This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Speedup DOM and tree-cfg a lot (partial of PR 15524)
- From: law at redhat dot com
- To: Andrew Pinski <pinskia at physics dot uc dot edu>
- Cc: Diego Novillo <dnovillo at redhat dot com>, Eric Christopher <echristo at redhat dot com>, "gcc-patches at gcc dot gnu dot org Patches" <gcc-patches at gcc dot gnu dot org>
- Date: Tue, 18 May 2004 21:43:59 -0600
- Subject: Re: [PATCH] Speedup DOM and tree-cfg a lot (partial of PR 15524)
- Reply-to: law at redhat dot com
In message <D962329B-A93C-11D8-A4A1-000393A6D2F2@physics.uc.edu>, Andrew Pinski
writes:
>On May 18, 2004, at 22:19, Diego Novillo wrote:
>
>> On Tue, 2004-05-18 at 22:12, Eric Christopher wrote:
>>
>>> I don't see how this can be much faster if at all. I've talked to
>>> Andrew
>>> about it and he's doing a timing. If it does turn out to be noticeably
>>> faster then perhaps we'll want to comment it with something like:
>>>
>> Didn't Andrew say that he had timed this and found it faster? The
>> patch
>> was innocent enough otherwise.
>
>Sorry about that.
>
>>> /* Compare for equal using tree_int_cst_compare since these
>>> are only ever INTEGER_CST. */
>>>
>>> or something?
>>>
>> Ah, good point.
>
>Well I just found that it actually slowed it down and I so I decided
>to look for an equal instead and there is one: tree_int_cst_equal.
>OK if faster and bootstraps and tests fine?
>
> * tree-cfg.c (find_case_label_for_value): Replace call to
> simple_cst_equal with tree_int_cst_equal.
This is fine. I've done similar things within DOM itself, I don't recall
if the improvement was measurable or not, but tree_int_cst_equal is
clearly simpler and ought to be more efficient than simple_cst_equal.
jeff