[m-rev.] For review by Fergus: Loop Invariant Hoisting

[Zoltan Somogyi]

On 30-Oct-2002, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> Probably the increase in code size is more costly (in terms of cache
> performance) for asm_fast.gc, which already has a high code size.

I agree.

> What's the uncertainty/variance in these numbers?

I doubt that any of the differences are significant, except in hlc.gc at -O3
and *maybe* in asm_fast.gc at -O6.

> Did you run the benchmarks at the same time that the nightly tests
> were running?

I looked at the raw numbers, and the CPU utilization was consistently
above 99%, so the answer is no.

> I think it should be disabled if --optimize-space is specified.

Agreed.

Zoltan.
--------------------------------------------------------------------------
mercury-reviews mailing list
post:  mercury-reviews at cs.mu.oz.au
administrative address: owner-mercury-reviews at cs.mu.oz.au
unsubscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-reviews-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------