d2e0bbff9a82e43f02ae3a46ab4b08ec3","status":"affected","version":"0ab115237025f5e379620bbcd56a02697d07b002","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["net/sunrpc/xprtrdma/backchannel.c","net/sunrpc/xprtrdma/rpc_rdma.c","net/sunrpc/xprtrdma/transport.c","net/sunrpc/xprtrdma/verbs.c","net/sunrpc/xprtrdma/xprt_rdma.h"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"5.3"},{"lessThan":"5.3","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"6.1.*","status":"unaffected","version":"6.1.178","versionType":"semver"},{"lessThanOrEqual":"6.6.*","status":"unaffected","version":"6.6.145","versionType":"semver"},{"lessThanOrEqual":"6.12.*","status":"unaffected","version":"6.12.97","versionType":"semver"},{"lessThanOrEqual":"6.18.*","status":"unaffected","version":"6.18.40","versionType":"semver"},{"lessThanOrEqual":"7.1.*","status":"unaffected","version":"7.1.5","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.2-rc1","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.1.178","versionStartIncluding":"5.3","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.6.145","versionStartIncluding":"5.3","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.12.97","versionStartIncluding":"5.3","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.18.40","versionStartIncluding":"5.3","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.1.5","versionStartIncluding":"5.3","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.2-rc1","versionStartIncluding":"5.3","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nxprtrdma: Decouple req recycling from RPC completion\n\nrl_kref formerly served two distinct lifetimes through a single\nrefcount: it gated when a Reply could wake its RPC task, and it\ngated when an rpcrdma_req could return to its free pool. The\nmarshal path took the Send-side reference only when SGEs needed\nDMA-unmap (sc_unmap_count > 0), which made a Send carrying only\npre-registered buffers an exception: the Reply handler dropped\nrl_kref from 1 to 0 and freed the req while the HCA might still\nbe DMA-reading from its send buffer.\n\nGive rl_kref a narrower job. The RPC layer takes one reference\nwhen slot allocation hands a req out. rpcrdma_prepare_send_sges()\ntakes a Send-side reference unconditionally after WR preparation\nsucceeds. xprt_rdma_free_slot() and xprt_rdma_bc_free_rqst() drop\nthe RPC-layer reference; rpcrdma_sendctx_unmap() drops the\nSend-side reference. The req returns to its free pool only after\nboth owners have signed off.\n\nThe existing kref_init(&req->rl_kref) call in\nrpcrdma_prepare_send_sges() is removed. Initialization moves to\nthe slot-allocation paths (xprt_rdma_alloc_slot and\nrpcrdma_bc_rqst_get), and the release callback re-arms rl_kref\nbefore the req returns to a free pool. A re-init in the marshal\npath would discard the RPC-layer reference that already exists\non entry.\n\nThree invariants follow:\n\n  - Any rpcrdma_req held by an rpc_rqst has rl_kref >= 1.\n    xprt_rdma_alloc_slot(), rpcrdma_bc_rqst_get(), and the\n    backlog-wake branch in xprt_rdma_alloc_slot() each kref_init\n    rl_kref before publishing the req. Without this invariant,\n    an RPC task that aborts between slot allocation and marshal\n    (gss_refresh failure or signal during call_connect, for\n    example) would drive xprt_release() ->\n    xprt_rdma_free_slot() -> kref_put against a refcount of\n    zero, saturating refcount_t and stranding the slot.\n\n  - The Send-side reference is taken only after WR prep\n    succeeds. A mapping failure in rpcrdma_prepare_send_sges()\n    runs rpcrdma_sendctx_cancel(), which DMA-unmaps the sendctx\n    and clears sc_req without touching rl_kref. The sendctx\n    ring walks in rpcrdma_sendctx_put_locked() and\n    rpcrdma_sendctxs_destroy() skip entries with sc_req == NULL,\n    so a burst of -EIO marshal failures cannot hold reqs off\n    rb_send_bufs.\n\n  - The release callback re-arms rl_kref so the next consumer\n    enters with the invariant satisfied.\n\nReplies now complete the RPC directly. rpcrdma_reply_handler()\ncalls rpcrdma_complete_rqst() in place of kref_put on the\nnon-LocalInv branch. The LocalInv branch already completes the\nRPC from frwr_unmap_async() and is unaffected.\n\nBecause Send-side references can now outlive RPC completion,\nconnection teardown drains sendctx entries whose unsignaled\nSends never had a later signaled completion to walk the ring.\nrpcrdma_sendctxs_destroy() walks the active range and runs\nrpcrdma_sendctx_unmap() on each entry with a non-NULL sc_req\nbefore the request buffers are reset, and is moved ahead of\nrpcrdma_reqs_reset() in rpcrdma_xprt_disconnect() so the reqs\nare still in their pre-reset state when the Send-side refs are\nreleased.\n\nThe drain creates a teardown-ordering hazard on the backchannel\npath. With the new lifetime, releasing a bc_prealloc req from\nrpcrdma_req_release() re-adds it to bc_pa_list. The disconnect\nin xprt_rdma_destroy() runs after xprt_destroy_backchannel() has\nalready emptied bc_pa_list, so the drained reqs would otherwise\nleak. xprt_rdma_destroy() now runs xprt_rdma_bc_destroy(xprt, 0)\na second time after the disconnect to reclaim them."}],"providerMetadata":{"dateUpdated":"2026-08-15T05:57:14.412Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/740975054a1970c0cf15f70ac39724a064f45847"},{"url":"https://git.kernel.org/stable/c/9f3d9b68c1c6c51746e5ecdb52b2e6a2901de37e"},{"url":"https://git.kernel.org/stable/c/e7632089523acddcdd8f090ad19e96fb3107b04d"},{"url":"https://git.kernel.org/stable/c/53442c7d0c888e51b8bc3da196970a669cc6b294"},{"url":"https://git.kernel.org/stable/c/8203f760a72bd39a3b66bc4eff0aa272a99fe22b"},{"url":"https://git.kernel.org/stable/c/e786233d2e0bbff9a82e43f02ae3a46ab4b08ec3"}],"title":"xprtrdma: Decouple req recycling from RPC completion","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-72473","datePublished":"2026-08-15T05:57:14.412Z","dateReserved":"2026-08-09T03:40:39.934Z","dateUpdated":"2026-08-15T05:57:14.412Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-08-15 06:22:21","lastModifiedDate":"2026-08-15 06:22:21","problem_types":[],"metrics":[],"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"72473","Ordinal":"1","Title":"xprtrdma: Decouple req recycling from RPC completion","CVE":"CVE-2026-72473","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"72473","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\nxprtrdma: Decouple req recycling from RPC completion\n\nrl_kref formerly served two distinct lifetimes through a single\nrefcount: it gated when a Reply could wake its RPC task, and it\ngated when an rpcrdma_req could return to its free pool. The\nmarshal path took the Send-side reference only when SGEs needed\nDMA-unmap (sc_unmap_count > 0), which made a Send carrying only\npre-registered buffers an exception: the Reply handler dropped\nrl_kref from 1 to 0 and freed the req while the HCA might still\nbe DMA-reading from its send buffer.\n\nGive rl_kref a narrower job. The RPC layer takes one reference\nwhen slot allocation hands a req out. rpcrdma_prepare_send_sges()\ntakes a Send-side reference unconditionally after WR preparation\nsucceeds. xprt_rdma_free_slot() and xprt_rdma_bc_free_rqst() drop\nthe RPC-layer reference; rpcrdma_sendctx_unmap() drops the\nSend-side reference. The req returns to its free pool only after\nboth owners have signed off.\n\nThe existing kref_init(&req->rl_kref) call in\nrpcrdma_prepare_send_sges() is removed. Initialization moves to\nthe slot-allocation paths (xprt_rdma_alloc_slot and\nrpcrdma_bc_rqst_get), and the release callback re-arms rl_kref\nbefore the req returns to a free pool. A re-init in the marshal\npath would discard the RPC-layer reference that already exists\non entry.\n\nThree invariants follow:\n\n  - Any rpcrdma_req held by an rpc_rqst has rl_kref >= 1.\n    xprt_rdma_alloc_slot(), rpcrdma_bc_rqst_get(), and the\n    backlog-wake branch in xprt_rdma_alloc_slot() each kref_init\n    rl_kref before publishing the req. Without this invariant,\n    an RPC task that aborts between slot allocation and marshal\n    (gss_refresh failure or signal during call_connect, for\n    example) would drive xprt_release() ->\n    xprt_rdma_free_slot() -> kref_put against a refcount of\n    zero, saturating refcount_t and stranding the slot.\n\n  - The Send-side reference is taken only after WR prep\n    succeeds. A mapping failure in rpcrdma_prepare_send_sges()\n    runs rpcrdma_sendctx_cancel(), which DMA-unmaps the sendctx\n    and clears sc_req without touching rl_kref. The sendctx\n    ring walks in rpcrdma_sendctx_put_locked() and\n    rpcrdma_sendctxs_destroy() skip entries with sc_req == NULL,\n    so a burst of -EIO marshal failures cannot hold reqs off\n    rb_send_bufs.\n\n  - The release callback re-arms rl_kref so the next consumer\n    enters with the invariant satisfied.\n\nReplies now complete the RPC directly. rpcrdma_reply_handler()\ncalls rpcrdma_complete_rqst() in place of kref_put on the\nnon-LocalInv branch. The LocalInv branch already completes the\nRPC from frwr_unmap_async() and is unaffected.\n\nBecause Send-side references can now outlive RPC completion,\nconnection teardown drains sendctx entries whose unsignaled\nSends never had a later signaled completion to walk the ring.\nrpcrdma_sendctxs_destroy() walks the active range and runs\nrpcrdma_sendctx_unmap() on each entry with a non-NULL sc_req\nbefore the request buffers are reset, and is moved ahead of\nrpcrdma_reqs_reset() in rpcrdma_xprt_disconnect() so the reqs\nare still in their pre-reset state when the Send-side refs are\nreleased.\n\nThe drain creates a teardown-ordering hazard on the backchannel\npath. With the new lifetime, releasing a bc_prealloc req from\nrpcrdma_req_release() re-adds it to bc_pa_list. The disconnect\nin xprt_rdma_destroy() runs after xprt_destroy_backchannel() has\nalready emptied bc_pa_list, so the drained reqs would otherwise\nleak. xprt_rdma_destroy() now runs xprt_rdma_bc_destroy(xprt, 0)\na second time after the disconnect to reclaim them.","Type":"Description","Title":"xprtrdma: Decouple req recycling from RPC completion"}]}}}