This is a bit off-topic but hoping someone may have some advice.

I have a PowerMac Quad Core G5 2.5ghz (late 2005 - updated Panasonic
cooler, not the problematic Delphi). I've had it since 2007 and in
that time it's always run Linux. I don't have the original MacOS
disks.

Right now it's running 3.0 Linux kernel. It's overheating. It will
boot into X11 and remain up pretty much indefinitely if the load
average remains under 1.5. If I start Firefox I'll get a spike
(log warnings from Windfarm that the temp is "overtemp" followed by
it returning to normal). If I really load the machine (run a prime
factoring load test) i'll get a "critical" warning from Windfarm
followed by an immediate power off.

Overheating is common and I'm aware of the process to try to fix it
but my concern is that the machine has also become totally unstable
in Open Firmware (OF).

Initially OF was stable but I noticed that the machine had stopped
chiming at power on and in the course of trying the various fixes for
this OF has become totally unstable.

If I power the machine on from cold (i.e powered off > 2 hours) it
will hang in Open Firmware after 2 or 3 commands (innocous commands
like printenv).

If I try "reset-nvram" it usually causes a hang right after it reports
"ok". "setenv" no longer updates the NVRAM (confirmed by rebooting and
running "printenv") but oddly if I boot into Linux and run "nvram" and
then boot into OF I can see the changed value.

I downloaded ASD 2.6.3 which is the correct version for the hardware,
converted the dmg to iso. If I boot with 'c' the screen goes all grey
(which seems normal) then the machine hangs. I can mount the ISO
from Linux using "-t hfsplus" and the contents seem correct.

I've tested the RAM (4 sticks of PC2-4200) in a PC with MemTestx86+
and they are fine. I've also tried swapping in different RAM.

I've tried all of the usual fixes, remove CMOS battery, power off
for a day, clear NVRAM, boot holding down power key to go thru
programmers tone into OF. This dinking is what made it worse :)

Has anyone experienced similar? Maybe I'm wrong but the OF hangs
don't strike me as the same overheating issue.

Tony

--
Tony Jones
SUSE Kernel Performance Team

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Hi Tony,

On 13.05.22 01:11, tonyj@suse.de wrote:

[...]
Overheating is common and I'm aware of the process to try to fix it
but my concern is that the machine has also become totally unstable
in Open Firmware (OF).

Have you already looked below the cover of the cooling system? Is there
any visible leakage (remove power before looking into that)? Even if
not, it could be that the water cooler is missing some liquid after that
much time which could limit its performance.

Other things that come to my mind:

* Have you checked that the radiator of the cooling system is not
cluttered with dust?

UPDATE: Or maybe the pump(s) is/are defective?

[...]
I've tried all of the usual fixes, remove CMOS battery, power off
for a day, clear NVRAM, boot holding down power key to go thru
programmers tone into OF. This dinking is what made it worse :)

Has anyone experienced similar? Maybe I'm wrong but the OF hangs
don't strike me as the same overheating issue.

Not sure about that. I currently have no 11,2 G5 at hand to verify - I
have a single CPU 2.3 Ghz one but it's in storage - but I assume the
CPUs could run unconstrained in OF. And if your cooling solution can't
remove the heat fast enough that could lead to hangs.

Say do you hear the fans speeding up over time?

You can "force" the fans to run on full speed by removing the
transparent cover. Maybe that could make a difference. If it runs better
or longer that way, I'd say there's definitely something wrong with your cooling system.

Cheers,
Frank

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 5/12/22 19:11, tonyj@suse.de wrote:

This is a bit off-topic but hoping someone may have some advice.

I have a PowerMac Quad Core G5 2.5ghz

I have a similar unit that runs flawlessly and seems to be fairly indestructable. The only issues happen when messing with a new Debian
installer image and then tracking down funky new features therein. With
a bit of work with good people it "just runs". Also I did build the new
Linux kernel 5.18.0-rc4 and that also works perfectly. I knew there were
issues with the "windfarm" in the past but those were solved a long time
ago somewhere in the Linux ver 4.x kernel.

Have you given consideration to a fresh up to date install of Debian?



--
Dennis Clarke
RISC-V/SPARC/PPC/ARM/CISC
UNIX and Linux spoken
GreyBeard and suspenders optional

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, May 13, 2022 at 03:58:08PM +0200, Frank Scheiner wrote:

Hi Tony,

Hi.

Have you already looked below the cover of the cooling system? Is
there any visible leakage (remove power before looking into that)?

Yes and no.

not, it could be that the water cooler is missing some liquid after that
much time which could limit its performance.

/sys/devices/platform/windfarm.0 reports

backside-fan: 1100
backside-temp: 57.000
cpu-current-0: 7.566
cpu-current-1: 8.054
cpu-current-2: 9.277
cpu-current-3: 9.886
cpufreq-clamp: 0
cpu-front-fan-0: 3104
cpu-front-fan-1: 3104
cpu-power-0: 9.291
cpu-power-1: 9.891
cpu-power-2: 12.706
cpu-power-3: 13.179
cpu-pump-0: 3584
cpu-pump-1: 3584
cpu-rear-fan-0: 3200
cpu-rear-fan-1: 3200
cpu-temp-0: 39.265
cpu-temp-1: 39.625
cpu-temp-2: 66.046
cpu-temp-3: 70.515
cpu-voltage-0: 1.228
cpu-voltage-1: 1.228
cpu-voltage-2: 1.369
cpu-voltage-3: 1.333


Clearly cpus 2/3 are running hotter. I need to try disabling
them (kernel cmd line or at runtime) and see if cpus 0/1 start
running as hot.

* Have you checked that the radiator of the cooling system is not
cluttered with dust?

Yes.

Say do you hear the fans speeding up over time?

No. The onty time I hear the fan speed up is if the machine hangs,
then about 45-60 seconds later it will go full.

You can "force" the fans to run on full speed by removing the
transparent cover.

I have tried this. No change.

or longer that way, I'd say there's definitely something wrong with
your cooling system.

Clearly there is. My concern however is that I have two issues and
the OF one is distinct from the cooling one. The machine can boot into
the existing Linux (from SATA) and stay up indefinitely (if load av <
1.5) but from a cold boot, will hang in OF in under 1 minute. Maybe
Forth is a real resource hog :-)

My hope in posting this was that I'd find someone on this list who was knowledgable about the underlying hardware. My concern is that I'll
burn the time fixing the cooling and still be left with this OF issue.

--
Tony Jones
SUSE Kernel Performance Team

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, May 13, 2022 at 08:56:00AM -0700, tonyj@suse.de wrote:

On Fri, May 13, 2022 at 03:58:08PM +0200, Frank Scheiner wrote:

Hi Tony,

Hi.

Have you already looked below the cover of the cooling system? Is
there any visible leakage (remove power before looking into that)?

Yes and no.

not, it could be that the water cooler is missing some liquid after that much time which could limit its performance.

/sys/devices/platform/windfarm.0 reports

backside-fan: 1100
backside-temp: 57.000
cpu-current-0: 7.566
cpu-current-1: 8.054
cpu-current-2: 9.277
cpu-current-3: 9.886
cpufreq-clamp: 0
cpu-front-fan-0: 3104
cpu-front-fan-1: 3104
cpu-power-0: 9.291
cpu-power-1: 9.891
cpu-power-2: 12.706
cpu-power-3: 13.179
cpu-pump-0: 3584
cpu-pump-1: 3584
cpu-rear-fan-0: 3200
cpu-rear-fan-1: 3200
cpu-temp-0: 39.265
cpu-temp-1: 39.625
cpu-temp-2: 66.046
cpu-temp-3: 70.515
cpu-voltage-0: 1.228
cpu-voltage-1: 1.228
cpu-voltage-2: 1.369
cpu-voltage-3: 1.333

Clearly cpus 2/3 are running hotter. I need to try disabling
them (kernel cmd line or at runtime) and see if cpus 0/1 start
running as hot.

* Have you checked that the radiator of the cooling system is not
cluttered with dust?

Yes.

Say do you hear the fans speeding up over time?

No. The onty time I hear the fan speed up is if the machine hangs,
then about 45-60 seconds later it will go full.

You can "force" the fans to run on full speed by removing the
transparent cover.

I have tried this. No change.

or longer that way, I'd say there's definitely something wrong with
your cooling system.

