Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

When (I think) going above 2GB for final installer, get "Internal compiler error #12345: error mmapping datablock to 32140454." #259

Open
pseudotensor opened this issue Aug 20, 2023 · 1 comment

Comments

@pseudotensor
Copy link

Love the project, using for h2oGPT: https://github.com/h2oai/h2ogpt

With torch being 4GB and some other stuff being another 1GB, my total build folder becomes order 5.5GB. The building of installler seems to nearly complete then fails like this. I know before adding another 600MB the final file was 1.9GB, so I assume going over made things fail.

I'm unable to delete any files, so can I increase the compression or otherwise make work? I have 64-bit system and bitness=64 is specified.

~~~ Running makensis ~~~
Processing config: C:\Program Files (x86)\NSIS\nsisconf.nsh
Processing script file: "build\nsis\installer.nsi" (ACP)

Internal compiler error #12345: error mmapping datablock to 32140454.

Note: you may have one or two (large) stale temporary file(s) left in your temporary directory (Generally this only happens on Windows 9x).
@takluyver
Copy link
Owner

It seems like this is a limitation of NSIS, the installer system that Pynsist builds on. From the documentation:

Installers can be as large as 2GB

This old forum thread and this Stackoverflow question have some suggestions of possible workarounds. Many of these probably won't work with Pynsist directly; you might want to generate the .nsi file with --no-makensis and then modify it before building it with NSIS directly.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants