{"id":21511,"date":"2026-09-12T04:31:17","date_gmt":"2026-09-11T20:31:17","guid":{"rendered":"https:\/\/www.54iter.cn\/?vuln=cve-2026-89589"},"modified":"2026-09-12T04:31:17","modified_gmt":"2026-09-11T20:31:17","slug":"cve-2026-89589","status":"publish","type":"vuln","link":"https:\/\/www.54iter.cn\/?vuln=cve-2026-89589","title":{"rendered":"CVE-2026-89589 &#8211; In the Linux kernel, the following vulnerability has been resolved:\n\nacpi\/apei\/ghes: Use r"},"content":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>acpi\/apei\/ghes: Use raw_spinlock_t for CXL CPER work locks<\/p>\n<p>The CXL CPER work registration and unregistration helpers acquire<br \/>\ncxl_cper_work_lock and cxl_cper_prot_err_work_lock with a spinlock<br \/>\nguard(), which leaves local interrupts enabled. The corresponding post<br \/>\npaths (cxl_cper_post_event(), cxl_cper_post_prot_err()) execute in hard<br \/>\nIRQ context (they are called from the GHES error notification path) and<br \/>\nacquire the same locks with an irqsave guard().<\/p>\n<p>If a CPU is holding one of these locks via a spinlock guard() when a GHES<br \/>\ninterrupt arrives on the same CPU, the IRQ handler spins on the held lock<br \/>\nwaiting for it to release, while the lock holder is preempted by the IRQ.<br \/>\nThe result is a deadlock.<\/p>\n<p>Convert both locks from spinlock_t to raw_spinlock_t and use guard() at<br \/>\nall call sites. On PREEMPT_RT kernels spinlock_t is backed by rt_mutex and<br \/>\nsleeping from hard IRQ context is not permitted; raw_spinlock_t is safe in<br \/>\nboth contexts.<\/p>\n<p>Add WARN_ONCE to both register functions to surface double-registration<br \/>\nbugs at runtime.<\/p>\n<p>Restructure both unregister functions to clear the global work pointer<br \/>\nunder the lock before calling cancel_work_sync(), closing the window<br \/>\nwhere a CPER interrupt could schedule work on a pointer about to be<br \/>\nfreed. Add kfifo_reset() after cancel_work_sync() so stale entries<br \/>\nare not replayed on next module load.<\/p>\n<p>Both kfifos are single-consumer: only one work_struct is registered at<br \/>\na time, enforced by the WARN_ONCE guard in the register functions.<br \/>\nkfifo_reset() is safe outside the lock because cancel_work_sync() has<br \/>\nalready quiesced the consumer, and no new consumer can register until<br \/>\nthe current module exit completes and a fresh module init runs.<\/p>\n<p>Remove the redundant cancel_work_sync() call from cxl_ras_exit() and<br \/>\ncxl_pci_driver_exit(). The CPER unregister functions now quiesce<br \/>\nthe work internally.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>acpi\/apei\/ghes: Use raw_spinlock_t for CXL CPER work locks<\/p>\n<p>The CXL CPER work registration and unregistration helpers acquire<br \/>\ncxl_c<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-21511","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/21511","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln"}],"about":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/types\/vuln"}],"wp:attachment":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=21511"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=21511"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=21511"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}