Newsgroups: | comp.compilers |
From: | Ronald Bodkin <rjbodkin@theory.lcs.mit.edu> |
Organization: | MIT Lab for Computer Science |
Date: | Tue, 4 Aug 1992 04:25:09 GMT |
Keywords: | performance, storage |
References: | 92-07-064 92-08-004 |
graham@maths.su.oz.au (Graham Matthews) writes:
... Your translator should have compiled the loop into
for( i = 0; i < 100; i++ )
object_access(p)[i] = object_access(q)[i];
...
You take a performance hit here but in the presence of GC surely you have
to take that hit no matter what IL you use?
This is simply untrue. In the presence of GC, you can STILL store
pointers that are part of the "seed" for live data in registers. The
problem is that C optimizers don't consider such a use of registers. If
you go to assembly output, you certainly can use registers and (safely)
garbage collect.
Ron
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.