PCChips M758LMR+ works for me but...

[Todd E. Johnson]

Hi again all,

I am able to boot and get X up.  It takes about 45 seconds, which is not
bad, but it is a bit more than I was expecting.  sshd takes up a bit of
this time though.  I have most everything else disabled.

One of the largest problems is that none of the onboard USB ports work.
Also, none of the PCI cards I am using appears to work as well (Which
all of the devices work fine with the factory bios).  Finally, I
unfortunately rely heavily on ACPI to startup and shutdown the system.
I have included a bit of output with this message, such as my config
file, output from `lspci -vvv`, and the output from `dmesg`.  Perhaps I
did something wrong, or I'm setting my expectations a bit too high.  I
appreciate anyone's time and effort that the may put into this, and I
thinking this pretty much rocks!  (This would explain why I bought
another motherboard and a few extra DOCs...)

Any and all ideas are appreciated!

Regards,

Todd E. Johnson
[e-mail]

[Ronald G Minnich]

On Thu, 3 Oct 2002, Todd E. Johnson wrote:
> I am able to boot and get X up.  It takes about 45 seconds, which is not
> bad, but it is a bit more than I was expecting.  sshd takes up a bit of
> this time though.  I have most everything else disabled.

I assume this is redhat 7. The slowness of the bootup scripts has managed
to almost eliminate the speed of boot you get with linuxbios. It's
depressing.

There is a tool that will optimize boot-time scripts you might want to
look at.
> One of the largest problems is that none of the onboard USB ports work.

I believe this has been solved, Andrew Ip and Ollie are the experts on
this one.
> Also, none of the PCI cards I am using appears to work as well (Which
> all of the devices work fine with the factory bios).

This sounds like an IRQ table problem. Did you get a chance to run getpir
and compare to linuxbios getpir?
> Finally, I unfortunately rely heavily on ACPI to startup and shutdown
> the system.

what features do you use? we're weak in this area and could use help.

Thanks for your patience ...

ron

[Todd E. Johnson]

Thanks for the response!
> I assume this is redhat 7. The slowness of the bootup scripts 
> has managed to almost eliminate the speed of boot you get with

linuxbios. It's
> depressing.

No, sorry, It's Slackware 8.1 and about half the boot scripts have been
remarked out.
> I believe this has been solved, Andrew Ip and Ollie are the experts on
> this one.

That's great, I have been relying on USB...
> This sounds like an IRQ table problem. Did you get a chance 
> to run getpir
> and compare to linuxbios getpir?

No, but I will certainly so this!
> what features do you use? we're weak in this area and could use help.

I primarily have been using ACPI for startup & shutdown.  Startup will
always work when there is momentary contact to the power header.  ACPI
fails to load which causes it to not see the power button.  There is not
an ACPI directory under /proc, so ACPID has nothing to watch for (ex:
Momentary contact on the power button to initiate `shutdown`.

Regards,

Todd E. Johnson
[e-mail]

[Ronald G Minnich]

On Fri, 4 Oct 2002, Todd E. Johnson wrote:
> I primarily have been using ACPI for startup & shutdown.  Startup will
> always work when there is momentary contact to the power header.  ACPI
> fails to load which causes it to not see the power button.  There is not
> an ACPI directory under /proc, so ACPID has nothing to watch for (ex:
> Momentary contact on the power button to initiate `shutdown`.

The trend in the free os community is to do ACPI functions in the OS --
this is smart, I think. You need to get ACPI tables from the existing BIOS
and then see how far the Linux ACPI emulator is and how well it works.

This is a bit of work but would be a huge help and we would love to have a
writeup.

ron

[Ian Molton]

On Fri, 4 Oct 2002 08:42:09 -0600 (MDT)
Ronald G Minnich  wrote:
> The trend in the free os community is to do ACPI functions in the OS
> -- this is smart, I think.

Even smarter would be to ditch ACPI and design a proper interface...

[Ronald G Minnich]

On Fri, 4 Oct 2002, Ian Molton wrote:
> Even smarter would be to ditch ACPI and design a proper interface...

agreed, but somebody has to take that on and then get the world to agree.

ron

[Todd E. Johnson]

Hello,

I would tend to agree that it _may_ help in a rather large cluster.  For
example, a node's fan or environmental control fails and the system
becomes too hot, ACPI may signal an event to gracefully shut down the
node.  Or perhaps look towards CPU throttling.  However, from some of
the messages I have seen to the LinuxACPI mailing list from Linus, it is
hard to gauge the outlook for ACPI in Linux, or I am mis-reading the
messages altogether.

For my development purposes, I have completely disabled ACPI in my
kernel config, however, this had no effect.  Also, when I run `getpir`
and generate a new irq_table.c, do I need to make any modification to
that source?  Or, do I simply replace the existing <freebios
home>/src/util/mainboard/pcchips/m758lmr+/irq_table.c ?

If the latter is the case, this had no effect.  I used the resulting
irq_table.c from `getpir` with "option DISABLE_INTERNAL_DEVICES=1" in my
config, and without.  No change: No USB, and the same IRQ problems.  The
only change is  with "option DISABLE_INTERNAL_DEVICES=1" set in the
config file, it now boots in 35 seconds, but of course, I didn't have
any devices working besides the FrameBuffer console...

