Re: Weird behaviour with new config method

ron minnich  writes:
> interesting. I am loading linux direct from flash and it works fine with 
> the new config method. Now I'll have to try etherboot and see what's up.
>
> this is complicated as my tree is in an intermediate state, but I hope to 
> work it out monday.

Something very peculiar is going on at the moment as I can't even get
a configuration to build.

Eric

bash-2.05a$ ./buildtarget arima/hdama/
make: `config.py' is up to date.
Configuring TARGET hdama
Creating directory arima/hdama/hdama
Will place Makefile, crt0.S, etc. in arima/hdama/hdama
Configuring ROMIMAGE fallback
Configuring DIR /config/Config.lb
Configuring DIR /lib/Config.lb
Configuring DIR /console/Config.lb
Configuring DIR /stream/Config.lb
Configuring DIR /devices/Config.lb
Configuring DIR /pc80/Config.lb
Configuring DIR /boot/Config.lb
Configuring PART mainboard, path arima/hdama
Configuring PART arch, path i386
Adding init file: config/crt0.base
Configuring DIR lib/Config.lb
Configuring DIR boot/Config.lb
Configuring DIR smp/Config.lb
End PART arch
WARNING: Option CONFIG_SMP using default value 0
Configuring PART northbridge, path amd/amdk8
End PART northbridge
Configuring PART southbridge, path amd/amd8111
End PART southbridge
Configuring PART southbridge, path amd/amd8131
End PART southbridge
Configuring PART superio, path NSC/pc87360
End PART superio
Configuring DIR /pc80/Config.lb
Warning, object/driver mc146818rtc previously defined
Configuring PART cpu, path p5
End PART cpu
Configuring PART cpu, path p6
End PART cpu
Configuring PART cpu, path k7
End PART cpu
Configuring PART cpu, path k8
End PART cpu
End PART mainboard
WARNING: Option AMD8111_DEV using default value 0x3800
WARNING: Option CONFIG_IDE_STREAM using default value 0
Traceback (most recent call last):
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 2137, in ?
    if (not parse('board', fp.read())):
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1782, in parse
    return wrap_error_reporter(P, rule)
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/yappsrt.py", line 162, in wrap_error_reporter
    try: return getattr(parser, rule)()
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1716, in board
    romstmts = self.romstmts()
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1701, in romstmts
    romimage = self.romimage()
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1680, in romimage
    romstmt = self.romstmt(1)
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1671, in romstmt
    mainboard = self.mainboard(C)
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1650, in mainboard
    if (C): mainboard(PATH)
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1034, in mainboard
    partpop()
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 1087, in partpop
    print "WARNING: Option %s using default value %s" % (i, getformated(i, curpart.image))
  File "/home/eric/projects/linuxbios/checkin/freebios2/util/newconfig/config.py", line 884, in getformated
    return (f % v.contents())
TypeError: an integer is required

heads up with new config method.

[ron minnich]

I am going to make a very minor tweak in a few minutes that should not 
affect any of you.

But we'll see. This tweak is aimed at letting us shrink the top-level 
Config.lb files to almost nothing.

ron

heads up with new config method.

ron minnich  writes:
> I am going to make a very minor tweak in a few minutes that should not 
> affect any of you.
>
> But we'll see. This tweak is aimed at letting us shrink the top-level 
> Config.lb files to almost nothing.

I now have a mysterious requirement to have a chip.h in
the arima/hdama mainboard directory.

This looks like a new a requirement, and the file was not added.

Eric

heads up with new config method.

[ron minnich]

On 4 Aug 2003, Eric W. Biederman wrote:
> I now have a mysterious requirement to have a chip.h in
> the arima/hdama mainboard directory.

my fault! I forgot to cvs add.

Fixed and committed.

FWIW this build is working well for me as of today for both etherboot and 
kernel.

ron

heads up with new config method.

ron minnich  writes:
> On 4 Aug 2003, Eric W. Biederman wrote:
> > I now have a mysterious requirement to have a chip.h in
> > the arima/hdama mainboard directory.
>
> my fault! I forgot to cvs add.
>
> Fixed and committed.
>
> FWIW this build is working well for me as of today for both etherboot and 
> kernel.

If you could please verify that is the only one because I think
I ran into at least one more chips.h that was not committed.

If I can get the thing to build I can comment a lot more fairly.

Eric

heads up with new config method.

[ron minnich]

On 4 Aug 2003, Eric W. Biederman wrote:
> If you could please verify that is the only one because I think
> I ran into at least one more chips.h that was not committed.

I'll look again but ... the clean checkout yesterday built fine, which is 
odd.

ron

heads up with new config method.

