Fatdog64-710 Final [4 Dec 2016]

A home for all kinds of Puppy related projects
Message
Author
Terry H
Posts: 708
Joined: Sun 29 Mar 2009, 16:48
Location: The Heart of Muskoka, ON Canada

#521 Post by Terry H »

I restarted without a savefile, then updated mesa only. On reboot the panels were displayed successfully. Completed other updates via gslapt, all still OK. I then copied the panel.conf and my mozilla .default file from the old save file. After several reboots, all working normally now with a new savefile. I checked the /usr/lib64 and missing files are there now.

Yoliano
Posts: 64
Joined: Mon 27 Feb 2012, 15:07

Pmcputemp

#522 Post by Yoliano »

@ Smokey01

Thanks for your suggestion, which I have followed. The content of pmcputemprc
on my PC is as shown in the image below. Any thoughts?
Attachments
pmcputemprc.png
(50.74 KiB) Downloaded 212 times

Yoliano
Posts: 64
Joined: Mon 27 Feb 2012, 15:07

Pmcputemp

#523 Post by Yoliano »

@ Smokey01

Forgot to mention your first suggestion! My apologies. I deleted /root/.config/pmcputemp and it regenerated instantly. But it changed nothing. Icon still disappearing. Did this before copying content of pmcputemprc shown in previous post. Thanks.

User avatar
smokey01
Posts: 2813
Joined: Sat 30 Dec 2006, 23:15
Location: South Australia :-(
Contact:

#524 Post by smokey01 »

@Yoliano, I'm running out of ideas.

I have Fatdog64-710 installed on an ext4 partition. Are you using something different? I can't see this being the culprit but you never know.

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#525 Post by SFR »

@Yoliano: Did you try to launch pmcputemp from terminal?
Any error messages when it stops working?

Greetings!
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

Yoliano
Posts: 64
Joined: Mon 27 Feb 2012, 15:07

Pmcputemp

#526 Post by Yoliano »

@smokey01

710 is installed on USB 3.1 memstick, (see earlier post), on ext4 partition, all in accordance with jb's instructions.

@SFR

Thanks for your intervention. After the temp icon had already disappeared in this session, I relaunched it from a terminal. A few minutes later it disappeared again. The terminal output is as follows:
Attachments
pmcputemp-terminal-message.png
(10.6 KiB) Downloaded 636 times

User avatar
smokey01
Posts: 2813
Joined: Sat 30 Dec 2006, 23:15
Location: South Australia :-(
Contact:

#527 Post by smokey01 »

@Yoliano, looks like it might be a hardware issue. Might have to look at 01micko's code. SFR is probably a better person to do that than me :lol:

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#528 Post by SFR »

@Yoliano: Looks like Mick set the max. allowed temperature to 99°, whereas it actually can get higher (e.g. on my Acer netbook TJMax is 105°).

I've increased it to 150°, see if it works reliably now.
Btw, there's also a min. limit (10°), but I didn't touch it, because it seems to be max's issue in your case.

Greetings!
Attachments
pmcputemp-0.63-x86_64-1.txz.gz
NOTE: Remove fake .gz extension!

MD5: 7e29b070e234e49dc909288b8a720206 pmcputemp-0.63-x86_64-1.txz
(11.8 KiB) Downloaded 95 times
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

Yoliano
Posts: 64
Joined: Mon 27 Feb 2012, 15:07

Pmcputemp

#529 Post by Yoliano »

@smokey01

Many thanks for your help, anyway. It encouraged me to keep trying, which was a very valuable stimulus.

@SFR

Thanks indeed for your gracious gift of adjusted software. I downloaded it and removed the false .gz. Using gslapt, I uninstalled the existing version. Then I installed your adjusted version and launched it in a terminal. The Temp Icon appeared in the tray. Left PC running and went away for twenty minutes to have a coffee. Returned to find that the Temp Icon had disappeared! The terminal output was exactly the same as that posted previously. Launched it again and commenced typing this post. I happened to glance at the Icon and saw it disappear, a few moments ago. So no luck, I regret to have to tell you.

For your information, my PC is a Toshiba Satellite 17" Laptop, Model: C75D-B7215, with AMD A8 Quadcore processor, 8GB Ram, 1 TB HD. It is about 2 years old. It's a good machine in every way and Fd64-710 sits on it very happily and delivers a smooth and fast performance, - except for this little minor glitch! The strange thing is that Pmcputemp functioned perfectly for about six weeks after first installing 710, (previously using 702), then suddenly started this random exiting. Something must have changed, although I have not observed any other signs of change in the performance of my PC.

I am surprised at the magnitude of the temperatures that you mention in your last post. In my observation of the temps. presented by the Pmcputemp Icon, I have seen:

Immediately after Boot: 39/40 degrees.
After heavy cpu load session: 53/55 degrees. (This is maximum).
Average session: 46/49 degrees.

