CVE-2024-58240

In the Linux kernel, the following vulnerability has been resolved:

tls: separate no-async decryption request handling from async

If we're not doing async, the handling is much simpler. There's no reference counting, we just need to wait for the completion to wake us up and return its result.

We should preferably also use a separate crypto_wait. I'm not seeing a UAF as I did in the past, I think aec7961916f3 ("tls: fix race between async notify and socket close") took care of it.

This will make the next fix easier.

This patch refactors TLS RX decryption to use a separate, stack-local crypto_wait for the non-async path and bypass the async reference-counting/notification logic. It doesn’t fix a user-triggerable bug by itself; it’s a correctness/simplification change that reduces coupling and prepares the ground for a follow-up fix (“tls: fix use-after-free on failed backlog decryption”). No externally exploitable behavior changes are introduced. The bug is actual for the older versions of Red Hat Enterprise Linux (before 9.3 and for all versions of the Red Hat Enterprise Linux 8) where patch aec7961916f3 "tls: fix race between async notify and socket close" not backported yet. The CVSS being calculated for worse case scenario where the previous patch aec7961916f3 not applied yet (that leads to the use after free possibility).

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

To mitigate this issue, prevent module tls from being loaded. Please see https://access.redhat.com/solutions/41278 for how to blacklist a kernel module to prevent it from loading automatically.