Re: Array prefetch patch

[Jan Hubicka <jh at suse dot cz>]

> 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!

Honza