But I can understand that rapid, transient, high temperatures could occur under certain conditions, but are so quickly cooled that they don't get presented by the Icon. Putting two fingers in front of the cooling air vent tells me that the cooling air temperature does indeed change very quickly, both up and down. Just an observation!

I don't know if there are any other avenues to explore in this matter, so I place myself in your hands. But thank you very much indeed for your dedicated help in providing this experiment. I appreciate it very much.

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#530 Post by SFR »

Hmm, this error message can appear only in one of two conditions: raw temperature value is less than 10000 or greater than 100000 (or 150000 with my patch).
So, maybe indeed it's a kind of unusually low/high spike in your hardware...
To know for sure, please run the attached debug packgage in terminal.
It's the original version with added debug statement, so we'll see what's the actual reading right before it exits.

Also, I've rewritten my previous patch, so now it won't exit anymore, even if the value is out of range (will be clipped instead).
Please try it after the debug one is finished.

Greetings!
Attachments
pmcputemp-0.63_debug-x86_64-1.txz.gz
NOTE: Remove fake .gz extension!

MD5: 018bfa872ce1734020d61531dd90e815 pmcputemp-0.63_debug-x86_64-1.txz
(11.69 KiB) Downloaded 72 times
pmcputemp-0.63-x86_64-1.txz.gz
NOTE: Remove fake .gz extension!

MD5: e62814551135b948c362638b8de05cd5 pmcputemp-0.63-x86_64-1.txz
(11.85 KiB) Downloaded 89 times
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

Yoliano
Posts: 64
Joined: Mon 27 Feb 2012, 15:07

Pmcputemp

#531 Post by Yoliano »

@SFS

Thank you very much for your new resources. Before using them, I deleted all previous pmcputemp files, using gslapt & pfind. Cleared the decks! Made a fresh start, installed only your debug version. Ran it from Terminal and waited. Icon disappeared. Debug output OK and last page copied below. It looks as if the temp. sensor stops sending it's signal. Perhaps the sensor has a dodgy circuit component that results in a random, intermittent connection. Maybe you have a better explanation.

After that, I uninstalled the debug version and installed your version with the second patch. Ran it from a Terminal. No messages returned. It is running now as I type. It has been running for about 35 mins. and continues to display the varying core temperature. My impression is that it is more stable. It has already run longer than any of the recent, previous versions. Which tends to negate my theory above! Clearly, if the temperature display remains faithful and does not disappear, I am going to have to conduct a longer term test and report back to you, perhaps in a few days.

Look forward to hearing your conclusions. Thanks again for your most valuable help.

Still displaying!

Cheers.
Attachments
pmcputemp-debug.png
(59.15 KiB) Downloaded 449 times

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#532 Post by SFR »

That's more or less what I was expecting, but the other way around - the "out of range" reading was 0 and not above 100000.
Hard to tell why's that happening, maybe some small glitch in the sensor..?

The 2nd revision of my patch (debugless, btw, that's why there's no output in the terminal) was designed not to exit, no matter what value it gets from sensors, so it can handle that 0 well and move on.

Anyway, glad we managed to sort it out.

Greetings!
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#533 Post by prehistoric »

@SFR,

Your debugless version is now steadily displaying in my tray, but the value displayed after startup is 10 degrees, which is implausible on fundamental grounds, and explains why it was exiting.

This is on a desktop with an AMD A10-5800K quad-core processor in an FM2 socket on an Asus F2A85-M board.

Sorry I don't have time to do much debugging. I'm getting ready to move house, and with my collection of stuff this is a major problem.

Sage
Posts: 5536
Joined: Tue 04 Oct 2005, 08:34
Location: GB

#534 Post by Sage »

The cpu temperature conundrum is as old as the hills/286, filling many tomes! They used to stick a flexi thermocouple under the base, inside the centre of the socket ,so that it brushed onto the underside of the cpu case. Some folks smeared it with HS compound to get better contact. Loads of reports, loads of different values, and, of course, not all cpu s built the same even back then. Furthermore, the recorded temperature only related to the outside of the cpu encapsulation. So one of the fabs (might've Texas?) had the great idea of putting the sensor inside the case; others followed. It wasn't clear what the nature of the sensor was, where it was sited, what measuring algorithm was in FW, and, most importantly, how and where it was interrogated, i.e some did it in BiOS, others in the OS SW, some both - giving different answers. Everyone always got different answers, anyway. Next they incorporated an extra diode within the main cpu design. There was a short hiatus during which most people either got the same answer or were happy with what they saw. Notwithstanding, most of the intangibles were still present to some extent. As Mr Moore's Law progressed and features became smaller, temperature became much more relevant. Intel had even more dramatic design failures. [Probably no excuse in their case as they doubtless have access to s.o.t.a. ir and other more direct means of sensing the internal temperatures in the lab, although there was a period when they released designs with amazing egg-boiling external temperatures.] Silver dag, solid copper HSF s weighing massive amounts, water-cooling, all saw service along the track until we had the advent of multiple cores, presently arriving at up to eight being switched in and out in FW according to function, demand and a plethora of determinants. I've lost sight of where the diode or diodes are located in which core or cores and how they're accessed and how the temperature or temperatures are interpreted or interpolated in SW??
Most of the intangibles still remain, my reported T s seem to be diminishing in both SW reports and by touch of the finger tip. But with features down to 14microns, they have to run cool(er) because those junctions are still alloys, by definition; silicon melts at 1683K, but alloys take a steep dive towards the eutectic low. If line widths get much smaller, either we'll hit the Uncertainty Principle (!) or need a bucket of liqHe. They say it might be cooler on Mars...

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#535 Post by SFR »

prehistoric wrote:@SFR,

Your debugless version is now steadily displaying in my tray, but the value displayed after startup is 10 degrees, which is implausible on fundamental grounds, and explains why it was exiting.

This is on a desktop with an AMD A10-5800K quad-core processor in an FM2 socket on an Asus F2A85-M board.

Sorry I don't have time to do much debugging. I'm getting ready to move house, and with my collection of stuff this is a major problem.
That 10° lasts only for some initial period of time right after startup or is it permanent?

If the latter, do you get normal readings if you use

Code: Select all

sensors
command?

No rush, take your time.

Greetings!
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

chiron²
Posts: 70
Joined: Tue 21 Jan 2014, 18:36

#536 Post by chiron² »

Hello,

Just did the step from FD631 to 710. Everything's OK so far.

Has someone tried to do a MyPaint (1.2) package for FD710? I am stuck with some unresolvable error messages, claiming that a pygtkcompat cannot be found, when it is actually there. Some version conflict within Python maybe? IDK, given up on it, it's not that important.

Is there a distribution where one could scavenge packages? Or is it more or less 'compile yourself'?

Hanvon graphics tablet works, after compiling the Hanvon driver. The included Hanwang.ko does not work (and never did for the ArtMaster tablet).

Stellarium and Celestia SFS from 631 work.

Cheers

Norman

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#537 Post by prehistoric »

SFR wrote:...
That 10° lasts only for some initial period of time right after startup or is it permanent?

If the latter, do you get normal readings if you use

Code: Select all

sensors
command?
...
The reading seems permanent. Here's what I get from sensors

Code: Select all

# sensors
radeon-pci-0008
Adapter: PCI adapter
temp1:         +2.0°C  (crit = +120.0°C, hyst = +90.0°C)

k10temp-pci-00c3
Adapter: PCI adapter
temp1:         +6.1°C  (high = +70.0°C)
                       (crit = +70.0°C, hyst = +69.0°C)

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#538 Post by SFR »

Such low readings seem to be quite common for k10temp:
https://bbs.archlinux.org/viewtopic.php?id=208303 wrote:k10temp-pci-00c3
Adapter: PCI adapter
temp1: +11.0°C (high = +70.0°C)
(crit = +80.0°C, hyst = +77.0°C)
https://forums.opensuse.org/showthread.php/497668-k10temp-reports-impossibly-low-temperatures wrote:k10temp-pci-00c3
Adapter: PCI adapter
temp1: +12.0°C (high = +70.0°C)
(crit = +90.0°C, hyst = +87.0°C)
https://askubuntu.com/questions/254305/what-is-my-cpu-temperature wrote:k10temp-pci-00c3
Adapter: PCI adapter
temp1: +7.4 C (high = +70.0 C)
(crit = +90.0 C, hyst = +87.0 C)
Greetings!
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

User avatar
prehistoric
Posts: 1744
Joined: Tue 23 Oct 2007, 17:34

#539 Post by prehistoric »

@SFR

Something wrong there when room temperature is 80F (27 C).

User avatar
SFR
Posts: 1800
Joined: Wed 26 Oct 2011, 21:52

#540 Post by SFR »

Well, it appears so, but (at least in case of k10temp; coretemp and even k8temp seem to be more realistic) those values have not much to do with an actual, physical temperature:
https://www.kernel.org/doc/Documentation/hwmon/k10temp wrote:Tctl is the processor temperature control value, used by the platform to control cooling systems. Tctl is a non-physical temperature on an arbitrary scale measured in degrees. It does _not_ represent an actual physical temperature like die or case temperature. Instead, it specifies the processor temperature relative to the point at which the system must supply the maximum cooling for the processor's specified maximum case temperature and maximum thermal power dissipation.
Bottom line is: as long as it's below high/crit, it's ok, regardless how weird it looks. They're just numbers...

Greetings!
[color=red][size=75][O]bdurate [R]ules [D]estroy [E]nthusiastic [R]ebels => [C]reative [H]umans [A]lways [O]pen [S]ource[/size][/color]
[b][color=green]Omnia mea mecum porto.[/color][/b]

Post Reply