feipoa wrote on 2020-04-14, 09:16:

Have other users tested DOS HWINFO using a Cyrix 5x86?

Yes, I have some reports from the past where it worked..

1  ║ Main Processor:           Cyrix 5x86, 100.1 MHz                          ║
2  ║ Math Co-Processor:        Internal, 98.5 MHz                             ║
3  ║ Bus Type:                 ISA + PCI Local Bus (V2.10)                    ║
4  ║ Mainboard Chipset:        SiS-496-497/A/B                      ║

I'm now preparing a special build that will output more information about what it's actually checking to narrow down the issue.

feipoa wrote on 2020-04-14, 09:16:

Have other users tested DOS HWINFO using a Cyrix 5x86?

Could you please try this build just like before and if it hangs, please take a screenshot where it did:
[pulled, newer build available]

I tried the new file with no ini file, with ini having mode 2 only, with ini having mode 1 and 3 only, and with ini having mode 1/2/3 all 0's.

hwinfo -r hangs as shown previously.

hwinfo hanges as shown previously in this image:

No fatal error warning this time.

That's all I have time for this evening. It is pretty late here and with schools closed, they are expecting parents to become at-home primary school teachers. Will have to pick this up tomorrow.

feipoa wrote on 2020-04-14, 10:14:

I tried the new file with no ini file, with ini having mode 2 only, with ini having mode 1 and 3 only, and with ini having mode […]
Show full quote

I tried the new file with no ini file, with ini having mode 2 only, with ini having mode 1 and 3 only, and with ini having mode 1/2/3 all 0's.

hwinfo -r hangs as shown previously.

hwinfo hanges as shown previously in this image:
Running.jpg
No fatal error warning this time.

That's all I have time for this evening. It is pretty late here and with schools closed, they are expecting parents to become at-home primary school teachers. Will have to pick this up tomorrow.

Thanks! Yes, the situation including schools and teaching parents is same here, and with 2 primary school kids at home I know how you feel 😁

This new build should be showing detection progress at bottom of the screen (in the grey bar) once you open the Mainboard info screen. Have you noticed anything reported there before it hung?
Also this CPU (rev 5) should be able to enable CPUID instruction and HWiNFO attempts to do so. Do you know if it's possible here, have you perhaps tried that and what does it provide then?
I'm attaching another build, where I updated 5x86 detection, let's see what this will do: [pulled, newer build available]

And here a new build for Win9x/NT test, which should fix the BSOD on NT4 and improve some items in Win95: www.hwinfo.com/beta/hwi32_625_4137.zip

I also have an inquire, when booting from a floppy disk with DOS (and a mouse driver), I get a "divide error" when exiting the hardware info screen (to get back to the menus).

Dusko wrote on 2020-04-15, 01:33:

I also have an inquire, when booting from a floppy disk with DOS (and a mouse driver), I get a "divide error" when exiting the hardware info screen (to get back to the menus).

Which version is that, 6.0.5? Are you using an XMS manager (HIMEM.SYS) ? Are you able to browse subsequent screens under that one without crashing?
If you don't have XMS loaded, please try this build, which should fix a crash when XMS manager is not loaded: Re: HWiNFO for DOS resurrected !

Mumak wrote on 2020-04-14, 10:26:

Thanks! Yes, the situation including schools and teaching parents is same here, and with 2 primary school kids at home I know ho […]
Show full quote

Thanks! Yes, the situation including schools and teaching parents is same here, and with 2 primary school kids at home I know how you feel :-D

This new build should be showing detection progress at bottom of the screen (in the grey bar) once you open the Mainboard info screen. Have you noticed anything reported there before it hung?
Also this CPU (rev 5) should be able to enable CPUID instruction and HWiNFO attempts to do so. Do you know if it's possible here, have you perhaps tried that and what does it provide then?
I'm attaching another build, where I updated 5x86 detection, let's see what this will do: HWINF605.ZIP

And here a new build for Win9x/NT test, which should fix the BSOD on NT4 and improve some items in Win95: www.hwinfo.com/beta/hwi32_625_4137.zip

It is pretty tough, I have 3 kids - pre-school, kindergarten (english), and 2nd grade (french). Mr. Pre-school just won't leave the other two alone. Noise and fighting all day long. Learning isn't going too well - I'm having to bribe one of them with chocolate to focus.

OK. I have tested the latest DOS HWINFO, but it didn't help the situation. The screen looks like the attached after going to Mainboard Info. Also tried with the -r flag, but hangs on Creating Report File. Also, I don't see a progress bar.

