{"api_version":"1","generated_at":"2026-09-12T02:47:34+00:00","cve":"CVE-2026-89532","urls":{"html":"https://cve.report/CVE-2026-89532","api":"https://cve.report/api/cve/CVE-2026-89532.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-89532","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-89532"},"summary":{"title":"svcrdma: Fix pcl_for_each_segment for empty chunks","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nsvcrdma: Fix pcl_for_each_segment for empty chunks\n\nWhen a parsed chunk list contains a chunk whose ch_segcount is zero,\npcl_for_each_segment computes its inclusive upper bound as\n&chunk->ch_segments[ch_segcount - 1]. ch_segcount is u32, so the\nsubtraction wraps to 0xFFFFFFFF and the bound lands far past the\nch_segments flex array. The loop body then walks unrelated memory at\nsizeof(struct svc_rdma_segment) stride until it faults.\n\nA zero-segcount chunk is reachable from the wire:\nxdr_check_write_chunk() only rejects segcount values greater than\nrc_maxpages, and pcl_alloc_write() links a freshly allocated chunk\nonto rc_write_pcl/rc_reply_pcl before its segment-fill loop runs,\nso a Write or Reply chunk advertising zero segments leaves\nch_segcount == 0 on the list. When the transport has negotiated\nSend-With-Invalidate, svc_rdma_get_inv_rkey() iterates all four\nPCLs with pcl_for_each_segment and dereferences segment->rs_handle\non each iteration, turning the underflow into an out-of-bounds read\nand a general protection fault.\n\n    xdr_check_write_list / xdr_check_reply_chunk\n      pcl_alloc_write()\n        chunk = pcl_alloc_chunk(...)  /* ch_segcount = 0 */\n        list_add_tail(&chunk->ch_list, &pcl->cl_chunks)\n        /* fill loop iterates zero times for wire segcount 0 */\n\n    svc_rdma_get_inv_rkey()\n      pcl_for_each_chunk(rc_write_pcl)\n        pcl_for_each_segment(segment, chunk)\n          pos <= &ch_segments[0u - 1u]  /* 0xFFFFFFFF */\n          segment->rs_handle            /* OOB read -> GPF */\n\nFix by switching the macro to a half-open upper bound that uses\nch_segcount directly. For ch_segcount == 0 the loop start equals the\nloop end and the body is skipped; for ch_segcount > 0 the iteration\nrange is unchanged. All six existing call sites in\nnet/sunrpc/xprtrdma/svc_rdma_recvfrom.c and\nnet/sunrpc/xprtrdma/svc_rdma_rw.c remain correct under the new bound,\nso no caller changes are needed.","state":"PUBLISHED","assigner":"Linux","published_at":"2026-09-11 20:19:36","updated_at":"2026-09-11 20:19:36"},"problem_types":[],"metrics":[],"references":[{"url":"https://git.kernel.org/stable/c/9c5a03c3dc505c0295339b0a1b9e4fe36447e482","name":"https://git.kernel.org/stable/c/9c5a03c3dc505c0295339b0a1b9e4fe36447e482","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/b7713a784c59515d0aba558c8f5df6a0164dd3a9","name":"https://git.kernel.org/stable/c/b7713a784c59515d0aba558c8f5df6a0164dd3a9","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/a1c954ca4977a4e6ec73ef92fe48073ec54f9fc8","name":"https://git.kernel.org/stable/c/a1c954ca4977a4e6ec73ef92fe48073ec54f9fc8","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/6d33a7e6bf6c6b293a266a617201285a1ad32c56","name":"https://git.kernel.org/stable/c/6d33a7e6bf6c6b293a266a617201285a1ad32c56","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-89532","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-89532","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 78147ca8b4a9b6cf0e597ddd6bf17959e08376c2 9c5a03c3dc505c0295339b0a1b9e4fe36447e482 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 78147ca8b4a9b6cf0e597ddd6bf17959e08376c2 a1c954ca4977a4e6ec73ef92fe48073ec54f9fc8 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 78147ca8b4a9b6cf0e597ddd6bf17959e08376c2 6d33a7e6bf6c6b293a266a617201285a1ad32c56 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 78147ca8b4a9b6cf0e597ddd6bf17959e08376c2 b7713a784c59515d0aba558c8f5df6a0164dd3a9 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 5.11","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 5.11 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.12.109 6.12.