Upstream explicitly say in their CONFIG.md file to build with -Os:
Make sure you build with your compiler's "optimize for size" option.
On `gcc` this is `-Os` [ ... ] HarfBuzz heavily uses inline functions
and the optimize-size flag can make the library smaller by 20% or
more. Moreover, sometimes, based on the target CPU, the optimize-size
builds perform *faster* as well, thanks to lower code footprint and
caching effects
Drop the patch to build just hb-subset-plan-layout.cc with -Os (which
was a workaround for a GCC bug), and pass -Os globally.
This manages to reduce the duration to harfbuzz:do_compile on my machine
from 75s to 47s, and has a big impact on the library sizes:
harfbuzz: PKGSIZE changed from 1769358 to 1237070 (-30%)
harfbuzz-dbg: PKGSIZE changed from 84920168 to 71203208 (-16%)
harfbuzz-subset: PKGSIZE changed from 1579247 to 940191 (-40%)
(From OE-Core rev: d795b3e16ed01d7273d4f3779684eb41cc16f809)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
===========
- fontconfig: Improve the add_font_file implementation
- fontconfig: Combine font features and style variants
- fontconfig: Make sure font faces stay alive
- win32: Drop some caching
- win32: Make sure font faces stay alive
- win32: Modernize and simplify the code
- win32: Stop synthesizing fonts
- win32: Implement list models
- coretext: Support synthetic small caps
- layout: Avoid assertions in line breaking
- build: Require GLib 2.82
(From OE-Core rev: c2658fb4b877fe5144446f5498cd3115418dbf15)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Antonin Godard <antonin.godard@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
libglvnd recipe was sent before the OE-Core changes reworking S /
UNPACKDIR rework. Drop S assignment which is now being rejected by OE /
bitbake.
Fixes: 621e9be9663e ("libglvnd: migrate from meta-oe")
(From OE-Core rev: ec3cd6a73468510e57c4cda874dc3cb20afef9c2)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Let the distro to be a central authority regarding GLVND feature.
Make mesa.inc enable glvnd PACKAGECONFIG if it is enabled in
DISTRO_FEATURES.
(From OE-Core rev: 25d29420919caadae4d35e7eb0291dd3504d8552)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
libglvnd is a vendor-neutral approach to handling OpenGL / OpenGL ES /
EGL / GLX libraries. It has been proposed and initially implemented by
NVIDIA in order to simplify coinstallation of GL drivers provided by
different vendors. Major Linux distributions (Debian, Ubuntu, Fedora,
etc.) have already switched to libglvnd.
Having libglvnd in OE-Core simplifies integration of this ABI into
distro and BSP layers (e.g. it will help meta-tegra, which currently
provides its own version of the recipe).
Import recipe for libglvnd from the meta-oe layer (changes: fix virtual
package names, add add the SUMMARY data, drop git@ from SRC_URI, require
glvnd DISTRO_FEATURE).
(From OE-Core rev: 621e9be9663e448dde3f6b6a02b413fd65983e6c)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The BCP for virtual runtime package names is virtual-foo. Make mesa
follow the established convention and change the names of the provied
packages to virtual-foo-icd.
Fixes: 9d3b4c9bc403 ("mesa: sort out PROVIDES for the glvnd case")
(From OE-Core rev: d69e06dbd8575e258a4542f34c5eaba413da7750)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some of them were introduced by mass-removal of S = WORKDIR/git assignments;
rather than try to fix up (or redo) just these, I've run this sed command over
the whole tree:
sed -i -z -E 's/([ \t\f\v\r]*\n){3,}/\n\n/g' `find . -name *.bb -o -name *.inc`
The rationale is that more than one empty line is wasting vertical screen space, and
does nothing for readability.
(From OE-Core rev: cedc4ff7c9bcfb22a20e43e47f9759f4007a4f1a)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There is no need to have two assignments; also setting S from
UNPACKDIR (and not vice versa) is more logical as unpack
task comes first, and tasks that come after unpack use S to access
unpack results.
Also adjust reference test data in selftest/recipeutils to reflect changes in test recipes.
(From OE-Core rev: f64b7e5fb3181734c8dde87b27e872a026261a74)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Removing all the S = ${WORKDIR}/git assignments works because BB_GIT_DEFAULT_DESTSUFFIX
is set to match S from bitbake.conf (which itself is set to match typical tarball
releases).
