| |
On Tue, 11 Jul 2000, Dave Baum wrote:
> I have had better luck with egcs-1.1.2 rather than gcc 2.95.2. I've
> heard that there's a problem with the way gcc 2.95.2 generates code for
> memcpy. I've also heard that the fix is to use an assembly version of
> memcpy instead of letting the compiler generate it. I just stuck with
> egcs myself and avoided the entire issue.
>
> Both compilers will fail while trying to build the libraries. The
> problem stems from the H8 target support assuming signed chars, and gcc
> hosted on ppc defaulting to unsigned chars. My fix was to add
> -fsigned-char to CFLAGS in the Makefiles for binutils and gcc. Here's
> my previous post in the matter:
>
> http://www.lugnet.com/robotics/rcx/legos/?n=1111
Ah, excellent. I saw this post when you posted it, but didn't think about
it much at the time. I'll give this a shot and report back. Thanks very
much.
--
"Do not go gentle into that good night.
Rage, rage against the dying of the light."
- Dylan Thomas
Mike Ash - <http://www.mikeash.com/>, <mailto:mail@mikeash.com>
|
|
Message is in Reply To:
 | | Re: LegOS on LinuxPPC
|
| In article (...) I have had better luck with egcs-1.1.2 rather than gcc 2.95.2. I've heard that there's a problem with the way gcc 2.95.2 generates code for memcpy. I've also heard that the fix is to use an assembly version of memcpy instead of (...) (24 years ago, 11-Jul-00, to lugnet.robotics.rcx.legos)
|
4 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
