Related articles |
---|
Optimization opportunities unexploited: examples sought pshuang@athena.mit.edu (1995-05-08) |
Re: Optimization opportunities unexploited: examples sought glew@ichips.intel.com (1995-06-24) |
Newsgroups: | comp.compilers,comp.arch |
From: | glew@ichips.intel.com (Andy Glew) |
Keywords: | optimize |
Organization: | Intel Corp., Hillsboro, Oregon |
References: | 95-05-061 |
Date: | Sat, 24 Jun 1995 12:30:33 GMT |
Status: | RO |
(1) this should have been cross-posted to comp.arch, not comp.arch.arithmetic.
(2) profiling feedback to compilers is standard technology nowadays -
although admittedly Intel's compiler team is the biggest advocate.
The real issue is not doing the feedback, but making easier and easier
the task of collecting the data (for which I espouse the use of
statistical feedback using interrupt sampling, measuring real users on
real machines in production, not just test).
I.e. it is more of an OS issue than a compiler issue.
A related research issue is using non-temporal performance metrics,
e.g. using measurements of where cache misses and branch
mispredictions are occurring.
--
Andy "Krazy" Glew, glew@ichips.intel.com, Intel,
M/S JF1-19, 5200 NE Elam Young Pkwy, Hillsboro, Oregon 97124-6497.
Place URGENT in email subject line for mail filter prioritization.
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.