Re: pkg/38674 (libarchive broken on IRIX)

["Stuart Shelton" <srcshelton%gmail.com@localhost>]

The following reply was made to PR pkg/38674; it has been noted by GNATS.

From: "Stuart Shelton" <srcshelton%gmail.com@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: joerg%netbsd.org@localhost, gnats-admin%netbsd.org@localhost, 
pkgsrc-bugs%netbsd.org@localhost, 
        tnn%NetBSD.org@localhost
Subject: Re: pkg/38674 (libarchive broken on IRIX)
Date: Tue, 3 Jun 2008 15:12:17 +0100

 > The problem might be the HAVE_VFORK ifdef (IRIX only has fork).
 > AC_FUNC_VFORK can be used to provide that.
 > Anyway, the local #includes need to be moved higher for that fallback
 > code segment to work.
 
 Actually, the manpage for fork() says:
 
 "
 NAME
      fork, vfork - create a new process
 
 C SYNOPSIS
      #include <sys/types.h>
      #include <unistd.h>
 
      pid_t fork (void);
 
      #if _XOPEN_SOURCE && _XOPEN_SOURCE_EXTENDED==1
      pid_t vfork (void);
      #endif
 
      Only when in the X/Open XPG4 Extended Unix name space are vfork and fork
      exactly the same in this system and any reference to fork in this
      document also applies to vfork.  It should also be noted that this vfork
      is NOT and DOES NOT have the same semantics as the original BSD vfork.
 "
 
 So it appears that *a* vfork() is available (with the correct symbols
 defined), but that it may not work as intended...
 
 All this seems to affect is whether external compression programs can
 be used - will this have any effect on the use of pkg_install?