On Sat, 26 Nov 2011, Richard Henderson wrote: > The m68k-linux failure for the various omp atomic tests > is due to the fact that BIGGEST_ALIGNMENT is 16 bits on > that platform. I think it's pretty reasonable to assume > that if something is aligned to BIGGEST_ALIGNEMENT, then > it can be considered "aligned".
BIGGEST_ALIGNMENT means aligned enough for normal access, but not necessarily for atomic access. This particular fix wouldn't do it for CRIS, for example, were BIGGEST_ALIGNMENT is 8 (bits), but an atomic access (inside the ll/sc sequence) such as for a futex requires not straddling a page boundary, i.e. "natural" alignment. Not that OMP support is imminent or critical for cris-linux or anything, but can we have a new macro? brgds, H-P
