Re: New CRC algorithm: Slicing by 8

Jeremy Drake <pgsql(at)jdrake(dot)com> writes:
> On Mon, 23 Oct 2006, Tom Lane wrote:
>> That's not a good workaround, because making mycrc expensive to access
>> means your inner loop timing isn't credible at all. Instead try making the
>> buffer array nonlocal --- malloc it, perhaps.

> That did not make any difference. The way I see it, the only way to
> convince the compiler it really needs to do this loop more than once is to
> make it think it is not overwriting the same variable every time. The
> subscript was the cheapest way I could think of to do that. Any other
> suggestions on how to do this are welcome.

Hmm. Maybe store the CRCs into a global array somewhere?

uint32 results[NTESTS];

for ...
{
INIT/COMP/FIN_CRC32...
results[j] = mycrc;
}

This still adds a bit of overhead to the outer loop, but not much.

Another possibility is to put the INIT/COMP/FIN_CRC32 into an external
subroutine, thereby adding a call/return to the outer loop overhead.

regards, tom lane