Age | Commit message (Collapse) | Author |
|
Change-Id: I7a516fb0b7449dd33f95724f311b63387cc42f34
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121869
Tested-by: René Engelhard <rene@debian.org>
Reviewed-by: René Engelhard <rene@debian.org>
|
|
we are loading a spreadsheet in parallel here, but the parallel threads
achievei very little actual concurrency because of heavy contention
in the SharedStringPool mutex.
So switch to a concurrent hash map. I looked at a couple of different
ones (including the Folly one), and this was the one with the simplest
resulting code.
This takes my load time from 12.5s to 8s
Change-Id: I04d6d8e11d613b510eb3bc981f3338819b7ac813
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121717
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ie7481a122436ae765e36642f27afa45b8a11bba3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121761
Reviewed-by: René Engelhard <rene@debian.org>
Tested-by: René Engelhard <rene@debian.org>
|
|
Updating to m91 or newer required handling the problem with Vulkan
crashing, but that's been finally fixed by the previous commit.
This Skia version also requires at least freetype-2.8.1, so disable
Skia if that's not available (LO's minimum is lower). Additionally
patch out Skia's use of TT_SUPPORT_COLRV1, which seems to be
an unstable freetype API from the git version and it doesn't even
compile with the latest stable 2.9.11 release
Change-Id: I4a17effb912468c43fc7cdb69fe453062e4d2447
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120967
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
This part focuses on allowing it on replacing arrayfunctor
By thefault it will try AVX512F (1,17%)
If not available will use AVX (94,77%)
Use of AVX2 (82,28%) has been avoided even if the code could been more compact
Source of hardware statistics: https://store.steampowered.com/hwsurvey
Change-Id: Iae737a565379e82c5f84f3fdee6321ac74f59d40
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115675
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I7a9abde4101164af8c47433acfa35f4f9d3b3d04
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120907
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: Ifad159f99a68cf06426cdd0554df1b68d71cd3fb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120915
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: Ia73117e06f843bcdd228da018fe3f34af6405653
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120621
Tested-by: Jenkins
Reviewed-by: David Tardon <dtardon@redhat.com>
|
|
dependencies need to be built with the msan flags so undo the
--with-system .a usage now that trying to squeeze into the 12 hour build
window isn't critical because the afl 6-8 hour build+check is disabled
Change-Id: I3729d37e0f028c5efe62ce5d66ea6fe636d3d0c2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120788
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I2d0fcf28aa8d2fe937f850a2a54125073a6913d4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120423
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
e.g. the example of
https://bugs.launchpad.net/ubuntu/+source/libreoffice/+bug/1524638
has the same output .urd using cpp with these flags as ucpp created
Change-Id: Iadfd2deba7d365c66c3260fd6736c031987e107c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120347
Tested-by: Jenkins
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I12e44bf2dbeebc0d18e48a22831099c3dad7b12d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120402
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
With PATH essentially serving the role of LD_LIBRARY_PATH under
Windows, there was the notion that this needs to be provided in
Windows notation, for win32 gnumake.
That was perhaps once true; currently we're always evaluating PATH
inside a shell, not the Makefile. So this since a while only worked
accidentally, due to cygwin transparently converting between DOS and
UNIX PATH vars. It did break though for corner-cases, e.g.
SRCDIR!=BUILDDIR, and BUILDDIR e.g. D:\FOO.
With that simplification, also GNUMAKE_WIN_NATIVE can go.
Change-Id: Ied5a0443dc70e7dc629c0c0620e6ce911d9a73d0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/119941
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <thorsten.behrens@allotropia.de>
|
|
This only builds the library, without using it.
Change-Id: I54e940bcaa61852b831dbe7ae5660a5daf5fe609
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/119356
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Switch to '-z $LIB_FUZZING_ENGINE' to check if LIB_FUZZING_ENGINE is
set. The previous version evaluates to false when LIB_FUZZING_ENGINE is
not set, meaning you would not be warned at configure time when using
-enable_fuzzers without setting LIB_FUZZING_ENGINE.
Original broken version landed in:
44b36a0602b04342566362bce3f6bed7d2b096e4
( https://gerrit.libreoffice.org/c/core/+/111691 )
Change-Id: Ic2174777ebf724e471cd605ba54b245e4e804705
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/118372
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
So we don't build 3.8.8rc1 anymore. I didn't look into 3.9.
Change-Id: Ife7d898c913b9b164168b0ef23a055deea55815f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117757
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Regression from commit 14069d84174ca7a4e60db4d75912903e9679b643
("configure: Improve help-options handling").
We now must forward the help and extension configure options, as
this is needed by the cross toolset to know, if it needs to build
the HelpIndexer and HelpLinker executables.
Also get rid of gb_Helper_optional_for_host. While it's in theory
more correct, it makes gb_Helper_optional more complex, so we now
rely on the fact, that the config_build.mk BUILD_TYPE is a subset
of the config_host.mk BUILD_TYPE.
This should fix the Windows Arm64 build, which currently seem to
be the only cross-build including help.
Change-Id: I8b90923c891b43cd2655b2578015c9bd579aae7d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117758
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: Ia1d0c358868e899e121b8331fad17ce36d454f10
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117668
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
|
|
This patch includes:
* A README.help.md as a general documentation for the different
help types, the LO code and help-related build options.
* Adds --disable-xmlhelp for removing the xmlhelp support from the
build. Disable for iOS, Android and Emscripten. This was partly
included in HAVE_FEATURE_DESKTOP before.
* Rename WITH_HELP define to WITH_HELPPACKS, which reflects the
actual usage AFAIK.
* Depend --with-omindex on --with-help=online and don't override
the --with-help setting. Error out on conflicting options.
* Depend --with-helppack-integration on build help variants, which
actually result in help packs.
Kind of reverts commit 2c38ea6d16b910294220cefaf8ae6a0683e6405a
("Building without --with-help is not supposed to disable help
functionality").
Change-Id: Ie4cb73905b3ed94e991d9f1bd75cfbd6de9da385
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116222
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
I missed that PostgreSQL can be build on MacOSX with OpenLDAP
support. Stephan had an incremental build failure because of it.
But instead of opting-in using test_openldap=yes on MacOSX, like
on Haiku, this drops the using_x11 dependency and defaults to
"yes", like it was before. This was more of a convenience then
reality anyway, and it makes really not much sense to disable
OpenLDAP in case of --without-x.
Change-Id: I959a3b30f1c369264174302f18267a4eb40a3a52
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117046
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: I22adfa913dad2a4ae5d6fbeab87daf22f5545282
|
|
Change-Id: If318cd33356c0082a63cbd941cdfc753c9d800e3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116977
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
since...
commit ffc6d564b91692cd9d99a0eb3b1ceaf54f7db89e
Date: Tue Jun 1 07:40:12 2021 +0200
configure: Refactor disabling DB drivers
Change-Id: I94e5e0aca54e1d6d0355d63b27d8fd73dd433083
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116964
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I330065f61d2d0fdfeeaeba4ee2e739e222d1c665
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116918
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
missing $ for test_gtk4 in...
commit 3d1f28dd1fbb2fe3f5b933a9d692fb4d033f08b6
Date: Sun May 30 18:19:40 2021 +0200
configure: Refactor platform defaults
Change-Id: I8f08ab5945fbd98e2ff9eeede83bfad324b1c44a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116558
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
dropped by...
commit 3d1f28dd1fbb2fe3f5b933a9d692fb4d033f08b6
Date: Sun May 30 18:19:40 2021 +0200
configure: Refactor platform defaults
so cairo and pixman get downloaded and built out of the box with --enable-gtk3
Change-Id: I5f0b05ccaca008b9a36c6456811c9e105f13e334
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116550
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Disabling the DB connectivity is incomplete. We can't generally
disable it for the cross-toolset, because svx / gengal then fails
to build. But at least we can always disable most DB drivers in
these cases and just use --disable-database-connectivity in the
cross-toolset when building without galleries (which also gives
this flag some testing with the Android cross-toolset).
This also moves the DB modules in the registry generation target
into a common DBCONNECTIVITY block, to prevent build failures.
Change-Id: Ib98e1e8918cccc3caf9bc04185bac533509329a2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116511
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
The main idea was to move complexity from source (and Makefiles)
into configure.ac, because otherwise these must replicate the
"same" branching, often resulting in diversions. Better to keep
the logic in one place (configure.ac) and set additional variables
to be used directly by source code and Makefiles.
Notably this introduce the "using_*" platform flags, which should
be considered constants. There is USING_X11, which actually tells,
if the build uses the platform's X11 (and I opted for the removal
of HAVE_FEATURE_X11). I also consider variables constant, after
they have been exported by AC_SUBST, which should never be
conditional, or some AC_DEFINE set them for a config header.
A large block of defaults depends on $using_x11, so we set them to
the same value, but just if the platform doesn't set it.
