On Tuesday, March 18, 2008 at 22:41:24 +0100, Tobias Frost wrote:
> Unfortuantly strace is not available on lenny. I Will install ist from
> sid, and then send the strace log too.
Much thanks, it helps:
>| open("/dev/port", O_RDWR) = 3
>| lseek(3, 112, SEEK_SET) = 112
>| write(3, "\212", 1 <unfinished ...>
>| +++ killed by SIGSEGV +++
Everything is normal, up to and including this first /dev/port write().
But writing to I/O port 0x70 segfaults. This seems to indicate that
there is no RTC there. Or rather that if there is an RTC, it is not
interfaced in the usual way on ports 0x70 and 0x71.
| $ grep rtc /proc/ioports
| 0070-007f : rtc
I wonder how hwclock behaves in such conditions. Could you please try:
| # hwclock --directisa
| # hwclock
I know nothing about the armel arch, but Google seems to confirm unusual
clock chips. It seems mandatory to use one of the new RTC device drivers
accessible thru /dev/rtc0. Could you please try to symlink /dev/rtc to
rtc0, and then retry:
| # adjtimex --compare=1 --verbose
Alain.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
