This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [autovect] [patch] vectorize in cases when number of iterations may be zero
- From: Sebastian Pop <sebastian dot pop at cri dot ensmp dot fr>
- To: Victor Kaplansky <VICTORK at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, pop at cri dot ensmp dot fr, sebpop at gmail dot com
- Date: Fri, 16 Jun 2006 12:13:47 +0200
- Subject: Re: [autovect] [patch] vectorize in cases when number of iterations may be zero
- References: <OF77B42963.6CBF8DA5-ONC225718E.0043E658-C225718E.0046A838@il.ibm.com>
Victor Kaplansky wrote:
>
> 1 int
> 2 foo (short *in, short *out)
> 3 {
> 4 unsigned i;
> 5 unsigned N;
> 6 unsigned M;
> 7 int acc;
> 8
> 9 for (N = 0; N < 16; N++)
> 10 {
> 11 M = 500 - N;
> 12
> 13 acc = 0;
> 14
> 15 for (i = 0; i < M; i++)
> 16 {
> 17 acc += in[i + M];
> 18 }
> 19
> 20 out[N] = (short) acc;
> 21 }
> 22
> 23 return acc;
> 24 }
>
At least for this example there is a way to get rid of the may_be_zero
condition (that is something like "N < 500"), as N is known to be
{0,+,1}_1 and the number of iterations in the outermost loop is 16.
So you can prove this way that the condition may_be_zero is satisfied.
I have started on such a patch but didn't finished it last weekend.
I'll try to get it done for Monday.
Sebastian