syszbot reported a warnon for XRSTOR raising #GP:
https://lore.kernel.org/r/[email protected]
with a syzcaller reproducer and a conclusive bisect result. It took a
while to destill a simple C reproducer out of it which allowed to pin point
the root cause: The recent addition of supervisor XSTATEs broke the signal
restore path for the case where the signal handler wreckaged the XSTATE on
stack because it does not sanitize the XSTATE header which causes a
subsequent XRSTOR to fail and #GP.
The following series addresses the problem and fixes related issues
which were found while inspecting the related changes and testing
signal restore and PKRU.
V2 of this can be found here:
https://lore.kernel.org/r/[email protected]
Changes vs. V2:
- Drop the cleanup patches for now, so the bug fixes make progress
- Fix the PKRU context switch code
- Address review comments
Thanks,
tglx
The series as a whole looks good to me. So, especially for the PKRU parts:
Acked-by: Dave Hansen <[email protected]>
On Tue, 2021-06-08 at 16:36 +0200, Thomas Gleixner wrote:
> The following series addresses the problem and fixes related issues
> which were found while inspecting the related changes and testing
> signal restore and PKRU.
Woah, you sure found some subtle stuff there.
Acked-by: Rik van Riel <[email protected]>
--
All Rights Reversed.