This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
RE: Remove splay_tree from gimplify.c
- From: Aditya K <hiraditya at msn dot com>
- To: Jakub Jelinek <jakub at redhat dot com>, Jeff Law <law at redhat dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Wed, 27 May 2015 18:59:01 +0000
- Subject: RE: Remove splay_tree from gimplify.c
- Authentication-results: sourceware.org; auth=none
- References: <BLU179-W1888AA8689CF2B3906A606B6C40 at phx dot gbl>,<5565F226 dot 1020107 at redhat dot com>,<20150527164155 dot GJ10247 at tucnak dot redhat dot com>
----------------------------------------
> Date: Wed, 27 May 2015 18:41:55 +0200
> From: jakub@redhat.com
> To: law@redhat.com
> CC: hiraditya@msn.com; gcc-patches@gcc.gnu.org
> Subject: Re: Remove splay_tree from gimplify.c
>
> On Wed, May 27, 2015 at 10:34:46AM -0600, Jeff Law wrote:
>> So the question here is whether or not the other uses are commonly looking
>> up elements we've already searched for -- that's the whole purpose of a
>> splay tree, to improve lookup performance for commonly hit items.
>
> First of all, this is only used for OpenMP/OpenACC/Cilk+ constructs,
> so it really isn't that performance criticial.
> The code probably dates back to Richard's and Diego's changes.
> And, I believe splay trees are the right thing to use here, while sometimes
> you lookup different vars, looking up the same var many times in a row is
> very common.
If that is the case then I guess we can abandon the patch. I do not have a way to measure the compile time for OpenMP.
Thanks for the review.
-Aditya
>
> Jakub