The other important flag is $using_freetype_fontconfig, if the
platform uses freetype and fontconfig. The headless plugin uses
cairo for its drawing operations and freetype+fontconfig for text,
so $test_cairo = $using_freetype_fontconfig. This is independent
from a the cairo canvas!
The OpenGL X11 code now depends on USING_X11, but it doesn't yet
reflect the filter in Library_vcl.mk protecting glx. I don't know
how correct this glx filter is, seeing that the source code just
checked for "UNX - some non-X11 targets".
Change-Id: Id1ba1a967a5d8d10ee217458be879ed00459f7e6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116440
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: Ibddfc30a5f0828ab77235ec1155f1c2e1eef24ee
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116506
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Re-cat's the build / cross-toolset warnings file at the end of
the host configure run and keeps them.
Change-Id: If46fa3660dbd04cd5a23a4d9cd79fd19067dcfbe
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116437
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
I was wondering why removing instdir stuff forced a rebuild of
the cross toolset. Turned out some cross-toolset bits were wrongly
depending on host build stuff. It even had FIXME...
As a consequence, gb_CPPU_ENV was replaced by config_host.mk flags
to provide an CPPU_ENV_FOR_BUILD and also uses the correct
OS_FOR_BUILD.
Change-Id: I50e8e8dca50ab1ad3164948a585a792a52e4a39a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116359
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
.. and convert Android to it. Will also be used by WASM.
It's also kind of a followup on commit
f5af2104fc490b90510e36bbf1d2adec8017c594.
Change-Id: I3a1b5bc2eae2692e706da10c6352534433c61e57
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116385
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
s/BUILD_TARGET/BUILD_TYPE/g
Probably just passed build, because default TLS is NSS, which
includes the same externals.
Thanks Caolan for catching that.
Fixes regression from 07556be594c77f9b7886ff31c2e1752f937cacd4.
Change-Id: I4fa2f2b53c1234bd1612966151792ae7b33358be
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116362
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
And define USE_CRYPTO_* macros to select it in code. This way we
can get rid of all the HAVE_FEATURE_NSS and _WIN32 variations.
This also reverts 1f6b98f21495f0ecc5ded493cb3273da03852191.
Change-Id: I101e4ae2f49cdb127d59bd49a4f1c86304ca2238
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116338
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Use --with-webdav instead, which supports neon, serf or no and is
available since commit dcae0509135c2067635ae3cf84b05dc9fb3d2d02
("Build webdav based on serf if neon is disabled.") from 2014.
Change-Id: I4b329b9c1d28e88961e86272a5554bfa4481f01e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116142
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
So we need either NSS or a system OpenLDAP.
Also add the $test_openldap flag.
Change-Id: I134d1ed3a0a9654e264ccc66cdbe993a355620cb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116109
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: Idce493e5f5ac045f7e977b4073152c2fe2668b56
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116111
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: Ic98a6db70355aa24e29590bb9cb1416c15852090
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116144
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: I1d7b82da3d0597aa6c5ffb251b9ad08bf464d2e2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116136
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I35cda87c35876469bf581be223bc608e29f07b09
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116105
Tested-by: Jenkins
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
|
|
Change-Id: I417c734484481223f1779e0c555b4f3e16e10b23
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116112
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
And move all the output handling into a single place.
Change-Id: Ia75440fc12a435b92239d7ec144be3d58e45d7d0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/116110
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: Iad586802e89b19701a20bebff06b238b617af2a9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115769
Tested-by: Jenkins
Reviewed-by: David Tardon <dtardon@redhat.com>
|
|
Thanks to Matt Whitlock!
(also reported in https://bugs.gentoo.org/show_bug.cgi?id=780432)
Change-Id: I38fec6faf9a929adf1f300b8c369ec6c16a91ed8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115716
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
|
|
Because developers (especially new ones) don't need this, or the
extra dependencies it tends to trigger
Update distro and jenkins configs so that the ones that were building
ODK before, are still building it after this.
Change-Id: I5dc71e70dc457b7921a146008d7d2317b199caab
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115647
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I77c77c88eac413649279d104fecea0799f5046d0
|
|
Change-Id: I32836175a877349777dcbb6eb7b0d813aa31199a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115479
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
|
|
Change-Id: Ibb26390e6cc13c925f499bf95cfc6177d8c9b735
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115355
Tested-by: Jenkins
Reviewed-by: David Tardon <dtardon@redhat.com>
|
|
this is not supposed to work or anything even close to that
Change-Id: I46b4fed6a1e6cfc885cb4f7c24660bb6438d5101
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115293
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: Ibc86f8a76080b55dd7c5a458e2b8fa7ce534a4b3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/115164
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
|