summaryrefslogtreecommitdiff
path: root/configure.ac
AgeCommit message (Collapse)AuthorFilesLines
2020-10-07bump product version to 6.4.7.2.0+mimo-6-4-branch-pointlibreoffice-6-4-7Christian Lohmaier1-1/+1
Change-Id: I45e7d2b3b7683f8892fe2873c17aab833ceba3ce
2020-09-28fix disable qrcodegen optionCaolán McNamara1-1/+1
Change-Id: Ic554f01125653022987c70d03c8c9d86fe3f547a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103204 Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de> Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com> Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org> Tested-by: Xisco Fauli <xiscofauli@libreoffice.org>
2020-09-23bump product version to 6.4.7.1.0+Xisco Fauli1-1/+1
Change-Id: I3e3756a44f28faf4c6d1ab4f03686c025ef1568a
2020-09-23add an explicit --disable-qrcodegen configure optionCaolán McNamara1-17/+35
Change-Id: If8e965fa955aecdb9e7011bdddc690de9cad0c4d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103120 Tested-by: Jenkins Reviewed-by: Caolán McNamara <caolanm@redhat.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103158 Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
2020-07-21bump product version to 6.4.7.0.0+Christian Lohmaier1-1/+1
Change-Id: I96bb42d9756033c9422ccf57b3f75655fe4f9b92
2020-06-10bump product version to 6.4.6.0.0+Christian Lohmaier1-1/+1
Change-Id: I0d8d1d617c14fed144b545c9c02d90e220b0f097
2020-05-17Fix typo.Yunusemre Şentürk1-1/+1
It was causing "./configure: line 9997: =no: command not found" when autogen.sh is used. Change-Id: Iee57fb43c7bfbe4ac64ea5f995af05ddc8a26ad4 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94004 Tested-by: Andras Timar <andras.timar@collabora.com> Reviewed-by: Andras Timar <andras.timar@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94375 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-05-17Add --with-latest-c++ to explicitly opt in for -std=c++20/-std:c++latestStephan Bergmann1-2/+16
Adaptation of this change to this branch. The point is to avoid compiling as C++2a as the code for externals has not been patched properly for that here. Here is the original commit message even if I assume it is a bit misleading in this branch: cd472d1d8489f30797f47d3f6dafede28c1feb90 "Compile as C++2a, where available" had started to unconditionally check for support of -std=c++2a (and later also -std=c++20) for Clang and GCC, but that can cause occasional issues especially for Linux distros, see e.g. 55c724b93dfd4c9a1afb10d60fbc2d7a9a66cf61 "replace boost::bimap in sdext pdfimport" or <https://bugzilla.redhat.com/show_bug.cgi?id=1818723> "/usr/include/boost/format/alt_sstream_impl.hpp incompatible with -std=c++20 (std::allocator::allocate hint argument)" (where 677c8de4fa79cd9b278b142013ba7f1c9e4e41c3 "external/boost: Adapt to std::allocator parts removed in C++20" is not picked up due to --with-system-boost). So better require an explicit opt-in via a new --with-latest-c++. And while at it, also make that enable -std:c++latest for MSVC. Change-Id: I2d1f03144fad9a7884562e56b1b76cab5eb8f080 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/92555 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/93204 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Tor Lillqvist <tml@collabora.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/93692 Tested-by: Tor Lillqvist <tml@collabora.com>
2020-04-29bump product version to 6.4.5.0.0+Christian Lohmaier1-1/+1
Change-Id: Ic81b4105c4ff1d9d419fdb49c5f65b6a778a70fa
2020-03-25bump product version to 6.4.4.0.0+Xisco Fauli1-1/+1
Change-Id: I6a2dd266ac9b1f048ed4ab9258a0b14c759a4e63
2020-03-25configure,gbuild: gla11y fails on Fedora 31Michael Stahl1-2/+6
The problem is that the LD_LIBRARY_PATH on the command line causes /usr/bin/python to find LO's libpython*.so*: 18269: find library=libpython3.7m.so.1.0 [0]; searching 18269: search path=.../instdir/program (LD_LIBRARY_PATH) 18269: trying file=.../instdir/program/libpython3.7m.so.1.0 Presumably LD_LIBRARY_PATH is used to find bundled libxml/libxslt. So let's try to disable the broken case where a bundled lxml is used with system python and bundled libxml/libxslt; this cannot work. (regression from 84ef6d82546b044990f4efd57e51e29c6c6565c8) Change-Id: I67aa8250691cae8f899d65f674aa9da23a9d1d7a Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90906 Reviewed-by: Samuel Thibault <sthibault@hypra.fr> Reviewed-by: Michael Stahl <michael.stahl@cib.de> Tested-by: Jenkins (cherry picked from commit 190f81e34d918da289310a90416f9b6b7be7295f) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90823 Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2020-03-24python3: upgrade to release 3.7.7Michael Stahl1-1/+1
Fixes CVE-2020-8315; this only affects Windows 7 and is a regression in Python 3.6. Change-Id: Ic1706e064a1b03ca1de6361794ed4586a89821d9 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90916 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@cib.de> (cherry picked from commit 74c811da0dedb205976eae69d8589fd91bbaefa2) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90824 Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
2020-02-28bump product version to 6.4.3.0.0+Christian Lohmaier1-1/+1
Change-Id: I8f102ccd43554b4088cc45290cd0b1911c4d25d6
2020-02-22tdf#128849 -- Add Sifr (Dark + SVG) to the icon theme listrizmut1-4/+4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/89243 Tested-by: Jenkins Reviewed-by: Rizal Muttaqin <riz_17_oke@yahoo.co.id> (cherry picked from commit 4617a5b7ded7f5d0c67087d204e05a991a50ee41) Change-Id: I314f925c54a5ed30cd74e4fbbfba065a1b70c947 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/89145 Tested-by: Jenkins Reviewed-by: Rizal Muttaqin <riz_17_oke@yahoo.co.id>
2020-02-15tdf#122218: Hack to avoid blurry text with macOS SDK 10.15Stephan Bergmann1-0/+12
...by setting the LC_VERSION_MIN_MACOSX load command's sdk value to n/a in the soffice executable. See <https://bugs.documentfoundation.org/show_bug.cgi?id=122218#c167> for how this helps, even though I have no idea why it helps. (Adding that -platform_version linker option appears to generate warnings like > ld: warning: passed two min versions (10.13.0, 10.13) for platform macOS. Using 10.13. but which are probably harmless.) (cherry picked from commit 645fe53be0dc36535dba0ed684e21ca4cda80d70) Plus cherry-pick of follow-up b7fd89100d8653dc73955780358fe31d38b68ebf "tdf#122218: Baseline Xcode 9.3 ld presumably doesn't support -platform_version" (and resolving the merge conflict in desktop/Executable_soffice_bin.mk). Change-Id: I043498c7ff2d148d4a7e1e0e9d46241b638f2eba Reviewed-on: https://gerrit.libreoffice.org/c/core/+/88667 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com> Reviewed-on: https://gerrit.libreoffice.org/c/core/+/88753 Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2020-02-05bump product version to 6.4.2.0.0+Christian Lohmaier1-1/+1
Change-Id: I4bc9c52f6d9f8e60ec1d4450e4f796735b53f219
2020-01-31Adapt o3tl::span to P1872R0Stephan Bergmann1-0/+16
