Chris Saari believes that NSPR needs to call WaitNextEvent() instead of Delay(). Additional code may be necessary to prevent certain events from getting lost, such as AppleEvents. Also, this might not be desired for all threads.

Reassigned the bug to gordon.

You won't loose events if you give WaitNextEvent a null event mask. In other words, you're not really retrieving the next event, you're yeilding time to the system which Delay() doesn't do. Delay() is evil. It is a spin loop on a non-preemptive system. Nothing can happen during a Delay() call.

Changed target milestone to M6.

NSPR now has its own Bugzilla product. Moving this bug to the NSPR product.

Fix is checked in. Note to QA - you can't really test this other than looking at the source for / mozilla/nsprpub/ps/src/md/mac/macio.c.

good enough for me. marking verified (code level fix).