Clearly there is. My concern however is that I have two issues and
the OF one is distinct from the cooling one. The machine can boot into
the existing Linux (from SATA) and stay up indefinitely (if load av <
1.5) but from a cold boot, will hang in OF in under 1 minute. Maybe
Forth is a real resource hog :-)

It is certainly not unusual for firmware to run on a single core at 100% (implementing idle handling and interrupts and such is just considered
too much hassle for something that isn't normally going to run).
So yeah it is actually not unlikely that openfirmware would cause the
CPU to run hot compared to linux at low load.

--
Len Sorensen

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 13 May 2022, tonyj@suse.de wrote:

Initially OF was stable but I noticed that the machine had stopped
chiming at power on and in the course of trying the various fixes for
this OF has become totally unstable.

The volume for the chime can be changed, also muted. I think it is an OF
NVRAM setting that does this, but I'm not sure if it's an obvious one or
e.g. a binary OF variable. There used to be an external (by an
independent developer) Mac OS X system settings panel that could be used
to set the "startup volume".

I downloaded ASD 2.6.3 which is the correct version for the hardware, converted the dmg to iso. If I boot with 'c' the screen goes all grey
(which seems normal) then the machine hangs. I can mount the ISO
from Linux using "-t hfsplus" and the contents seem correct.

My experience with .dmg's as images for boot media is that they are best created under Mac OS X. I've had many corrupted CDs and I also had some ATA/ATAPI optical drives that just wouldn't work, even with
known-working media. In any case the original Apple partition layout
matters, thus a converted ISO may not work.

In my experience it almost always worked to dd the raw contents of a
partition from an image/disc to a partition on the internal hard disk.
Setting the correct APM (Apple Partition Map) partition type is key
though, also make sure the partitions are the exact same size in blocks.
(I'd use mac-fdisk to accomplish this.) In my experience it is then
always bootable, and when holding the Option (Ctrl) key after turning on
the Power Mac (at the chime) it is listed as a valid boot option.
My point is: You could try to boot ASD from the internal hard disk...

I've tried all of the usual fixes, remove CMOS battery, power off
for a day, clear NVRAM, boot holding down power key to go thru
programmers tone into OF. This dinking is what made it worse :)

You did replace the PRAM battery, right? In my experience Power Macs act
up when a) no battery is installed and b) a weak/empty battery is used.
Anyway, if you didn't already, replace the battery with a new one. (And
make sure the new one isn't empty as well!)


Long story short, I also think that the liquid cooling needs checking
first, especially before trying anything like thermal calibration for
the CPUs/fans from ASD...

My Power Mac G5s are air cooled. The dual-CPU 2003 model also has one
hotter and one cooler CPU. I think that's normal. The dual-core
single-CPU 2005 model is also air cooled, and when the PRAM battery died
it would simply spin up the fans and hang with no OF and no boot.


I hope you can fix it. Good luck!
Linux User #330250

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Hi Tony!

On 5/13/22 01:11, tonyj@suse.de wrote:

Right now it's running 3.0 Linux kernel. It's overheating. It will
boot into X11 and remain up pretty much indefinitely if the load
average remains under 1.5. If I start Firefox I'll get a spike
(log warnings from Windfarm that the temp is "overtemp" followed by
it returning to normal). If I really load the machine (run a prime
factoring load test) i'll get a "critical" warning from Windfarm
followed by an immediate power off.

There have been quite a lot of PowerPC- and G5-related fixes in newer
kernels as compared to 3.0, so I would suggest using a more recent
version.

Do you know that we have current Debian unstable images for 32- and 64-bit PowerPC available?

https://cdimage.debian.org/cdimage/ports/snapshots/2022-03-28/

Overheating is common and I'm aware of the process to try to fix it
but my concern is that the machine has also become totally unstable
in Open Firmware (OF).

Initially OF was stable but I noticed that the machine had stopped
chiming at power on and in the course of trying the various fixes for
this OF has become totally unstable.
(...)
I've tested the RAM (4 sticks of PC2-4200) in a PC with MemTestx86+
and they are fine. I've also tried swapping in different RAM.

I've tried all of the usual fixes, remove CMOS battery, power off
for a day, clear NVRAM, boot holding down power key to go thru
programmers tone into OF. This dinking is what made it worse :)

Sounds like bad capacitors [1] to me. Have you checked for any suspiciously looking capacitors on the mainboard?

Adrian

PS: We're colleagues, I also work at SUSE ;).