..."span should have size_type, not index_type" (<http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p1872r0.pdf>), as implemented by libc++ since <https://github.com/llvm/llvm-project/commit/ 1466335cf4b2854a0be1defcf279fe50772bad6f> "[libc++][P1872] span should have size_type, not index_type." All uses of index_type had been added to mitigate the previous std::span change from signed (ptrdiff_t) to unsigned (size_t) index_type, see 6ef8420fdbf8dff16de13147c5ab833bc5e01121 "Adapt o3tl::span to updated C++2a std::span". There is no easy solution to transparently support all three std::span variants currently out there (signed index_type, unsigned index_type, unsigned size_type), without causing compilation failures due to CPPUNIT_ASSERT_EQUAL with arguments of different types, or compiler warnings about mixed signed/unsigned comparisons. So rule out the oldest std::span variant (signed index_type) in configure.ac (so that o3tl::span will use its own hand-rolled code in that case) and simplify the uses of index_type to std::size_t (as had already been mentioned in 6ef8420fdbf8dff16de13147c5ab833bc5e01121). Change-Id: I6ddf424ffb7941da3f69ad66fd29ecd35f09afae Reviewed-on: https://gerrit.libreoffice.org/84652 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com> (cherry picked from commit 8e6865188242bccb3d8aa857ddc990d72a058d3d) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87757 Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-01-22Breeze & Sifr: tdf#128849 & Sifr: tdf#129846rizmut1-4/+4
- Breeze & Sifr: Add SVG variant of dark version - Sifr: Add more 32px icons Change-Id: I8ce5ff12b1178d215d4ca756a512ff01754fbff9 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87188 Tested-by: Jenkins Reviewed-by: Rizal Muttaqin <riz_17_oke@yahoo.co.id>
2020-01-13python3: bundle libffi for GNU/Linux buildsMichael Stahl1-0/+3
CPython commit f40d4ddff3c800b3c956a5e8820aabe3aa87cddd "Closes #27979: Remove bundled copy of libffi" causes a bit of a problem because it turns out that libffi isn't all that stable; there's libffi.so.5 on CentOS 6, libffi.so.6 on CentOS 7 and libffi.so.7 on lo_daily_update_gandalf tinderbox. So we have to bundle it in LO; it's only used on GNU/Linux currently. CPython commit 32119e10b792ad7ee4e5f951a2d89ddbaf111cc5 "bpo-35947: Update Windows to the current version of libffi (GH-11797)" also removes the libffi for MSVC, so in a future python upgrade we will have to build libffi for MSVC too. The libffi fork for MacOSX is still in CPython git master. (regression from b10be5d48433076f0b7238d818020f708553e114) Change-Id: Ibc20cf8cd3614cf9941b6970662bd930496776b2 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86493 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@cib.de> (cherry picked from commit 79084665f0e351a3f83fdee88071919f05ec9cc3) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86500 Reviewed-by: Caolán McNamara <caolanm@redhat.com>
2020-01-08bump product version to 6.4.1.0.0+Christian Lohmaier1-1/+1
Change-Id: I4dd8377f4c16f73362d397a054a5b57441efb365
2020-01-08python3: upgrade to release 3.7.6Michael Stahl1-2/+2
* external/python3/python-3.3.3-aix.patch.1: most of it doesn't apply and AIX port isn't maintained anyway so remove it for now * external/python3/ubsan.patch.0: apparently one of the files was removed * 0001-3.6-bpo-17239-Disable-external-entities-in-SAX-parse.patch.1: fixed upstream * python3-osx-avoid-new-10.13.patch.1: replace with simply passing ac_cv_func_utimensat=no to configure * external/python3/python-3.5.4-ssl.patch.1: project files to build OpenSSL removed upstream * There have been changes to how python locates OpenSSL; new variables OPENSSL_INCLUDES etc; it turns out that you have to pass one directory to --with-openssl, as the variables cannot be passed * libuuid.so.1 is a new dependency of the _uuid module * libffi.so.6 is a new dependency of the _ctypes module (the bundled copy of libffi for non-Darwin platforms was removed) * python-3.3.0-pythreadstate.patch.1: the PyThreadState functions have been changed such that CppunitTest_services asserts when there is a PyThreadAttach on top of PyThreadDetach on top of PyThreadAttach, i.e., 2 PyThreadState per thread (PyGILState_Check() fails). Instead of patching in additional workarounds, change PyThreadAttach so that it re-uses an existing PyThreadState if one exists for the thread. Change-Id: I24c19d79b43a30709261fd9db66312b2e3872fd9 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/84765 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@cib.de> (cherry picked from commit b10be5d48433076f0b7238d818020f708553e114) Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86398 Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
