-
Notifications
You must be signed in to change notification settings - Fork 77
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
make more portable #55
Comments
You are also monkey-patching epoll_wait and stuff. Maybe there should be a "light" version that only provides malloc, realloc, calloc and free, along with a sized free and a few mesh_* functions. |
most of those are pulled in indirectly. let me see what I can do. very open to a light version. |
its not much shorter yet, but our full list of symbols is now:
|
most of the symbols we depend on are still included transitively through linking against static libstdc++. I'm not sure why -- I would have expected passing |
once emeryberger/Heap-Layers#26 is merged into Heap-Layers, I'll update our dep, and Mesh should build under musl (I've been testing in a local alpine docker container). |
We are pulling in a bunch of wide character support. Not of this should be necessary. |
note that this commit (emeryberger/Heap-Layers#26) was merged. |
a) I think we should build against musl at this point, and b) I've made a bit more progress in general. the symbols we now depend on are down to:
I'll note two things -- first, a bunch of these symbols are simply syscall wrappers ( We could consider switching to our own wrappers around Hope this helps! let me know if you see any smaller/easier answers here! |
one more note, |
https://github.com/daanx/mimalloc-bench is interested as well :) |
mesh uses many symbols from glibc, and I didn't manage to build it against musl either. I was interested in investigating using it as an allocator for zig, which does not use libc (and implements its own, as llvm does require some libc symbols).
The text was updated successfully, but these errors were encountered: