This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] [4.3. projects] Vectorize int to float conversions
- From: "Richard Guenther" <richard dot guenther at gmail dot com>
- To: "Tehila Meyzels" <TEHILA at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, "Ira Rosen" <IRAR at il dot ibm dot com>, "Dorit Nuzman" <DORIT at il dot ibm dot com>
- Date: Tue, 23 Jan 2007 16:55:54 +0100
- Subject: Re: [PATCH] [4.3. projects] Vectorize int to float conversions
- References: <OF05C46923.CBA2D496-ONC225726C.00486D2C-C225726C.005207B4@il.ibm.com>
On 1/23/07, Tehila Meyzels <TEHILA@il.ibm.com> wrote:
This patch adds the necessary target hooks and vectorizer changes to
support vectorization of statements that include
casting from int to float, In addition, it implements the hook for altivec.
Testcase attached.
Bootstrapped and tested on the vectorizer testcases on powerpc-linux.
Bootstrapped with vectorization enabled and tested on the vectorizer
testcases on i386-linux.
Full makecheck in progress.
OK for mainline, once testing completes?
I would like it more if you did a generic vectorize_conversion hook (to also
support float->int conversion) and if you added the necessary changes to
support mixed-vector operations to for example enable vectorization of
lrint (which is a float->int conversion, too). This patch looks really too
special.
I suppose we can work on more generic support for type (and width) changing
operations.
Thanks,
Richard.