2019-12-18bump product version to 6.4.0.1.0+Christian Lohmaier1-1/+1
Change-Id: I21b5e66df3283425a678f5a204b4496007dafbf8
2019-12-12Qt5 fix missing XCB_ICCCM_WM_HINT_WINDOW_GROUPJan-Marek Glogowski1-2/+9
This is the application level equivalent of the Qt5 fix for bug QTBUG-46626 / commit 0de4b32 ("xcb: fix issue with dialogs hidden by other windows"), which was broken since Qt 5.4 and is just fixed since Qt 5.12. It is needed for some window managers, which don't know about the WM_CLIENT_LEADER property. Both settings are the same, but just the latter is set by older Qt5 releases. This probably isn't a real problem, as GNOME or XFCE would use the gtk VCL plugin, but since I already wrote the code when debugging tdf#129071, there is also no reason to drop it (except: more code, more bugs...). This fix is optional and needs development headers for xcb-icccm, which can actually be compiled into Qt5. If missing configure will just print a warning, since it's a runtime requirement and we explicitly drop the linked Qt version symbol, so the potential build Qt version won't matter. Change-Id: Ifc5a8f8a40ee13779a911efb53e8b8b868614d0b Reviewed-on: https://gerrit.libreoffice.org/84299 Tested-by: Jenkins Reviewed-by: Michael Weghorn <m.weghorn@posteo.de> Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de> (cherry picked from commit fe2baf9e84e0ca9aeaa683e37076f57fa3f38dca) Reviewed-on: https://gerrit.libreoffice.org/84810
2019-11-19python3: upgrade to release 3.5.9Michael Stahl1-1/+1
Fixes CVE-2019-9948 CVE-2019-9740 CVE-2019-10160 CVE-2019-16056 and expat CVE-2019-15903. python-3.3.5-pyexpat-symbols.patch.1 fails to apply, and it's a mystery why --with-system-expat is used everywhere but on MacOSX, where 292af048ace2d4b455b2da3a22c784cb05db1d09 disabled it for no obvious reason, so try to remove the special case and get rid of the patch. Change-Id: I5ba4532eb6e7c2fb90daba95d132dcc7c9013d96 Reviewed-on: https://gerrit.libreoffice.org/83117 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@cib.de> (cherry picked from commit b0930d56130fdddfe65e92b081a8afad77974076) Reviewed-on: https://gerrit.libreoffice.org/83187
2019-11-13bump product version to 6.4.0.0.beta1+Xisco Fauli1-1/+1
Change-Id: I2b4cc64dd863d8be755f639c71c540179f0ebdb1
2019-11-04find Clang headers also if it's Clang built from sourceLuboš Luňák1-0/+11
Change-Id: I2253bdce982028277b30d7bf911201675be45ca4 Reviewed-on: https://gerrit.libreoffice.org/81919 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2019-11-01Accept iOS SDK 13.2Tor Lillqvist1-2/+2
Change-Id: I5d42a60a257661f39d1c9af6299ca3278f783d2b Reviewed-on: https://gerrit.libreoffice.org/81870 Tested-by: Jenkins Reviewed-by: Tor Lillqvist <tml@collabora.com> Tested-by: Tor Lillqvist <tml@collabora.com>
2019-10-30fix vcldemo lookup of iconsLuboš Luňák1-0/+1
Icon themes are in [srcdir]/icon-themes, which is not necessarily the same as $PWD/icon-themes. Change-Id: Id2c5037afcbea4ea7dd511a9e10e19e05fa52a5a Reviewed-on: https://gerrit.libreoffice.org/81701 Tested-by: Jenkins Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk> Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2019-10-30Upgrade to ICU 65.1Eike Rathke1-2/+2
