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
Issue while compiling for Apple M1 #631
Comments
pytorch/pytorch#63558 |
@pgwadapool thanks for checking. As @junjihashimoto mentions there's two requirements - upgrading the libtorch binary and GHC. From past experience, we've tended to be on the conservative side with GHC upgrades (probably lagging ~ 6 months post release) as the heavy use of type-level functions unearths bugs that haven't been caught elsewhere and it's painful if we're one of a few that encounters it. We'll get there with PT 1.10 and GHC 9.2, but it may be a few months to make sure the latter is stable. Happy to take PRs if it's something you want to explore. |
@austinvhuang any chance for GHC 9.2 support please? It's sufficiently mature now. |
GHC 9.2 is already supported. |
@junjihashimoto The build fails for me with
I assume |
You should install the c library of tokenizers_haskell. |
You can install it with makefile, apt or homebrew. |
Thanks @junjihashimoto! |
|
I tried Apple M2 with ghc928. Catching a exception of c++ does not work. |
Now, the master branch supports ghc964, so the above issue of exception is fixed. |
I am trying to compile this with ghc 8.10.7 and I get the below error with warnings of stdc++.dylib not found on Apple M1
libHSlibtorch-ffi-1.9.0.0-inplace-ghc8.10.7.dylib, 0x0005): symbol not found in flat namespace '__ZN2at10__lshift__ERKNS_6TensorERKN3c106ScalarE'
Looks like ghc 9.2.1 is not supported.
Any plans to support Apple M1?
The text was updated successfully, but these errors were encountered: