crvs wrote on 2021-04-09, 08:10:

Mumak wrote on 2021-04-08, 17:01:

Thanks. Could you please repeat that sequence with debug mode (-d switch) and attach a new DBG file? Maybe it will tell me more why BIOS PnP services are causing such issue.

Of course, here is the result.

Thanks. Looks like you'll need to disable the "Skip PnP BIOS Calls" option in Setup -> Configuration menu.

Mumak wrote on 2021-04-09, 08:20:

Thanks. Looks like you'll need to disable the "Skip PnP BIOS Calls" option in Setup -> Configuration menu.

This motherboard has a graphical "windows-style" AMI BIOS, and unfortunately it doesn't contain such option.

Diving deeper into the matter, I have found an unofficial M919 guide and another Vogons thread, which build an impression that the "original" M919 BIOS (1995) may not actually support PNP, and other applications may be also confused by this MB. Among the tools for M919, there is a driver PNPFIX.DRV described as the fix for issues with Creative PNP sound cards. After adding it into "CONFIG.SYS", HWINFO reports that PNP is not supported by the BIOS, and "Peripherals info" screen opens flawlessly.

Basing on the screenshots from another thread, newer M919 BIOSes (after 1996) already have PNP support, but since "stock" BIOS chip in this MB is non-flashable I won't be able to verify this in near future.

I'm not sure whether it's possible to implement proper logic with older BIOSes, but it would be nice if the tool could be improved not to freeze or fall over in such cases. M919 boards are somewhat collectible nowadays, due both infamous 'fake cache' and decent performance, so other people may also come across similar behavior.

In any case, thank for your responses, they helped me to identify the root cause.

crvs wrote on 2021-04-09, 17:27:

This motherboard has a graphical "windows-style" AMI BIOS, and unfortunately it doesn't contain such option. […]
Show full quote

Mumak wrote on 2021-04-09, 08:20:

Thanks. Looks like you'll need to disable the "Skip PnP BIOS Calls" option in Setup -> Configuration menu.

This motherboard has a graphical "windows-style" AMI BIOS, and unfortunately it doesn't contain such option.

Diving deeper into the matter, I have found an unofficial M919 guide and another Vogons thread, which build an impression that the "original" M919 BIOS (1995) may not actually support PNP, and other applications may be also confused by this MB. Among the tools for M919, there is a driver PNPFIX.DRV described as the fix for issues with Creative PNP sound cards. After adding it into "CONFIG.SYS", HWINFO reports that PNP is not supported by the BIOS, and "Peripherals info" screen opens flawlessly.

Basing on the screenshots from another thread, newer M919 BIOSes (after 1996) already have PNP support, but since "stock" BIOS chip in this MB is non-flashable I won't be able to verify this in near future.

I'm not sure whether it's possible to implement proper logic with older BIOSes, but it would be nice if the tool could be improved not to freeze or fall over in such cases. M919 boards are somewhat collectible nowadays, due both infamous 'fake cache' and decent performance, so other people may also come across similar behavior.

In any case, thank for your responses, they helped me to identify the root cause.

Sorry for not being clear, the option I mentioned above is in HWiNFO.
I'm not sure what exactly is wrong on that system, but it would require much more testing with multiple test builds to determine more precisely the issue and implement a fix for it. My assumption is that the BIOS might be some early attempt to support PnP services, which might be incomplete of faulty.
Maybe if you could send me the BIOS content I could have a look inside if I can see something. You can use the attached tool to capture the actual BIOS content, but please avoid loading the PNPFIX driver in this case so I can see the original BIOS structures.

UPDATE: I just had a look at the PNPFIX driver and what it actually does 😀 It disables BIOS write-protection (or shadow BIOS WP) and then clears the "PnP$" identifier in BIOS. That means PnP support in the BIOS is broken and best solution is to avoid it completely. I think there's no need for the BIOS image anymore, but if you could send me a full HWINFO log file + Debug File (you can run HWINFO -r -d) then I should be able to add a workaround for this case.

Mumak wrote on 2021-04-09, 17:40:

Maybe if you could send me the BIOS content I could have a look inside if I can see something. You can use the attached tool to capture the actual BIOS content, but please avoid loading the PNPFIX driver in this case so I can see the original BIOS structures.
GETBIOS.COM.zip

UPDATE: I just had a look at the PNPFIX driver and what it actually does 😀 It disables BIOS write-protection (or shadow BIOS WP) and then clears the "PnP$" identifier in BIOS. That means PnP support in the BIOS is broken and best solution is to avoid it completely. I think there's no need for the BIOS image anymore, but if you could send me a full HWINFO log file + Debug File (you can run HWINFO -r -d) then I should be able to add a workaround for this case.

See created files in the attached archive - please note that HWINFO got stuck on "Creating Report File / Plug-and-Play Devices", I had to reboot the machine. Just in case if BIOS image may still be useful to you, I have also included the dump created by GETBIOS, and BIOS backup taken by UNIFLASH.

crvs wrote on 2021-04-09, 21:59:

Mumak wrote on 2021-04-09, 17:40:

Maybe if you could send me the BIOS content I could have a look inside if I can see something. You can use the attached tool to capture the actual BIOS content, but please avoid loading the PNPFIX driver in this case so I can see the original BIOS structures.
GETBIOS.COM.zip

UPDATE: I just had a look at the PNPFIX driver and what it actually does 😀 It disables BIOS write-protection (or shadow BIOS WP) and then clears the "PnP$" identifier in BIOS. That means PnP support in the BIOS is broken and best solution is to avoid it completely. I think there's no need for the BIOS image anymore, but if you could send me a full HWINFO log file + Debug File (you can run HWINFO -r -d) then I should be able to add a workaround for this case.