A few recipes are setting S to a sub-directory of the git tree and need
to be adjusted accordingly.
bzip2 recipe is fetching a tarball and separately cloning tests;
adjust the recipe to put the latter into 'bzip2-tests', instead of 'git'.
devupstream.bbclass no longer needs to rewrite S, and is adjusted accordingly.
Adjust scripts/lib/recipetool/append.py to not hardcode 'git' as unpack
destination.
Adjust kernel-yocto.bbclass to use the git unpack variable instead
of hardcoding 'git' (there's also removal of repetition of
string constants and a correction of workdir/unpackdir mismatch in
one of the if-else branches).
Ensure build-appliance-image recipe does not use 'git' as checkout directory for
poky repo, but rather explicitly name it 'poky'.
Ensure reproducible.py code that looks for git repositories does not
hardcode 'git' but uses the destination set by BB_GIT_DEFAULT_DESTSUFFIX.
Ensure recipetool does not write out unneeded S settings into newly
created recipes that fetch from git.
Adjust selftest to not hardcode 'git' as unpack directory.
(From OE-Core rev: f80c07019ddadaf9c5fb890faabfda7920ecd15e)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
- Hardened the libjpeg API against hypothetical calling applications that may
erroneously change the value of the 'data_precision' field in
'jpeg_compress_struct' or 'jpeg_decompress_struct' after calling
'jpeg_start_compress()' or 'jpeg_start_decompress()'.
(From OE-Core rev: 49a69532bc2b4a7d6ae8136123e92b10b5e657f5)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
While at it, also add libpam as a dependency to the vnc pkgconf, to
avoid the following error:
| Run-time dependency pam found: NO (tried pkgconfig and cmake)
|
| ../weston-14.0.2/libweston/meson.build:95:15: ERROR: C shared or static library 'pam' not found
Changelog:
- frontend: Fix crash in output resize handler
- libweston: fix bitshift in weston_idalloc_get_id()
- libweston: fix realloc in weston_idalloc
- libweston: refactor update_lowest_free_bucket()
- libweston-desktop: don't try to move child surfaces to not existing layer
- libweston/input.c: Fix weston crash with the mouse event
- windowed-output-api.h: Provide ARRAY_LENGH() for windowed-output-api
- drm: Fix underlay test
- compositor: Mark pnode accordingly when buffer type is direct
- compositor: re-order paint node placeholder checks
- gl-renderer: Take direct-display into consideration
- shared: fix binding-modifier none
- vnc: Allow neatvnc in version 0.9.0
- compositor: Prevent startup crash when hdcp mode is set on display
- libweston/desktop: Avoid a potential crash on invalid resource
- libweston/desktop: Don't destroy the xdg_surface
(From OE-Core rev: 0ebf7dc82498e7dfabb45fd4e6f7b80375a93835)
Signed-off-by: Gyorgy Sarvari <skandigraun@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The Nouveau driver is used for NVIDIA GPUs and thus doesn't really need
to be always enabled when building gallium-llvm drivers.
So let's guard it with a nouveau PACKAGECONFIG.
The only intended change is nouveau not being build on target mesa on
non-x86/i686 machines when building gallium-llvm drivers as I assume
NVIDIA GPUs aren't that common on other CPU architectures (tegra already
bringing in the nouveau driver should handle the NVIDIA SoCs that would
benefit from nouveau driver and thus are not impacted by this change).
(From OE-Core rev: 22e7f562fb4b5becc342fe7d60c37c70e42ccf6b)
Signed-off-by: Quentin Schulz <quentin.schulz@cherry.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It doesn't make much sense to always build the r300 AMD gallium LLVM
driver as that's HW-specific.
Instead, let's guard it with the amd PACKAGECONFIG.
This will only disable R300 for target mesa on non-x86/i686 machines
where gallium-llvm PACKAGECONFIG is selected but not amd, otherwise
behavior is left unchanged.
(From OE-Core rev: 28d41017ffd03d0bcc4ef0272a95d19412488624)
Signed-off-by: Quentin Schulz <quentin.schulz@cherry.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
six was dropped as a dependency five years ago:
288e7b3b74
(From OE-Core rev: 915acbfb2e10a3e50d1ddc984b8e9c363dac6f78)
Signed-off-by: Guðni Már Gilbert <gudni.m.g@gmail.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
tests: Add test for GL_EXT_shader_clock
tests: Add test for GL_EXT_shader_realtime_clock
vulkan: Add cmat tests using constant as accumulator
tests: test UBO and SSBO backed by the same buffer
arb_direct_state_access: fix internaformat query behavior
khr_texture_compression_astc: normalize error-color
tests/util: provide simple array for cleanup handlers
ext_external_objects: use piglit_set_destroy_func instead of atexit
ext_external_objects: fix timeline semaphore validation error
ext_external_objects: fix validation error with barriers
ext_external_objects: fix validation error on cmdbuf usage
ext_external_objects: fix validation error releasing resources
ext_external_objects: fix validation error for cmdbuffer usage
ext_external_objects: fix validation error with semaphore value
arb_shading_language_420pack: test uniform block matching
gl: test push pop attribute with color material
GL_EXT_texture_integer: basic validation tests
cmake: Fix mis-matching arguments warning
fix invalid typecast
(From OE-Core rev: c09a0cec08e1212ba82d0702686e0e2991cfd03a)
Signed-off-by: Gyorgy Sarvari <skandigraun@gmail.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mesa 25.1.3 Release Notes / 2025-06-07
Mesa 25.1.3 is a bug fix release which fixes bugs found since the 25.1.2 release.
Mesa 25.1.3 implements the OpenGL 4.6 API, but the version reported by glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used. Some drivers don’t support all the features required in OpenGL 4.6. OpenGL 4.6 is only available if requested at context creation. Compatibility contexts may report a lower version depending on each driver.
Mesa 25.1.3 implements the Vulkan 1.4 API, but the version reported by the apiVersion property of the VkPhysicalDeviceProperties struct depends on the particular driver being used.
New features
None
Bug fixes
25.1.2 - instant crash on upgrade, very first attempt at a new surface
Mesa 25.1.2 Release Notes / 2025-06-04
Mesa 25.1.2 is a bug fix release which fixes bugs found since the 25.1.1 release.
Mesa 25.1.2 implements the OpenGL 4.6 API, but the version reported by glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used. Some drivers don’t support all the features required in OpenGL 4.6. OpenGL 4.6 is only available if requested at context creation. Compatibility contexts may report a lower version depending on each driver.
Mesa 25.1.2 implements the Vulkan 1.4 API, but the version reported by the apiVersion property of the VkPhysicalDeviceProperties struct depends on the particular driver being used.
New features
None
Bug fixes
Confidential issue #13281
anv, regression: Missing terrain in It Takes Two on BMG
“breaking-limit” benchmark will Freeze before starting.
rusticl: CL_MEM_HOST_READ_ONLY cl_image2d reads zeroes from host on AMD Vega8
Bug (bad code optimization?) in the GLSL (compute) shader compiler
nvk: Broken lighting in Trine 5
RX9070 hard crash with Mafia Definitive Edition
RADV: Potential bug with vulkan fragment shader interpolation (on outputs from mesh shaders?)
In the game “Foundation” a buildings areas of effect is missing
ANV: Dota 2 May 22 2025 update crashing in vkCmdBindDescriptorSets with no validation error
Vulkan Video engages during playback of format which is not supported by my Fiji GPU
Weston hangs on virtio (non-accelerated version) on QEMU with Mesa main
KVM/qemu: GDM fails to start / gnome-shell crashes after update to mesa-dri-drivers-25.0.3
(From OE-Core rev: da2c318f530a29b42c36fe8d5015fcdd3e1c1420)
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This was being pulled in through python by accident. It no longer needs
it but exposes this missing dependency.
(From OE-Core rev: 6334fac0a0b0783298957e2ccbe3a27490f7da09)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
PIGLIT_BUILD_DIR was set to source tree location relative to B as a
reproducibility workaround some years ago. Aafter some investigation
I found where the actual issue is (addressed by the patch).
Also, setting PIGLIT_BUILD_DIR relative to ${B} in a hardcoded way
is problematic as S and B are not necessarily related that way (e.g. when
sources are in UNPACKDIR or somewhere else entirely).
(From OE-Core rev: 7625176fedacede86fda84a47dd76f62a14cb8e0)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Development has moved to freedesktop.org, so switch SRC_URI.
Drop an upstreamed patch.
(From OE-Core rev: 13f5ba6b9d5cee5d704e669fb80e953bd97b25f1)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Support for OSMesa has been removed in the latest Mesa versions, drop
corresponding PACKAGECONFIG option from mesa-demos. Note, there is no
need to pass -Dosmesa=disabled, meson will detect that the library is
not available.
