This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: rs6000 patch 2 of 4 (altivec)
- From: Aldy Hernandez <aldyh at redhat dot com>
- To: Alan Modra <amodra at bigpond dot net dot au>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: 24 Apr 2004 00:32:36 -0400
- Subject: Re: rs6000 patch 2 of 4 (altivec)
- References: <20040423115223.GF2531@bubble.modra.org>
>>>>> "Alan" == Alan Modra <amodra@bigpond.net.au> writes:
> Fairly obvious patch. The old test was equivalent to
> if (info->vrsave_mask != 0)
> since ALTIVEC_REG_BIT(i) is non-zero for all values of i in the
> loop.
We really should look at the archives to see why all this
ALTIVEC_REG_BIT was added. IIRC, it was to keep VRSAVE sets from
being clobbered by flow. In the end we made VRSAVE sets unspec
volatiles, thus rendering all this vrsave_mask magic useless.
Until then...
OK.
Thanks.
Aldy