Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.
Sign upSomewhat working i386 port #115
Conversation
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
andrewrk
Feb 9, 2016
Member
Thanks for the contribution! I do have a few notes which may be of interest in doc/targets.md.
I went ahead and fixed the sign compares and other GCC 5 incompatibilities. what was -Wno-error=return-type for?
I'll start working on cross-compiling and conditional compilation now, and we'll see if we can get this code integrated.
|
Thanks for the contribution! I do have a few notes which may be of interest in doc/targets.md. I went ahead and fixed the sign compares and other GCC 5 incompatibilities. what was I'll start working on cross-compiling and conditional compilation now, and we'll see if we can get this code integrated. |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
andrewrk
Feb 11, 2016
Member
I just pushed 2bf6c28 which adds the ability to cross compile. I was able to build hello_libc.zig both statically and dynamically for windows and run it on a windows box.
Next up, expose the target information via @compile_var and then I think we'll be able to integrate your code.
|
I just pushed 2bf6c28 which adds the ability to cross compile. I was able to build hello_libc.zig both statically and dynamically for windows and run it on a windows box. Next up, expose the target information via |
andrewrk
closed this
in
592210a
Feb 12, 2016
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
andrewrk
Feb 12, 2016
Member
@sanxiyn I didn't test on a i386 machine yet but I will tomorrow. At any rate the x86_64 tests all pass with this conditional compilation code so it's at least a step toward i386 support. Cheers!
|
@sanxiyn I didn't test on a i386 machine yet but I will tomorrow. At any rate the x86_64 tests all pass with this conditional compilation code so it's at least a step toward i386 support. Cheers! |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
sanxiyn
commented
Feb 12, 2016
|
Thanks. |
sanxiyn
deleted the
sanxiyn:i386
branch
Feb 12, 2016
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
andrewrk
Feb 12, 2016
Member
It's not quite working yet due to undefined references which would be resolved by compiler-rt/libgcc.
I'm working on a solution that avoids the cross compiling nightmare.
|
It's not quite working yet due to undefined references which would be resolved by compiler-rt/libgcc. I'm working on a solution that avoids the cross compiling nightmare. |
This comment has been minimized.
Show comment
Hide comment
This comment has been minimized.
andrewrk
Feb 13, 2016
Member
Alright, I just tested i386 on ubuntu and it now works out of the box and passes all the tests.
|
Alright, I just tested i386 on ubuntu and it now works out of the box and passes all the tests. |
sanxiyn commentedFeb 9, 2016
These tests fail: enum type sizeof (alignment difference?), 5 more tests which I haven't investigated, and self hosted tests. Other tests pass.
Obviously this can't be merged, but if you have any idea how to handle architecture ports, I am willing to revise this towards that.