[1] https://en.wikipedia.org/wiki/Capacitor_plague

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer
`. `' Physicist
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, May 13, 2022 at 11:21:47AM -0700, Ken Cunningham wrote:

No doubt you know ... will need to re-run thermal calibration

To do that you will need to get that machine to boot from that ASD
CD, one way or another.

Yes, I do.

I'll take the motherboard out this weekend. I'll check the capacitors
also. I have a Sencore LC103 [who doesn't :-)]

Thanks for all the replies.

--
Tony Jones
SUSE Kernel Performance Team

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

No doubt you know, but I don’t see anyone has said (may have missed it) that you will need to re-run thermal calibration after monkeying with the CPUs as otherwise they are not set up right.

To do that you will need to get that machine to boot from that ASD CD, one way or another.

Ken

On May 13, 2022, at 11:12 AM, tonyj@suse.de wrote:

On Fri, May 13, 2022 at 07:51:27PM +0200, Linux User #330250 wrote:

I downloaded ASD 2.6.3 which is the correct version for the hardware,
converted the dmg to iso. If I boot with 'c' the screen goes all grey
(which seems normal) then the machine hangs. I can mount the ISO
from Linux using "-t hfsplus" and the contents seem correct.

My experience with .dmg's as images for boot media is that they are best
created under Mac OS X. I've had many corrupted CDs and I also had some

Yeah, this had crossed my mind. I can mount it using -t hfsplus but something may be wrong preventing it from booting. I'll see if I can
find a local mac user who can burn the dmg natively from MacOS.

You did replace the PRAM battery, right? In my experience Power Macs act
up when a) no battery is installed and b) a weak/empty battery is used.
Anyway, if you didn't already, replace the battery with a new one. (And
make sure the new one isn't empty as well!)

Yup, new one.

Long story short, I also think that the liquid cooling needs checking
first, especially before trying anything like thermal calibration for
the CPUs/fans from ASD...

I think the first step is to apply new thermal paste to the cpus and
the Northbridge (backside of board). Draining, flushing and
refilling the cooling is an option but it seems like maybe the last
step. Maybe there are some intermediate steps between these two.
I'll do some research.

I hope you can fix it. Good luck!

Thanks! Appreciate the replies.

--
Tony Jones
SUSE Kernel Performance Team

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, May 13, 2022 at 07:51:27PM +0200, Linux User #330250 wrote:

I downloaded ASD 2.6.3 which is the correct version for the hardware, converted the dmg to iso. If I boot with 'c' the screen goes all grey (which seems normal) then the machine hangs. I can mount the ISO
from Linux using "-t hfsplus" and the contents seem correct.

My experience with .dmg's as images for boot media is that they are best created under Mac OS X. I've had many corrupted CDs and I also had some

Yeah, this had crossed my mind. I can mount it using -t hfsplus but
something may be wrong preventing it from booting. I'll see if I can
find a local mac user who can burn the dmg natively from MacOS.

You did replace the PRAM battery, right? In my experience Power Macs act
up when a) no battery is installed and b) a weak/empty battery is used. Anyway, if you didn't already, replace the battery with a new one. (And
make sure the new one isn't empty as well!)

Yup, new one.

Long story short, I also think that the liquid cooling needs checking
first, especially before trying anything like thermal calibration for
the CPUs/fans from ASD...

I think the first step is to apply new thermal paste to the cpus and
the Northbridge (backside of board). Draining, flushing and
refilling the cooling is an option but it seems like maybe the last
step. Maybe there are some intermediate steps between these two.
I'll do some research.

I hope you can fix it. Good luck!

Thanks! Appreciate the replies.

--
Tony Jones
SUSE Kernel Performance Team

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Hi Tony,


tonyj@suse.de wrote:

Right now it's running 3.0 Linux kernel. It's overheating. It will
boot into X11 and remain up pretty much indefinitely if the load
average remains under 1.5. If I start Firefox I'll get a spike
(log warnings from Windfarm that the temp is "overtemp" followed by
it returning to normal). If I really load the machine (run a prime
factoring load test) i'll get a "critical" warning from Windfarm
followed by an immediate power off.

partially perhaps unrelated, but I run 5.x series kernel on my iMac G5,
which is of course not liquid cooled and single-CPU only, but I too get
weird messages from the windfarm.
Up to now I never got critical, but I get very frequent overtemp messages.

I never did run MacOS on it, so i cannot compare. I got this machine
with MorphOS which was running fans at full speed always, like a vacuum cleaner.