[ron minnich]

On Tue, 5 Aug 2003, ron minnich wrote:
> I'll look again but ... the clean checkout yesterday built fine, which is 
> odd.

I mean odd as in the sense of 'works for me, but not for you'. I actually 
still can't do a real build on RH 7.3 as romcc still fails on that 
platform.

ron

heads up with new config method.

[ron minnich]

On Tue, 5 Aug 2003, ron minnich wrote:
> I mean odd as in the sense of 'works for me, but not for you'. I actually 
> still can't do a real build on RH 7.3 as romcc still fails on that 
> platform.

yep it was missing a chip.h.

Try it now.

ron

Weird behaviour with new config method

[ron minnich]

On 4 Aug 2003, Eric W. Biederman wrote:
> WARNING: Option AMD8111_DEV using default value 0x3800
> WARNING: Option CONFIG_IDE_STREAM using default value 0

you've got something different. We're building from cvs here with no 
trouble.

The order of some of your messages differs from ours.

Are you sure it's a 'stock hdama' or ...

ron

Weird behaviour with new config method

ron minnich  writes:
> On 4 Aug 2003, Eric W. Biederman wrote:
> > WARNING: Option AMD8111_DEV using default value 0x3800
> > WARNING: Option CONFIG_IDE_STREAM using default value 0
>
> you've got something different. We're building from cvs here with no 
> trouble.
>
> The order of some of your messages differs from ours.
>
> Are you sure it's a 'stock hdama' or ...

100%.  I just ran a cvs update -d -P in my freebios2 directory and I have run
a cvs diff and I while I have one or two modified files, none of them are the
new configuration files.

I did do a rebuild on config.py before I got started which may affect something.

Eric

Weird behaviour with new config method

[ron minnich]

fresh checkout:

bash-2.05b$ python /tmp//freebios2/util/newconfig/config.py  Config.lb 
/tmp///freebios2

Configuring TARGET hdama
Creating directory hdama
Will place Makefile, crt0.S, etc. in hdama
Configuring ROMIMAGE fallback
Configuring DIR /config/Config.lb
Configuring DIR /lib/Config.lb
Configuring DIR /console/Config.lb
Configuring DIR /stream/Config.lb
Configuring DIR /devices/Config.lb
Configuring DIR /pc80/Config.lb
Configuring DIR /boot/Config.lb
Configuring PART mainboard, path arima/hdama
Configuring PART arch, path i386
Adding init file: config/crt0.base
Configuring DIR lib/Config.lb
Configuring DIR boot/Config.lb
Configuring DIR smp/Config.lb
End PART arch
WARNING: Option CONFIG_SMP using default value 0
Configuring PART northbridge, path amd/amdk8
End PART northbridge
Configuring PART southbridge, path amd/amd8111
End PART southbridge
Configuring PART southbridge, path amd/amd8131
End PART southbridge
Configuring PART superio, path NSC/pc87360
End PART superio
Configuring DIR /pc80/Config.lb
Warning, object/driver mc146818rtc previously defined
Configuring PART cpu, path p5
End PART cpu
Configuring PART cpu, path p6
End PART cpu
Configuring PART cpu, path k7
End PART cpu
Configuring PART cpu, path k8
End PART cpu
End PART mainboard
WARNING: Option USE_NORMAL_IMAGE using default value 0
WARNING: Option CONFIG_IDE_STREAM using default value 0
WARNING: Option AMD8111_DEV using default value 0x3800
End ROMIMAGE
Build ROM size 1048576
Creating directory hdama/fallback
Creating hdama/fallback/static.c
Creating hdama/fallback/Makefile.settings
Creating hdama/fallback/crt0_includes.h
Creating hdama/fallback/Makefile
Creating hdama/fallback/ldoptions
Creating hdama/Makefile.settings
Creating hdama/Makefile

=========================
BUT: greg points out you are using buildtarget maybe.

