glibc 2.42 #24640
glibc 2.42 #24640
Conversation
| extern int __pthread_mutexattr_destroy (pthread_mutexattr_t *__attr); | ||
| libc_hidden_proto (__pthread_mutexattr_destroy) | ||
|
|
||
| extern int __pthread_mutexattr_settype (pthread_mutexattr_t *__attr, |
There was a problem hiding this comment.
Looks like __pthread_mutexattr_settype went away, so this might need an Zig ifdef guard for older versions? I'm not confident about how this symbol would actually get referenced though, so maybe not? If you do bring it back, its also not clear to me if the new libc_hidden_proto should apply, or if the block of weak_extern/pragma below would need to be applied to this symbol?
|
|
||
| #ifdef __USE_GNU | ||
| /* Return the Linux TID for THREAD_ID. Returns -1 on failure. */ | ||
| extern pid_t pthread_gettid_np (pthread_t __thread_id); |
There was a problem hiding this comment.
I believe this symbol is new for glibc 2.42, so Zig should probably have the #ifdef guard around it:
#if (__GLIBC__ == 2 && __GLIBC_MINOR__ >= 42) || __GLIBC__ > 2
| #endif | ||
|
|
||
| #if __GLIBC_USE (ISOC2Y) | ||
| extern unsigned int uabs (int __x) __THROW __attribute__ ((__const__)) __wur; |
There was a problem hiding this comment.
This block of uabs/ulabs/'ullabs' prototypes should probably also get a Zig header, as I believethey were introduced in v2.42. Ah, but they're behind IOSC2Y` (which I understand to be "next version" of ISO C), so maybe Zig can wait until they get promoted out of that?
We usually add these version checks on an as-needed basis when people need them, not proactively. PRs to do so are definitely welcome as follow-up work, but I'll go ahead and merge this one in any case. The long-term solution to this problem is going to be something like universal-headers (currently vaporware). |
2 participants
Closes #24579.