[9fans] SATA boot failure

[9fans] SATA boot failure

[Peter A. Cejchan]

hi,

using 9atom.iso from Apr 19, installation completes but boot freezes.
here is the transcript:
cpu0: ...
pat: ...
ELCR: CC28
pcirouting: ignoring south bridge PCI 0.31.0 8086/2810
oni 0x732 phyno 1, mccv=0x38000000, phyv=0x002
#l0: rtl8169: 100Mb parport 0x6800 irq 10 tu 1514:001a9272efe7
#S/sdE: ahci ahci port 0xe000080: sss 1 ncs 31 coal 1 mport 3 led 1 clo 1 ems 1
#S/sdE ahci: Sata II with 4 ports
2047M memory, 256M kernel data, 1791m user, 2416 swap

<--- here it freezes

any hint?
thanks, regards,
++pac

Re: [9fans] SATA boot failure

[erik quanstrom]

> using 9atom.iso from Apr 19, installation completes but boot freezes.
> here is the transcript:
> cpu0: ...
> pat: ...
> ELCR: CC28
> pcirouting: ignoring south bridge PCI 0.31.0 8086/2810
> oni 0x732 phyno 1, mccv=0x38000000, phyv=0x002
> #l0: rtl8169: 100Mb parport 0x6800 irq 10 tu 1514:001a9272efe7
> #S/sdE: ahci ahci port 0xe000080: sss 1 ncs 31 coal 1 mport 3 led 1 clo 1 ems 1
> #S/sdE ahci: Sata II with 4 ports
> 2047M memory, 256M kernel data, 1791m user, 2416 swap
>
> <--- here it freezes
>
> any hint?

thanks for the report.  i don't have anything at the moment.
i see this sometimes on a reboot through /dev/reboot, but
never on a cold boot.

you don't have pata drives mapped to sata, do you?

- erik

Re: [9fans] SATA boot failure

[Peter A. Cejchan]

no PATA drives, just SATA-II, it worked before, but stopped to, I
thought it was damaged but surface scan was OK, I reinstalled, but the
problem persists..... it well may be also caused by the fact that I,
in the meantime, used that box with linux IDE discs, so I fiddled with
BIOS, sorrrrry...

peter.

Re: [9fans] SATA boot failure

[erik quanstrom]

On Tue Apr 20 09:23:42 EDT 2010, tyapca7@gmail.com wrote:
> no PATA drives, just SATA-II, it worked before, but stopped to, I
> thought it was damaged but surface scan was OK, I reinstalled, but the
> problem persists..... it well may be also caused by the fact that I,
> in the meantime, used that box with linux IDE discs, so I fiddled with
> BIOS, sorrrrry...

does resetting bios help?

- erik

Re: [9fans] SATA boot failure

[EBo]

> > no PATA drives, just SATA-II, it worked before, but stopped to, I
> > thought it was damaged but surface scan was OK, I reinstalled, but the
> > problem persists..... it well may be also caused by the fact that I,
> > in the meantime, used that box with linux IDE discs, so I fiddled with
> > BIOS, sorrrrry...

This sounds exactly like the trouble I initially had on my AMD machine.  It
turned out there is a bios bug and it is highly sensitive to how the SATA is
configured, and whether in ACHI or native IDE mode.  I also found it necessary
to remove power to clear nvram or memory or something which seemed to maintain
state unless power was removed.

Re: [9fans] SATA boot failure

[Peter A. Cejchan]

>
> does resetting bios help?

no. i tried (almost) all combinations that came on my mind.....there
is also some JMicron SATA option that I recall it caused me problems
with this same disk installation some half-a-year ago...Aug 11 2009
actually, the disk stopped to boot with the original native 9
installation, i first thought that the surface was damaged, as i
removed it from the box, but the surface check was ok (under windoze,
in a supplier company), than i thought that the bootloader in MBR got
damaged, so i completely reinstalled from 9atom.iso, as of Apr 19.
now i am almost convinced that i must fiddle more with bios, but i
have no idea what more to change.
regards, peter,
aka ++pac

Re: [9fans] SATA boot failure

[erik quanstrom]

