On 8/5/22 10:24, Dave Hansen wrote:
On 8/3/22 20:16, Kyle Huey wrote:
When management of the PKRU register was moved away from XSTATE, emulation of PKRU's existence in XSTATE was added for APIs that read XSTATE, but not for APIs that write XSTATE. This can be seen by running gdb and executing `p $pkru`, `set $pkru = 42`, and `p $pkru`. On affected kernels (5.14+) the write to the PKRU register (which gdb performs through ptrace) is ignored.
Do you happen to have a reproducer for this sitting around? I'd love to get an addition to the pkeys selftest/ in place to make sure we don't break this again. PKRU is a very special snowflake.
Let me put this another way: I'm much more likely to quickly merge fixes that come with a selftest that demonstrates the breakage and the fix. An in-kernel test ensures:
1. There is a problem now 2. The patch fixes the problem 3. The problem does not recur