On 07/13/2011 04:34 PM, Richard Guenther wrote: > On Wed, Jul 13, 2011 at 3:13 PM, Andreas Krebbel > <kreb...@linux.vnet.ibm.com> wrote: >> Hi, >> >> the widening_mul pass might increase the number of multiplications in >> the code by transforming >> >> a = b * c >> d = a + 2 >> e = a + 3 >> >> into: >> >> d = b * c + 2 >> e = b * c + 3 >> >> under the assumption that an FMA instruction is not more expensive >> than a simple add. This certainly isn't always true. While e.g. on >> s390 an fma is indeed not slower than an add execution-wise it has >> disadvantages regarding instruction grouping. It doesn't group with >> any other instruction what has a major impact on the instruction >> dispatch bandwidth. >> >> The following patch tries to figure out the costs for adds, mults and >> fmas by building an RTX and asking the backends cost function in order >> to estimate whether it is whorthwhile doing the transformation. >> >> With that patch the 436.cactus hotloop contains 28 less >> multiplications than before increasing performance slightly (~2%). >> >> Bootstrapped and regtested on x86_64 and s390x. > > Ick ;)
Just to defend myself I did not invent this. I basically copied it over from ivopts. > Maybe this is finally the time to introduce target hook(s) to > get us back costs for trees? For this case we'd need two > actually, or just one - dependent on what finegrained information > we pass. Choices: Having another target hook for costs actually doesn't look that much better to me. Duplicating the logic is not really practical for backend developers. We rather should have a generic interface for calculating tree costs based on the rtx_cost hook. Bye, -Andreas-
