Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <432e997617a0669886cd9ea5ceac7c1a2173044b.camel@decadent.org.uk>
 <1562273427-204-1-git-send-email-Cedric_Hombourger@mentor.com> <1562273427-204-2-git-send-email-Cedric_Hombourger@mentor.com>
In-Reply-To: <1562273427-204-2-git-send-email-Cedric_Hombourger@mentor.com>
From:   Masahiro Yamada <yamada.masahiro@socionext.com>
Date:   Sun, 7 Jul 2019 17:58:15 +0900
Message-ID: <CAK7LNARgFmSm+kCngidevQ3HnNjqKDKfGzYXrFPzhgBRHmccXw@mail.gmail.com>
Subject: Re: [PATCH v3 1/1] builddeb: generate multi-arch friendly
 linux-libc-dev package
To:     Cedric Hombourger <Cedric_Hombourger@mentor.com>
Cc:     Ben Hutchings <ben@decadent.org.uk>, isar-users@googlegroups.com,
        Michal Marek <michal.lkml@markovi.net>,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Riku Voipio <riku.voipio@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Jul 5, 2019 at 5:50 AM Cedric Hombourger
<Cedric_Hombourger@mentor.com> wrote:
>
> Debian-based distributions place libc header files in a machine
> specific directory (/usr/include/<libc-machine>) instead of
> /usr/include/asm to support installation of the linux-libc-dev
> package from multiple architectures. Move headers installed by
> "make headers_install" accordingly using Debian's tuple from
> dpkg-architecture.
>
> Signed-off-by: Cedric Hombourger <Cedric_Hombourger@mentor.com>
> ---
>  scripts/package/builddeb | 8 ++++++++
>  scripts/package/mkdebian | 5 +++--
>  2 files changed, 11 insertions(+), 2 deletions(-)
>
> diff --git a/scripts/package/builddeb b/scripts/package/builddeb
> index b03dd56a4782..15a034e18c37 100755
> --- a/scripts/package/builddeb
> +++ b/scripts/package/builddeb
> @@ -132,6 +132,14 @@ fi
>  if [ "$ARCH" != "um" ]; then
>         $MAKE -f $srctree/Makefile headers_check
>         $MAKE -f $srctree/Makefile headers_install INSTALL_HDR_PATH="$libc_headers_dir/usr"
> +       if [ -n "$debarch" ]; then
> +               # move asm headers to /usr/include/<libc-machine>/asm to match the structure
> +               # used by Debian-based distros (to support multi-arch) but only if ARCH was
> +               # translated to Debian's (debarch) - this is done by mkdebian
> +               host_arch=$(dpkg-architecture -a$debarch -qDEB_HOST_MULTIARCH)
> +               mkdir $libc_headers_dir/usr/include/$host_arch
> +               mv $libc_headers_dir/usr/include/asm $libc_headers_dir/usr/include/$host_arch/
> +       fi
>  fi
>
>  # Install the maintainer scripts
> diff --git a/scripts/package/mkdebian b/scripts/package/mkdebian
> index 8351584cb24e..2d670ae2cabc 100755
> --- a/scripts/package/mkdebian
> +++ b/scripts/package/mkdebian
> @@ -197,6 +197,7 @@ Architecture: $debarch
>  Description: Linux support headers for userspace development
>   This package provides userspaces headers from the Linux kernel.  These headers
>   are used by the installed headers for GNU glibc and other system libraries.
> +Multi-Arch: same
>
>  Package: $dbg_packagename
>  Section: debug
> @@ -212,11 +213,11 @@ cat <<EOF > debian/rules
>  srctree ?= .
>
>  build:
> -       \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} \
> +       \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} debarch=${debarch} \
>         KBUILD_BUILD_VERSION=${revision} -f \$(srctree)/Makefile
>
>  binary-arch:
> -       \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} \
> +       \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} debarch=${debarch} \
>         KBUILD_BUILD_VERSION=${revision} -f \$(srctree)/Makefile intdeb-pkg


Is there any smarter way (any command) to get debarch
from the builddeb script?

scripts/package/Makefile passes -a flag, like follows:

dpkg-buildpackage -r$(KBUILD_PKG_ROOTCMD) -a$$(cat debian/arch)
$(DPKG_FLAGS) -b -nc -uc

So, debarch=${debarch} looks somewhat redundant to me.


-- 
Best Regards
Masahiro Yamada