This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Exploiting dual mode operation, implementation.
- From: Richard Henderson <rth at redhat dot com>
- To: Leehod Baruch <LEEHOD at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Mircea Namolaru <NAMOLARU at il dot ibm dot com>, sleehodb at t2 dot technion dot ac dot il, stevenb at suse dot de
- Date: Sun, 31 Jul 2005 10:48:10 -0700
- Subject: Re: [PATCH] Exploiting dual mode operation, implementation.
- References: <20050627234722.GA18719@redhat.com> <OF722D851B.433D85B4-ONC225704F.003F040D-C225704F.00556F3E@il.ibm.com>
On Sun, Jul 31, 2005 at 06:30:46PM +0300, Leehod Baruch wrote:
> In other words, Is it good enough to handle only webs that has one type
> of extension (SIGN_EXTEND or ZERO_EXTEND and
Yes.
> only one pair of modes for source and dest of the extension)?
I'm not sure this is difficult at all. Extend to the max mode,
when all modes are not larger than word_mode. Extensions to
larger modes generally aren't free, and so aren't really the
target of your operation.
> In order to make the gcse/lcm work I need that the sign-extension
> source will be of the same register.
Ah, so it'd be fixed if you set up the lcm yourself, as suggested.
r~