T2 build hangs at stage 0, glibc

[bkauler-2]

Recent versions of Puppy Linux have all the requirements to do a T2 build
(note, although Puppy uses Busybox, all cases where T2 requires the full
utility, it has been substituted for the Busybox applet).

Recently I compiled the 2.6.24-rc4 kernel, and for the very first time I
turned on the 'libata' PATA support for IDE drives. So, /dev/hda on my
laptop is now /dev/sda. I also turned on SMP, but mostly left the config
settings as before.

Running a Puppy system identical to before, just the new kernel, a T2
build gets to the 'glibc' package in Stage 0, and that's it, stays there
forever.
When I ran the build, I waited for about 5 hours -- that's how long it was
stuck on 'glibc', and I noticed the 'glibc.out' had grown to 50MB, CPU
constantly very busy.

So, I booted the Puppy with the older Puppy (2.6.21.7 kernel) so now back
with /dev/hda, and the build is running right now, going full steam ahead.
'glibc.log' in stage 0 is only 206KB.

I was wondering, does this ring a bell with anyone? A problem with the
libata emulation layer on IDE drives, with respect to building T2?

I should have kept the 'glibc.out', will do it again and upload it, if
that is the recommended next step to solving this problem. I only looked
briefly at the file, and noticed lots of 'rm' commands near the end.

Regards,
Barry Kauler



-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2

[bkauler-2]

Problem solved, well, avoided anyway.

I rolled back to the 2.6.23.11 kernel, compiled with same config
settings, and I tested T2, it builds fine. So, it has nothing
to do with libata or smp, seems to be some quirk in the
2.6.24-rc4.

Regards,
Barry Kauler

>
> Recent versions of Puppy Linux have all the requirements to do a T2 build
> (note, although Puppy uses Busybox, all cases where T2 requires the full
> utility, it has been substituted for the Busybox applet).
>
> Recently I compiled the 2.6.24-rc4 kernel, and for the very first time I
> turned on the 'libata' PATA support for IDE drives. So, /dev/hda on my
> laptop is now /dev/sda. I also turned on SMP, but mostly left the config
> settings as before.
>
> Running a Puppy system identical to before, just the new kernel, a T2
> build gets to the 'glibc' package in Stage 0, and that's it, stays there
> forever.
> When I ran the build, I waited for about 5 hours -- that's how long it was
> stuck on 'glibc', and I noticed the 'glibc.out' had grown to 50MB, CPU
> constantly very busy.
>
> So, I booted the Puppy with the older Puppy (2.6.21.7 kernel) so now back
> with /dev/hda, and the build is running right now, going full steam ahead.
> 'glibc.log' in stage 0 is only 206KB.
>
> I was wondering, does this ring a bell with anyone? A problem with the
> libata emulation layer on IDE drives, with respect to building T2?
>
> I should have kept the 'glibc.out', will do it again and upload it, if
> that is the recommended next step to solving this problem. I only looked
> briefly at the file, and noticed lots of 'rm' commands near the end.
>
> Regards,
> Barry Kauler
>
>
>
> -----------------------------------------------------------
> If you wish to unsubscribe from this mailing, send mail to
> lists@... with a subject of: unsubscribe t2
>

-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@... with a subject of: unsubscribe t2