Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Romain Naour <romain.naour@openwide.fr>
To: buildroot@busybox.net
Subject: [Buildroot] [RFC 00/37] efl bump to 1.14.2
Date: Tue, 4 Aug 2015 01:10:16 +0200	[thread overview]
Message-ID: <55BFF4D8.6090204@openwide.fr> (raw)
In-Reply-To: <CAAMcf8BWy-NzBFHKsbyx3SRz9RC98_KUiCoq-7C5C-nRS+jmGQ@mail.gmail.com>

Hi Vicente,

Le 03/08/2015 20:55, vj a ?crit :
> Hello Romain
> 
> I've tested your patch series successfully!

That's great, thanks for testing :)

> 
> Just a few notes:
> 
> 1)
> It's been tested with the musl c library. To do so, I had to remove the libc
> restriction you added.
> When using musl, there are some packages (at least pulseaudio) that can not be
> build, so the BR2_PACKAGE_EFL_RECOMMENDED_CONFIG had to be disabled.
> So, as it build and worked fine, could you remove the glibc requirement or add
> musl to the restricted list of supported libcs ?

Yes, you're right. efl can be build with a musl libc.

I've mainly tested this series with a glibc toolchain and I had a link issue
with a uClibc-ng toolchain due to missing mkstemps/mkostemps().

Before sending the series, I tried to build with a musl toochain and I had the
same issue with pulsaudio and give up. (hence this patches series [1])

Also, I've sent this patch [2] to the uclibc-ng mailing list to add missing
mkstemps/mkostemps().

Since then, I've successfully built enlightenment with a musl toolchain (not
runtime tested though).

But I'm wonder if I should remove BR2_PACKAGE_EFL_RECOMMENDED_CONFIG and make
sure that all recommended dependencies are mandatory. Because it's mainly
untested or even not built at all by efl maintainers.

So, ok I'll enable musl for the next respin.

[1] http://patchwork.ozlabs.org/patch/502724/
[2] http://mailman.uclibc-ng.org/pipermail/devel/2015-August/000432.html
> 
> 2)
> When BR2_PACKAGE_EFL_RECOMMENDED_CONFIG is disabled the efl dependency on
> BR2_PACKAGE_UTIL_LINUX_LIBMOUNT is lost but it is still a requirement, so this
> patch (or a variation of it) is needed:
> --- a/package/efl/Config.in
> +++ b/package/efl/Config.in
> @@ -12,6 +12,8 @@ config BR2_PACKAGE_EFL
>      select BR2_PACKAGE_LIBCURL # eina_module, runtime dependency
>      select BR2_PACKAGE_LUA # Evas lua 5.1 or 5.2
>      select BR2_PACKAGE_LIBXKBCOMMON if BR2_PACKAGE_WAYLAND
> +    select BR2_PACKAGE_UTIL_LINUX
> +    select BR2_PACKAGE_UTIL_LINUX_LIBMOUNT
>      help
>        Enlightenment Foundation Libraries

Thanks will fix, but I haven't really tested without recommended config.

>  
> 3)
> To test it I've added the terminology package which depends on efl and
> configured buildroot like:
> BR2_arm=y
> BR2_cortex_a8=y
> BR2_ARM_EABIHF=y
> BR2_ARM_FPU_NEON=y
> BR2_OPTIMIZE_3=y
> BR2_TOOLCHAIN_BUILDROOT_MUSL=y
> BR2_BINUTILS_VERSION_2_25_X=y
> BR2_TOOLCHAIN_BUILDROOT_CXX=y
> BR2_GCC_ENABLE_LTO=y
> BR2_GCC_ENABLE_GRAPHITE=y
> BR2_TARGET_OPTIMIZATION="-march=armv7-a -mtune=cortex-a8 -mcpu=cortex-a8
> -mfpu=neon -mfloat-abi=hard"
> BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
> BR2_PACKAGE_TERMINOLOGY=y
> # BR2_PACKAGE_EFL_RECOMMENDED_CONFIG is not set
> BR2_PACKAGE_EFL_FB=y
> BR2_PACKAGE_EFL_PNG=y
> BR2_PACKAGE_EFL_JPEG=y
> BR2_PACKAGE_LUA_5_3=y
> BR2_TARGET_ROOTFS_CPIO=y
> # BR2_TARGET_ROOTFS_TAR is not set

I haven't sent the complete series yet, just the bump of current efl related
packages.
I have pushed my working branch here, if you want to look at it:

https://github.com/RomainNaour/buildroot/tree/efl-1.14.2-v4

There is already a terminology package and more ;-)

> 
> 4)
> Unrelated to efl: to make it work with an external toolchain when the
> arhitecture is "armhf" instead of "arm" this patch is also required:
> --- a/toolchain/toolchain-external/toolchain-external.mk
> <http://toolchain-external.mk>
> +++ b/toolchain/toolchain-external/toolchain-external.mk
> <http://toolchain-external.mk>
> @@ -496,6 +496,8 @@ endef
>  ifeq ($(BR2_TOOLCHAIN_EXTERNAL_MUSL),y)
>  ifeq ($(BR2_i386),y)
>  MUSL_ARCH = i386
> +else ifeq ($(BR2_ARM_EABIHF),y)
> +MUSL_ARCH = armhf
>  else
>  MUSL_ARCH = $(ARCH)
>  endif

Can you send a proper patch for this please ?

See the manual if needed:
http://buildroot.uclibc.org/downloads/manual/manual.html#submitting-patches

> 
> Thanks for updating efl!

You're welcome.

I hope you'll send a Tested-by and/or Reviewed-by tag for the next respin ;-)

Best regards,
Romain

> 
> Regards,
>   Vicente.
> 
> 
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot
> 

  reply	other threads:[~2015-08-03 23:10 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-03 18:55 [Buildroot] [RFC 00/37] efl bump to 1.14.2 vj
2015-08-03 23:10 ` Romain Naour [this message]
  -- strict thread matches above, loose matches on Subject: below --
2015-08-21  0:30 vj
2015-08-21  9:51 ` Romain Naour
2015-08-21 10:18   ` Romain Naour
2015-08-21 17:56   ` vj
2015-08-24 21:42     ` Romain Naour
2015-08-24 22:09       ` vj
2015-09-05 14:55         ` Romain Naour
2015-09-27  0:37           ` vj
2015-09-28 21:00             ` Romain Naour
2015-07-26 18:55 Romain Naour
2015-08-06  8:45 ` Romain Naour

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=55BFF4D8.6090204@openwide.fr \
    --to=romain.naour@openwide.fr \
    --cc=buildroot@busybox.net \
    /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