Hi,
I have an Asus Commando motherboard, p965 chipset with Marvell
88E8056 and 88E8001 gigabit lan onboard. skge and sky2 are compiled
in.
When booting linux-2.6.24-rc7-git6, everything seemed fine, but when
attempting to reboot, the machine would freeze on POST, right before
detecting AHCI drives.
In order to get it to POST, a reset or shutdown from the power button
is not enough, I actually have to cut power to the motherboard using
the PSU's power switch!
I first thought that the BIOS or CMOS may have gotten corrupted, but
eventually I fund that linux was the culprit. After a git bisect, I
found the bad commit below
root@ops-desktop:/home/ops/linux-2.6# git bisect bad
84cd2dfb04d23a961c5f537baa243fa54d0987ac is first bad commit
commit 84cd2dfb04d23a961c5f537baa243fa54d0987ac
Author: Stephen Hemminger <[email protected]>
Date: Thu Jan 10 16:14:13 2008 -0800
sky2: remove check for PCI wakeup setting from BIOS
The driver checks status of PCI power management to mark
default setting of Wake On Lan. On some systems this works, but often
it reports a that WOL is disabled when it isn't.
This patch gets rid of that check and just reports the wake on
lan status based on the hardware capablity.
Signed-off-by: Stephen Hemminger <[email protected]>
Signed-off-by: Jeff Garzik <[email protected]>
:040000 040000 291580ccfb2bae4a20f73198852b3824ba7ba9e0
5a0ae3b3c740899315347f2529b582c97ff7d488 M drivers
lspci and dmesg of bad kernel are attached
Please CC me when replying as I'm not signed up to the list
Regards
On Tue, 15 Jan 2008 13:01:47 +0200
"Ioan Ionita" <[email protected]> wrote:
> Hi,
>
> I have an Asus Commando motherboard, p965 chipset with Marvell
> 88E8056 and 88E8001 gigabit lan onboard. skge and sky2 are compiled
> in.
> When booting linux-2.6.24-rc7-git6, everything seemed fine, but when
> attempting to reboot, the machine would freeze on POST, right before
> detecting AHCI drives.
> In order to get it to POST, a reset or shutdown from the power button
> is not enough, I actually have to cut power to the motherboard using
> the PSU's power switch!
>
> I first thought that the BIOS or CMOS may have gotten corrupted, but
> eventually I fund that linux was the culprit. After a git bisect, I
> found the bad commit below
>
Is Wake On Lan enabled in the BIOS?
--
Stephen Hemminger <[email protected]>
On Jan 15, 2008 8:09 PM, Stephen Hemminger <[email protected]> wrote:
> On Tue, 15 Jan 2008 13:01:47 +0200
> "Ioan Ionita" <[email protected]> wrote:
>
>
> Is Wake On Lan enabled in the BIOS?
I don't have a Wake on Lan option in BIOS, only wake on pci, i assume
it's the same. It was disabled. I enabled it and the regression
behaved the same way. Wouldn't POST after shutdown or soft reset.
Regards
On Tue, 15 Jan 2008 21:00:13 +0200
"Ioan Ionita" <[email protected]> wrote:
> On Jan 15, 2008 8:09 PM, Stephen Hemminger <[email protected]> wrote:
> > On Tue, 15 Jan 2008 13:01:47 +0200
> > "Ioan Ionita" <[email protected]> wrote:
> >
>
> >
> > Is Wake On Lan enabled in the BIOS?
>
> I don't have a Wake on Lan option in BIOS, only wake on pci, i assume
> it's the same. It was disabled. I enabled it and the regression
> behaved the same way. Wouldn't POST after shutdown or soft reset.
>
I assume if you do:
ethtool -s eth0 wol d
it will POST okay. Okay, for now I'll have Jeff revert the patch.
Looks like the BIOS in your system is broken for WOL.
--
Stephen Hemminger <[email protected]>