All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Seiderer <ps.report@gmx.net>
To: buildroot@busybox.net
Subject: [Buildroot] [RFC v4 3/3] package/qt5/qt5wayland: fix compile for rpi3
Date: Sat, 11 Apr 2020 17:24:54 +0200	[thread overview]
Message-ID: <20200411172454.707a25ac@gmx.net> (raw)
In-Reply-To: <20200411150347.2dd79d94@windsurf.home>

Hello Thomas,

On Sat, 11 Apr 2020 15:03:47 +0200, Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote:

> On Fri,  3 Apr 2020 23:23:57 +0200
> Peter Seiderer <ps.report@gmx.net> wrote:
> 
> > Add three patches to fix compile for rpi3.
> > 
> > Signed-off-by: Peter Seiderer <ps.report@gmx.net>  
> 
> I'm not super at ease with the patches. Especially, I'm always worried
> about a commit that fixes things for one particular platform: patches 1
> and 3 don't seem at all related to Rpi3 specifically.

That is why this is sent as an extra patch (and as RFC)...should have marked
it as WIP ;-)
 
> 
> > diff --git a/package/qt5/qt5wayland/0001-fix-libdrm-related-configure-test-failure.patch b/package/qt5/qt5wayland/0001-fix-libdrm-related-configure-test-failure.patch
> > new file mode 100644
> > index 0000000000..b4d0a2c302
> > --- /dev/null
> > +++ b/package/qt5/qt5wayland/0001-fix-libdrm-related-configure-test-failure.patch
> > @@ -0,0 +1,56 @@
> > +From ba93ccba2aa70bd76f0af43675fa5864e166e3c3 Mon Sep 17 00:00:00 2001
> > +From: Peter Seiderer <ps.report@gmx.net>
> > +Date: Fri, 3 Apr 2020 21:52:30 +0200
> > +Subject: [PATCH 1/3] fix libdrm related configure test failure  
> 
> Nit: use git format-patch -N to generate patches.

Ups..., will do....

> 
> > +
> > +Fixes:
> > +
> > +  Project ERROR: Test config.qtwayland_client.tests.dmabuf-server-buffer tries to use undeclared library 'drm'
> > +  Project ERROR: Test config.qtwayland_compositor.tests.dmabuf-server-buffer tries to use undeclared library 'drm'
> > +  Project ERROR: Test config.qtwayland_compositor.tests.dmabuf-client-buffer tries to use undeclared library 'drm'
> > +
> > +Signed-off-by: Peter Seiderer <ps.report@gmx.net>  
> 
> Should the "drm" library be declared somewhere instead? How is this
> building for upstream?

Seems they always build with libdrm available....

Did not find the right way to fix it for the qt configure framework,
but opened an upstream bug report [1], [2]...but no feedback yet...

> 
> > diff --git a/package/qt5/qt5wayland/0002-brcm-egl-needs-wayland-scanner-client-wayland-protoc.patch b/package/qt5/qt5wayland/0002-brcm-egl-needs-wayland-scanner-client-wayland-protoc.patch
> > new file mode 100644
> > index 0000000000..f14c374e76
> > --- /dev/null
> > +++ b/package/qt5/qt5wayland/0002-brcm-egl-needs-wayland-scanner-client-wayland-protoc.patch
> > @@ -0,0 +1,36 @@
> > +From 4770af94884be5a603934b36e6804eed7d18d43c Mon Sep 17 00:00:00 2001
> > +From: Peter Seiderer <ps.report@gmx.net>
> > +Date: Fri, 3 Apr 2020 22:42:23 +0200
> > +Subject: [PATCH 2/3] brcm-egl: needs
> > + wayland-scanner-client-wayland-protocol-include
> > +MIME-Version: 1.0
> > +Content-Type: text/plain; charset=UTF-8
> > +Content-Transfer-Encoding: 8bit
> > +
> > +Fixes:
> > +
> > +  wayland-brcm-client-protocol.h: In function ?wl_buffer* qt_brcm_create_buffer(qt_brcm*, int32_t, int32_t, wl_array*)?:
> > +  wayland-brcm-client-protocol.h:115:29: error: ?wl_buffer_interface? was not declared in this scope; did you mean ?qt_brcm_interface??
> > +    115 |     QT_BRCM_CREATE_BUFFER, &wl_buffer_interface, NULL, width, height, data);
> > +        |                             ^~~~~~~~~~~~~~~~~~~
> > +        |                             qt_brcm_interface
> > +
> > +Signed-off-by: Peter Seiderer <ps.report@gmx.net>  
> 
> A bit more explanation on the relationship between the build issue and
> the fix would be good.

