Definitions for i686-unknown-linux-musl, arm and asmjs musl #122
Conversation
|
Looking at the build failures now. |
brson
force-pushed
the
musl
branch
3 times, most recently
from
a2ccd82 to
6d132c4
Compare
|
Remaining failure looks bogus. |
| pub const __SIZEOF_PTHREAD_RWLOCK_T: usize = 56; | ||
| #[cfg(any(target_arch = "x86", | ||
| target_arch = "arm", | ||
| target_arch = "asmjs"))] |
There was a problem hiding this comment.
Could these #[cfg] blocks be avoided by splitting this into a musl/$arch.rs-style configuration like with Linux?
|
Yeah I think build failures are unrelated, I'm prepping a PR that should bring them all in line |
|
I've moved some of the definitions around as requested. |
| __reserved: [::c_long; 3], | ||
| } | ||
| // This definition is slightly different on aarch64 | ||
| #[cfg(not(target_arch = "aarch64"))] |
There was a problem hiding this comment.
Do we actually have aarch64 support?
Stylistically I also tend to prefer copying things around instead of using #[cfg] and/or #[cfg(not)] (even though it's a little wordier), but doesn't matter too much here, it can just be cleaned up later if it becomes a problem.
There was a problem hiding this comment.
No, but I noticed in passing these definitions are weird and just added these defensively. I can remove them.
|
Updated for most recent reviews. |
| @@ -108,7 +111,8 @@ s! { | |||
| pub dli_saddr: *mut ::c_void, | |||
| } | |||
|
|
|||
| #[cfg_attr(any(target_arch = "x86", target_arch = "x86_64"), | |||
| #[cfg_attr(any(all(target_arch = "x86", not(target_env = "musl")), | |||
| target_arch = "x86_64"), | |||
There was a problem hiding this comment.
This line should be lined up with the a in all, right?
|
I'm fine merging whenever (just minor nits), but CI doesn't like two of the constants for MUSL |
I've tested x86 against C locally, but not arm or asmjs. I added the arm definitions because asmjs's C is derived from arms. Mysteriously, my locally-built musl does not contain a definition for _SC_2_C_VERSION, so I just removed it.
Definitions for i686-unknown-linux-musl, arm and asmjs musl
|
📦 |
Here's another go at adding emscripten support. This needs to wait again on new [libc definitions](rust-lang/libc#122) landing. To get the libc definitions right I had to add support for i686-unknown-linux-musl, which are very similar to emscripten's, which are derived from arm/musl. This branch additionally removes the makefile dependency on the `EMSCRIPTEN` environment variable by not building the unused compiler-rt. Again, this is not sufficient for actually compiling to asmjs since it needs additional LLVM patches. r? @alexcrichton
* added _mm_cvtps_pd * added _mm_set_sd * added _mm_set1_pd * added _mm_set_pd1 * added _mm_set_pd * added _mm_setr_pd * added _mm_setzero_pd
I've tested x86 against C locally, but not arm or asmjs.
I added the arm definitions because asmjs's C is derived from arms.
Mysteriously, my locally-built musl does not contain a
definition for _SC_2_C_VERSION, so I just removed it.
Some of these conditions, particularly in notbsd, could be pushed further down the tree.