Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Gary Bisson <gary.bisson@boundarydevices.com>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH 3/3] weston: add weston-imx variant when using imx-gpu-viv
Date: Thu, 17 May 2018 15:08:10 +0200	[thread overview]
Message-ID: <20180517130810.GA6584@g751.home> (raw)
In-Reply-To: <2d58cd37-effa-a5bf-5f9b-6d1e143062db@mind.be>

Hi Arnout,

On Wed, May 16, 2018 at 11:24:17PM +0200, Arnout Vandecappelle wrote:
> 
> 
> On 16-05-18 17:52, Gary Bisson wrote:
> > This variant contains various optimizations for i.MX processors.
> 
>  Argh, this is annoying...

Yeah I know, I felt the same way.

> > For instance, on i.MX6/7 devices with GPU, the gl-renderer needs to be
> > enabled for the fbdev-backend which was removed from upstream weston
> > long time ago.
> > 
> > Also, weston-imx adds support for G2D which is enabled by default, this
> > patch makes sure to disable it when imx-gpu-g2d isn't selected.
> > 
> > The tag version rel_imx_4.9.51_8mq_ga proved to work fine on both
> > i.MX6Q/DL and i.MX8MQ processors.
> > 
> > Here are the commands used to start weston on i.MX6Q:
> > - Using 3D GPU (gl-renderer):
> >  # weston --tty=1 --device=/dev/fb0
> > - Using 2D GPU (G2D):
> >  # weston --tty=1 --device=/dev/fb0 --use-g2d=1
> > 
> > Upstream repository:
> > https://source.codeaurora.org/external/imx/weston-imx/
> > 
> > Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
> > ---
> >  package/weston/weston.hash |  2 ++
> >  package/weston/weston.mk   | 15 +++++++++++++++
> >  2 files changed, 17 insertions(+)
> > 
> > diff --git a/package/weston/weston.hash b/package/weston/weston.hash
> > index 55d87ac3b1..c854e66cea 100644
> > --- a/package/weston/weston.hash
> > +++ b/package/weston/weston.hash
> > @@ -3,3 +3,5 @@ md5 33709aa4d5916f89643fca0fc0064b39  weston-4.0.0.tar.xz
> >  sha1 df1da4a880920c515162e95b18f3709a46690be7  weston-4.0.0.tar.xz
> >  sha256 a0fc0ae7ef83dfbed12abfe9b8096a24a7dd00705e86fa0db1e619ded18b4b58  weston-4.0.0.tar.xz
> >  sha512 0af41016ff4eae85779f95b5c5e44b9683f4ef681a8e52256efeebfa38073082b83e039d0db3c94ac22f22f8d8314c9d6cd16611144b260b353fc5bfdd1ded19  weston-4.0.0.tar.xz
> > +# locally computed
> > +sha256 0f0de7b7b1f65870139c95dde7abc19ed305631ae7c5d37c386db40cde108632  weston-rel_imx_4.9.51_8mq_ga.tar.gz
> > diff --git a/package/weston/weston.mk b/package/weston/weston.mk
> > index efe12bc01e..1248e4a6ef 100644
> > --- a/package/weston/weston.mk
> > +++ b/package/weston/weston.mk
> > @@ -4,9 +4,16 @@
> >  #
> >  ################################################################################
> >  
> > +ifeq ($(BR2_PACKAGE_IMX_GPU_VIV_OUTPUT_WL),y)
> 
>  It is not entirely impossible that people use Buildroot to generate a rootfs
> that should be able to boot on several CPUs, in which case they may not want to
> use the NXP fork of weston. So, I'd make an explicit Config.in option for this:
> 
> if BR2_PACKAGE_WESTON
> config BR2_PACKAGE_WESTON_IMX
> 	bool "weston-imx"
> 	default y
> 	depends on BR2_PACKAGE_IMX_GPU_VIV_OUTPUT_WL
> 	help
> 	  On i.MX6/7/8 platforms, a fork of weston can be used that enables
> 	  various optimisations for this platform. Blah blah blah.
> 
> 	  http://some-url-that-explains-it
> endif
> 
>  But maybe I'm exaggerating.

I'm ok with that approach too. I'll let the others comment before making
the V2.

Yann, Thomas, what do you prefer?

> > +WESTON_VERSION = rel_imx_4.9.51_8mq_ga
> > +WESTON_SITE = https://source.codeaurora.org/external/imx/weston-imx
> > +WESTON_SITE_METHOD = git
> > +WESTON_AUTORECONF = YES
> > +else
> >  WESTON_VERSION = 4.0.0
> >  WESTON_SITE = http://wayland.freedesktop.org/releases
> >  WESTON_SOURCE = weston-$(WESTON_VERSION).tar.xz
> > +endif
> >  WESTON_LICENSE = MIT
> >  WESTON_LICENSE_FILES = COPYING
> >  
> > @@ -50,6 +57,14 @@ else
> >  WESTON_CONF_OPTS += --disable-weston-launch
> >  endif
> >  
> > +ifeq ($(BR2_PACKAGE_IMX_GPU_VIV_OUTPUT_WL),y)
> > +ifeq ($(BR2_PACKAGE_IMX_GPU_G2D),y)
> 
> WESTON_CONF_OPTS += --enable-imxg2d

Ok, I forgot you wanted that --enable-xxx options to be explicit
although it ends up in the 'unrecognized options'.

>  With that:
>  Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

Thanks for the review. I'll add that in the V2 once it is decided what
to do about the config option.

Regards,
Gary

  reply	other threads:[~2018-05-17 13:08 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-16 15:51 [Buildroot] [PATCH 0/3] imx: add Wayland GPU libs / Weston-imx support Gary Bisson
2018-05-16 15:51 ` [Buildroot] [PATCH 1/3] imx-gpu-viv: add Wayland backend option Gary Bisson
2018-05-19 20:06   ` Thomas Petazzoni
2018-05-16 15:52 ` [Buildroot] [PATCH 2/3] weston: remove dependency on mesa3d to enable egl Gary Bisson
2018-05-19 20:46   ` Thomas Petazzoni
2018-05-16 15:52 ` [Buildroot] [PATCH 3/3] weston: add weston-imx variant when using imx-gpu-viv Gary Bisson
2018-05-16 21:24   ` Arnout Vandecappelle
2018-05-17 13:08     ` Gary Bisson [this message]
2018-05-17 21:19       ` Arnout Vandecappelle
2018-06-04  9:52         ` Gary Bisson
2018-06-04 10:05           ` Thomas Petazzoni
2019-02-06 16:42             ` Arnout Vandecappelle

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=20180517130810.GA6584@g751.home \
    --to=gary.bisson@boundarydevices.com \
    --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