LIBGCC_PATH was used before we initialized it, somehow it magically worked ...
Change-Id: I0a9a748ffbfc641f60736b0cf85d23f499b64e66
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Thumb instructions trigger a compiler bug in e.g. Qt Script
when used with armv5 in the recent versions of the toolchain.
We already disabled this for the 4.8 toolchain, but the bug
is also present in 4.9, so we need to disable it there as
well.
Change-Id: I6cadc7efd2b59b9a2ffec038559edd0e7782a4b9
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
it makes no sense to let every spec do that separately, as it's fixed
by the generator+shell.
putting it into a file which is loaded regardless of the spec also
allows us to remove the hardcoded fallbacks from qmake.
if somebody overrode the values in their spec for some weird reasons,
they'll need to override spec_post.prf.
shell-{unix,win32}.conf are now dummies and print warnings.
Task-number: QTBUG-37269
Change-Id: I66c24fb4072ce4d63fdbfc57618daa2a48fa1d80
Reviewed-by: Jochen Seemann <seemann.jochen@gmail.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@theqtcompany.com>
If we set ANDROID_NDK_PLATFORM env variable, QMAKE_DEFAULT_XXXDIRS still contains
old ANDROID_PLATFORM, so we need to overwrite it.
Change-Id: I917e24caa11bd589966b3fb11be3a9f3c4370b3e
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@theqtcompany.com>
On some devices, a miscompilation of libc.so has caused it to
return the wrong value. Instead of returning the dest pointer,
it returns dest + n. When compiling with optimizations turned
on, gcc may use this return value for subsequent accesses to
dest after the memmove() call, causing memory corruption.
This caused problems e.g. in QVector::prepend() which would
overwrite the whole vector with the new value.
Setting -fno-builtin-memmove disables the optimization and
works around this bug with very little risk or impact.
More information in:
http://code.google.com/p/android/issues/detail?id=81692
[ChangeLog][Android] Fixed device-specific crash on Samsung
Galaxy Tab 3 Lite 7" and some other devices.
Task-number: QTBUG-34984
Change-Id: I0c1347149eb5fe1c298758fe7de81aca4137f652
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
The new default compler is gcc 4.9, it is needed to compile
64 architectures.
Change-Id: I7ccbac7615b6dc20f5b0441908590de7d4a2e8cb
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@theqtcompany.com>
It's needed by androiddeployqt tool to run "zipalign" tool
and to set it to gradle properties.
Task-number:QTBUG-40481
Change-Id: I3dd665a7461a4e981867cdad75a50940e46a5ae6
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
that is, make it match the shell used by the make command.
this is unlike DIRLIST_SEPARATOR and DIR_SEPARATOR, which always match
the actual system shell.
Change-Id: I0fb277d75b32be029808623a3b77e1358c4e265e
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
On Android, there's a limitation set on the names of the libraries
you deploy that they must start with "lib" and end with ".so", so
Android apps will link against and deploy with the unversioned
libQt5FooBar.so libraries. When cross-compiling on Windows however,
due to the lack of symbolic links, the only installed library
used to be the main library target "libQt5FooBar.so.X.Y.Z" (for
version X.Y.Z.) This has been worked around in packaging, but
breaks building add-on modules on top of Qt, and is clearly
wrong.
This patch introduces a new "unversioned_libname" configuration
in qmake which is currently only supported for the Unix makefile
generator and only enabled for Android builds. When it is enabled,
only the unversioned library "libQt5FooBar.so" will be created.
Task-number: QTBUG-38347
Change-Id: Ia8897ca7a23a62e2a526d0e02854899b02eb19dc
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
The 4.8 compiler in the Android NDK sometimes produces uncompilable
code for armv5. In change b4252802b3
we reintroduced thumb instructions on armv5 with a work-around for
a particular case of this bug. However, the compiler now breaks
for Qt Script. There's no time to try to find a work-around for
this case, so we need to disable thumb again. It actually also
seems that this is completely broken with this compiler, so it would
probably be best to keep this disabled until the compiler is fixed.
Especially since armv5 will become part of the CI builds now, so the
compiler bug can end up blocking critical changes in the future if it's
re-enabled :/
Task-number: QTBUG-37376
Change-Id: I9e1b050ce596717ba1fa7ec2f5e8a3ce3581a3af
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
There is no real printing support though, it just prepares
the land for bringing printing support for API-19+.
Change-Id: Iabf78499f8fe1e4d2527e339cee5746acb8f3b3f
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
* The explicitly linked support libraries like gnustl_shared, etc. can be
linked as private libs, there's no need for them to show up in the .prl files of our libraries.
* Removed the explicit linkage against libsupc++, which is a static library whos symbols
are also available in libgnustl_shared. It is only needed when linking against gnustl_static,
which we fortunately don't do.
For QtQml on Android this is more than just cleanup. Without the first change, the libgnustl_shared
comes early on in the link line, because it is a dependency of for example Qt5Network. Anything that
qml.pro itself adds to LIBS comes afterwards. That is not intended, we want libgnustl_shared to come
at the end of the link line, in order to make sure that the linker finds an overriding symbol from
another library earlier in the link line first.
The explicit linkage against libsupc++ affects the same, as that's the library
that contains the symbol we want to override locally (__cxa_end_cleanup).
(needed for QTBUG-33892)
Change-Id: Id6dff733d6610ae8b15aa72f9cf60ae2c7834194
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: BogDan Vatra <bogdan@kde.org>
This outputs a json file with the necessary settings from qmake
so that an external build tool can easily get the settings without
having to parse the entire .pro source. Used by the androiddeployqt
tool.
Task-number: QTBUG-32856
Change-Id: I5d3ac0ab6a0350162d06b0a0bf0c9bcbd90d8b5a
Reviewed-by: Daniel Teske <daniel.teske@digia.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Qt Creator needs to parse the qmake.conf in order to get the
architecture for this Qt build when the Qt build is added.
At this point it does not know that it's reading an android
project, so it can't set the right environment variables. By
moving the error to the bottom, we let Creator parse the whole
file before getting an error, so that it can detect the options
it needs to define the Qt version.
Change-Id: I119523cb21e330dc5e957a6992476c4c3d4ab7b5
Reviewed-by: Daniel Teske <daniel.teske@digia.com>
Gcc 4.8 uses dwarf-4 by default which is not supported
by the GDB that is officially shipped with the NDK.
Change-Id: I913a038e095df52b0defd5d3da2606ef2e5456b7
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Set the SONAME to the library name without the major
version number appended, as android does not have the versioned
symlinks.
Change-Id: I41c504869019a393a3f112b2f7fc81c7ad5afa1c
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: BogDan Vatra <bogdan@kde.org>
When building in thumb mode for armv5 applications will crash
with SIGILL on startup. This has been observed on armv7 devices
and emulators. It could be a bug in the gcc 4.4.3 toolchain, but
since the other toolchains in the NDK have other bugs that
make it impossible to use them for building, we need to disable
thumb until the cross-compiler has been fixed.
Task-number: QTBUG-31338
Change-Id: I22dd228158ef8c43b0b1d6e549d5725c1930536b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
We'll use nm to get the listing of symbols in the next commit.
The -P option is "portable", which sounds like a good idea. I don't
have access to any of the commercial Unix systems, but I do remember
them printing a different format than GNU binutils's nm.
Change-Id: If6f80624bedaf2b1dabf608e16aa097d9910d739
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
... in mingw-using specs because msysgit
doesn't provide install.exe and Windows
doesn't care about Unix permissions anyway.
Task-number: QTBUG-31147
Change-Id: Ic8032ca1a970ef41381852b6c5c372b805a124f1
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
There is no need to, and it confuses the generation of cmake files.
Change-Id: I00c8751990e707cf32652babcb9af3e4b681561a
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Need to set MINGW_IN_SHELL to 1 and QMAKE_DIR_SEP to /
Change-Id: If470f1a4617555d6bc551e8cdf917779d0e64e62
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Use the appropriate way to load the .pri file which also
checks for its existence before loading it.
Change-Id: I7d36da1593bb7fa1b5f6fd4d10b69e20c6aaa836
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Replace all tabs with proper space characters and consistently align
the '=' characters. The default alignment for the '=' of 25 characters
has been left as is to get a minimal diff. Lines with the '=' further
to the right and those belonging to 'proper code (TM)' have not been
touched.
The work was mostly done using the following python script (might
come in handy again...):
import sys, re
indent_eq = 25 + 0*4 # 25 characters was the most widely used indentation for the '=' character
p = re.compile(r'(\w+)[ \t]*([\-\+]?)(=$|= )[ \t]*(.*$)')
for fn in sys.argv[1:]:
with open(fn, 'r+') as f:
lines = []
nl_count = 0
continuity_indent = None
for l in f:
m = p.match(l)
nl = l
if m:
n_spaces = max(m.start(3), indent_eq - 1) - len(m.group(2)) - len(m.group(1))
if m.group(2) and m.start(2) >= indent_eq-1 and m.start(2) % 4 == 0:
n_spaces -= 1 # left-shift '+=' by one if the '+' is aligned to a multiple of 4
n_spaces = max(1, n_spaces) # we want at least one space before '='/'+='
nl = m.group(1) + ' '*n_spaces + ''.join(m.group(2,3,4)) + '\n'
continuity_indent = nl.find('= ') + 2 if l[-2] == '\\' else None # remember indent on '\\$'
elif continuity_indent:
nl = ' '*continuity_indent + l.lstrip()
if l[-2] != '\\': # check when to stop the continuation
continuity_indent = None
elif l.startswith('#'):
nl = l.expandtabs(2)
if l != nl:
nl_count += 1
lines.append(nl)
if nl_count > 0:
print fn, nl_count, len(lines)
f.seek(0)
f.writelines(lines)
f.truncate()
Change-Id: I1d2870d0a2fe2e30d398c140fe523e69dd20c81b
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Based on the Necessitas project by Bogdan Vatra.
Contributors to the Qt5 project:
BogDan Vatra <bogdan@kde.org>
Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
hjk <hjk121@nokiamail.com>
Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Paul Olav Tvete <paul.tvete@digia.com>
Robin Burchell <robin+qt@viroteck.net>
Samuel Rødal <samuel.rodal@digia.com>
Yoann Lopes <yoann.lopes@digia.com>
The full history of the Qt5 port can be found in refs/old-heads/android,
SHA-1 249ca9ca2c7d876b91b31df9434dde47f9065d0d
Change-Id: Iff1a7b2dbb707c986f2639e65e39ed8f22430120
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
