| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Change-Id: I15a655045d43ea5f2cfe1158016db6426ded0224
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
https://bugs.webkit.org/show_bug.cgi?id=181827
rdar://problem/36393031
Reviewed by Daniel Bates.
* wtf/spi/darwin/XPCSPI.h:
git-svn-id: http://svn.webkit.org/repository/webkit/trunk@227212 268f45cc-cd09-0410-ab3c-d52691b4dbfc
Change-Id: I808f2a26d9d7f108675617553439fbba2cd504e2
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change-Id: Idfb37cd43929536d4c67d1fa5d8cb598e9c0ad7e
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I38406c67ba681f7693a45b0fa2d8c2e44a6d9a0e
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I3d4c1bf98fe1d1aaad868f4a6210f241aa43a2eb
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: Id24f973470620b1c7a36576606b81f5df0fd75d0
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I1069f65fada85860f5c3da6c37a5ab5a8daecd51
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I3f9320f43d5d1fc5169a6c1b9dcea454974d6578
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
|
|
|
|
|
|
| |
The QT_OPENGL_ES_2 macro is a QtGui feature, so it's not provided by
qglobal.h, but instead qtguiglobal.h (a header new in 5.8.0).
Using the QtGui/ prefix here to make sure it includes even if
WTF/Platform.h is compiled without -I$QTDIR/include/QtGui.
Change-Id: I8d96dea9955d4c749b99fffd14cde9c566436381
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: Iccd335ea312d91e826885857fee6d0da3e913b8c
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: Ifdedb7bc3162434686201813dc1d994cf5ae7e70
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I480668a0cb8114dccf7a1195190a993282875759
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: Ied338d9a16b44b5da35d484e80a21acc284eb625
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I7e96b8e5ba5eef9b0c6c0835e200f770200573ed
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I7df106cef8ce93ce33e49ad6fb0d202cd066d87c
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I8886595114569f61168aed76b23ad7288c5cb34c
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I12a5c8f98aeaac1b542ce20c0ff297c57b5d84d1
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: Idc1156f4a64df4e7d89e90d4e03451f004c8ae6d
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I8a2348dd9eb59df840dde6b66141d562aff29bf4
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: Iccbb1e8bd8b7f72322614224c203fc509a8dbb79
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I9f66467721cf923468af87915c260c1bb44e013e
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I46d453477a779a8db544ffc50d088fa13331da9b
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I8dda875c38075d43b76fe3a21acb0ffa102bb82d
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
| |
Change-Id: I3b1d8a2808782c9f34d50240000e20cb38d3680f
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
|
|
|
|
|
|
|
|
| |
ICC's supplied math.h has #defines for isnan and signbit. So work around
that.
Change-Id: I149e0540c00745fe8119fffd1463d16ae0047cb1
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
|
|
|
|
|
|
|
|
|
| |
Add a __cplusplus condition to an already existing OpenBSD defined part
to activate the code only on older gcc compilers.
Change-Id: I3ae3c4f792af12868e5efe1c3536f9743de78ce3
Reviewed-by: Konstantin Tokarev <annulen@yandex.ru>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
| |
Change-Id: I32d6029139ac5b8118c3a350b68d6e55c8093a1a
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|\
| |
| |
| | |
Change-Id: I1df1e59d4a52a2667941105f3336f745a688bf94
|
| |
| |
| |
| |
| |
| |
| |
| | |
As it is rightly said in https://bugs.webkit.org/show_bug.cgi?id=125920#c2,
result should be updated if we want to return the 2nd half of the mmaped area.
Change-Id: I83dc6548d80138934565e53a4646d472dfd30210
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|\ \
| |/ |
|
| |\
| | |
| | |
| | | |
Change-Id: Iceb5c402acfd83632007cb5f7673b4d6cda08e6b
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Fixes building qtwebkit with MSVC 2015
Change-Id: I94b085ab822b3cdfa911814467e051c11c93d79e
Task-number: QTBUG-46344
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
|
|/ /
| |
| |
| |
| |
| |
| |
| | |
Ensure we define NSAutoreleasePool the same as in Qt to avoid conflicts
Change-Id: I8c3b1638e055400badf8df4b09003cd006c723a8
Reviewed-by: Liang Qi <liang.qi@theqtcompany.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@theqtcompany.com>
|
|\ \
| |/
| |
| | |
Change-Id: I824a984e8552340c2f9f631ac951f2c9eaa90b74
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
On Linux x32 ABI, pointer size is different than on classic x86_64
systems, so the current assembly is wrong.