sberg says: On Windows, implicit --enable-extras first causes a build breaker in workdir/UnpackedTarball/icu/source/extras/scrptrun when linking, because Windows link.exe doesn't understand -o. But even with a patch > --- source/extra/scrptrun/Makefile.in > +++ source/extra/scrptrun/Makefile.in > @@ -74,7 +74,7 @@ > && CONFIG_FILES=$(subdir)/$@ CONFIG_HEADERS= $(SHELL) ./config.status > > $(TARGET) : $(OBJECTS) > - $(LINK.cc) -o $@ $^ $(LIBS) > + $(LINK.cc) $(OUTOPT)$@ $^ $(LIBS) > $(POST_BUILD_STEP) > > invoke: linking would still fail with a missing ../../lib/icuucdd.lib, which is apparently expanded from $(LIBS) there, but I have no idea where it should be built but isn't. Lets hope that --disable-extras is sufficient for our needs. Change-Id: I6d0117b230caa41abf488fcd069028e3474700f8 Reviewed-on: https://gerrit.libreoffice.org/81632 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-28Upgrade to ICU 64.2Eike Rathke1-2/+2
As an interim step to upgrade to ICU 65.1 Adds new scripts and Unicode blocks from Unicode 12. Change-Id: Idc4a6b29ffb04bcb424522fcbd29a8db0428c056 Reviewed-on: https://gerrit.libreoffice.org/81611 Reviewed-by: Eike Rathke <erack@redhat.com> Tested-by: Jenkins
2019-10-24android: Allow specification of the API level.Jan Holesovsky1-1/+14
Change-Id: Icf33e2703f42a7866ce895437cf5f276066eeebe Reviewed-on: https://gerrit.libreoffice.org/81227 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Jan Holesovsky <kendy@collabora.com> (cherry picked from commit 59eee888bf7174114e5749855d95e8ff7dd15013) Reviewed-on: https://gerrit.libreoffice.org/81327 Tested-by: Jenkins
2019-10-24android: Allow using SDK and NDK directly from the Android Studio.Jan Holesovsky1-2/+2
Just specify: --with-android-ndk=$HOME/Android/Sdk/ndk-bundle --with-android-sdk=$HOME/Android/Sdk in your autogen.input, install the appropriate components via Android Studio and you are done. Change-Id: Ic99790b781b9017eb4e642380e230d6f7b49e9b7 Reviewed-on: https://gerrit.libreoffice.org/81228 Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com> Reviewed-by: Jan Holesovsky <kendy@collabora.com> (cherry picked from commit 246f1b5b4485b7db9f9584e4b3b819c87e331c0e) Reviewed-on: https://gerrit.libreoffice.org/81328 Tested-by: Jenkins
2019-10-21We do not need a C++ GNU dialectStephan Bergmann1-9/+1
1cf7ab61a71d4b7295942ff5c855896e60c15081 "use -std=gnu++0x rather than -std=c++0x" appears to have started this, but the only rationale it gives is that it keeps things in sync with GCC's default behavior when no -std= is given. But it apparently works fine to build with a -std=c++... standard dialect. This allows to get rid of the check introduced with 50cd28e5728b6a64c1e605567540739ea6ef42ca "Ensure configuration that defines math_errhandling in <cmath>". (It kept bothering me to say "I observe this-and-this with -std=c++2a" when what configure.ac made me actually use was -std=gnu++2a. And truthfully saying "-std=gnu++2a" would have been a distraction, as what is relevant for such an observation is most likely the "2a" and not the "gnu" part.) Change-Id: I7c213a702ffb7df6f4c2c4a421008e30e2712a51 Reviewed-on: https://gerrit.libreoffice.org/81176 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
2019-10-18tdf#128189: fix couple issues in configure.acJulien Nabet1-8/+8
Patch from David L. Craig License statement in lo-dev forum 2019/10/18 Change-Id: I772807b66e096c0abba1cf464aaced432209451f Reviewed-on: https://gerrit.libreoffice.org/81005 Tested-by: Jenkins Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
2019-10-16bump product version to 6.4.0.0.alpha1+Christian Lohmaier1-1/+1
Change-Id: I4eaf798b1b43fe8a78d1697d9ebc207ff8ace492