What happens to me is that on boot up to promp, all is fine, no messages
and I hear the fan spinning "very low", the computer is quiet and nice.

Now I can get several scenarios

1) I do nothing serious, just use the shell or even leave it there. I
can get an overtemp messages, which *usually* clears after a few seconds
of fan at full speed(=vacuum cleaner). Sometimes not, the fan gets stuck
high.
2) I do serious stuff, I compile stuff like ArcticFox which takes
several hours. Here two there are several scenarios
- Nothing happens for minutes, hours, fans stay low!
- promptly an overtemp message comes, fan full speed.. and it may
disappear and fan go normal after a few seconds and compilation continues
- fans may remain stuck high and stay so even after compilation finished!

Together with the fan burst, also the message of clocking down appears.
Is there a way to monitor easily the CPU clock like with "apm" equivalent?

so... I wonder if anybody with an air-cooled iMac with mac knows if
there is any intermediate fan setting - I am used that PCs since many
many years have variable fan speeds, same goes for many Suns, etc.

I'm suspicious, if a burst of just fan and a second of clock down is
enough... it is stupid thermal regulation. It could be even unreliable,
that is shutting down too early, sicne sometimes it never does!
I also fear that this way performance is suboptimal.

Riccardo

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, 13 May 2022 at 19:09, John Paul Adrian Glaubitz < glaubitz@physik.fu-berlin.de> wrote:

On 5/13/22 01:11, tonyj@suse.de wrote:

https://cdimage.debian.org/cdimage/ports/snapshots/2022-03-28/

Overheating is common and I'm aware of the process to try to fix it
but my concern is that the machine has also become totally unstable
in Open Firmware (OF).

Initially OF was stable but I noticed that the machine had stopped
chiming at power on and in the course of trying the various fixes for
this OF has become totally unstable.
(...)
I've tested the RAM (4 sticks of PC2-4200) in a PC with MemTestx86+
and they are fine. I've also tried swapping in different RAM.

I've tried all of the usual fixes, remove CMOS battery, power off
for a day, clear NVRAM, boot holding down power key to go thru
programmers tone into OF. This dinking is what made it worse :)

Sounds like bad capacitors [1] to me. Have you checked for any suspiciously looking capacitors on the mainboard?

My first thought reading this thread was also capacitors. I'd check none
are bulging/have leaked. Could also be capacitors inside the PSU so testing
a known working PSU could also be quite a good idea.

Ed

<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, 13 May 2022 at 19:09, John Paul Adrian Glaubitz &lt;<a href="mailto:glaubitz@physik.fu-berlin.de">glaubitz@physik.fu-berlin.de</a>&gt; wrote:<

</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="gmail-im">On 5/13/22 01:11, <a href="mailto:tonyj@suse.de" target="_blank">tonyj@suse.de</a> wrote:</span><br>

&gt; <a href="https://cdimage.debian.org/cdimage/ports/snapshots/2022-03-28/" rel="noreferrer" target="_blank">https://cdimage.debian.org/cdimage/ports/snapshots/2022-03-28/</a><br>

&gt; Overheating is common and I&#39;m aware of the process to try to fix it <br>
&gt; but my concern is that the machine has also become totally unstable<br> &gt; in Open Firmware (OF). <br>
&gt; <br>
&gt; Initially OF was stable but I noticed that the machine had stopped <br> &gt; chiming at power on and in the course of trying the various fixes for <br> &gt; this OF has become totally unstable.<br>
&gt; (...)<br>
&gt; I&#39;ve tested the RAM (4 sticks of PC2-4200) in a PC with MemTestx86+ <br>
&gt; and they are fine.  I&#39;ve also tried swapping in different RAM.<br> &gt; <br>
&gt; I&#39;ve tried all of the usual fixes,  remove CMOS battery, power off <br>
&gt; for a day, clear NVRAM,  boot holding down power key to go thru<br>
&gt; programmers tone into OF.  This dinking is what made it worse :)<br>

Sounds like bad capacitors [1] to me. Have you checked for any suspiciously<br> looking capacitors on the mainboard?<br></blockquote><div><br></div><div>My first thought reading this thread was also capacitors. I&#39;d check none are bulging/have leaked. Could also be capacitors inside the PSU so testing a known working PSU could
also be quite a good idea.<br><br></div><div>Ed</div><br></div></div>

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)