{"id":21678,"date":"2026-09-12T04:31:30","date_gmt":"2026-09-11T20:31:30","guid":{"rendered":"https:\/\/www.54iter.cn\/?vuln=cve-2026-89756"},"modified":"2026-09-12T04:31:30","modified_gmt":"2026-09-11T20:31:30","slug":"cve-2026-89756","status":"publish","type":"vuln","link":"https:\/\/www.54iter.cn\/?vuln=cve-2026-89756","title":{"rendered":"CVE-2026-89756 &#8211; In the Linux kernel, the following vulnerability has been resolved:\n\nmm\/migrate: report RC"},"content":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>mm\/migrate: report RCU-tasks quiescent states in migrate_pages_batch()<\/p>\n<p>migrate_pages_batch() unmaps each folio before moving it, and every<br \/>\nunmap runs the mmu_notifier invalidate callbacks.  On KVM hosts<br \/>\ntry_to_migrate() ends up in kvm_mmu_notifier_invalidate_range_start() -&gt;<br \/>\ntdp_mmu_zap_leafs(), which is expensive, so unmapping a large batch keeps<br \/>\nthe CPU busy for a long time.<\/p>\n<p>The loop already calls cond_resched(), but on PREEMPTION kernels that is<br \/>\na no-op, and involuntary preemption is not a Tasks-RCU quiescent state.<\/p>\n<p>A long batch therefore never reports a quiescent state, and the<br \/>\nmigrating task (e.g. kcompactd) becomes a Tasks-RCU holdout, stalling the<br \/>\nTasks-RCU grace period for minutes, which is common at Meta fleet:<\/p>\n<p>  INFO: rcu_tasks detected stalls on tasks:<br \/>\n  0000000055349ecc: .. nvcsw: 1157401\/1157401 holdout: 1 idle_cpu: -1\/56 task:kcompactd0      state:R  running task<br \/>\n  Call Trace:<br \/>\n   tdp_mmu_zap_leafs<br \/>\n   tdp_mmu_next_root<br \/>\n   gfn_to_pfn_cache_invalidate_start<br \/>\n   kvm_mmu_notifier_invalidate_range_start<br \/>\n   __mmu_notifier_invalidate_range_start<br \/>\n   try_to_migrate_one<br \/>\n   try_to_migrate<br \/>\n   migrate_pages_batch<br \/>\n   migrate_pages<br \/>\n   compact_zone<br \/>\n   compact_node<br \/>\n   kcompactd<br \/>\n   kthread<\/p>\n<p>Use cond_resched_tasks_rcu_qs() so a quiescent state is reported even<br \/>\nwhen cond_resched() does nothing.<\/p>\n<p>This has also been discussed at [1]<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>mm\/migrate: report RCU-tasks quiescent states in migrate_pages_batch()<\/p>\n<p>migrate_pages_batch() unmaps each folio before moving it, a<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-21678","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/21678","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=21678"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=21678"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=21678"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}