[SRU][J:linux][PATCH 0/1] CVE-2024-46787

[Philip Cox]

CVE-2024-46787

SRU Justification:

[Impact]
In the Linux kernel, the following vulnerability has been resolved:
userfaultfd: fix checks for huge PMDs Patch series 
“userfaultfd: fix races around pmd_trans_huge() check”.

The pmd_trans_huge() check is racy and can lead to a
BUG_ON() (if you hit the right two race windows), or on older
kernels (before 6.5), you’d just have to win a single fairly wide
race to hit this. 


[Backport]
Fixed in upstream commit 71c186efc1b2cf1aeabfeff3b9bd5ac4c5ac14d8

The patch does not apply clean due to the fact pmd_read_atomic()
was renamed to pmdp_get_lockless() in upstream commit
dab6e717429e5ec795d558a0e9a5337a1ed33a3d.

I did not pick dab6e717429e as a prerequesit for this change
because it does not apply cleanly either, and would require
more changes to be picked, further increasing the regression risk.

I strongly feel that is much safer resolve the cherry-pick merge
conflict by renaming pmdp_get_lockless() to pmd_read_atomic()
especially in the ESM kernels.


[Test]
compile and boot tested.

[What could go wrong]
The main risk would be that further cherry-picks may have merge
conflicts due to the new changes, and may have incorrectly resolved
merge conflicts.  This shold be minimal, and caught by code review,
but the risk does exist due to the fact that this change has not
been merged into the stable branches that these backports are
targetting. 

--