lkml.org 
[lkml]   [1998]   [Nov]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: kernel time *UNSAFE* during interrupts
Michael Krause wrote:
> The page says you printk()'ed the time from both get_fast_time() and
> do_gettimeofday() -- however, on a Pentium system with TSC enabled,
> get_fast_time() also ends up in do_gettimeofday() [do_get_fast_time
> gets overwritten at the bottom of arch/i386/kernel/time.c] so I don't
> quite see how you can get different results.

Ah, but I don't! do_gettimeofday and get_fast_time produce the same
results at interrupt time, both of them show the -10ms jump. It is only
from user space (or non-interrupt time?) that gettimeofday is safe.

> What a machine do you have, and is CONFIG_CPU_IS_SLOW enabled?

I have tried this on a PII 266MHz with 196MB of ram, and an old P75.
Both of these will have TSC.

I could not find the CONFIG_CPU_IS_SLOW option anywhere, where does it
get set or cleared?

I would very much like it if you could try the tcpdump experiment, so
that I can say that its not only me that sees this bug!

> Hmm. I was sure it's only a matter of time until someone finds a bug
> in the new time.c code :)

Ah well, this is the nature of things. :)

Stephen.
--
-----------------------------------------------------------------------
Stephen Donnelly (BCMS) email: sfd@cs.waikato.ac.nz
Waikato ATM Group Lab G1.31 phone +64 7 838 4466 x6728
Computer Science Department, University of Waikato, New Zealand
-----------------------------------------------------------------------

-
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/

\
 
 \ /
  Last update: 2005-03-22 13:45    [W:0.415 / U:0.016 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site