site_archiver@lists.apple.com Delivered-To: darwin-dev@lists.apple.com Please repot it with code to reproduce it. -- Terry I hope this is the right group, perhaps someone will advise me if not. When I switch the order of the binds, it is always the second one that hangs. -- Jay Reynolds Freeman --------------------- Jay_Reynolds_Freeman@mac.com http://web.mac.com/jay_reynolds_freeman (personal web site) _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/tlambert%40apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.appl... On Feb 14, 2010, at 2:59 AM, Jay Reynolds Freeman <jay_reynolds_freeman@mac.com

wrote: I have a situation in which I am doing two separate binds of completely

different Unix sockets ( socket( AF_UNIX, SOCK_STREAM, 0) ), very close together in time. The first one works fine, but the second one hangs; that is, it does not return an error, rather it never returns at all. But when I put a one-second sleep, or an OSMemoryBarrier, between the two binds, all is well. The fact that OSMemoryBarrier fixes the problem (and be it noted that I am running on an 8-core Mac Pro, so real parallelism is possible), makes me suspect that I have a reportable issue, and I will indeed kludge up a test case and file a bug, but I thought I would query the list first, to see if anyone has anything particular to suggest by way of test, analysis, or precedent. I have nosed around on the web a bit, but can't find anything particularly relevant. This email sent to tlambert@apple.com This email sent to site_archiver@lists.apple.com