Reaching Lyudmil Antonov

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

cma6
Posts: 219
Joined: Thu May 29, 2014 5:58 pm

_Imp_virtualAlloca failed"

Post by cma6 »

Lyudmil: I posted this issue yesterday on github but cannot find it today:

Today, for the first time, asmFish_2017-0122_bmi2.exe and ...popcnt.exe had problems when used in Aquarium 2016 over a LAN (or directly via infinite analysis): using RTHomeserver over the LAN on the slave systems to run Aquarium's IDeA, one got the following error with both executables on 4 slave systems:
"_Imp_virtualAlloca failed"
so that IDeA did not run on slave systems.

Do you plan to fix this error on the next commit?
User avatar
lantonov
Posts: 216
Joined: Sun Apr 13, 2014 5:19 pm

Re: _Imp_virtualAlloca failed"

Post by lantonov »

cma6 wrote:Lyudmil: I posted this issue yesterday on github but cannot find it today:

Today, for the first time, asmFish_2017-0122_bmi2.exe and ...popcnt.exe had problems when used in Aquarium 2016 over a LAN (or directly via infinite analysis): using RTHomeserver over the LAN on the slave systems to run Aquarium's IDeA, one got the following error with both executables on 4 slave systems:
"_Imp_virtualAlloca failed"
so that IDeA did not run on slave systems.

Do you plan to fix this error on the next commit?
Art, exactly this problem was fixed with the latest patch, I wrote above. It arose with a guy with Fritz GUI, Mohammed Li made a fix patch, and I applied it, all within 20 minutes. I made a second compile of 2017-01-22 on top of the first one late yesterday evening. The problem was that a call to virtualAlloca sometimes failed and a safety proc is now introduced to bypass this. This problem was specific to Windows OS. It was reported also by a second user from Germany who reported it solved with the second compile.
If you try again the 2017-01-22 compile, chances are that it will work now. The guy who reported it first wrote today that "She runs like a charm. Everything's sorted now!"

Please give feedback about you.
cma6
Posts: 219
Joined: Thu May 29, 2014 5:58 pm

_Imp_virtualAlloca failed"

Post by cma6 »

Lyudmil:
Thank you for the prompt response. I should have added that I use Windows 7 (and Windows 10) on all my 5 systems. I have reverted back to the 1/14/2017 commit, which works great.
But will again try the new commit fairly soon.
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: _Imp_virtualAlloca failed"

Post by reflectionofpower »

This is what I am getting for an error.

Image
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
User avatar
lantonov
Posts: 216
Joined: Sun Apr 13, 2014 5:19 pm

Re: _Imp_virtualAlloca failed"

Post by lantonov »

It looks like either your UCI engines settings file *.ini or (or *.eng in Fritz as far as I remember) is wrong / non-existent or that you are using Winboard protocol instead of UCI.
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: _Imp_virtualAlloca failed"

Post by reflectionofpower »

lantonov wrote:It looks like either your UCI engines settings file *.ini or (or *.eng in Fritz as far as I remember) is wrong / non-existent or that you are using Winboard protocol instead of UCI.
I don't use Winboard. UCI engines settings are in the C:\Users\reflectionofpower\AppData\Roaming\ChessBase\Engines.UCI\ directory as UCI extension which is a text file that can be altered and is only made when you save a certain configuration within Fritz GUI.

I can not even make one as it will not go through the "Create UCI Engine" process and then I get the error.

Fritz does not use ENG anymore. It is embedded within the program's GUI. (ChessProgram14.exe)
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
User avatar
lantonov
Posts: 216
Joined: Sun Apr 13, 2014 5:19 pm

Re: _Imp_virtualAlloca failed"

Post by lantonov »

Strange indeed. Let's start at the beginning, without any GUI. Click twice on the asmFish executable and see if it starts immediately. Some Windows systems have defenses which require user confirmation before starting any new executable. What is the size of the downloaded file? Should be 112,128 bytes (114,688 bytes on disk).
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: _Imp_virtualAlloca failed"

Post by reflectionofpower »

lantonov wrote:Strange indeed. Let's start at the beginning, without any GUI. Click twice on the asmFish executable and see if it starts immediately. Some Windows systems have defenses which require user confirmation before starting any new executable. What is the size of the downloaded file? Should be 112,128 bytes (114,688 bytes on disk).
36,864 bytes ??? So maybe that is the problem. O, wait. My mouse is not user-friendly, probably the person operating it.

I was "saving link as" instead of clicking the file and then the download comes up. I thought it strange that the EXE was 36KB. I thought, wow, really tight programming :lol:

No problems now. Thanks for the response.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
User avatar
lantonov
Posts: 216
Joined: Sun Apr 13, 2014 5:19 pm

Re: _Imp_virtualAlloca failed"

Post by lantonov »

You are welcome. Glad we found the "bug" :)
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: _Imp_virtualAlloca failed"

Post by reflectionofpower »

lantonov wrote:You are welcome. Glad we found the "bug" :)
Yes, the "LB virus" - Lonnie Bug.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie