...similar to how this is done for the other $(SODEST)/... targets above (and
which has presumably been missing accidentally from
eb050ae015 "tdf#123290 fix android tinderbox
builds/properly package libc++_shared.so".
Saw <https://ci.libreoffice.org/job/gerrit_android_x86/830/> fail once with
> cp: cannot create regular file ‘/home/tdf/lode/jenkins/workspace/android_x86/android/jniLibs/x86/libc++_shared.so’: No such file or directory
> ../Bootstrap/Makefile.shared:75: recipe for target '/home/tdf/lode/jenkins/workspace/android_x86/android/jniLibs/x86/libc++_shared.so' failed
> make[2]: *** [/home/tdf/lode/jenkins/workspace/android_x86/android/jniLibs/x86/libc++_shared.so] Error 1
> make[2]: *** Waiting for unfinished jobs....
> /home/tdf/lode/jenkins/workspace/android_x86/android/CustomTarget_lo_android.mk:17: recipe for target '/home/tdf/lode/jenkins/workspace/android_x86/workdir/CustomTarget/android/source/done' failed
> make[1]: *** [/home/tdf/lode/jenkins/workspace/android_x86/workdir/CustomTarget/android/source/done] Error 2
Change-Id: Ief6b6274ad789d441c2692b941ba0aa8df276e40
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/106119
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
commit 0dffc65236
Date: Fri Jun 5 08:59:26 2020 -0400
android: fix the build output directory
submitted on 2020-07-07 had changed the output directory
for native libraries, so the directory used by
Android Viewer was empty, and native libs no longer
contained in the package/APK.
Store the actual/new directory in a variable and
use that one in build.gradle to make things work
again.
Change-Id: I5a051342278ab14e71edf47dd93e8dacdb9f3cb6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98970
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
This is similar to commit c6dadf5035
("commit android: Fix linking of liblo-native-code.so on aarch64.").
Build previously failed with
Linking obj/local/x86_64/liblo-native-code.so
.../x86_64-linux-android/bin/ld.gold: error: cannot find -landroid_support
clang++: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [../Bootstrap/Makefile.shared:62: obj/local/x86_64/liblo-native-code.so] Error 1
make[1]: *** [.../android/CustomTarget_lo_android.mk:18: .../workdir/CustomTarget/android/source/done] Error 2
make: *** [Makefile:282: build] Error 2
I quickly checked that the app actually runs on an x86_64 AVD
and is able to open and display documents after successfully
building with this commit in place.
Change-Id: I92b3759af6a86a6717c287d035b6a1add3a9af7a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/85204
Tested-by: Jenkins
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
Forces gold as the linker; but according to the switches that were used,
I suppose gold was used previously anyway, so hopefully not really a
problem.
Change-Id: I7bc54abf0d3b4b3d86d53e4ea6f01146cf196216
Reviewed-on: https://gerrit.libreoffice.org/81905
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
Tested-by: Jan Holesovsky <kendy@collabora.com>
(cherry picked from commit fa145a035fbab652ec8f3ee870e44e43dd1d770a)
Reviewed-on: https://gerrit.libreoffice.org/81915
While it might not be the original problem, at least that is the current
problem the tinderbox provided builds have and should make it work again
Change-Id: Ifaedb3304622286cfe7a6cb49e862a359b53f892
This partially reverts a7f6338875. The
problem is that with that approach, the problems with unwinding as
described in 312eeeee42 were plaguing the
Online-based Android app, leading to mysterious crashes with stack
traces like:
#00 pc 0005f2b6 /apex/com.android.runtime/lib/bionic/libc.so (abort+166) (BuildId: 68c87e04526a60689ecb5deb329804a0)
#01 pc 05caaae7 /data/app/libreoffice.debug-fBEOIUwPHY3_k3eQid159Q==/lib/arm/liblo-native-code.so (BuildId: 1c3d0e6336ddfc3578e1c186944160252a750b8b)
...