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.18.50 6.18.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2.4 7.2.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.3-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":["include/linux/sunrpc/svc_rdma_pcl.h"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"9c5a03c3dc505c0295339b0a1b9e4fe36447e482","status":"affected","version":"78147ca8b4a9b6cf0e597ddd6bf17959e08376c2","versionType":"git"},{"lessThan":"a1c954ca4977a4e6ec73ef92fe48073ec54f9fc8","status":"affected","version":"78147ca8b4a9b6cf0e597ddd6bf17959e08376c2","versionType":"git"},{"lessThan":"6d33a7e6bf6c6b293a266a617201285a1ad32c56","status":"affected","version":"78147ca8b4a9b6cf0e597ddd6bf17959e08376c2","versionType":"git"},{"lessThan":"b7713a784c59515d0aba558c8f5df6a0164dd3a9","status":"affected","version":"78147ca8b4a9b6cf0e597ddd6bf17959e08376c2","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["include/linux/sunrpc/svc_rdma_pcl.h"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"5.11"},{"lessThan":"5.11","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"6.12.*","status":"unaffected","version":"6.12.109","versionType":"semver"},{"lessThanOrEqual":"6.18.*","status":"unaffected","version":"6.18.50","versionType":"semver"},{"lessThanOrEqual":"7.2.*","status":"unaffected","version":"7.2.4","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.3-rc1","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.12.109","versionStartIncluding":"5.11","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.18.50","versionStartIncluding":"5.11","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.2.4","versionStartIncluding":"5.11","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.3-rc1","versionStartIncluding":"5.11","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nsvcrdma: Fix pcl_for_each_segment for empty chunks\n\nWhen a parsed chunk list contains a chunk whose ch_segcount is zero,\npcl_for_each_segment computes its inclusive upper bound as\n&chunk->ch_segments[ch_segcount - 1]. ch_segcount is u32, so the\nsubtraction wraps to 0xFFFFFFFF and the bound lands far past the\nch_segments flex array. The loop body then walks unrelated memory at\nsizeof(struct svc_rdma_segment) stride until it faults.\n\nA zero-segcount chunk is reachable from the wire:\nxdr_check_write_chunk() only rejects segcount values greater than\nrc_maxpages, and pcl_alloc_write() links a freshly allocated chunk\nonto rc_write_pcl/rc_reply_pcl before its segment-fill loop runs,\nso a Write or Reply chunk advertising zero segments leaves\nch_segcount == 0 on the list. When the transport has negotiated\nSend-With-Invalidate, svc_rdma_get_inv_rkey() iterates all four\nPCLs with pcl_for_each_segment and dereferences segment->rs_handle\non each iteration, turning the underflow into an out-of-bounds read\nand a general protection fault.\n\n    xdr_check_write_list / xdr_check_reply_chunk\n      pcl_alloc_write()\n        chunk = pcl_alloc_chunk(...)  /* ch_segcount = 0 */\n        list_add_tail(&chunk->ch_list, &pcl->cl_chunks)\n        /* fill loop iterates zero times for wire segcount 0 */\n\n    svc_rdma_get_inv_rkey()\n      pcl_for_each_chunk(rc_write_pcl)\n        pcl_for_each_segment(segment, chunk)\n          pos <= &ch_segments[0u - 1u]  /* 0xFFFFFFFF */\n          segment->rs_handle            /* OOB read -> GPF */\n\nFix by switching the macro to a half-open upper bound that uses\nch_segcount directly. For ch_segcount == 0 the loop start equals the\nloop end and the body is skipped; for ch_segcount > 0 the iteration\nrange is unchanged. All six existing call sites in\nnet/sunrpc/xprtrdma/svc_rdma_recvfrom.c and\nnet/sunrpc/xprtrdma/svc_rdma_rw.c remain correct under the new bound,\nso no caller changes are needed."}],"providerMetadata":{"dateUpdated":"2026-09-11T19:44:11.495Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/9c5a03c3dc505c0295339b0a1b9e4fe36447e482"},{"url":"https://git.kernel.org/stable/c/a1c954ca4977a4e6ec73ef92fe48073ec54f9fc8"},{"url":"https://git.kernel.org/stable/c/6d33a7e6bf6c6b293a266a617201285a1ad32c56"},{"url":"https://git.kernel.org/stable/c/b7713a784c59515d0aba558c8f5df6a0164dd3a9"}],"title":"svcrdma: Fix pcl_for_each_segment for empty chunks","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-89532","datePublished":"2026-09-11T19:44:11.495Z","dateReserved":"2026-09-11T19:38:34.720Z","dateUpdated":"2026-09-11T19:44:11.495Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-09-11 20:19:36","lastModifiedDate":"2026-09-11 20:19:36","problem_types":[],"metrics":[],"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"89532","Ordinal":"1","Title":"svcrdma: Fix pcl_for_each_segment for empty chunks","CVE":"CVE-2026-89532","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"89532","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\nsvcrdma: Fix pcl_for_each_segment for empty chunks\n\nWhen a parsed chunk list contains a chunk whose ch_segcount is zero,\npcl_for_each_segment computes its inclusive upper bound as\n&chunk->ch_segments[ch_segcount - 1]. ch_segcount is u32, so the\nsubtraction wraps to 0xFFFFFFFF and the bound lands far past the\nch_segments flex array. The loop body then walks unrelated memory at\nsizeof(struct svc_rdma_segment) stride until it faults.\n\nA zero-segcount chunk is reachable from the wire:\nxdr_check_write_chunk() only rejects segcount values greater than\nrc_maxpages, and pcl_alloc_write() links a freshly allocated chunk\nonto rc_write_pcl/rc_reply_pcl before its segment-fill loop runs,\nso a Write or Reply chunk advertising zero segments leaves\nch_segcount == 0 on the list. When the transport has negotiated\nSend-With-Invalidate, svc_rdma_get_inv_rkey() iterates all four\nPCLs with pcl_for_each_segment and dereferences segment->rs_handle\non each iteration, turning the underflow into an out-of-bounds read\nand a general protection fault.\n\n    xdr_check_write_list / xdr_check_reply_chunk\n      pcl_alloc_write()\n        chunk = pcl_alloc_chunk(...)  /* ch_segcount = 0 */\n        list_add_tail(&chunk->ch_list, &pcl->cl_chunks)\n        /* fill loop iterates zero times for wire segcount 0 */\n\n    svc_rdma_get_inv_rkey()\n      pcl_for_each_chunk(rc_write_pcl)\n        pcl_for_each_segment(segment, chunk)\n          pos <= &ch_segments[0u - 1u]  /* 0xFFFFFFFF */\n          segment->rs_handle            /* OOB read -> GPF */\n\nFix by switching the macro to a half-open upper bound that uses\nch_segcount directly. For ch_segcount == 0 the loop start equals the\nloop end and the body is skipped; for ch_segcount > 0 the iteration\nrange is unchanged. All six existing call sites in\nnet/sunrpc/xprtrdma/svc_rdma_recvfrom.c and\nnet/sunrpc/xprtrdma/svc_rdma_rw.c remain correct under the new bound,\nso no caller changes are needed.","Type":"Description","Title":"svcrdma: Fix pcl_for_each_segment for empty chunks"}]}}}