but that works too. Dammit!
=========================

bash-2.05b$ sh buildtarget arima/hdama/
make: `config.py' is up to date.
Configuring TARGET hdama
Will place Makefile, crt0.S, etc. in arima/hdama/hdama
Configuring ROMIMAGE fallback
Configuring DIR /config/Config.lb
Configuring DIR /lib/Config.lb
Configuring DIR /console/Config.lb
Configuring DIR /stream/Config.lb
Configuring DIR /devices/Config.lb
Configuring DIR /pc80/Config.lb
Configuring DIR /boot/Config.lb
Configuring PART mainboard, path arima/hdama
Configuring PART arch, path i386
Adding init file: config/crt0.base
Configuring DIR lib/Config.lb
Configuring DIR boot/Config.lb
Configuring DIR smp/Config.lb
End PART arch
WARNING: Option CONFIG_SMP using default value 0
Configuring PART northbridge, path amd/amdk8
End PART northbridge
Configuring PART southbridge, path amd/amd8111
End PART southbridge
Configuring PART southbridge, path amd/amd8131
End PART southbridge
Configuring PART superio, path NSC/pc87360
End PART superio
Configuring DIR /pc80/Config.lb
Warning, object/driver mc146818rtc previously defined
Configuring PART cpu, path p5
End PART cpu
Configuring PART cpu, path p6
End PART cpu
Configuring PART cpu, path k7
End PART cpu
Configuring PART cpu, path k8
End PART cpu
End PART mainboard
WARNING: Option USE_NORMAL_IMAGE using default value 0
WARNING: Option CONFIG_IDE_STREAM using default value 0
WARNING: Option AMD8111_DEV using default value 0x3800
End ROMIMAGE
Build ROM size 1048576
Creating arima/hdama/hdama/fallback/static.c
Creating arima/hdama/hdama/fallback/Makefile.settings
Creating arima/hdama/hdama/fallback/crt0_includes.h
Creating arima/hdama/hdama/fallback/Makefile
Creating arima/hdama/hdama/fallback/ldoptions
Creating arima/hdama/hdama/Makefile.settings
Creating arima/hdama/hdama/Makefile
bash-2.05b$ 
=========================

Python variability?

Python 2.2.1 (#1, Aug 30 2002, 12:15:30) 
[GCC 3.2 20020822 (Red Hat Linux Rawhide 3.2-4)] on linux2

This on RH 8.0

ron

Weird behaviour with new config method

ron minnich  writes:
> fresh checkout:
>
> Python variability?

Quite possible as I just confirmed the issue on a fresh checkout.
> Python 2.2.1 (#1, Aug 30 2002, 12:15:30) 
> [GCC 3.2 20020822 (Red Hat Linux Rawhide 3.2-4)] on linux2
>
> This on RH 8.0

Python -V
Python 2.1.3

This is on a debian system.

Hmm. Other versions of python are easy to upgrade so:

Python 2.3b2 works
Python 2.2.3 works
Python 2.1.3 fails
Python 1.5   fails

Previously when we had this conversation we tried for a minimum python
of 1.5 and the old config tool manages that.  We may need to bump the
python requirement but we this needs investigation to see why the earlier
versions of python have problems.

Eric

Weird behaviour with new config method

[ron minnich]

On 4 Aug 2003, Eric W. Biederman wrote:
> Previously when we had this conversation we tried for a minimum python
> of 1.5 and the old config tool manages that.  We may need to bump the
> python requirement but we this needs investigation to see why the earlier
> versions of python have problems.

we're going to have to bump minimums anyway. romcc fails for me on RH 7.3.

ron

Weird behaviour with new config method

[Greg Watson]

At 2:56 PM -0600 4/8/03, ron minnich wrote:
> On 4 Aug 2003, Eric W. Biederman wrote:
> > Previously when we had this conversation we tried for a minimum python
> >  of 1.5 and the old config tool manages that.  We may need to bump the
> >  python requirement but we this needs investigation to see why the earlier
> >  versions of python have problems.
>
> we're going to have to bump minimums anyway. romcc fails for me on RH 7.3.
>
> ron

It'll have to be version 2.0 at a minimum since yapps (and config.g) 
depends on some 2.0 features. It maybe a later version depending on 
what Eric's problem turns out to be.

Greg

Weird behaviour with new config method

Greg Watson  writes:
> At 2:56 PM -0600 4/8/03, ron minnich wrote:
> > On 4 Aug 2003, Eric W. Biederman wrote:
> > > Previously when we had this conversation we tried for a minimum python
> > >  of 1.5 and the old config tool manages that.  We may need to bump the
> > >  python requirement but we this needs investigation to see why the earlier
> > >  versions of python have problems.
> >
> > we're going to have to bump minimums anyway. romcc fails for me on RH 7.3.

Hmm.  I will take a look later.  But my default compiler is 2.95.4 and I believe
romcc still compiles with that.  Dropping egcs-2.91.66 is unfortunate but
livable.  With compilers and even more so with binutils pushing the
requirements is not a big deal as it is necessary to support the
newest hardware.

We have no innate requirement for python which makes me worry much
more about pushing the version.
> It'll have to be version 2.0 at a minimum since yapps (and config.g) depends on
> some 2.0 features. It maybe a later version depending on what Eric's problem
> turns out to be.

Hopefully we can track that down.  A few weeks ago it worked with
my version 2.1 of python.

Eric