#06 pc 07452e43 /data/app/libreoffice.debug-fBEOIUwPHY3_k3eQid159Q==/lib/arm/liblo-native-code.so (BuildId: 1c3d0e6336ddfc3578e1c186944160252a750b8b)
#07 pc 0005a37c /apex/com.android.runtime/lib/bionic/libc.so (__restore_rt) (BuildId: 68c87e04526a60689ecb5deb329804a0)
#08 pc 0005f2b2 /apex/com.android.runtime/lib/bionic/libc.so (abort+162) (BuildId: 68c87e04526a60689ecb5deb329804a0)
#09 pc 07565d3b /data/app/libreoffice.debug-fBEOIUwPHY3_k3eQid159Q==/lib/arm/liblo-native-code.so (BuildId: 1c3d0e6336ddfc3578e1c186944160252a750b8b)
...
#11 pc 07565641 /data/app/libreoffice.debug-fBEOIUwPHY3_k3eQid159Q==/lib/arm/liblo-native-code.so (BuildId: 1c3d0e6336ddfc3578e1c186944160252a750b8b)
#12 pc 0001e394 /apex/com.android.runtime/lib/bionic/libm.so (__gnu_Unwind_RaiseException+108) (BuildId: 0ee3a75927e0da4fa0fe29469ccf8b39)
#13 pc 0001eee8 /apex/com.android.runtime/lib/bionic/libm.so (_Unwind_RaiseException+20) (BuildId: 0ee3a75927e0da4fa0fe29469ccf8b39)
Change-Id: Ie4e7b20380c9104770229a8d30036ad76093283d
Reviewed-on: https://gerrit.libreoffice.org/81485
Tested-by: Jenkins
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
support for targeting API 14 and 15 was removed in NDK 18, so set
minimum version to 16
mips support was removed in NDK 17
Clang now takes care about correct linking with libc++ shared or
static, so don't manually specify them anymore.
Same with __ANDROID_API_LEVEL__ define and the sysroot / isystem
handling, that is all covered by a single -target <triple><version>
simplifying things quite a bit.
also align ownloud sdk values with main build.gradle
Change-Id: Ib3ae4484e52214677e826270b731ecf7c5c15445
Reviewed-on: https://gerrit.libreoffice.org/77104
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
Tested-by: Jenkins
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Useful when you switch sdk to a different location.
Change-Id: I95301cc9adf50a0bbd918cc7562b8871093625b8
Reviewed-on: https://gerrit.libreoffice.org/67873
Tested-by: Jenkins
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
It had been left out in 4082a18406 "android: use
unified headers and llvm-c++ STL (x86) with NDK 16" because "arm unfortunately
crashes with llvm-c++, so keep with gnustl for now/fix that later".
Making armeabi-v7a work with libc++ etc. required a number of changes, listed
below, in this commit and in preceding ones. At least 32-bit x86 already worked
with libc++ etc. prior to these changes in view mode, though it crashed in the
experimental editing mode (enabled with strippedUIEditing in
android/soruce/Makefile) as soon as one types in something, But it is not
entirely clear to me why 32-bit x86 view mode didn't also fail similar to how I
saw armeabi-v7a fail. (On 32-bit x86, these changes appear to neither improve
nor worsen the current state, view mode still appears to work fine while editing
still crashes upon typing anything. With these changes, editing mode on
armeabi-v7a appears to work fine. But I tested armeabi-v7a only with a real
device and 32-bit x86 only with an emulator, in case that might make a
difference.)
