Messages in this thread | | | Date | Fri, 27 Mar 2015 13:31:05 -0700 | Subject | Re: ia32_sysenter_target does not preserve EFLAGS | From | Linus Torvalds <> |
| |
On Fri, Mar 27, 2015 at 1:16 PM, Andy Lutomirski <luto@amacapital.net> wrote: > > Does it matter on 32-bit kernels? There's no swapgs, so IRQs should > still be safe, and we have a real stack pointer before sysexit.
Fair enough. On 32-bit, the only worry is the race between "return to user space" and "something set a thread flag", resulting in delayed signals and/or higher scheduling latency etc. So on 32-bit, the bug is much less of an issue, I agree.
So yeah, using sysretl instead of sti+sysexit on 64-bit sounds more reasonable given the potential worry about sti+sysexit atomicity in the presense of nmi's.
Linus
| |