Re: [patch] fastcall-2.3.32-B6, SYSENTER/SYSEXIT support

Linus Torvalds (torvalds@transmeta.com)
Tue, 14 Dec 1999 16:42:10 -0800 (PST)


On Wed, 15 Dec 1999, Chris Wedgwood wrote:
>
> Surely many such applications should be able to use the processor
> timestamp with suitable calibration?

Yes. But they should not KNOW. They should call "gettimeofday()" like a
well-written portable program shall, and that call should use the
timestamp IF IT IS AVAILBLE. And with the right scaling, so that there is
never any question of the program running fine on one CPU but breaking
when run on a faster/slower one. Talk about nasty debugging.

And this is exactly why the magical kernel mapping is such a win.

Linus

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/