See created files in the attached archive - please note that HWINFO got stuck on "Creating Report File / Plug-and-Play Devices", I had to reboot the machine. Just in case if BIOS image may still be useful to you, I have also included the dump created by GETBIOS, and BIOS backup taken by UNIFLASH.

Thanks. Can you please try the attached build if it doesn't crash now and also recognizes your mainboard?

Mumak wrote on 2021-04-09, 23:33:

Thanks. Can you please try the attached build if it doesn't crash now and also recognizes your mainboard?
HWINF613.ZIP

Thank you very much!

This build is not freezing anymore, however

• When I'm running it with "-r -d", it creates reports and immediately terminates - is this expected behavior? It's not mentioned in "HWINFO.DOC"
• When I'm launching it just with "-d" or without the parameters, both "BIOS information", "PNP information" and "Peripherals information" screens are opening flawlessly. However, PnP is displayed as supported (w/o any remarks), could it be possible to report identified 'inconistencies' of the BIOS (or other hardware)? I wouldn't have imagined that my rack may have a weird issue with PnP if HWINFO had being running smoothly from the beginning, like now. PS: Your CPU errata test is really helpful, BTW.

+Attached application logs for both options, just in case if you may need any other details

crvs wrote on 2021-04-11, 10:39:

Thank you very much! […]
Show full quote

Mumak wrote on 2021-04-09, 23:33:

Thanks. Can you please try the attached build if it doesn't crash now and also recognizes your mainboard?
HWINF613.ZIP

Thank you very much!

This build is not freezing anymore, however

• When I'm running it with "-r -d", it creates reports and immediately terminates - is this expected behavior? It's not mentioned in "HWINFO.DOC"
• When I'm launching it just with "-d" or without the parameters, both "BIOS information", "PNP information" and "Peripherals information" screens are opening flawlessly. However, PnP is displayed as supported (w/o any remarks), could it be possible to report identified 'inconistencies' of the BIOS (or other hardware)? I wouldn't have imagined that my rack may have a weird issue with PnP if HWINFO had being running smoothly from the beginning, like now. PS: Your CPU errata test is really helpful, BTW.

+Attached application logs for both options, just in case if you may need any other details

Thanks for the feedback.
Yes, running with -r is expected to automatically create the report file and terminate.
HWiNFO now doesn't use BIOS PnP services on your board to avoid the issue, but it's using its own direct low-level PnP serial isolation to enumerate PnP-capable boards. That seemed to work in the manual mode and detected your ESS PnP-capable card. Not sure why it didn't work in the automated mode tho..

Mumak wrote on 2021-04-11, 12:37:

Thanks for the feedback.
Yes, running with -r is expected to automatically create the report file and terminate.
HWiNFO now doesn't use BIOS PnP services on your board to avoid the issue, but it's using its own direct low-level PnP serial isolation to enumerate PnP-capable boards. That seemed to work in the manual mode and detected your ESS PnP-capable card. Not sure why it didn't work in the automated mode tho..

My pleasure, it's a great tool, glad to help making it even better. If I manage to update a BIOS on this MB, I'll test it again and post here a comment, intrigued if there will be a difference.

Great program! I'm enjoying using it in my 'new' 486DX2-66.

It doesnt seem to detect the 256k cache on my Soyo SY-25P2 motherboard. Report attached

trixster wrote on 2021-04-28, 10:19:

Great program! I'm enjoying using it in my 'new' 486DX2-66.

It doesnt seem to detect the 256k cache on my Soyo SY-25P2 motherboard. Report attached

Thanks for your report. Can you please attach the HWiNFO DBG file (run with -d switch) so I can see more details why it doesn't recognize the L2 cache?

Sure, thanks for the reply. Attached

trixster wrote on 2021-04-28, 12:42:

Sure, thanks for the reply. Attached

Thanks. I will try to fix this in the next version.

Somewhere between 6.0.2 and 6.1.0, HWINFO lost the ability to correctly recognize Celeron 266.
Works fine in 6.0.2, but in 6.1.0+ it's misrecognized as Pentium II:

Grzyb wrote on 2021-05-01, 20:27:

Somewhere between 6.0.2 and 6.1.0, HWINFO lost the ability to correctly recognize Celeron 266. Works fine in 6.0.2, but in 6.1.0 […]
Show full quote

Somewhere between 6.0.2 and 6.1.0, HWINFO lost the ability to correctly recognize Celeron 266.
Works fine in 6.0.2, but in 6.1.0+ it's misrecognized as Pentium II:
cel1.png
cel2.png

Can you please attach the HWiNFO Debug File (run HWINFO -d) so I can check the details?

Mumak wrote on 2021-05-01, 21:08:

Can you please attach the HWiNFO Debug File (run HWINFO -d) so I can check the details?

Here you are...

Grzyb wrote on 2021-05-02, 11:52:

Mumak wrote on 2021-05-01, 21:08:

Can you please attach the HWiNFO Debug File (run HWINFO -d) so I can check the details?

Here you are...

Thanks. Does this attached build fix the issue?

Mumak wrote on 2021-05-02, 14:13:

Thanks. Does this attached build fix the issue?

No, it's still Pentium II.

Can you please attach a new DBG file?

Mumak wrote on 2021-05-02, 15:02:

Can you please attach a new DBG file?

Attached...

Thanks, seems like it was identified different now. Can you please also post the first two screens as before? You can press F2 to save them into LOG file.