Re: Direct access to hardware

From: Khimenko Victor (khim@sch57.msk.ru)
Date: Tue Jul 25 2000 - 09:31:27 EST


In <200007251316.e6PDGRN06413@pincoya.inf.utfsm.cl> Horst von Brand (vonbrand@inf.utfsm.cl) wrote:
> James Sutherland <jas88@cam.ac.uk> said:
>> On Sun, 23 Jul 2000, Horst von Brand wrote:

> [...]

>> > Yep. It should also carefully check that those damn ELF thingies being
>> > thrown at the CPU better contain valid code!

>> It does. How do you think the kernel works around CPU bugs - replicate
>> nanites to rewire the CPU?

> Please show me where Linux checks every byte of an executable before
> running it to make sure it doesn't f00f or coma your CPU.

This F00F is GREAT example of how such problem should be resolved.

1. Linux MUST NOT try to protect hardware from itself by default (linux does
not scan executable to find "bad bytes").
2. When bug in hardware is found manufacturer ADMIT that it's indeed his bug
and offer workaround - perhaps developed with linux developers (F00F bugfix
was incorporated in Linux very fast when it was developed by Intel).

And no, you Intel not tried to sue anyone for executing program with this
problematic F0 0F XX XX sequence. Quite on contratary: Intel was happy to
avoid counter-claims for selling broken hardware.

-
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/



This archive was generated by hypermail 2b29 : Mon Jul 31 2000 - 21:00:19 EST