VIRUS_VFORK -- define it!

Ted Nolan ted at usceast.UUCP
Sat Aug 4 16:17:16 AEST 1984

<all v's are forked>

In configuring a kernal for 2.9, the guidelines tell you that defining
VIRUS_VFORK may not be a good idea on systems that swap heavily, so
we decided not to define it.  

It turns out however, that csh uses a vfork and that csh is somehow tied
up in the shell escapes of a whole lot of programs (even those that give 
you a shell according to your SHELL environment var).  This has the
unpleasant consequence of rendering shell escapes from any of the
newer programs impossible.  Even worse, if you use csh, you will be
unable to do any non builtin command.  One fix would be to recompile
csh with a fork instead of a vfork, but given that the documentation
states that using vforks changes the way processes look in memory and
given the way the shells are known to play with memory themselves, we
decided the safest thing to do was go ahead and define VIRUS_VFORK.
You will have to do one or the other.

Ted Nolan                               ...decvax!mcnc!ncsu!ncrcae!usceast!ted
6536 Brookside Circle                   ...akgua!usceast!ted
Columbia, SC 29206
      ("We pray for one last landing on the globe that gave us birth..")

More information about the Comp.bugs.2bsd mailing list