To fix this, drop the quad qualifier and let the compiler infer the
argument size. This way it will work on both x86_64 and x32.
Done-with: Adam Borowski <kilobyte@angband.pl>
Change-Id: I7a5f9edbba08705397ce7e771b1567f2a140fb1b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lisandro Damián Nicanor Pérez Meyer <perezmeyer@gmail.com>
|
|/
|
|
|
|
|
| |
Improved warnings in gcc 5.0 has exposed some unclear expressions.
Change-Id: I2c269528f6246319dab1a83d929d55c8d8e5a17d
Reviewed-by: Andras Becsi <andras.becsi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Since the x86-64 assembler generator depends on jumps being expressable
as 32bit offsets we need to use the fixed size executable allocator and
not the on demand allocator. This patch removes its exclusion from being
used on Windows.
Task-number: QTBUG-42417
Change-Id: Ie6b127e9b4291869566fd99cc7efe05b566c8e66
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
Use the GCC feature defines to check for support for ARMv6 style
unaligned access support. This fixes configurations where it was
disabled and the ARMv6-M processors where it is not supported.
Change-Id: I00c9bb19ba582b64ecb9eaa6da9887ba0df8fd3c
Reviewed-by: Julien Brianceau <jbriance@cisco.com>
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In fixing QTBUG-35681 we disabled LLInt when hard fp is not set. This
is however too strong a condition as we only need to be allowed to use
floating point instructions, not use the hard floating point ABI.
LLInt is also disabled on Android because it currently doesn't build.
Task-number: QTBUG-35681
Change-Id: Ibe109b0f8fe6acaec90d42305c6151a440141ea9
Reviewed-by: Julien Brianceau <jbriance@cisco.com>
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
| |
Change-Id: I9848c2b384e6fa85a5224ce0691456105d2fe8a1
Reviewed-by: BogDan Vatra <bogdan@kde.org>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@digia.com>
|
|
|
|
|
| |
Change-Id: Ieecee7c05e78117e8c9fc335f60970e2144ede81
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
All the JIT implementations targets the little-endian versions of ARM,
MIPS and SH4 respectively. So in case anyone builds on those platforms
we should not enable JIT.
Task-number: QTBUG-41896
Change-Id: I5be167511e2c3eac6c63e603b82da84fb6b29eaa
Reviewed-by: Julien Brianceau <jbriance@cisco.com>
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
| |
We get negative values in monotonic increasing time because we forget
to initialize the start time.
Change-Id: I292792f8f996025e7f0254bb90391b7af6873877
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
| |
Change-Id: I868224d189fc34af2517bfbb8931591e30dcbaa0
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Fix method now available in QNX 6.6, lack of std::move even
with C++11 enabled, binary structure of stack frames on
QNX x86 and the wrongly advertised existence of MADV_FREE and
MADV_REUSE.
Change-Id: I0dacbd19ed932ec6b3b2cc1d625a347e169615a5
Reviewed-by: Milian Wolff <milian.wolff@kdab.com>
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|
|
|
|
|
|
| |
Patch taken from https://bugs.webkit.org/show_bug.cgi?id=125920
Change-Id: I6583ea37e00f4dff80971bd44906936229540171
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@digia.com>
|
|
|
|
|
|
|
| |
Solves building WebKit on HPPA.
Change-Id: I00a455ad03d3f896de589aaf6e6f4dc0e71398ac
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
The fix for the stack layout in applications that are loading .NET
libraries in addition did not update the assertions that make sure
that the actual end of the stack is where it should be.
Task-number: QTBUG-38067
Change-Id: Ib8be1228bc9f10baf8d3484bb280a3a5c8021ab8
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Under .NET, on WinXP and sometimes on Win7 we will have two blocks
of uncommited memory before the guard page, so look one block further.
In debug mode it would crash due to assert
guardPage.Protect & PAGE_GUARD
In release mode it would crash due to only using a 4K stack.
Change-Id: If9f2c84122ba95a18403a8f413d080dd5cc58250
Reviewed-by: Michael Bruning <michael.bruning@digia.com>
|
|\
| |
| |
| | |
refs/staging/stable
|