On Mon, 29 May 2000, David Balazic wrote:
> if ( You are saying : "Adding this assert() fixes the problem" )
> then {
> This means that you are triggering some compiler bug.
> At least it looks like that to me.
> Rationale : The added code doesn't really do anything, yet
> the behaviour is different.
> }
Yes. It really looks like that. But. I have used two completely different
compilers. Both of the can generate either m68k or powerpc code. Both
versions work fine on m68k but the ppc side causes problems. I don't think
both SAS/C and GCC have the same kind of compiler bugs. If you read my
second report, there was quite a lot of memory that has been trashed. I
don't know how adding assertions could "seem" to "fix" the problem but
obviously sometimes it does and it's not necessarily the problem in the
compiler itself. Of course everything is possible and because I can't have
verified these two compilers I can't say much.
I have heard that there are also other people having this kind of problems
with PPC version of lame. I wonder what compilers they have been using...
Kimmo
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
