Version: 1998-09-04 and before Platform: Macintosh, MacOS 7.x, 8.x Module: NSPR, machine dependent code The following code works fine, most of the times: (http://cvs-mirror.mozilla.org/webtools/lxr/source/nsprpub/pr/src/md/mac/macio.c #266) 266 if (op == READ_ASYNC) 267 (void) PBReadAsync(&pbAsync); 268 else 269 (void) PBWriteAsync(&pbAsync); 270 271 WaitOnThisThread(me, PR_INTERVAL_NO_TIMEOUT); Sometimes, however, it waits on this thread forever. Couldn't figger out why. Not yet. The following modifications work fine for me, but eliminate the context switch: 266 if (op == READ_ASYNC) 267 (void) PBReadSync(&pbAsync); /* now synchronous */ 268 else 269 (void) PBWriteSync(&pbAsync); /* now synchronous */ 270 271 /* WaitOnThisThread(me, PR_INTERVAL_NO_TIMEOUT); not needed */ 272 me->io_pending = PR_FALSE; /* nothing pending, io done */ Cheers, Peter Hanappe hanappe@ircam.fr

Setting all current Open/Normal to M4.

per leger, assigning QA contacts to all open bugs without QA contacts according to list at http://bugzilla.mozilla.org/describecomponents.cgi?product=Browser

Reassigning Mac NSPR bugs to gordon.

Changed target milestone to M6.

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

This was addressed by the change to not call WaitForThisThread() if the I/O call completes immediately (part of bug #3536).

sounds good to me -- marking verified.