Newsgroups: | comp.arch,comp.compilers |
From: | glew@ichips.intel.com (Andy Glew) |
Keywords: | optimize, design |
Organization: | Intel Corp., Hillsboro, Oregon |
References: | 94-10-108 94-11-021 |
Date: | Sun, 13 Nov 1994 02:16:42 GMT |
Now arguing that structure field rearrangement could make processor memory
management task easier is nice, but you haven't clearly demonstrated a
benefit. If the source code needs fields in particular locations, the
optimization of rearrangement will break the application. If the locations
are not important, the application programmer may have already hand
optimized the structure, so how much better is the compiler going to do
this???
The almost 2X benefit demonstrated by Alvy LeBeck isn't good enough
for you?
(Personally, I have spent much of my career optimizing other people's
code. Rearranging data structures is part of my standard book of
tricks. I am only trying to make it possible for compilers to do what
a skilled human programmer does as a matter of course.)
--
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.