mailing list archives
Re: SIGSEGV signal on Darwin OS X (Intel) when using version matching (nmap 4.01)
From: "Kurt Grutzmacher" <grutz () jingojango net>
Date: Fri, 10 Mar 2006 11:12:50 -0800
I've been able to track this down to the Mac's GCC compiler and the
Perform optimizations where we check to see if a jump branches to
location where another comparison subsumed by the first is found.
If so, the first branch is redirected to either the destination
the second branch or a point immediately following it, depending
whether the condition is known to be true or false.
Enabled at levels -O2, -O3, -Os, -Oz (APPLE ONLY).
Process of elimination finally gave it away. I set CFLAGS to -O1 and
manually set each -f that -O2 enables and plucked my way through them until
I found the flag(s) that caused nmap to SIGSEV. No matter what
-fthread-jumps always caused a SIGSEV.
Fixing this is way beyond my knowledge. Seems like it's a problem with
Using built-in specs.
-enable-werror --prefix=/usr --mandir=/share/man
--with-arch=pentium-m --with-tune=prescott --program-prefix=
Thread model: posix
gcc version 4.0.1 (Apple Computer, Inc. build 5250)
Sent through the nmap-dev mailing list