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

Release 5.9.6 to fix segfaults #337

Open
trufae opened this issue Dec 11, 2024 · 11 comments
Open

Release 5.9.6 to fix segfaults #337

trufae opened this issue Dec 11, 2024 · 11 comments

Comments

@trufae
Copy link
Contributor

trufae commented Dec 11, 2024

ref flathub/org.radare.iaito#98 (comment)

@wargio
Copy link
Owner

wargio commented Dec 11, 2024

Are you asking to just make the tag? because the CI is not happy with the last changes.

@lexiwitch
Copy link

lexiwitch commented Dec 11, 2024

Looking at the commit list it seems like CI is happy now for some reason ? (The little green checkmark is there at least)

That being said, what I came here to say is if the CI isn't happy for some reason with the most recent commit, 4944f07 would be a viable candidate for tagging that fixes the crash but doesn't make the CI hiccup.

Also, I honestly didn't expect the response to be so fast, so thank you for this!

@wargio
Copy link
Owner

wargio commented Dec 11, 2024

seems that commit was not liked by macos.
For some reasons the CI was stuck :/

@lexiwitch
Copy link

seems that commit was not liked by macos. For some reasons the CI was stuck :/

Do you mean windows? I just checked again to see if i could maybe help out and the same commit had both a failing and passing build in CI for windows (I saw a checkmark earlier which led me to believe it was fixed but now it's showing a cross to say it's failing — not sure what Github is doing there...). The macos builds appear to have been commented out in CI since last year though.

@wargio
Copy link
Owner

wargio commented Dec 11, 2024

What i'm referring to is #335

@wargio
Copy link
Owner

wargio commented Dec 11, 2024

i'm ok to tag it, but windows seems to be broken due VS being updated.

@trufae
Copy link
Contributor Author

trufae commented Dec 11, 2024

just tested master on macOS, it's segfaulting again

@wargio
Copy link
Owner

wargio commented Dec 12, 2024

since r2 integrates qjs, is there a way to link against it?

@trufae
Copy link
Contributor Author

trufae commented Dec 12, 2024

qjs in r2 shouldnt be exposing the apis either, so this is partly a bug in r2 too. and actually my rewrite to es6 of r2dec was just an r2js plugin, so just shipping a .js file that implements a core plugin isntead of a core plugin in C that loads the same runtime. this could be another approach, and i was using r2frida-compile to generate the single esm blob.

i can try to fix this thing in r2 i have some ideas:

  • dlopen the plugin by not allowing to resolve the symbols outside
  • fix the build of qjs to not expose those symbols to the public
  • Add a linker script to rename those symbols or strip them down after compiling

@trufae
Copy link
Contributor Author

trufae commented Jan 14, 2025

can you try again with r2 from git?

@wargio
Copy link
Owner

wargio commented Jan 15, 2025

I will retry this weekend

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

3 participants