CVE-2026-64069

*** Pendiente de traducción *** In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> netfs: Fix cancellation of a DIO and single read subrequests<br /> <br /> When the preparation of a new subrequest for a read fails, if the<br /> subrequest has already been added to the stream-&gt;subrequests list, it can&amp;#39;t<br /> simply be put and abandoned as the collector may see it. Also, if it<br /> hasn&amp;#39;t been queued yet, it has two outstanding refs that both need to be<br /> put. Both DIO read and single-read dispatch fail at this; further, both<br /> differ in the order they do things to the way buffered read works.<br /> <br /> Fix cancellation of both DIO-read and single-read subrequests that failed<br /> preparation by the following steps:<br /> <br /> (1) Harmonise all three reads (buffered, dio, single) to queue the subreq<br /> before prepping it.<br /> <br /> (2) Make all three call netfs_queue_read() to do the queuing.<br /> <br /> (3) Set NETFS_RREQ_ALL_QUEUED independently of the queuing as we don&amp;#39;t<br /> know the length of the subreq at this point.<br /> <br /> (4) In all cases, set the error and NETFS_SREQ_FAILED flag on the subreq<br /> and then call netfs_read_subreq_terminated() to deal with it. This<br /> will pass responsibility off to the collector for dealing with it.