Re: unalligned sparc memory access problem (memcpy vs bcopy)
Bernd Eckenfels <lists@lina.inka.de> writes:
> just to keep you informed, I have a test program and opened a bug against
> gcc since i believe it is an error with optimizer, It happens for memcpy if
> the source pointer is some "&p->m" construction, not "p+i", even if both
> point to the same (unalligned) address. It never happens with bcopy.
I looked at #136659, but can't seem to find your program. Would you
mind posting it?
Last week I encountered vague problems with g++ -O2 on powerpc (2.95.4
as well as 3.0.4), but haven't been able to create a small test
progam. This could well be it, I've reported strange memcpy problems
with g++ for powerpc before (which have been fixed).
Jan.
--
Jan Nieuwenhuizen <janneke@gnu.org> | GNU LilyPond - The music typesetter
http://www.xs4all.nl/~jantien | http://www.lilypond.org
Reply to: