This is the mail archive of the
mailing list for the GCC project.
Re: [patch] Improve prefetch emitting
- From: Roger Sayle <roger at eyesopen dot com>
- To: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Sat, 11 Nov 2006 07:46:38 -0700 (MST)
- Subject: Re: [patch] Improve prefetch emitting
On Sat, 11 Nov 2006, Zdenek Dvorak wrote:
> * tree-ssa-loop-prefetch.c (schedule_prefetches): Cleanup and improve
> (issue_prefetch_ref): Move assignment to write_p out of loop.
> (determine_unroll_factor): Do not take PARAM_MAX_UNROLL_TIMES and
> SIMULTANEOUS_PREFETCHES into account.
> (loop_prefetch_arrays): Do not pass ahead to determine_unroll_factor.
> * lambda-code.c (lcm): Export.
> * tree-flow.h (lcm): Declare.
> * gcc.dg/tree-ssa/prefetch-3.c: New test.
This is OK for mainline.
At the risk of repeating myself, do you have any performance figures
for these changes?
Now that "lcm" is a public function, it might make sense to rename
it to "lowest_common_multiple" to avoid potential conflicts, or being
mistaken for "lazy code motion", or something similar.
Thanks in advance,