* Preceding <https://gerrit.libreoffice.org/#/c/64964/> "Move NSSLIBS to a more
sensible place on the linker command line" plus this change's addition of
-lunwind to the liblo-native-code.so linker command line make sure that
liblo-native-code.so uses _Unwind_* functions from libunwind.a, instead of
erroneously picking up the ones from libgcc.a that happen to be included in
NSSLIB's nspr4 (-lgcc is automatically added to the end of the linker command
line by the invoking compiler, that's how libgcc.a's _Unwind_* end up in
NSSLIB's nspr4; it is neither clear to me why NSSLIB's nspr4, being a pure C
library, uses _Unwind_* functions, nor why exception handling in
liblo-native-code.so fails when using _Unwind_* functions from libgcc.a
instead of from libunwind on armeabi-v7a, nor why that would work on 32-bit
x86, but that's what I observed: ModuleManager::identify
(framework/source/services/modulemanager.cxx) throws a
css::lang::IllegalArgumentException, which calls __cxa_throw ->
_Unwind_RaiseException, which ultimately lead to odd misbehavior and
std::abort during stack unwinding when using _Unwind_RaiseException from
libgcc.a instead of from libunwind). (There is no libunwind.* in
android-ndk-r16b for 32-bit x86 at least, so is presumably using _Unwind_*
functions from libgcc.a. It doesn't appear to make a difference if it
indirectly uses those _Unwind_* functions from NSSLIB's nspr4, or directly
from libgcc.a included in liblo-native-code.so if the
$(if $(filter armeabi-v7a,$(ANDROID_APP_ABI)),-lunwind)
had a ",-lgcc" else branch.)
* Preceding <https://gerrit.libreoffice.org/#/c/64965/> "Export RTTI symbols
from liblo-native-code.so, for binary UNO bridge" makes sure that excpetions
thrown from the binary UNO bridge can be caught by compiled catch clauses.
Not sure why the corresponding state of
bridges/source/cpp_uno/gcc3_linux_intel shouldn't have run into the same
issue.
* Preceding <https://gerrit.libreoffice.org/#/c/64966/> "Adapt gcc3_linux_arm
__cxa_exception to NDK 18 libc++abi" makes sure that our version of
__cxa_exception matches the version from libc++abi. This is clearly not
relevant for 32-bit x86. (The comment there android-ndk-r18b, but the
additional member is already present in
android-ndk-r16b/sources/cxx-stl/llvm-libc++abi/src/cxa_exception.hpp, too.)
The remainder of this change just drops old armeabi-v7a--specific workarounds
that are no longer needed/no longer work.
Change-Id: Ief4c2d562c5032abe6c3b94ca3b3394be6fcd4d3
Reviewed-on: https://gerrit.libreoffice.org/64973
Tested-by: Stephan Bergmann <sbergman@redhat.com>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
511ae02c64 "Android: Enable HAVE_FEATURE_NSS and
package the NSS libraries with apk" had added them to WHOLELIBS probably just
because that already had the $(addprefix -l,...), even though --whole-archive
doesn't make any sense for shared libraries. Better place them later on the
linker command line (after all our own archives and compiler support libraries),
so that switching armeabi-v7a to libc++/libc++abi/libunwind (coming soon) will
be able to override erroneously picking _Unwind_* symbols from NSSLIBS's nspr4
instead of libunwind.
Change-Id: Ie0c0b7a55da3eabe1bb427232d698b2a4af63e78
Reviewed-on: https://gerrit.libreoffice.org/64964
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
gnustl (and others) are to be removed in future versions of the ndk
also bump gradle and build-tools to current versions along with it
arm unfortunately crashes with llvm-c++, so keep with gnustl for now/fix
that later
Change-Id: Ic794c3293b599b77ec48096bf3283a99c09cbb79
Reviewed-on: https://gerrit.libreoffice.org/45163
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
packing gdbserver into apk conflicts with extractNativeLibs="false", as
the gradle pugin compresses it (which could be disabled using
aaptOptions), but furthermore it doesn't page-align it, breaking
installation.
So instead let the user manually push the gdbserver tool to device and
remove the hardcoded values that were only there to please the ndk-gdb
scripts.
Using lldb from within Android Studio is more comfortable anyway :-)
Change-Id: I31c3af4847a479c56b3fcd6b5bed114e004bf0d2
Reviewed-on: https://gerrit.libreoffice.org/41950
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
This commit enables HAVE_FEATURE_NSS for Android and fixes the svl/ vcl/
and xmlsecurity/ module to use NSS.
xmlsecurity/ wasn't built for Android previously, this commit enables
building xmlsecurity/ for Android and disables the support of gpgme
in the same module(Only for Android).
