On Wed, Nov 16, 2022 at 3:31 PM Dave Hansen dave.hansen@intel.com wrote:
On 11/15/22 15:09, Kyle Huey wrote:
Following last week's discussion I've reorganized this patch. The goal remains to restore the pre-5.14 behavior of ptrace(PTRACE_SET_REGSET, NT_X86_XSTATE) for the PKRU register (which was equivalent to a hardware XRSTOR instruction).
The new version looks great. I've applied it.
I did remove the stable@ tags for now. There were a couple reasons for that. First, most of the x86 stuff marked for stable@ goes via our tip/urgent branch and this doesn't seem super urgent. It also touches code that's exposed in at least three separate UABIs, so I want a bit more soak time than x86/urgent normally provides.
I have zero objections if anyone wants to submit it to stable@ after it hits Linus's tree.
Works for me, thanks.
- Kyle