This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH] Merge powerpc slowpow.c into generic code
- From: Siddhesh Poyarekar <siddhesh at redhat dot com>
- To: Andreas Jaeger <aj at suse dot com>
- Cc: libc-alpha at sourceware dot org
- Date: Wed, 6 Mar 2013 15:38:51 +0530
- Subject: Re: [PATCH] Merge powerpc slowpow.c into generic code
- References: <20130225101832.GH21039@spoyarek.pnq.redhat.com><1440669.3Xse0NFY2p@byrd>
On Mon, Feb 25, 2013 at 11:31:12AM +0100, Andreas Jaeger wrote:
> On Monday, February 25, 2013 15:48:32 Siddhesh Poyarekar wrote:
> > Hi,
> > This patch adds the powerpc-specific tweak of using long double as an
> > alternative before falling back to multiprecision, into generic code
> > and removes the duplicated files from the powerpc sysdeps. The bits
> > are built only when a macro USE_LONG_DOUBLE_FOR_MP is defined (only on
> > powerpc for now). The generated code is identical on x86_64 and
> > powerpc (--with-cpu=power7). Checked by verifying that the sha1sum
> > is different (different debuginfo) but objdump -d is the same.
> > OK to commit?
> This looks fine to me,
Should I go ahead and commit this? The generated code is unchanged