The problems you are seeing are solved by this little patch:
--- arch/alpha/kernel/osf_sys.c.org Mon Mar 29 17:29:02 1999
+++ arch/alpha/kernel/osf_sys.c Thu Apr 1 13:41:36 1999
@@ -1408,8 +1408,7 @@
offsetof(struct timex32, time)))
return -EFAULT;
- if ((ret = do_adjtimex(&txc)))
- return ret;
+ ret = do_adjtimex(&txc);
/* copy back to timex32 */
if (copy_to_user(txc_p, &txc, offsetof(struct timex32, time)) ||
@@ -1418,5 +1417,5 @@
(put_tv32(&txc_p->time, &txc.time)))
return -EFAULT;
- return 0;
+ return ret;
}
What is does is that it removes the overoptimisation of the adjtimex
syscall someone put in somewhere in the 2.1.x series and wich breaks
xntpd. I've had lots of 'fun' tracking this little bug down.
This fix is in Alan's tree, so I hope it will end up in Linus' tree also.
~~ Wieger
> Tim.
> */
>
>
> -
> 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/
>
~~ Wieger
-
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/