Re: -fno-strength-reduce
Arvind Sankar (arvinds@mit.edu)
Fri, 22 Jan 1999 17:47:36 -0500
On Fri, Jan 22, 1999 at 07:06:37PM +0000, Philipp Rumpf wrote:
> On Fri, Jan 22, 1999 at 10:12:49AM -0500, Arvind Sankar wrote:
> > specs be? The defaults are bigger than what are currently specified
> > (currently they are 32 bit alignment for loops, jumps and functions).
> > The default is 64-bit on i486 and up, and if gas >= 2.8 is used (likely)
> > then it is either 64-bit or 128-bit if you are less than 64 bits from
> > a 128-bit boundary. Wouldn't 64-bit alignment be faster for Ppros?
>
> The reason for 128 bit boundaries are the 32 byte cache lines. It is much more
> likely to get split when one uses 64 bits. [while using 256 bit = 32 byte align-
> ment would be overkill probably]. I may be wrong though.
I probably wasn't very clear as to what I meant. I was referring to the fact that
64-bit alignment would be faster than the current 32-bit alignment specified
in the arch/i386/Makefile. If 128 is even faster, so be it.
-- arvind
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/