On Wed, Feb 12, 2003 at 12:00:13PM -0500, Greg Troxel wrote: > > I wonder if emacs is doing some sort of asynchronous IO that isn't > handled correctly in the NetBSD kernel. >
I think it is. If you disable the use of SIGIO when you build emacs then you may find things more stable, Just #undef SIGIO in one of the headers (syssignal.h I think). This has worked for me for a long while. I did log this as a bug with the emacs people but they were at a loss as to what was going wrong. From the digging I did it appears that emacs gets into a loop where it is expecting to get a SIGIO but the signal is never delivered - if you kill -IO the emacs process then it will probably wake up and continue on. -- Brett Lymn
