public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: "H.J. Lu" <hjl.tools@gmail.com>
To: Andrew Senkevich <andrew.n.senkevich@gmail.com>
Cc: libc-alpha <libc-alpha@sourceware.org>
Subject: Re: [PATCH] Install libm.a as linker script [BZ #20539]
Date: Mon, 24 Oct 2016 22:02:00 -0000	[thread overview]
Message-ID: <CAMe9rOqirjhp+cpWGTVeQKd8p-yNdsO=mQ+-Ww-B4ULSjfUSWA@mail.gmail.com> (raw)
In-Reply-To: <CAMXFM3szvCP13C1P_Ed8NVV_FTt4Q8avz4uW-a=UbpWqMnF5aw@mail.gmail.com>

On Thu, Oct 20, 2016 at 12:20 PM, Andrew Senkevich
<andrew.n.senkevich@gmail.com> wrote:
> Hi,
>
> to avoid static link fail w/o passing additional -lmvec option (while
> building with GCC 6.1 and above) this patch makes libm.a installed as
> linker script.
> Tested on x86_64.  Ok for trunk?

A couple issues.  Don't use cut/paste since it adds some newlines
and the patch won't apply.

> 2016-10-20  Andrew Senkevich  <andrew.senkevich@intel.com>
>
>         [BZ #20539]
>         * Makerules (install-lib.a): filter out libm.a in case of libmvec build.
>         * math/Makefile: added rules for libm.a installation as linker script.
>
> diff --git a/Makerules b/Makerules
> index e865782..290f927 100644
> --- a/Makerules
> +++ b/Makerules
> @@ -1190,7 +1190,9 @@ $(addprefix $(inst_sbindir)/,$(install-sbin)): \
>         $(do-install-program)
>  endif
>  ifdef install-lib
> -install-lib.a := $(filter lib%.a,$(install-lib))
> +ifeq ($(build-mathvec),yes)
> +install-lib.a := $(filter-out libm.a,$(filter lib%.a,$(install-lib)))
> +endif

Should we use $(install-lib-ldscripts) instead of libm.a here?

>  install-lib-non.a := $(filter-out lib%.a,$(install-lib))
>  ifdef install-lib-non.a
>  $(addprefix $(inst_libdir)/$(libprefix),$(install-lib-non.a)): \
> diff --git a/math/Makefile b/math/Makefile
> index 2aa6a34..a96ef30 100644
> --- a/math/Makefile
> +++ b/math/Makefile
> @@ -128,8 +128,8 @@ generated += $(foreach s,.c .S,$(call
> type-foreach, $(calls:s_%=m_%$(s))))
>  routines = $(call type-foreach, $(calls))
>
>  ifeq ($(build-mathvec),yes)
> -# We need to install libm.so as linker script
> -# for more comfortable use of vector math library.
> +# We need to install libm.so and libm.a as linker scripts
> +# for transparent use of vector math library.
>  install-lib-ldscripts := libm.so
>  install_subdir: $(inst_libdir)/libm.so
>  $(inst_libdir)/libm.so: $(common-objpfx)format.lds \
> @@ -142,6 +142,19 @@ $(inst_libdir)/libm.so: $(common-objpfx)format.lds \
>               'AS_NEEDED ( $(libdir)/libmvec_nonshared.a
> $(slibdir)/libmvec.so$(libmvec.so-version) ) )' \
>         ) > $@.new
>         mv -f $@.new $@
> +
> +install-others = $(inst_libdir)/libm.a
> +$(inst_libdir)/libm.a: $(common-objpfx)format.lds \
> +                      $(libm) \
> +                      $(libmvec) \
> +                      $(+force)
> +       (echo '/* GNU ld script'; echo '*/';\
> +        cat $<; \
> +        echo 'GROUP ( $(libdir)/libm-$(lib-version).a ' \
> +             'AS_NEEDED ( $(libdir)/libmvec.a) )' \
                  ^^^^^^^^^^^^^ Remove it.
> +       ) > $@.new
> +       cp $(objpfx)libm.a $(libdir)/libm-$(lib-version).a
                                        ^^^^^ It should be inst_libdir.
> +       mv -f $@.new $@
>  endif
>
>  # Rules for the test suite.
>

Do you need to update rules for building static math tests with
libmvec.a?

-- 
H.J.

  reply	other threads:[~2016-10-24 22:02 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-20 19:21 Andrew Senkevich
2016-10-24 22:02 ` H.J. Lu [this message]
2016-11-14 19:16   ` Andrew Senkevich
2016-11-14 19:22     ` H.J. Lu
2016-12-05 19:40   ` Andrew Senkevich
2016-12-05 20:01     ` H.J. Lu
2016-12-06 12:17       ` Andrew Senkevich
2016-12-06 14:17         ` Andreas Schwab
2016-12-06 14:49           ` Andrew Senkevich
2016-12-06 15:21             ` Andreas Schwab
2016-12-06 15:50               ` Andrew Senkevich
2016-12-06 21:49                 ` H.J. Lu
2016-12-07 10:34                   ` Andrew Senkevich
2016-12-07 16:22                     ` H.J. Lu
2016-12-07 20:04                       ` Andrew Senkevich
2016-12-07 21:50                         ` H.J. Lu
2016-12-09 15:39                         ` Andreas Schwab
2016-12-09 17:42                           ` Andrew Senkevich
2016-12-09 18:23                             ` Andreas Schwab
2016-12-12 15:45                               ` Andrew Senkevich
2016-12-12 16:18                                 ` Andreas Schwab
2016-12-12 16:29                                   ` Andrew Senkevich
2016-12-12 16:42                                     ` Andreas Schwab
2016-12-12 17:58                                       ` Andrew Senkevich
2016-12-13  8:41                                         ` Andreas Schwab
2016-12-13 14:58                                           ` Andrew Senkevich

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAMe9rOqirjhp+cpWGTVeQKd8p-yNdsO=mQ+-Ww-B4ULSjfUSWA@mail.gmail.com' \
    --to=hjl.tools@gmail.com \
    --cc=andrew.n.senkevich@gmail.com \
    --cc=libc-alpha@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).