home *** CD-ROM | disk | FTP | other *** search
- x-gateway: rodan.UU.NET from help-lucid-emacs to alt.lucid-emacs.help; Thu, 10 Sep 1992 23:49:36 EDT
- Date: Thu, 10 Sep 1992 20:47:46 PDT
- Message-ID: <9209110347.AA02711@thalidomide.lucid>
- X-Windows: Form follows malfunction.
- From: jwz@lucid.com (Jamie Zawinski)
- Sender: jwz%thalidomide@lucid.com
- Subject: Re: 19.3 installation on decstation/ultrix4.2 with native cc
- References: <9209110328.AA00345@ntu.ac.sg>
- Newsgroups: alt.lucid-emacs.help
- Path: sparky!uunet!wendy-fate.uu.net!help-lucid-emacs
- Lines: 24
-
- In message <9209110328.AA00345@ntu.ac.sg> Frank Siebenlist wrote:
- >
- > The compiler is not very descriptive, but the offending line reads:
- >
- > char* real_name = (char*)alloca (length + 2);
- >
- > Not sure what is wrong with that (?),
-
- How about running that file through -E to see what cpp is really turning
- that line in to.
-
- > The gmalloc.c which came with 19.2 had no problem but seems "very"
- > different.
- > 19.3 seems to work with the 19.2 version of gmalloc.c though,
- > but there must be a reason for having a new version (?),
- > so I decided not to use gmalloc at all but the system's one.
-
- GCC2 didn't like the version of gmalloc that we were using in 19.2, so I
- grabbed a new version of gmalloc from the FSF v19 sources. I don't know how
- recent it is. I don't understand why it's so unportable. (I understand why
- it's hard, but I would have expected it to be more portable than it is by
- now.) I humbly await enlightenment...
-
- -- Jamie
-