Re: [RFC] x86, ia32entry: Use sysretl to return from sysenter

[Ingo Molnar <mingo@kernel.org>]

* Andy Lutomirski <luto@kernel.org> wrote:

> Sysexit is scary on 64-bit kernels -- sysexit must be invoked with
> usergs and IRQs on.  That means that we rely on sti to correctly
> mask interrupts for one instruction.  This is okay by itself, but
> the semantics with respect to NMIs are unclear.

At least judging by profiling output I think NMIs observe the STI 
window of one instruction non-execution as well. (But I'm not 100% 
sure.)

> Avoid the whole issue by using sysretl instead.  For background,
> Intel CPUs don't allow syscall from compat mode, but they do allow
> sysret back to compat mode.  Go figure.
> 
> Oddly this seems to be 30 cycles or so faster.  Avoiding popfq and
> sti will account for under half of that, I think, so my best guess
> is that Intel just optimizes sysret much better than sysexit.
> 
> Cc: stable@vger.kernel.org

I like it, but no way is this automatic -stable material ... if proven 
upstream we can forward it as a fix for SYSEXIT fragility, but not 
automatically, IMHO.

Thanks,

	Ingo