On Mon, 22 June 2009 23:20:39 +0100, David Woodhouse wrote: > On Mon, 2009-06-22 at 23:41 +0200, Jörn Engel wrote: > > Four loops doing the same increment with different data types: long, > > u64, we32 (wrong-endian) and we64. Compile with _no_ optimizations. > > That's a bit of a poor test then. Especially on architectures with a > load-and-swap instruction where it really shouldn't be any slower at > all. > > (Although since GCC doesn't have an __attribute__((littleendian)) I'm > not entirely sure how to entice it into _using_ said instruction for the > purpose of the test... I think the kernel does manage somehow though, if > you get the sources _just_ right.) Feel free to improve the test. It is admittedly crap and designed to support Chris' argument. But seeing that it still fails to do so and Arnd has already shown one improvement that weakened Chris' argument, I guess we can all agree that further improvments won't change the conclusion, can we? ;) Jörn -- It's just what we asked for, but not what we want! -- anonymous -- To unsubscribe from this list: send the line "unsubscribe linux-embedded" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