If I set the CPUID methods all to 0, I am able to press the Quit button, but am left without a command prompt:

feipoa wrote on 2020-04-15, 07:00:

It is pretty tough, I have 3 kids - pre-school, kindergarten (english), and 2nd grade (french). Mr. Pre-school just won't leave […]
Show full quote

Mumak wrote on 2020-04-14, 10:26:

Thanks! Yes, the situation including schools and teaching parents is same here, and with 2 primary school kids at home I know ho […]
Show full quote

Thanks! Yes, the situation including schools and teaching parents is same here, and with 2 primary school kids at home I know how you feel 😁

This new build should be showing detection progress at bottom of the screen (in the grey bar) once you open the Mainboard info screen. Have you noticed anything reported there before it hung?
Also this CPU (rev 5) should be able to enable CPUID instruction and HWiNFO attempts to do so. Do you know if it's possible here, have you perhaps tried that and what does it provide then?
I'm attaching another build, where I updated 5x86 detection, let's see what this will do: HWINF605.ZIP

And here a new build for Win9x/NT test, which should fix the BSOD on NT4 and improve some items in Win95: www.hwinfo.com/beta/hwi32_625_4137.zip

It is pretty tough, I have 3 kids - pre-school, kindergarten (english), and 2nd grade (french). Mr. Pre-school just won't leave the other two alone. Noise and fighting all day long. Learning isn't going too well - I'm having to bribe one of them with chocolate to focus.

OK. I have tested the latest DOS HWINFO, but it didn't help the situation. The screen looks like the attached after going to Mainboard Info. Also tried with the -r flag, but hangs on Creating Report File. Also, I don't see a progress bar.
Just_after.jpg

If I set the CPUID methods all to 0, I am able to press the Quit button, but am left without a command prompt:
Quit_returns_this.jpg

Wow, this is really weird. Have you perhaps noticed some information being shown in the right bottom part of screen (just where it's initially showing "Reading Data ..." during startup?
Are you able to open the Video screen?

In that section of the screen, there is a very brief display of "Test Cache". It is there for less than a second. I had to record a video of it to be sure.

I have another system based on the Cyrix 5x86 setup which I also tested. The system uses a Biostar MB-8433UUD motherboard and runs an IBM 5x86c at 133 MHz with a 66 MHz FSB. This system also had trouble running HWiNFO, but the symptoms are a little different.

Using CPU Method 0 or 2, the system hangs on Mainboard Info. Using CPU Method 1,3: system resets, but displays Resetting CPU (Method 3) briefly before reset. If I edit the ini file to use just CPU Method 1, the system will display Resetting CPU (Method 1) just before restarting.

feipoa wrote on 2020-04-15, 08:17:

I have another system based on the Cyrix 5x86 setup which I also tested. The system uses a Biostar MB-8433UUD motherboard and runs an IBM 5x86c at 133 MHz with a 66 MHz FSB. This system also had trouble running HWiNFO, but the symptoms are a little different.

Using CPU Method 0 or 2, the system hangs on Mainboard Info. Using CPU Method 1,3: system resets, but displays Resetting CPU (Method 3) briefly before reset. If I edit the ini file to use just CPU Method 1, the system will display Resetting CPU (Method 1) just before restarting.

And here a new test build, I removed attempt to enable CPUID, perhaps that was causing the issue:
[pulled, newer build available]

1Fatal Error
2Cannot Continue
3Quit

Hangs, cannot Quit.

Should I have CPU methods all at 0? I left it at default.

feipoa wrote on 2020-04-15, 09:34:

Hangs, cannot Quit. […]
Show full quote

Copy code to clipboard

1Fatal Error
2Cannot Continue
3Quit

Hangs, cannot Quit.

Should I have CPU methods all at 0? I left it at default.

0 for would be best to exclude a problem with the reset.. Are you able to catch it also reporting "Get freq" in the status before it freezes?

All it flashes is Test Cache.

Oh this is totally weird. I will need to think about what to try next, but for now let's have some rest...

If AMD Am5x86 chips work, can you try treating the Cyrix 5x86 as an Am5x86?

feipoa wrote on 2020-04-15, 09:52:

If AMD Am5x86 chips work, can you try treating the Cyrix 5x86 as an Am5x86?

I don't think that would help in this case. I'm rather thinking about writing some sort of Debug File (or other debug info) during the run, that would help me to narrow down the issue.

There's also another interesting factor. The CPU detection (except resetting) runs also right after you click OK at the welcome screen. This seems to work as you can reach the menu.
But the second time it runs (with resetting if enabled) in the Mainboard info screen it hangs.. So if you have all Reset Methods disabled, it should be the same...

Not sure what to say about that. Symptoms occur on both Cx5x86 systems. I did try it with all 0's in the ini file, but it still hangs.

On a different note, I saw this comment,

Mumak wrote on 2020-03-19, 15:25:

Thanks for the feedback. ... It would be even better if I could tell the TI486DLC from Cx486S, but I'm not sure if this is possible at all.

And recalled that the Cx486s has 2 KB of cache, while the DLC has 1 kb. Can you use the cache size difference as an aid here?

feipoa wrote on 2020-04-15, 10:33:

Not sure what to say about that. Symptoms occur on both Cx5x86 systems. I did try it with all 0's in the ini file, but it stil […]
Show full quote

Not sure what to say about that. Symptoms occur on both Cx5x86 systems. I did try it with all 0's in the ini file, but it still hangs.

On a different note, I saw this comment,

Mumak wrote on 2020-03-19, 15:25:

Thanks for the feedback. ... It would be even better if I could tell the TI486DLC from Cx486S, but I'm not sure if this is possible at all.

And recalled that the Cx486s has 2 KB of cache, while the DLC has 1 kb. Can you use the cache size difference as an aid here?

Yes, that should be possible, I will add that.
I have just another idea for the Cx5x86 to dump other debug info, can post a new build in a few minutes..