Please do not reply directly to this email. All additional comments should be made in the comments box of this bug.
https://bugzilla.redhat.com/show_bug.cgi?id=357941
John Ellson john.ellson@comcast.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Flag| |needinfo?(tim@niemueller.de | |)
--- Comment #8 from John Ellson john.ellson@comcast.net 2008-10-02 19:48:17 EDT --- I've forgotten what my conclusion was on this, but it looks as though I made one change. limits.h is now included unconditionally by arith.h before testing #ifndef INT_MAX, so it looks to me like this should work if limit.h is included again either before or after arith.h.
If limits.h doesn't exist on SGI's then i don't understand why I haven't heard complaints about this.
Could you test with rpms from www.graphviz.org to see if this problem still exists?