Fixes #31229 - fixes up BSD type mismatch errors and updates libc dependency. DEAD, see PR #31263 #31230
Conversation
|
r? @brson (rust_highfive has picked a reviewer for you, use r? to override) |
| @@ -224,7 +224,7 @@ impl DirEntry { | |||
| fn name_bytes(&self) -> &[u8] { | |||
| unsafe { | |||
| ::slice::from_raw_parts(self.entry.d_name.as_ptr() as *const u8, | |||
| self.entry.d_namelen as usize) | |||
| self.entry.d_namlen as usize) | |||
There was a problem hiding this comment.
Can you unify this with the block above? after this change they're all the same
There was a problem hiding this comment.
unified now. thanks, i missed that.
|
@dhuseby could you merge back openbsd definition of There is a small difference in `HW_NCPU' meaning, but I think it shouldn't be a problem (under OpenBSD it is "number of cpus being used" whereas for NetBSD, FreeBSD and DragonFly it is "number of cups"). |
|
@semarie on OpenBSD, is the sysconf(_SC_NPROCESSORS_ONLN) the same as the sysctl(HW_NCPUS)? Technically, sysconf is the POSIX way to get the number of available CPUs whereas the sysctl function isn't. I'm not against switching back, but I would like to better understand why. I think switching to sysconf(_SC_NPROCESSORS_ONLN) is more accurate and more portable. |
|
@dhuseby yes |
|
Could you also tweak the title of the PR and the commit message to expand a little more what's going on? Helps out those who skim the git logs! |
dhuseby
changed the title
|
wtf? i didn't mean to close it. the latest commit isn't being picked up. i'll start another pr. |
dhuseby
changed the title
Something went haywire with github last night and the old PR #31230 got closed somehow. This new PR is to replace the old one. This incorporates all of the feedback from the other PR. @alexcrichton I incorporated the suggestion from @semarie and the result is cleaner and clearer. I think this is ready to go.
fixes the type mismatches breaking the build.