Re: Scheduling without profiling

Stephan Ceram <>
6 Mar 2009 14:44:03 GMT

          From comp.compilers

Related articles
Scheduling without profiling (Stephan Ceram) (2009-02-19)
Re: Scheduling without profiling (Matthias Blume) (2009-02-19)
Re: Scheduling without profiling (Sid Touati) (2009-02-20)
Re: Scheduling without profiling (Stephan Ceram) (2009-03-06)
Re: Scheduling without profiling (touati) (2009-03-10)
| List of all articles for this month |

From: Stephan Ceram <>
Newsgroups: comp.compilers
Date: 6 Mar 2009 14:44:03 GMT
Organization: Compilers Central
References: 09-02-098 09-02-101
Keywords: code, optimize
Posted-Date: 06 Mar 2009 21:19:49 EST

> I would say the contrary: scheduling based on profiling information has
> no practical sense, since profiling information depends on the chosen
> data input. For numerical applications, this would not be really a
> problem, and profiling helps. But for general purpose applications,
> profiling does not necessarily helps to produce better codes for any
> data input. This is why static code optimisation does not rely on
> profiling.

OK, so taking the danger of changing hot paths through different inputs
into account, do you think that it might be a promising idea to mix
instructions from several paths? This would possibly help the scheduler
to reorder instructions such that the pipelines/functional units are
maximally exploited but on the other hand no particular path is favored/
neglected too heavily.

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.