This is the mail archive of the
mailing list for the GCC project.
Re: Array prefetch patch
- From: Jan Hubicka <jh at suse dot cz>
- To: Janis Johnson <janis187 at us dot ibm dot com>
- Cc: Jan Hubicka <jh at suse dot cz>, Richard Henderson <rth at redhat dot com>, gcc-patches at gcc dot gnu dot org
- Date: Tue, 11 Dec 2001 09:54:59 +0100
- Subject: Re: Array prefetch patch
- References: <20011129130113.A4811@us.ibm.com> <20011130211819.B19261@atrey.karlin.mff.cuni.cz> <20011130134036.A16007@us.ibm.com> <20011201005437.B4901@atrey.karlin.mff.cuni.cz> <20011130164228.A20710@us.ibm.com> <20011130173521.E32390@redhat.com> <20011130175134.A10061@us.ibm.com> <20011201203646.A520@redhat.com> <20011210171753.B419@atrey.karlin.mff.cuni.cz> <20011210210519.A11421@us.ibm.com>
> I've been playing with Honza's prefetch optimization patch, too, and
> have some changes to the one he just sent in:
> Use of the function remove_constant_addition causes a failure in the
> test gcc.c-torture/execute/920501-6.c when it is compiled with
> "-O3 -fprefetch-loop-arrays". With the call to it commented out it's
> OK. I can look into it further, since this is undoubtedly supposed
> to be doing something useful.
Interesting, I went trought testsuite with -fprefetch-loop-arrays and it
did work for me. What kind of failure you get?
> The comments about heuristics didn't match the ones that are actually
> used. I've parameterized the individual heuristics so it's easier to
> turn them on and off for benchmarking purposes; it's also possible to
> select different rules for different targets that way.
> It's toplev.c, not loop.c, that includes insn-flags.h and needs to
> have its Makefile entry updated.
> The line in toplev.c with the call to loop_optimize is too long.
> Trailing periods were recently removed from warnings; this applies to
> the new warnings in toplev.c.
> There are some "???" strings where I find things confusing.
I will clarify/fix these and send updated patch after finishing the
-march/-mcpu tweeks we need for prefetch support on x86. Thanks!