[lkml]   [2005]   [Dec]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH 2.6] i8xx_tco - Add SuperMicro ICH[56] Pentium 4 watchdog support
Patch Devil <> wrote:
> Hello friends of LK:
> It seems that SuperMicro P4-based motherboards DO actually have working
> watchdog hardware (be sure jumpers *and* BIOS settings are configured
> appropriately), however the setup of their internal ports and registers
> deviate from canonical Intel specification.
> Because it seems impossible/difficult to do a proper auto-detection for these
> deviations, I've added a module_param to the TCO module to provide minor
> changes to the existing i8xx TCO code. I've given it some testing on i845
> (Northwood P4) and E7520 (Nacona/EM64T SMP-Xeon) and it does indeed activate
> watchdog (reset/reboot) functionality and provide WDT resets to keep the
> system up.
> SuperMicro had provided me with some documentation for their Pentium III
> motherboards, however I was not able to ever get it functioning on my SSE370+,
> however I've incorporated the logic as documented as a bit of a "bookmark" for
> future implementors who MAY be lucky to have P3 SuperMicros with working WDT
> functionality.
> My patch activates the watchdog even if the BIOS setting disables it, however
> it cannot override the jumper on the motherboard.
> Naturally, users should do two-way testing before deploying this on production
> hardware: 1) Verify that the watchdog IS actually able to reset the machine --
> so you'll need to deliberately lockup or kill the WDT refresh and make sure
> the machine reboots, and then 2) Verify that the watchdog timer is being
> properly reset.
> This patch was made against, and I provide some background detail in
> the patched source.
> I've had this working in production for a couple of months now without any
> issues. Of course, these very reliable systems never lockup (typical uptimes
> are over 300-400 days, and are rebooted for kernel upgrades only). But hey, I
> was a bit annoyed that an advertised feature wasn't available to me.
> Cheers,
> Robert

Thanks for doing this work.

Unfortunately we shouldn't be merging nontrivial patches from new
contributors with names like "Robert the Patch Devil" ;)

Please review Section 11 of Documentation/SubmittingPatches.txt and, if you
find the terms there to be acceptable, resend the patch with a
Signed-off-by: and your real name, thanks.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-12-17 21:41    [W:0.037 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site