linux/meta-freescale.git/dynamic-layers, branch dunfell

Move from codeaurora to github

2023-03-31T20:50:30+00:00

Fixes: #1469 Signed-off-by: Tom Hochstein

qt4-imx-support: fix typo

2022-10-07T11:41:39+00:00

Signed-off-by: Mauro Salvini (cherry picked from commit 3abc09fe62ca4de5b16a55a813cc24a9d7c1e919)

qt4-imx-support: change format of patches added by commit 4aed4a5

2022-09-29T14:24:04+00:00

Patches are the same but now are in git format ad have Signed-off-by and From fields as required (cherry picked from commit 24a68cbea9424b7ebc888c402068a09996a5ba17)

qt4-imx-support: fix build for imxgpu2d SOCs

2022-09-28T15:55:21+00:00

Before this change, when imxgpu2d is selected, the qt4 do_configure fails with the following error: | EGL (EGL/egl.h) auto-detection... () | compiling egl.cpp ... | /home/wrk/yocto/build-qt4/tmp/work/imx8mmevk-fslc-linux/qt4-embedded/4.8.7-r0/recipe-sysroot/usr/include/EGL/eglplatform.h:150:10: fatal error: X11/Xlib.h: No such file or directory | 150 | #include | | ^~~~~~~~~~~~ | compilation terminated. | Makefile:178: recipe for target 'egl.o' failed | make: *** [egl.o] Error 1 | EGL (EGL/egl.h) disabled. | EGL (GLES/egl.h) auto-detection... () ... | /home/wrk/yocto/build-qt4/tmp/work/imx8mmevk-fslc-linux/qt4-embedded/4.8.7-r0/recipe-sysroot/usr/include/EGL/eglplatform.h:150:10: fatal error: X11/Xlib.h: No such file or directory | 150 | #include | | ^~~~~~~~~~~~ | compilation terminated. | Makefile:178: recipe for target 'egl4gles1.o' failed | make: *** [egl4gles1.o] Error 1 | EGL (GLES/egl.h) disabled. | The EGL functionality test failed! This is because the -DLINUX=1 and -DEGL_API_FB=1 flags added to QT_CONFIG_FLAGS are not pushed down during these configuration tests. This commit adds these two flags directly to the related .pro files in Qt4 sources. Since it's been a long time that meta-qt4 is not updated anymore, the patch can be backported to LTS releases. Signed-off-by: Mauro Salvini (cherry picked from commit 4aed4a58335af8e96815124f17f47060d1053c48)

recipes: use https protocol and add explicit branch parameter

2021-11-23T15:29:24+00:00

Due to https://github.blog/2021-09-01-improving-git-protocol-security-github/#no-more-unauthenticated-git it is required to use https protocol for github repo accessing. Update created with oe-core/scripts/contrib/convert-srcuri.py (see [0]) Fixes: WARNING: /work/meta-freescale/recipes-bsp/u-boot/u-boot-fslc-mxsboot_2021.07.bb: URL: git://github.com/Freescale/u-boot-fslc.git;branch=2021.07+fslc uses git protocol which is no longer supported by github. Please change to ;protocol=https in the url. [0] - https://git.openembedded.org/openembedded-core/tree/scripts/contrib/convert-srcuri.py Signed-off-by: Pierre-Jean Texier (cherry picked from commit bef00d6e4f25b4a9d3272e0d69db7545590ed204)

opencv: Drop removal of v4l support

2020-09-14T16:26:46+00:00

In the past, i.MX gstreamer camera didn't work well with OpenCV v4l, so v4l was disabled. Now that i.MX gstreamer is no longer using OpenCV, v4l support can be restored. Fixes #475 Signed-off-by: Tom Hochstein (cherry picked from commit 30234f745c55e5f832c282d51bb44604f151e909)

qtbase: Define empty PACKAGECONFIG_VULKAN

2020-06-05T11:07:10+00:00

In cases where non imxgpu and non-fsl SOCs are target do_configure results in ERROR: qtbase-5.15.0+gitAUTOINC+f6fe4bbab7-r0 do_configure: QA Issue: qtbase: in valid PACKAGECONFIG: ${@bb.utils.contains('DISTRO_FEATURES', [invalid-packagecon fig] Therefore having empty value helps in avoiding this case Signed-off-by: Khem Raj

qtbase: Drop upstreamed patch

2020-06-02T12:18:25+00:00

Drop patch 0015-Add-eglfs-to-IMX-GPU.patch. It is upstream and no longer necessary. Signed-off-by: Tom Hochstein

qtbase: Add i.MX 8 configuration

2020-06-02T12:18:25+00:00

- 8M Quad has 3D but no 2D, so fix the override for GPU patches - Add gbm and kms for all i.MX 8 with GPU - Add vulkan for all i.MX 8 with GPU except 8M Mini Signed-off-by: Tom Hochstein

qtbase: Fix patch fuzz for mkspecs/linux-oe-g++/qmake.conf

2020-05-05T14:58:53+00:00

The current version of the patch causes a fuzz because of context change. Signed-off-by: antznin