On Tue Apr 20 05:21:28 EDT 2010, tyapca7@gmail.com wrote:
> hi,
>
> using 9atom.iso from Apr 19, installation completes but boot freezes.
> here is the transcript:
> cpu0: ...
> pat: ...
> ELCR: CC28
> pcirouting: ignoring south bridge PCI 0.31.0 8086/2810
> oni 0x732 phyno 1, mccv=0x38000000, phyv=0x002
> #l0: rtl8169: 100Mb parport 0x6800 irq 10 tu 1514:001a9272efe7
> #S/sdE: ahci ahci port 0xe000080: sss 1 ncs 31 coal 1 mport 3 led 1 clo 1 ems 1
> #S/sdE ahci: Sata II with 4 ports
> 2047M memory, 256M kernel data, 1791m user, 2416 swap

if you have an ide cdrom, could you try booting without?

i spend a few hrs the other day poking around at a similar
problem.  my atom board will not reboot via
	echo reboot /386/9pccpu>/dev/reboot
if i have an ide cdrom attached.  it hangs in approximately
the same place.  (though it's hard telling, since there are many
things that happen between the memory line and version ... time.)

here's what i know so far
1.  the typical warning about a spurious ipi (inter-processor
interrupt) on reboot is missing.
	cpu0: lapicerror: 0x00000000
i don't know what that means yet.
2.  compiling sdata out of the kernel results in proper booting.
rebooting after the ide interface has been touched  is fatal, even
if ide is compiled out of the reboot kernel.
3.  the boot process is started but the machine will lock solid
on the first sleep.  (typically the wait(2) for ip/ipconfig to run.)

- erik

Re: [9fans] SATA boot failure

[Peter A. Cejchan]

On Sun, Apr 25, 2010 at 9:01 PM, erik quanstrom <quanstro@quanstro.net> wrote:
> On Tue Apr 20 05:21:28 EDT 2010, tyapca7@gmail.com wrote:
>> hi,
>>
>> using 9atom.iso from Apr 19, installation completes but boot freezes.
>> here is the transcript:
>> cpu0: ...
>> pat: ...
>> ELCR: CC28
>> pcirouting: ignoring south bridge PCI 0.31.0 8086/2810
>> oni 0x732 phyno 1, mccv=0x38000000, phyv=0x002
>> #l0: rtl8169: 100Mb parport 0x6800 irq 10 tu 1514:001a9272efe7
>> #S/sdE: ahci ahci port 0xe000080: sss 1 ncs 31 coal 1 mport 3 led 1 clo 1 ems 1
>> #S/sdE ahci: Sata II with 4 ports
>> 2047M memory, 256M kernel data, 1791m user, 2416 swap
>
> if you have an ide cdrom, could you try booting without?
>
> i spend a few hrs the other day poking around at a similar
> problem.  my atom board will not reboot via
>        echo reboot /386/9pccpu>/dev/reboot
> if i have an ide cdrom attached.  it hangs in approximately
> the same place.  (though it's hard telling, since there are many
> things that happen between the memory line and version ... time.)
>
> here's what i know so far
> 1.  the typical warning about a spurious ipi (inter-processor
> interrupt) on reboot is missing.
>        cpu0: lapicerror: 0x00000000
> i don't know what that means yet.
> 2.  compiling sdata out of the kernel results in proper booting.
> rebooting after the ide interface has been touched  is fatal, even
> if ide is compiled out of the reboot kernel.
> 3.  the boot process is started but the machine will lock solid
> on the first sleep.  (typically the wait(2) for ip/ipconfig to run.)
>
> - erik
>
>

Blanking of HD's 9fat, as you recommended before, helped. Everything
went smooth then. And yes, I physically removed all IDE devices. Even
now, although I can boot with IDE HD connected,  I can see no /dev/hd*
devices. It may be a matter of BIOS, which behaves very strange, e.g.,
I must do 'Load Setup Defaults', after disconnecting the SATA  disks,
otherwise it is not able to find IDE disks at all. It is some version
(not up to date) of AMI BIOS.

Erik thank you for all you valuable work and help, wishing you a great day!

Peter.

Re: [9fans] SATA boot failure

[erik quanstrom]

> Blanking of HD's 9fat, as you recommended before, helped. Everything
> went smooth then. And yes, I physically removed all IDE devices. Even
> now, although I can boot with IDE HD connected,  I can see no /dev/hd*
> devices. It may be a matter of BIOS, which behaves very strange, e.g.,

do you see /dev/sd* ?

- erik

Re: [9fans] SATA boot failure

[Peter A. Cejchan]

> do you see /dev/sd* ?

yes!

++pac