6016457011
this implementation is rather heavy-weight, but it's the first solution i've found that's actually correct. all waiters actually wait twice at the barrier so that they can synchronize exit, and they hold a "vm lock" that prevents changes to virtual memory mappings (and blocks pthread_barrier_destroy) until all waiters are finished inspecting the barrier. thus, it is safe for any thread to destroy and/or unmap the barrier's memory as soon as pthread_barrier_wait returns, without further synchronization. |
||
|---|---|---|
| .. | ||
| madvise.c | ||
| mlock.c | ||
| mlockall.c | ||
| mmap.c | ||
| mprotect.c | ||
| mremap.c | ||
| msync.c | ||
| munlock.c | ||
| munlockall.c | ||
| munmap.c | ||
| posix_madvise.c | ||
| shm_open.c | ||
| shm_unlink.c | ||