I was unsure about this one, partly because I don't understand it well
enough. I'll back out the change. In part, my thinking was that at
least if I stuck it in, the issue wouldn't be forgotten and someone who
does understand it might fix it properly.
Yes, I'm glad somebody did something about it, let's hope we can come
to some good solution.
Judging by the vfork(2) man
page on Linux, I'm assuming it is just being used because at some point
in the past the small performance gain was worth it.
Yes, but I wonder whether there was any system where this code wasn't
broken from the beginning. The intention of vfork seems to be quite
the contrary to how it is used in the code.
Unfortunately I have too little experience on non-linux systems to
know if a general change from vfork to fork is reasonable.
Harald
_______________________________________________
Nmh-workers mailing list
Nmh-workers(_at_)nongnu(_dot_)org
http://lists.nongnu.org/mailman/listinfo/nmh-workers