Openembedded Devel Discussions
 help / color / mirror / Atom feed
From: Martin Jansa <martin.jansa@gmail.com>
To: openembedded-devel@lists.openembedded.org
Subject: Re: [meta-browser][PATCH 0/4] Introduce EGL/GLES and component build patches for Chromium
Date: Mon, 28 Apr 2014 06:05:18 +0200	[thread overview]
Message-ID: <20140428040518.GI2486@jama> (raw)
In-Reply-To: <1398597711-17825-1-git-send-email-dv@pseudoterminal.org>

[-- Attachment #1: Type: text/plain, Size: 5006 bytes --]

On Sun, Apr 27, 2014 at 01:21:47PM +0200, Carlos Rafael Giani wrote:
> This set of patches extends the chromium recipe to allow for EGL&GLESv2 based 2D rendering,
> considerably improving performance. This also adds hardware-accelerated WebGL support.
> 
> Three new PACKAGECONFIG for chromium are introduced:

Also fails when PACKAGECONFIG is empty:

+ tee -a log.world.20140428_030644.log/bitbake.world.log
NOTE: Started PRServer with DBfile: /home/jenkins/oe/shr-core-branches/shr-core/cache/prserv.sqlite3, IP: 127.0.0.1, PORT: 52990, PID: 32499
Parsing recipes...ERROR: Error executing a python function in <code>:

The stack trace of python calls that resulted in this exception/failure was:
File: '<code>', lineno: 9, function: <module>
     0005:__anon_247__home_jenkins_oe_shr_core_branches_shr_core_openembedded_core_meta_classes_package_ipk_bbclass(d)
     0006:__anon_20__home_jenkins_oe_shr_core_branches_shr_core_openembedded_core_meta_classes_debian_bbclass(d)
     0007:__anon_33__home_jenkins_oe_shr_core_branches_shr_core_openembedded_core_meta_classes_devshell_bbclass(d)
     0008:__anon_80__home_jenkins_oe_shr_core_branches_shr_core_openembedded_core_meta_classes_sstate_bbclass(d)
 *** 0009:__anon_46__home_jenkins_oe_shr_core_branches_shr_core_meta_browser_recipes_browser_chromium_chromium_35_0_1883_0_bb(d)
File: '__anon_46__home_jenkins_oe_shr_core_branches_shr_core_meta_browser_recipes_browser_chromium_chromium_35_0_1883_0_bb', lineno: 3, function: __anon_46__home_jenkins_oe_shr_core_branches_shr_core_meta_browser_recipes_browser_chromium_chromium_35_0_1883_0_bb
     0001:def __anon_46__home_jenkins_oe_shr_core_branches_shr_core_meta_browser_recipes_browser_chromium_chromium_35_0_1883_0_bb(d):
     0002:    pkgconfig = d.getVar('PACKAGECONFIG', True)
 *** 0003:    if 'use-egl' in pkgconfig.split():
     0004:        d.appendVar("DEPENDS", " virtual/egl virtual/libgles2")
     0005:
Exception: AttributeError: 'NoneType' object has no attribute 'split'

> * "component-build" : If set, Chromium is built in component mode. Each subsystem of Chromium
>   ends up in a separate shared object. The alternative is the regular mode, which links everything
>   into one big binary. Component builds are useful for development and testing, and consumes
>   significantly less RAM during building. It is essential when building Chromium on a 32-bit OS.
> 
> * "use-egl" : Ordinarily, Chromium will try to use GLX and regular (= Desktop) OpenGL for 2D
>   acceleration. If that fails, software rendering is used. If use-egl is set, it will use EGL
>   instead of GLX and OpenGL ES 2.0 instead of regular OpenGL. This is very useful if the
>   target platform provides hardware acceleration for this API. Chromium will then render all 2D
>   components as well as WebGL contents with hardware acceleration, Canvas being an exception
>   (see below).
> 
> * "ignore-lost-context" : Without this flag, HTML Canvas will not be accelerated, even with the
>   "use-egl" flag set. The reason for this is a conceptual problem: HTML Canvas expects the
>   backing store of the canvas to persist until the web page goes away. But if the backing store
>   is a GLES texture, then in theory the associated EGL context could get lost (for example, during
>   a power management cycle), and the texture along with it. This violates requirements for canvas.
>   So, one has to choose: either remain fully standards compliant, and have no acceleration, or
>   have acceleration, and risk losing the Canvas content. Note that context losses do not happen on
>   all platforms (they haven't been observed with Vivante GPUs for example, and internals of its
>   galcore module do not show anything that could cause a lost context). But since these recipes
>   do not depend on a specific platform, this option must be explicitely enabled.
> 
> Carlos Rafael Giani (4):
>   Add component-build PACKAGECONFIG option
>   Add variable to contain extra command-line arguments
>   Add use-egl PACKAGECONFIG to let chromium use EGL automatically
>   Add ignore-lost-context PACKAGECONFIG to ignore lost EGL contexts
> 
>  .../chromium/chromium/component-build.gypi         |  5 ++++
>  recipes-browser/chromium/chromium/google-chrome    | 11 +++++---
>  .../remove-linux-accel-canvas-from-blacklist.patch | 32 ++++++++++++++++++++++
>  recipes-browser/chromium/chromium_35.0.1883.0.bb   | 27 ++++++++++++++++--
>  4 files changed, 69 insertions(+), 6 deletions(-)
>  create mode 100644 recipes-browser/chromium/chromium/component-build.gypi
>  create mode 100644 recipes-browser/chromium/chromium/remove-linux-accel-canvas-from-blacklist.patch
> 
> -- 
> 1.8.3.2
> 
> -- 
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-devel

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

      parent reply	other threads:[~2014-04-28  4:05 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-04-27 11:21 [meta-browser][PATCH 0/4] Introduce EGL/GLES and component build patches for Chromium Carlos Rafael Giani
2014-04-27 11:21 ` [meta-browser][PATCH 1/4] Add component-build PACKAGECONFIG option Carlos Rafael Giani
2014-04-27 11:21 ` [meta-browser][PATCH 2/4] Add variable to contain extra command-line arguments Carlos Rafael Giani
2014-04-28 15:17   ` Otavio Salvador
2014-04-27 11:21 ` [meta-browser][PATCH 3/4] Add use-egl PACKAGECONFIG to let chromium use EGL automatically Carlos Rafael Giani
2014-04-28 15:14   ` Otavio Salvador
2014-04-27 11:21 ` [meta-browser][PATCH 4/4] Add ignore-lost-context PACKAGECONFIG to ignore lost EGL contexts Carlos Rafael Giani
2014-04-28  3:01 ` [meta-browser][PATCH 0/4] Introduce EGL/GLES and component build patches for Chromium Martin Jansa
2014-04-28  4:05 ` Martin Jansa [this message]

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=20140428040518.GI2486@jama \
    --to=martin.jansa@gmail.com \
    --cc=openembedded-devel@lists.openembedded.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