CVE-2024-50263

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> fork: only invoke khugepaged, ksm hooks if no error<br /> <br /> There is no reason to invoke these hooks early against an mm that is in an<br /> incomplete state.<br /> <br /> The change in commit d24062914837 ("fork: use __mt_dup() to duplicate<br /> maple tree in dup_mmap()") makes this more pertinent as we may be in a<br /> state where entries in the maple tree are not yet consistent.<br /> <br /> Their placement early in dup_mmap() only appears to have been meaningful<br /> for early error checking, and since functionally it&amp;#39;d require a very small<br /> allocation to fail (in practice &amp;#39;too small to fail&amp;#39;) that&amp;#39;d only occur in<br /> the most dire circumstances, meaning the fork would fail or be OOM&amp;#39;d in<br /> any case.<br /> <br /> Since both khugepaged and KSM tracking are there to provide optimisations<br /> to memory performance rather than critical functionality, it doesn&amp;#39;t<br /> really matter all that much if, under such dire memory pressure, we fail<br /> to register an mm with these.<br /> <br /> As a result, we follow the example of commit d2081b2bf819 ("mm:<br /> khugepaged: make khugepaged_enter() void function") and make ksm_fork() a<br /> void function also.<br /> <br /> We only expose the mm to these functions once we are done with them and<br /> only if no error occurred in the fork operation.