{"id":41261,"date":"2026-10-19T00:32:53","date_gmt":"2026-10-18T16:32:53","guid":{"rendered":"https:\/\/www.54iter.cn\/?vuln=cve-2026-98789"},"modified":"2026-10-19T00:32:53","modified_gmt":"2026-10-18T16:32:53","slug":"cve-2026-98789","status":"publish","type":"vuln","link":"https:\/\/www.54iter.cn\/?vuln=cve-2026-98789","title":{"rendered":"CVE-2026-98789 &#8211; In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix objects stuc"},"content":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>bpf: Fix objects stuck in free_by_rcu_ttrace<\/p>\n<p>do_call_rcu_ttrace() returns early when call_rcu_ttrace_in_progress is set<br \/>\nand leaves the objects in free_by_rcu_ttrace. __free_rcu() frees<br \/>\nwaiting_for_gp_ttrace only and clears the flag. Hence the objects that<br \/>\nfree_bulk() or __free_by_rcu() added while RCU tasks trace GP was in flight<br \/>\nstay in free_by_rcu_ttrace until free_bulk() or alloc_bulk() is called for<br \/>\nthe same bpf_mem_cache again, which may never happen. The number of such<br \/>\nobjects is not bounded.<\/p>\n<p>Turn call_rcu_ttrace_in_progress into three states:<br \/>\n0 &#8211; idle<br \/>\n1 &#8211; __free_rcu() is queued<br \/>\n2 &#8211; __free_rcu() is queued and free_by_rcu_ttrace got more objects since<\/p>\n<p>do_call_rcu_ttrace() sets 2. __free_rcu() does cmpxchg(1 -&gt; 0) and starts<br \/>\nthe next GP when it fails. It cannot clear the flag first and check<br \/>\nfree_by_rcu_ttrace later, since bpf_mem_alloc_destroy() frees bpf_mem_cache<br \/>\nwithout waiting for RCU callbacks when the flag is zero.<\/p>\n<p>Now __free_rcu() queues itself, so the one that didn&#8217;t see &#8216;draining&#8217; may<br \/>\ndo call_rcu_tasks_trace() after rcu_barrier_tasks_trace() in<br \/>\nfree_mem_alloc(). Queue it under rcu_read_lock() and do synchronize_rcu()<br \/>\nbefore the barriers. Calling rcu_barrier_tasks_trace() twice works too, but<br \/>\ncreating and destroying hash maps in a loop on many cpus slows down to one<br \/>\nfree_mem_alloc() per GP and kworkers pile up.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>bpf: Fix objects stuck in free_by_rcu_ttrace<\/p>\n<p>do_call_rcu_ttrace() returns early when call_rcu_ttrace_in_progress is set<br \/>\nand leaves<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-41261","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/41261","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=41261"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=41261"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=41261"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}