2019-10-15move HAVE_FEATURE_DESKTOP/OPENCL to their dedicated headersLuboš Luňák1-0/+2
HAVE_FEATURE_OPENCL is included by a common Calc header and HAVE_FEATURE_DESKTOP is included by a common Writer header, causing pretty much their full rebuilds if any feature changes. Change-Id: If29bf78bd4fd70b37981e0826a577777fd255c89 Reviewed-on: https://gerrit.libreoffice.org/80776 Tested-by: Jenkins Reviewed-by: Tor Lillqvist <tml@collabora.com>
2019-10-14configure: remove sanitizer flags from default COMPILER_PLUGINS_CXXMichael Stahl1-1/+1
When COMPILER_PLUGINS_CXX is default initialised from $CXX, filter out sanitizer flags, because: a) linking fails with unresolved symbols currently b) typically the slowdown is unhelpful in a build c) if anybody wants a sanitized plugin they can set COMPILER_PLUGINS_CXX manually (init from CXX was added in ad5cbcf6ba0afdc1d8d7405c2641cce8de4a360b) Change-Id: I64dc48fae5f7a23f87b0eee0545add9a1ebd5672 Reviewed-on: https://gerrit.libreoffice.org/80655 Tested-by: Jenkins Reviewed-by: Michael Stahl <michael.stahl@cib.de>
2019-10-12aarch64 callVirtualFunction needs to be compiled w/o -fstack-clash-protectionStephan Bergmann1-0/+11
At least when doing an aarch64 Flatpak build against org.freedesktop.Sdk//19.08, which uses GCC 9.2.0 and passes in `CXXFLAGS=-O2 -g -pipe -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-strong -grecord-gcc-switches -fasynchronous-unwind-tables -fstack-clash-protection`, callVirtualMethod (in bridges/source/cpp_uno/gcc3_linux_aarch64/callvirtualfunction.cxx) would decrement the stack pointer another 16 bytes after the stackargs = alloca(...); and before the asm block, so in the called virtual function, arguments read from the stack would read garbage (and CustomTarget_testtools/uno_test would fail with SIGSEGV at > #0 0x0000ffffb733eb48 in rtl::OUString::operator= (this=0xaaaaf9c1ac30, str=...) at /run/build/libreoffice/include/rtl/ustring.hxx:453 > #1 0x0000ffffb733a7bc in bridge_object::assign (rData=..., bBool=true, cChar=64 u'@', nByte=17 '\021', nShort=4660, nUShort=65244, nLong=305419896, nULong=4275878552, nHyper=0, nUHyper=187651311381888, fFloat=17.0814991, fDouble=3.1415926358999999, eEnum=-1698898192, rStr=..., xTest=..., rAny=...) at /run/build/libreoffice/testtools/source/bridgetest/cppobj.cxx:99 > #2 0x0000ffffb733a87c in bridge_object::assign (rData=..., bBool=true, cChar=64 u'@', nByte=17 '\021', nShort=4660, nUShort=65244, nLong=305419896, nULong=4275878552, nHyper=0, nUHyper=187651311381888, fFloat=17.0814991, fDouble=3.1415926358999999, eEnum=-1698898192, rStr=..., xTest=..., rAny=..., rSequence=...) at /run/build/libreoffice/testtools/source/bridgetest/cppobj.cxx:115 > #3 0x0000ffffb733ade4 in bridge_object::Test_Impl::setValues (this=0xaaaaf9c1abb0, bBool=1 '\001', cChar=64 u'@', nByte=17 '\021', nShort=4660, nUShort=65244, nLong=305419896, nULong=4275878552, nHyper=0, nUHyper=187651311381888, fFloat=17.0814991, fDouble=3.1415926358999999, eEnum=-1698898192, rStr=..., xTest=..., rAny=..., rSequence=..., rStruct=...) at /run/build/libreoffice/testtools/source/bridgetest/cppobj.cxx:548 > #4 0x0000ffffb740bff4 in callVirtualFunction (function=281473755360772, gpr=0xffffd1ab1f28, fpr=0xffffd1ab1f68, stack=0xffffd1ab1d40, sp=8, ret=0xffffd1ab22c0) at /run/build/libreoffice/bridges/source/cpp_uno/gcc3_linux_aarch64/callvirtualfunction.cxx:63 > #5 0x0000ffffb740ca70 in (anonymous namespace)::call (proxy=0xaaaaf9c291c0, slot=..., returnType=0xaaaaf9c00770, count=17, parameters=0xaaaaf9c3a210, returnValue=0xffffd1ab22c0, arguments=0xffffd1ab2230, exception=0xffffd1ab2370) at /run/build/libreoffice/bridges/source/cpp_uno/gcc3_linux_aarch64/uno2cpp.cxx:178 > #6 0x0000ffffb740d4c4 in bridges::cpp_uno::shared::unoInterfaceProxyDispatch (pUnoI=0xaaaaf9c291c0, pMemberDescr=0xaaaaf9c55950, pReturn=0xffffd1ab22c0, pArgs=0xffffd1ab2230, ppException=0xffffd1ab2370) at /run/build/libreoffice/bridges/source/cpp_uno/gcc3_linux_aarch64/uno2cpp.cxx:361 > #7 0x0000ffffb740720c in (anonymous namespace)::call (proxy=0xaaaaf9c549c0, description=..., returnType=0xaaaaf9c00770, count=17, parameters=0xaaaaf9c3a210, gpr=0xffffd1ab2510, fpr=0xffffd1ab2550, stack=0xffffd1ab2590, indirectRet=0xffffb7d24790) at /run/build/libreoffice/bridges/source/cpp_uno/gcc3_linux_aarch64/cpp2uno.cxx:120 > #8 0x0000ffffb74079a0 in (anonymous namespace)::vtableCall (functionIndex=40, vtableOffset=0, gpr=0xffffd1ab2510, fpr=0xffffd1ab2550, stack=0xffffd1ab2590, indirectRet=0xffffb7d24790) at /run/build/libreoffice/bridges/source/cpp_uno/gcc3_linux_aarch64/cpp2uno.cxx:291 > #9 0x0000ffffb7407b00 in (anonymous namespace)::vtableSlotCall (gpr0=187651311618536, gpr1=1, gpr2=64, gpr3=17, gpr4=4660, gpr5=65244, gpr6=305419896, gpr7=4275878552, fpr0=5.4321266044931319e-315, fpr1=3.1415926358999999, fpr2=0, fpr3=4.0039072046065485, fpr4=0, fpr5=4.003911019303815, fpr6=8.9589789687541617e+102, fpr7=-4.4588500238274385e-308) at /run/build/libreoffice/bridges/source/cpp_uno/gcc3_linux_aarch64/cpp2uno.cxx:348 > #10 0x0000ffffb739e60c in bridge_test::performTest (xContext=..., xLBT=..., noCurrentContext=false) at /run/build/libreoffice/testtools/source/bridgetest/bridgetest.cxx:378 > #11 0x0000ffffb73a3d58 in bridge_test::TestBridgeImpl::run (this=0xaaaaf9c18550, rArgs=...) at /run/build/libreoffice/testtools/source/bridgetest/bridgetest.cxx:1162 > #12 0x0000aaaad292a3ec in sal_main () at /run/build/libreoffice/cpputools/source/unoexe/unoexe.cxx:509 > #13 0x0000aaaad29297a0 in main (argc=8, argv=0xffffd1ab31b8) at /run/build/libreoffice/cpputools/source/unoexe/unoexe.cxx:349 .) By experiment, I found the problematic thing to be -fstack-clash-protection, which can apparently be cancelled with a subsequent -fno-stack-clash-protection at least on that GCC 9.2.0. (And -f[no-]stack-clash-protection appears to only be available since GCC 8, and not at all for Clang, so check for it with HAVE_GCC_STACK_CLASH_PROTECTION.) Change-Id: If667fdf704b1ba20a04593b38d2d1f079280df41 Reviewed-on: https://gerrit.libreoffice.org/80701 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-12-fstack-protector-strong is long since availableStephan Bergmann1-12/+0
...on our baselines, since <https://gcc.gnu.org/git/ ?p=gcc.git;a=commitdiff;h=b156ec373ccf27f4fcce7972de5e043d35acea43> (GCC 4.9?) and <https://github.com/llvm/llvm-project/commit/ e0fc1a80cba8b91e3943f3287e7dcf68c6bb9b7f> "[stackprotector] Add command line option -fstack-protector-strong" (Clang 3.5?) Change-Id: I48237b2304a1ee273cc66f0bb458e890a5a2f21a Reviewed-on: https://gerrit.libreoffice.org/80700 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-10Set RTL_OS to "iOS" for iOSTor Lillqvist1-0/+1
We accidentally had left it as "MacOSX". Affects at least the "generator" metadata stored in documents. Change-Id: I72eeefdbe192409084f7ab7a24adbc39dcafb624
2019-10-09With Cygwin, AC_PATH_PROG needs Cygwin-style pathsStephan Bergmann1-1/+5
(And instead directly specifying CLANGDIR as a Cygwin-style path in my clang-cl build's autogen.input doesn't work, as compilerplugins/Makefile-clang.mk spells a dependency on $(CLANGDIR)/bin/clang$(CLANG_EXE_EXT) which Make on Windows requires to be a Windows-style path.) Change-Id: I20ee3a2dfff0a3db66e1388cd6fc01084a6fd812 Reviewed-on: https://gerrit.libreoffice.org/80471 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-09Bump the minimum iOS run-time version to 12.2Tor Lillqvist1-2/+2
Change-Id: I082a7d62e222e625d1d921bea39b45578118d225
