From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from euclid.skiles.gatech.edu (list@euclid.skiles.gatech.edu [130.207.146.50]) by melb.werple.net.au (8.7.5/8.7.3) with ESMTP id IAA06551 for ; Wed, 8 May 1996 08:46:44 +1000 (EST) Received: (from list@localhost) by euclid.skiles.gatech.edu (8.7.3/8.7.3) id SAA24628; Tue, 7 May 1996 18:30:47 -0400 (EDT) Resent-Date: Tue, 7 May 1996 18:30:47 -0400 (EDT) Message-Id: <199605072229.SAA20258@ts2.galt.com.> X-Mailer: exmh version 1.6.2 7/18/95 To: zsh-workers@math.gatech.edu X-url: http://www.galt.com/~erik/ Subject: zsh-2.5.03-2.6b17 under Solaris 2.5 trouble with mult() Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 07 May 1996 18:29:32 -0400 From: Erik Ogan Resent-Message-ID: <"VU-YO2.0.k06.MyyZn"@euclid> Resent-From: zsh-workers@math.gatech.edu X-Mailing-List: archive/latest/1003 X-Loop: zsh-workers@math.gatech.edu Precedence: list Resent-Sender: zsh-workers-request@math.gatech.edu I just sub'ed yesterday, but I've poked around the archives and I haven't found anything that seems even remotely related to this problem. The following occurs on all of our Solaris 2.5 boxes for zsh versions 2.5.03 - 2.6b17 (ok, I didn't try EVERY beta...:) but does NOT occur under SunOS 4.1.x. Everything compiles fine, but in the when it (gcc-2.7.2) tries to link: gcc -o zsh builtin.o compat.o cond.o exec.o glob.o hashtable.o hist.o init.o input.o jobs.o lex.o linklist.o loop.o math.o mem.o params.o parse.o signals.o subst.o text.o utils.o watch.o zle_bindings.o zle_hist.o zle_main.o zle_misc.o zle_move.o zle_refresh.o zle_tricky.o zle_utils.o zle_vi.o zle_word.o -lnsl -ltermcap generates the following warnings: Warning: size of symbol `mult' changed from 4 to 284 in /lib/libmp.so.1 Warning: type of symbol `mult' changed from 1 to 2 in /lib/libmp.so.1 And the program will often dump core due to an illegal instruction when searching command hash tables in a normal interactive incarnation (works fine in a script, or in gdb (Yes, I know smart bugs can smell programmer fear a mile a way)). In light of that, the only other useful information I've found is by doing a back trace on the core file in gdb. Doing so confirms that the illegal instruction in question occurs in the mult() function, called from the iscom() function (exec.c:388) Being relatively new (read: forcibly dragged kicking and screaming) to Solaris, I'm having some trouble with this, but under SunOS, (I would assume its the same) this behavior would imply running out of memory, or an illegal mult() operation. The fact that I seem to be the only Solaris 2.5 user who's had this problem disturbs me, and makes me look cross-eyed at our installation of gcc. Any ideas would be greatly appreciated, by me as well as my co-workers, most of whom are getting tired of listening to me swear up and down at tcsh. -------- Erik R. Ogan Webmaster/Systems Editor GALT Technologies, Inc.