Suggested-by: Khem Raj <raj.khem@gmail.com>
(From OE-Core rev: 27e7e3bff40a884044a19e52d0eaa35a80880464)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fixes build with clang-20 as it updates fmt to latest
complete changes are here [1]
[1] eca122602a...d27205d14d
(From OE-Core rev: a0df43a0f117a53e9008ca6e970a3e1d1ff55367)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Extend mesa PACKAGECONFIG with new option, teflon, enabling building of
the TFLite delegate, libteflon.so. Currently it supports only
VeriSilicon NPUs.
Suggested-by: Fathi Boudra <fathi.boudra@linaro.org>
(From OE-Core rev: d639f2e3e6922efe845cf84fb35d729167693adb)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
XA tracker is unmaintained and deprecated. It has been removed in 25.2
branch and it is going to be disabled by default in the 25.1.2 release.
Be slightly more proactive, pick up the patch disabling XA by default
and drop support for XA tracker in the recipe.
(From OE-Core rev: e8bba3efd72bf90d70e674b0686d13639a44f0ad)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
osmesa has been removed from recently upgrade mesa recipes
remove it from the fallback for non-x11 systems
(From OE-Core rev: 7ee8e99a52044e18cd35bde8d280274ce44e26d4)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Clang complains/warns about mixing -ffp-contract=fast and -ffp-model=precise
lower this to warning only, this does not impact code generation, we still
get the diagnostic messge but wont break the build
(From OE-Core rev: 0fc79e9b7aea87de87f461eebaaa2c00c9ec4a77)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The GL/internal/dri_internal.h header is not a part of the OpenGL API.
Instead it defines Mesa interface with DRI drivers. Move it to mesa-dev
package to reside next to dri.pc.
(From OE-Core rev: 72ad9950f0b6c44068e52be3bb3a015bd3affa06)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If GLVND is enabled, Mesa provides two ICD libaries, libEGL_mesa.so.0
and libGLX_mesa.so.0. During the installation it also creates two extra
symlinks, libEGL_mesa.so and libGLX_mesa.so, however those symliks are
unnecessary and useless as nothing will ever link to those ICD archives.
Remove them from the install dir, which also makes package
libglx-mesa-dev disappear (it contained only this symlink).
(From OE-Core rev: badb1ee13ec87923aa6c4cdeb90756cc370e10bc)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If mesa is built with GLVND enabled, it doesn't provide GL / GL ES / EGL
libraries directly. Instead it provides two ICD libraries: libEGL_mesa.0
and libGLX_mesa.so.0. Remove virtual provides from the glvnd case
(dropping incorrect virtual/libglx provider while we are at it) and
replace those with runtime providers (to be used by libglvnd in order to
pull corresponding ICDs).
(From OE-Core rev: 9d3b4c9bc40392ba87f110ec5db0dedf381b8c4a)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If glvnd is enabled, mesa packages do not provide full library
implementations (instead it provides two ICDs, one for EGL, one for
GLX). This means that there is no more conflict between mesa packages
and other vendor packages. Stop setting those extra tags for mesa
packages.
(From OE-Core rev: b1b8a0f69dafe23c992754cffb7aaf575753e564)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With Clover being gone, there are no more separate gallium-pipe
libraries. Drop corresponding parts of the recipe.
(From OE-Core rev: b668f51f0384029892e8a87935c5fad26dd4e690)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The libopencl-mesa package provides OpenCL ICDs (Installable Client
Drivers). As such, there is no conflict between several packages
providing ICDs for different vendors. Split the loop that modifies
package metadata and stop adding extra RPROVIDES / RCONFLICTS /
RREPLACES tags to the libopencl-mesa package.
(From OE-Core rev: 6849af6df74f4a30ab7820a698598932832498dc)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Thre is no "development" package for libopencl-mesa. The main package
contains Installable Client Drivers, for which there are no headers and
the extra .so file isn't supposed to be linked with anything.
Drop the useless and confusing libopencl-mesa-dev package.
(From OE-Core rev: 2f9ee74eb7dfce569e9c49dca54e0ed4b5b8596d)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
OpenGL ES 3 is closely tied to the OpenGL ES 2: there is no separate
library, one can use GL ES 2 headers for GL ES 3 app, etc. Instead of
having separate packages, merge GL ES 3 packages into GL ES 2 ones.
Suggested-by: Böszörményi Zoltán <zboszor@gmail.com>
(From OE-Core rev: 7c0e9c170db93d9af7cc505dcd71c4f3d6c4e9c5)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