2019-10-09Accept iOS SDK 13.1Tor Lillqvist1-2/+2
Change-Id: I02870b35f67dd9ca47061311186d74dfec823aa7
2019-10-08My Windows clang-cl build still doesn't use LO_CLANG_SHARED_PLUGINSStephan Bergmann1-1/+3
...so disable the new configure.ac checks introduced with ad5cbcf6ba0afdc1d8d7405c2641cce8de4a360b "try to autodetect flags needed to build Clang plugins" that are only relevant when using LO_CLANG_SHARED_PLUGINS and would fail miserably for my clang-cl build Change-Id: I58f7f1f4608f1a615175f0c0d0d98c03c442a36c Reviewed-on: https://gerrit.libreoffice.org/80477 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-08Clarify COMPILER_PLUGINS_CXX_LINKFLAGS vs. COMPILER_PLUGINS_LINKFLAGSStephan Bergmann1-3/+3
COMPILER_PLUGINS_CXX_LINKFLAGS was introduced with 39e7a72b3e328e6b3d87479d693b01315610457b "Support loplugin in clang-cl" to augment COMPILER_PLUGINS_CXX. Due to MSVC cl.exe command-line processing, certain linker-related arguments must come at the very end of the command line, so cannot be included in COMPILER_PLUGINS_CXX. COMPILER_PLUGINS_CXX_LINKFLAGS is specified in autogen.input (along with COMPILER_PLUGINS_CXX) and configure.ac merely passes it on to its use in compilerplugins/Makefile-clang.mk. ad5cbcf6ba0afdc1d8d7405c2641cce8de4a360b "try to autodetect flags needed to build Clang plugins" now needs a configure.ac-internal variable to store the output of `llvm-config --ldflags ...`, similarly to how COMPILER_PLUGINS_CXXFLAGS stores the output of `llvm-config --cxxflags`. It reused COMPILER_PLUGINS_CXX_LINKFLAGS for that, but that makes it hard for my clang-cl build to pass my specification of COMPILER_PLUGINS_CXX_LINKFLAGS from autogen.input to its use in compilerplugins/Makefile-clang.mk. So rename this new variable to COMPILER_PLUGINS_LINKFLAGS (matching COMPILER_PLUGINS_CXXFLAGS). Change-Id: I93b0b50ba94803041773757d9978222e2726f9b0 Reviewed-on: https://gerrit.libreoffice.org/80473 Reviewed-by: Luboš Luňák <l.lunak@collabora.com> Tested-by: Jenkins
2019-10-08CXX is now always set before CXX_X64_BINARY on WindowsStephan Bergmann1-1/+1
...since ea3d4e806cbdde18173da92187329f1ac2177e14 "Fix CXX_BASE for clang-cl builds on Windows", as a side effect, moved the conditional CXX=$MSVC_CXX further up. This addresses the comment in the commit message of 463a79cbc16c1b4aba1775d7f8ae0324753c322c "CXX_X64_BINARY must be clang-cl not cl when building with clang-cl": "Ideally, the code would be reorganized so that CXX_X64_BINARY is only set after CXX has been set." Change-Id: Iaa011aeff88669ddd5d33fc5b1109abf02edff54 Reviewed-on: https://gerrit.libreoffice.org/80468 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-08Fix CXX_BASE for clang-cl builds on WindowsStephan Bergmann1-6/+4
...where the configure messages confusingly mentioned cl.exe instead of clang.exe (but the actual checks correctly used $CXX being clang) Change-Id: I9e0c6e1ab8ba64c45f752b413c3e6c22182506ac Reviewed-on: https://gerrit.libreoffice.org/80442 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2019-10-08update help text of --disable-guiAndras Timar1-4/+1
Change-Id: Idd01b11acee3683bbc3d81d0c2ccf8be6c4fbb69 Reviewed-on: https://gerrit.libreoffice.org/80428 Reviewed-by: Rene Engelhard <rene@debian.org> Tested-by: Rene Engelhard <rene@debian.org>
2019-10-08remove leftover debug output in configureLuboš Luňák1-1/+0
Change-Id: If2b1d7ca89f30544c4e8750119927701f9df5264
2019-10-07make the clang plugins configure check fasterLuboš Luňák1-5/+5
Use a header which is not so expensive to parse/compile. Change-Id: I4197fb16938b19c18fed541dbf94bf2c97a60e66 Reviewed-on: https://gerrit.libreoffice.org/80301 Tested-by: Jenkins Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-04-28 11:27:10 +0000