about summary refs log tree commit diff
path: root/src/linux
diff options
context:
space:
mode:
authorRich Felker <dalias@aerifal.cx>2011-09-27 13:50:29 -0400
committerRich Felker <dalias@aerifal.cx>2011-09-27 13:50:29 -0400
commit60164570111873175111cf8a5b973375e492eee9 (patch)
tree1557931c4e0bacd2be217965be08031d5e3a4e51 /src/linux
parent3f39c9b3130cd6c142d358159879b799370a6663 (diff)
downloadmusl-60164570111873175111cf8a5b973375e492eee9.tar.gz
musl-60164570111873175111cf8a5b973375e492eee9.tar.xz
musl-60164570111873175111cf8a5b973375e492eee9.zip
process-shared barrier support, based on discussion with bdonlan
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.
Diffstat (limited to 'src/linux')
0 files changed, 0 insertions, 0 deletions