It also enables the linking of NSS shared libraries with
liblo-native-code.so and adds a rule to package them along with the apk.
Change-Id: I7d0341688ac979ae92e9145c37dd107670417fe1
Reviewed-on: https://gerrit.libreoffice.org/41308
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
and of course also for the Java part
Using build-id linker flag allows lldb to map the installed .so to the
non-stripped version on the buildhost.
Also ndk-gdb supports specifying a different package name on the
commandline, so no need for the error in configure anymore.
Change-Id: If6887a27cc8ab15ee6ab612502cacf0a22ade737
move preparation tasks (creating assets/processing files) from the
makefile into the gradle script
This allows much easier integration into android-studio (just open
android/source with android studio after running make to compile the
native library)
Change-Id: I9a9d6832797c24a7e195a1c7954cd6d20f5a8496
replace with corresponding Math functions and bump android sdk target to
23 to be able to use current appcompat-v7
pre-JIT FloatMath was faster, but since then was wrapper to Math anyway
http://developer.android.com/reference/android/util/FloatMath.html
Change-Id: I19e316fb7d32e616162a70fca8c565fa320a8c1a
Has all been obsoleted by LibreOfficeKit.
Only some MOBILE_* constant #defines are now left in touch.h, but probably
those are used only by dead code.
Change-Id: I646945c4408b4e6cd5510da535cfc12088dd391c
Also, if we are at it:
- clean up 'run' as well: since the doc browser is the default activity,
no need to pass the test doc path anymore
- make 'install' not depend on build: a full build would need a toplevel
'make' anyway
Change-Id: Ia55d52f767ab3e0be02a753a95b2aac02f8491cc
Previously we included the android support library v4 for some
GUI elements like GridView. This commit in addition adds the v7
appcompat library which is needed for the new Lollipop style
Toolbar and many other new GUI elements.
The appcompat v7 library is not distributed as only a jar file
but needs to be build (as it includes additional resources) and
included as a library project. So to do this the content is copied
from SDK and build with the build system. The files also include
the v4 so it doesn't need to be copied from SDK anymore.
The target had to be raised to v21 (Lollipop), however the minimum
SDK version remains unchanged.
Change-Id: I4f1a6ce69e7f6c3f9df784a6835f376a01d4dfdb
Otherwise FSStorageFactory::createInstanceWithArguments() would throw,
resulting in a css::configuration::CorruptedConfigurationException
later, that makes LO throw up its hands in Desktop::Main() and say that
the instset is simply corrupted, there is no point in continuing
further.
Change-Id: I3a401ee77f4fbf1a42a09c5fedd7681b4f32e952
This now also allows to specify the version number; now you want to use:
cd android/
make versionCode=<previous_version_num+1> key=<key_name> release-apk
Change-Id: I078e8dbbe671969fc3b228ac987cdb9a4a53b281
Note that getPackageName() also throws NameNotFoundException, so in the
unlikely situations in case:
- package info (class containing the package version) is not found or
- the package version is not in an "a/b" form
We still just don't show anything.
Also, mark the new TextView as android:textIsSelectable, so it's
possible to copy&paste the version for bugreport purposes.
Change-Id: I63b53cca4126da17bfbda0293d7c98e8524ef41a
It can happen that this directory doesn't exist and the copy
script fails (especially on a clean install). The script has been
modified to force create this directory before trying to copy
stuff into it.
Change-Id: Iedf3caef07e6896405750aea9e8f211b1e80dc3a
...and had started to have different values in instdir (ProductMajor=450, ProductMinor=0)
vs. installation sets (ProductMajor=50, ProductMinor=)
Change-Id: I4db2c07b5f7b011218833fc355a3097eb13d0cd4
Stephan Bergmann