CVE-2026-43417

*** Pendiente de traducción *** In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> sched/mmcid: Handle vfork()/CLONE_VM correctly<br /> <br /> Matthieu and Jiri reported stalls where a task endlessly loops in<br /> mm_get_cid() when scheduling in.<br /> <br /> It turned out that the logic which handles vfork()&amp;#39;ed tasks is broken. It<br /> is invoked when the number of tasks associated to a process is smaller than<br /> the number of MMCID users. It then walks the task list to find the<br /> vfork()&amp;#39;ed task, but accounts all the already processed tasks as well.<br /> <br /> If that double processing brings the number of to be handled tasks to 0,<br /> the walk stops and the vfork()&amp;#39;ed task&amp;#39;s CID is not fixed up. As a<br /> consequence a subsequent schedule in fails to acquire a (transitional) CID<br /> and the machine stalls.<br /> <br /> Cure this by removing the accounting condition and make the fixup always<br /> walk the full task list if it could not find the exact number of users in<br /> the process&amp;#39; thread list.