A quick side note, in the config file, there is no difference between
"option DISABLE_INTERNAL_DEVICES=1" && "option
DISABLE_INTERNAL_DEVICES=0"

Regards,

Todd E. Johnson
[e-mail]
> -----Original Message-----
> From: [e-mail] 
> [mailto:[e-mail]] On Behalf Of Ronald 
> G Minnich
> Sent: Friday, October 04, 2002 10:42 AM
> To: Todd E. Johnson
> Cc: [e-mail]
> Subject: RE: PCChips M758LMR+ works for me but...
>
> On Fri, 4 Oct 2002, Todd E. Johnson wrote:
> > I primarily have been using ACPI for startup & shutdown.
>
> Startup will
> > always work when there is momentary contact to the power
>
> header.  ACPI
> > fails to load which causes it to not see the power button.
>
> There is not
> > an ACPI directory under /proc, so ACPID has nothing to
>
> watch for (ex:
> > Momentary contact on the power button to initiate `shutdown`.
>
> The trend in the free os community is to do ACPI functions in 
> the OS --
> this is smart, I think. You need to get ACPI tables from the 
> existing BIOS
> and then see how far the Linux ACPI emulator is and how well it works.
>
> This is a bit of work but would be a huge help and we would 
> love to have a
> writeup.
>
> ron
>
> Linuxbios mailing list
> [e-mail]
> http://www.clustermatic.org/mailman/listinfo/linuxbios

[Gregg C Levine]

Hello from Gregg C Levine
Um, Todd, way to go choosing Slackware, that's what I use. Now what's
this about  the slowness of the bootup scripts? I use Slackware here,
and about all I can say, it's a bit strange about those scripts. Of
course, this is version, 8.0, I haven't ordered copies of 8.1 on CD yet.
-------------------
Gregg C Levine [e-mail]
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke."  Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
> -----Original Message-----
> From: [e-mail] [mailto:linuxbios-
> [e-mail]] On Behalf Of Todd E. Johnson
> Sent: Friday, October 04, 2002 8:24 AM
> To: 'Ronald G Minnich'
> Cc: [e-mail]
> Subject: RE: PCChips M758LMR+ works for me but...
>
> Thanks for the response!
> > I assume this is redhat 7. The slowness of the bootup scripts
> > has managed to almost eliminate the speed of boot you get with
>
> linuxbios. It's
> > depressing.
>
> No, sorry, It's Slackware 8.1 and about half the boot scripts have

been
> remarked out.
> > I believe this has been solved, Andrew Ip and Ollie are the experts

on
> > this one.
>
> That's great, I have been relying on USB...
> > This sounds like an IRQ table problem. Did you get a chance
> > to run getpir
> > and compare to linuxbios getpir?
>
> No, but I will certainly so this!
> > what features do you use? we're weak in this area and could use

help.
> I primarily have been using ACPI for startup & shutdown.  Startup will
> always work when there is momentary contact to the power header.  ACPI
> fails to load which causes it to not see the power button.  There is

not
> an ACPI directory under /proc, so ACPID has nothing to watch for (ex:
> Momentary contact on the power button to initiate `shutdown`.
>  Regards,
>
> Todd E. Johnson
> [e-mail]
>
> Linuxbios mailing list
> [e-mail]
> http://www.clustermatic.org/mailman/listinfo/linuxbios

[Todd E. Johnson]

Actually,

Previously I moved most of the boot scripts off to a backup directory...
There was really no change.  I have since moved all boot scripts to a
directory called "/etc/rc.d/HOLD".  Heh... This was kind of a mistake.
For those interested, I have attached a copy of my serial post to this
message.  I'm probably going to re-build from scratch soon since I have
been doing a lot of un-organized development on this system.

BTW, it seems that there is no booting action (Based on the Serial
output) until the HDD spins up.  Is this a result of me keeping my root
file system on the HDD?

Regards,

Todd E. Johnson
[e-mail]
> -----Original Message-----
> From: Gregg C Levine [mailto:[e-mail]] 
> Sent: Friday, October 04, 2002 10:17 PM
> To: [e-mail]; 'Ronald G Minnich'
> Cc: [e-mail]
> Subject: RE: PCChips M758LMR+ works for me but...
>
> Hello from Gregg C Levine
> Um, Todd, way to go choosing Slackware, that's what I use. Now what's
> this about  the slowness of the bootup scripts? I use Slackware here,
> and about all I can say, it's a bit strange about those scripts. Of
> course, this is version, 8.0, I haven't ordered copies of 8.1 
> on CD yet.
> Gregg C Levine [e-mail]
>
> "The Force will be with you...Always." Obi-Wan Kenobi
> "Use the Force, Luke."  Obi-Wan Kenobi
> (This company dedicates this E-Mail to General Obi-Wan Kenobi )
> (This company dedicates this E-Mail to Master Yoda )

[Ronald G Minnich]

On Fri, 4 Oct 2002, Todd E. Johnson wrote:
> BTW, it seems that there is no booting action (Based on the Serial
> output) until the HDD spins up.  Is this a result of me keeping my root
> file system on the HDD?

you can't do anything until the HDD spins up, and yes it's because you've
god file system on the HDD. There's not much to be done for this.

ron

[Andrew Ip]

Todd,
> One of the largest problems is that none of the onboard USB ports work.
> Also, none of the PCI cards I am using appears to work as well (Which
> all of the devices work fine with the factory bios).  Finally, I
> unfortunately rely heavily on ACPI to startup and shutdown the system.
> I have included a bit of output with this message, such as my config
> file, output from `lspci -vvv`, and the output from `dmesg`.  Perhaps I
> did something wrong, or I'm setting my expectations a bit too high.  I
> appreciate anyone's time and effort that the may put into this, and I
> thinking this pretty much rocks!  (This would explain why I bought
> another motherboard and a few extra DOCs...)

For newer m758lr+, it has different irq routing table.  Try to generate a newer
one using getirq.  Also, don't forget to enable

option DISABLE_INTERNAL_DEVICES=1

-Andrew