Re: Porting gcc
- To: Joseph Gaffney <gaffney@xxxxxxxxxxxxxx>
- Subject: Re: Porting gcc
- From: Mark Walker <mwalker@xxxxxxxxxxxxxxxxxxxxx>
- Date: Fri, 23 Jan 1998 13:02:17 -0400
- Cc: MISC
- In-Reply-To: Your message of "Thu, 22 Jan 1998 18:13:33 MST." <3.0.1.32.19980122181333.007ac8e0@www.thethinker.com>
- Old-Date: Fri, 23 Jan 1998 10:03:53 -0700
- Posted-Date: Fri, 23 Jan 1998 10:03:53 -0700
- ReSent-Date: Sun, 25 Jan 1998 00:44:15 -0500 (EST)
- ReSent-From: Penio Penev <penev@xxxxxxx>
- ReSent-Message-ID: <Pine.SGI.3.96.980125004415.15007I@venezia.rockefeller.edu>
- ReSent-To: MISC
Joseph,
>>Are you planning to incorporate your work into the existing body of
>>cross-compilation (xgcc) facilities for gcc?
>
>I didn't know of a project to create a cross-compiler...
>
>Where is the information on this? It would be easier to help there then to
>do it alone =)
There's a copy of the CrossGCC FAQ at:
http://www.ssd.sscc.ru/misc/gcc/crossgcc-FAQ.html
An instance of an archive of the crossgcc maillist can be accessed at:
http://eli.elilabs.com/xgccarc/
I had some trouble accessing any of the links that refer to: eli.wariat.org
due to a name service error. It may be a temporary condition.
A number of years ago we had need of cross assembly/compilation capability
to replace an aging crossassembler we were using for a microprocessor course.
One of our student employees found a usenet news posting that contained a
patch for an additional crossgcc target. The new target was for building
embedded systems rather than applictions for an existing OS. This was just
what we needed as the other crossgcc targets wanted the OS runtime libraries
from the target system, and were unhappy that we didn't/couldn't provide them.
>From a quick glance, it appears Cygnus is quite active in the current crossgcc
arena. Also, I don't have a pointer, but I'm sure I saw info on setting up
a new target. It may have been in the crossgcc distribution, but if not
a little time spent on the web or in the gcc/xgcc/crossgcc related
maillists/newsgroups (dejanews?) would probably turn up plenty of info.
I suspect the task is significantly less work that you envisioned initially.
But, then again, I don't do compilers...
--
Mark Walker UNM/EECE, 7-3688, mwalker@eece.unm.edu
"An operating system without virtual memory, is an operating
system without virtue."
-- Ancient Inca Proverb
(As retold by Valentin Pepelea <valentin@netcom.com>)