All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Paul Rolland" <rol@as2917.net>
To: <linux-kernel@vger.kernel.org>
Cc: "'Paul Rolland'" <rol@as2917.net>
Subject: i810-tco : odd behavior, odd driver ?
Date: Wed, 2 Apr 2003 20:30:50 +0200	[thread overview]
Message-ID: <028c01c2f945$fcc0d230$3f00a8c0@witbe> (raw)

Hello,

I'm using a Super P4SGE motherboard, that does include the i810 
watchdog.
I did compile a 2.4.20 Linux kernel with support for this watchdog,
but the results are rather strange, as the machine keeps rebooting
every 5-6 minutes :
last says :
....
reboot   system boot  2.4.20-watchdog  Wed Apr  2 17:16          (00:52)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 17:09          (00:59)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 17:03          (01:05)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 16:59          (01:09)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 16:54          (01:14)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 16:49          (01:19)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 16:44          (01:23)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 16:38          (01:30)

reboot   system boot  2.4.20-watchdog  Wed Apr  2 16:30          (01:38)

...

The code to stop the watchdog is :
int getWatchdogFd(void)
{
  int fd;

  fd = open("/dev/watchdog", O_RDWR);
  if (-1 == fd)
    printf("Error opening /dev/watchdog : %s\n", strerror(errno));

  return(fd);
}
...
        fd = getWatchdogFd();
        option = WDIOS_DISABLECARD;
        if (-1 == ioctl(fd, WDIOC_SETOPTIONS, &option)) {
          printf("Error disabling watchdog [%d]\n", errno);
        } else {
          printf("Watchdog disabled\n");
        } /* endif */
        write(fd,"V",1);
        close(fd);

Calling this code results in some OK, and at least, by writing 'V', I
don't have the "Unexpected close, not stopping watchdog!" message.

By changing the driver code, I now know that the hardware I have is 
identified as :
PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801E_0
in the PCI table list.

So, I went to Intel documentation, and I got two big points :
Doc 273599-001, Page 278 :
TCO1_TMR : The timer is clocked approximately 0.6 seconds, and this
allows
time-outs ranging from 2.4 s to 38 s.

-> How is it possible that my system waits from 5-6 minutes before
restarting,
when the driver initialized this time-out to 30 seconds, as it is the
default
value ???

Same doc, page 281 :
TCO1_CNT is a 16bits register.
While I was at school, I was told that accessing a 16bits reg must be
done using 16bit operands.
The code, however does some awful :
unsigned char val;
val = inb(TCO1_CNT + 1)
and does access the 8higher bit by reading an 8 bit value at the 
register address + 1....
Is this correct ?

What can I do / test / ??? to have this watchdog running ?

Regards,
Paul, rol@as2917.net


                 reply	other threads:[~2003-04-02 18:19 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='028c01c2f945$fcc0d230$3f00a8c0@witbe' \
    --to=rol@as2917.net \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.