Will try give an better explanation...., positive feedback from
upstream [3], [4]..., only the target branch seems questionable...

> 
> > diff --git a/package/qt5/qt5wayland/0003-examples-fix-wayland-texture-sharing-custom-composit.patch b/package/qt5/qt5wayland/0003-examples-fix-wayland-texture-sharing-custom-composit.patch
> > new file mode 100644
> > index 0000000000..1f0f919c93
> > --- /dev/null
> > +++ b/package/qt5/qt5wayland/0003-examples-fix-wayland-texture-sharing-custom-composit.patch
> > @@ -0,0 +1,42 @@
> > +From f7d96ddd00f868810ed2ca260acd41435717e6b8 Mon Sep 17 00:00:00 2001
> > +From: Peter Seiderer <ps.report@gmx.net>
> > +Date: Fri, 3 Apr 2020 22:40:29 +0200
> > +Subject: [PATCH 3/3] examples: fix wayland/texture-sharing/custom-compositor
> > +MIME-Version: 1.0
> > +Content-Type: text/plain; charset=UTF-8
> > +Content-Transfer-Encoding: 8bit
> > +
> > +Fixes:
> > +
> > +  main.cpp:83:33: error: ?GL_RGBA8? was not declared in this scope; did you mean ?GL_RGBA4??
> > +     83 |             *glInternalFormat = GL_RGBA8;
> > +        |                                 ^~~~~~~~
> > +        |                                 GL_RGBA4
> > +  main.cpp:120:33: error: ?GL_RGBA8? was not declared in this scope; did you mean ?GL_RGBA4??
> > +    120 |             *glInternalFormat = GL_RGBA8;
> > +        |                                 ^~~~~~~~
> > +        |                                 GL_RGBA4
> > +
> > +Signed-off-by: Peter Seiderer <ps.report@gmx.net>
> > +---
> > + examples/wayland/texture-sharing/custom-compositor/main.cpp | 4 ++++
> > + 1 file changed, 4 insertions(+)
> > +
> > +diff --git a/examples/wayland/texture-sharing/custom-compositor/main.cpp b/examples/wayland/texture-sharing/custom-compositor/main.cpp
> > +index a39c8c3..af37502 100644
> > +--- a/examples/wayland/texture-sharing/custom-compositor/main.cpp
> > ++++ b/examples/wayland/texture-sharing/custom-compositor/main.cpp
> > +@@ -63,6 +63,10 @@
> > + 
> > + #include "QtWaylandCompositor/private/qwltexturesharingextension_p.h"
> > + 
> > ++#ifndef GL_RGBA8
> > ++#define GL_RGBA8 0x8058
> > ++#endif  
> 
> This feels like a hack. What should normally be providing the GL_RGBA8
> define? The OpenGL headers?

Same 'hack' is used at some other Qt locations and patch is
upstream accepted [5], [6] as is ;-)

Will wait some time for more feedback...., and provide an updated
buildroot patch soon...

Regards,
Peter

[1] https://bugreports.qt.io/browse/QTBUG-83303
[2] https://codereview.qt-project.org/c/qt/qtwayland/+/296213
[3] https://bugreports.qt.io/browse/QTBUG-83305
[4] https://codereview.qt-project.org/c/qt/qtwayland/+/296215
[5] https://bugreports.qt.io/browse/QTBUG-83304
[6] https://codereview.qt-project.org/c/qt/qtwayland/+/296214

> 
> Thomas

  reply	other threads:[~2020-04-11 15:24 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-03 21:23 [Buildroot] [RFC v4 1/3] package/qt5: bump version to 5.13.2 Peter Seiderer
2020-04-03 21:23 ` [Buildroot] [RFC v4 2/3] package/qt5: bump version to 5.14.2 Peter Seiderer
2020-04-11 13:01   ` Thomas Petazzoni
2020-04-03 21:23 ` [Buildroot] [RFC v4 3/3] package/qt5/qt5wayland: fix compile for rpi3 Peter Seiderer
2020-04-11 13:03   ` Thomas Petazzoni
2020-04-11 15:24     ` Peter Seiderer [this message]
2020-04-11 13:01 ` [Buildroot] [RFC v4 1/3] package/qt5: bump version to 5.13.2 Thomas Petazzoni
2020-04-11 15:09   ` Peter Seiderer

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=20200411172454.707a25ac@gmx.net \
    --to=ps.report@gmx.net \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.