Unable to set screen size

[keith]

Sorry++,

8.4b1008 still hangs displaying the same "Testing multiprocessor support..."

We looked into all changes that were made from v8.3 -> v8.4 and found a memory related issue that may have corrupted the memory used by the UEFI firmware. This may explain why certain benign UEFI functions like clearing the screen to hang, when it worked in the previous v8.3 builds. We have created a new thread to track the v8.4 builds.

https://www.passmark.com/forum/memte...-4-pre-release

Please give the new build a try at your convenience.

Does app not have a debug/trace log flag I can enable to add more stuff to log file?

The MemTest86.log already contains all debug output that can be written by the program, so we need to update the build if we want to increase the verbosity of the logs. Sorry for the constant back and forth, and thanks for all your time and effort to help us debug.

[sugob]

Sorry++,

8.4b1008 still hangs displaying the same "Testing multiprocessor support..."
[app really should display another progress message saying something like "Setting console/graphics mode..." so it would more accurately describe where it was in startup processing]

Log file ends
2020-04-29 18:36:21 - MP test passed. Setting default CPU mode to PARALLEL
2020-04-29 18:36:21 - Applying configurations
2020-04-29 18:36:21 - [CONFIG] This platform has 16 logical processors of which 16 are enabled.
2020-04-29 18:36:22 - Applying configurations complete
2020-04-29 18:36:22 - Console size = 80 x 25
2020-04-29 18:36:22 - Checking for graphics mode support
2020-04-29 18:36:22 - Graphics mode available
2020-04-29 18:36:22 - Enabling graphics mode


Does app not have a debug/trace log flag I can enable to add more stuff to log file?

[pictureon]

Thanks for that. We've removed the board from the blacklist.

Do you happen to still have the logs from the run?

Hi, you will find the log file in the attachment. I have aborted the test after a while as I was running out of time but you can see in the log file that it was running.

[keith]

Hi
Sorry, 8.4b1007 still hangs for me in same spot "Testing multiprocessor support..."

Thanks for the logs. It appears to indicate that it is hanging on a single line of code that clears the screen. It is a bit odd because this hasn't changed from v8.3.

However, we have removed the line to see if we can proceed further.

Can you give this build a try:
https://www.passmark.com/temp/memtes...-8.4b.1008.zip

[sugob]

Hi
Sorry, 8.4b1007 still hangs for me in same spot "Testing multiprocessor support..."

[keith]

It still works when I remove the board from the blacklist.

Thanks for that. We've removed the board from the blacklist.

Do you happen to still have the logs from the run?

[pictureon]

It still works when I remove the board from the blacklist.

[keith]

Hi Keith,

build 1006 worked for me now. In this version you have added my motherboard to the blacklist. So I can't say if the blacklisting fixed it or any changes in the code.
But nevermind, it's up and running now, thanks!

Good to hear, thanks for the feedback.

Can you try removing your board from the blacklist and see if it still hangs.

[pictureon]

Hi Keith,

build 1006 worked for me now. In this version you have added my motherboard to the blacklist. So I can't say if the blacklisting fixed it or any changes in the code.
But nevermind, it's up and running now, thanks!

[keith]

Thanks sugob for the quick turnaround.

We are a bit puzzled with what we are seeing in the logs, and what we have changed since v8.3. There doesn't seem to be any indication of any definitive cause to the problem.

There is a possibility the logs might not be up-to-date to when the hang occurred (due to the log entries being buffered and not written to disk).

If possible, can you try this new build which should output more log entries.
https://www.passmark.com/temp/memtes...-8.4b.1007.zip

[sugob]

Hi

Tried the new v8.4b006 version and it hung at same point as before [output to a 1920 x 1200 screen] but the log has more info at end than v8.4b1004
2020-04-22 11:15:36 - [CONFIG] This platform has 16 logical processors of which 16 are enabled.
2020-04-22 11:15:36 - [CONFIG] SelCPUNo = 0
2020-04-22 11:15:36 - [CONFIG] CurLang = 0
2020-04-22 11:15:36 - [CONFIG] ConsoleMode = 0 (MaxMode = 4) # This was last line in v8.4b1004 log
2020-04-22 11:15:36 - [CONFIG] BitFadeSecs = 300
2020-04-22 11:15:36 - [CONFIG] MaxErrCount = 10000
2020-04-22 11:15:37 - [CONFIG] TFTPStatusSecs = 60
2020-04-22 11:15:37 - [CONFIG] Setting remaining values
2020-04-22 11:15:37 - Applying configurations complete
2020-04-22 11:15:37 - Console size = 80 x 25

[keith]

Thanks for the logs.

As expected, it appears to hang when attempting to set the screen resolution via the UEFI firmware. We have made some changes to avoid changing the resolution. However at the end, we may have to blacklist the board.

Can you give the following build a try:

https://www.passmark.com/temp/memtes...-8.4b.1006.zip

[pictureon]

Thanks for the new build.
Unfortunately the same behavior as with 1004.
See the log file attached.

[keith]

Thanks for the log files.

We are still trying to isolate the issue.
If possible, please give the following build a try and post the resulting log file.

https://www.passmark.com/temp/memtes...-8.4b.1005.zip

[pictureon]

Thanks for sharing the new build Keith. I've tried it version 8.4.1004 today.
No black secreen any more, instead the computer reboots at this point