{"id":"CVE-2026-93229","published":"2026-09-24T16:17:18.387","lastModified":"2026-09-25T05:17:00.707","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nnfsd: add missing read barrier to rpc_status_get dumpit seqcount retry\n\nThe hand-rolled seqcount-like protocol in nfsd_nl_rpc_status_get_dumpit()\nis missing a read memory barrier (smp_rmb) before its second counter\ncheck.  The standard kernel read_seqcount_retry() includes smp_rmb()\nto ensure that all data reads complete before the counter is re-checked.\n\nWithout this barrier, on weakly-ordered architectures (ARM, POWER),\nthe CPU may reorder field reads past the second counter check, making\nthe retry logic ineffective: it could observe a consistent counter pair\nwhile reading fields that have been concurrently modified by the writer.\n\nAdd smp_rmb() before the second counter check to order the field reads\nahead of it, matching the barrier semantics of the standard seqcount\nread-side.  The begin-side smp_load_acquire() already pairs with the\nsmp_store_release() in nfsd_dispatch(); with the smp_rmb() now ordering\nthe field reads, the retry check no longer needs acquire semantics and\nreads the counter with a plain READ_ONCE(), as read_seqcount_retry()\ndoes.\n\n[ cel: Use READ_ONCE instead of smp_load_acquire() ]","cvssScore":7.1,"cvssSeverity":"HIGH","cvssVector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:H","cwes":[],"vendors":[],"products":[],"references":[{"url":"https://git.kernel.org/stable/c/1aea0482b98ecd7d0249204665f2ad4ad517f66b","tags":[]},{"url":"https://git.kernel.org/stable/c/9b5f6475006cd8e3b5b99b8eb3cd74dbb1ce9df8","tags":[]},{"url":"https://git.kernel.org/stable/c/a71f161a857117e8e0264deb7d14fff5c98adcf5","tags":[]},{"url":"https://git.kernel.org/stable/c/f501f2f4ec1d2dfe39e21c98630314074a9b30b0","tags":[]}],"exploitRefs":[],"hasPoc":false,"ai":null}