This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] PPC64: First in the series of patches implementing POWER8 vector math.
- From: GT <tnggil at protonmail dot com>
- To: "libc-alpha at sourceware dot org" <libc-alpha at sourceware dot org>
- Date: Mon, 25 Feb 2019 22:03:13 +0000
- Subject: Re: [PATCH] PPC64: First in the series of patches implementing POWER8 vector math.
- References: <nN0kG0GXbx7wYro2oE34bwvcauTJw_WuDyP-obLhVMc-zwTK071-mQXBD8hiITyOHI6WoxiwvNezhvebbsjFnNlAk8mCi5yX6NYpge4SAas=@protonmail.com> <alpine.DEB.2.21.1902190117000.480@digraph.polyomino.org.uk> <b7VIpuRTdy3ZwQBiTlgUL3a5zz5kQNT6OUuyOfyxrAhNG_1Xp8LolqtRZ1CssqxB5HAUcZ2-FYf64L8sAjgmfW4qiwng59NPWvSc8K0JQyo=@protonmail.com> <alpine.DEB.2.21.1902251714590.25244@digraph.polyomino.org.uk>
- Reply-to: GT <tnggil at protonmail dot com>
> However, I don't think that's your issue. You should have
> libmvec_nonshared.a; it's needed to deal with cases when the compiler
> generates calls to Z*__*_finite because of bits/math-finite.h being in
> effect together with bits/math-vector.h (see
> sysdeps/x86_64/fpu/svml_finite_alias.S and
> https://gcc.gnu.org/ml/gcc/2015-06/msg00173.html).
>
1. svml_finite_alias.S deals only with log/logf, exp/expf and pow/powf.
So does the discussion linked at the given url. At this time only the
cos function has been implemented. Do we then still require
libmvec_nonshared.a to be created?
2. I find a single reference in the entire source tree where
libmvec_nonshared.a doesn't appear in the dependency section of
a rule. That's in math/Makefile. Does this mean libmvec_nonshared.a
is created with 'make install'? I rather expected it would be built
by the initial 'make' which generates most other objects.