Re: [patch V4 10/15] x86/entry: Use generic syscall entry function

[Thomas Gleixner]

Kees,

Kees Cook <keescook@xxxxxxxxxxxx> writes:
> On Tue, Jul 21, 2020 at 12:57:16PM +0200, Thomas Gleixner wrote:
>
> This doesn't look very expensive, and they certain indicate really bad
> conditions. Does this need to be behind a CONFIG? (Whatever the answer,
> we can probably make those changes in a later series -- some of these
> also look not arch-specific...)

The most expensive part is native_save_flags(). The extra branches could
be visible in benchmarks, but its a good question whether this should be
just enabled.

I just keep it as is for now as I would need to make the flags and the
on_thread_stack() part x86 specific anyway, but yes we can revisit that
once we agreed on the general direction of this.

Thanks,

        tglx