Age | Commit message (Collapse) | Author |
|
removed epoxy.android.patch: fixed upstream
(https://github.com/anholt/libepoxy/pull/263/commits/063c1df27974f58b0f3c0a57417732da068923cf)
Change-Id: If0c6c438014121bc480efeaeb1018b6b74e51fcc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140359
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
...with recent Clang 16 trunk since
<https://github.com/llvm/llvm-project/commit/7068aa98412ade19a34b7ed126f4669f581b2311>
"Strengthen -Wint-conversion to default to an error", causing
> workdir/UnpackedTarball/epoxy/src/gl_generated_dispatch.c:114547:77: error: incompatible integer to pointer conversion passing 'uintptr_t' (aka 'unsigned long') to parameter of type 'GLhandleARB' (aka 'void *') [-Wint-conversion]
> GEN_THUNKS(glAttachObjectARB, (GLhandleARB containerObj, GLhandleARB obj), ((uintptr_t)containerObj, (uintptr_t)obj))
> ^~~~~~~~~~~~~~~~~~~~~~~
> workdir/UnpackedTarball/epoxy/src/dispatch_common.h:150:40: note: expanded from macro 'GEN_THUNKS'
> GEN_GLOBAL_REWRITE_PTR(name, args, passthrough) \
> ^~~~~~~~~~~
> workdir/UnpackedTarball/epoxy/src/dispatch_common.h:95:14: note: expanded from macro 'GEN_GLOBAL_REWRITE_PTR'
> name passthrough; \
> ^~~~~~~~~~~
etc.
That only hit on macOS because of the different
> #ifdef __APPLE__
> typedef void *GLhandleARB;
> #else
> typedef unsigned int GLhandleARB;
> #endif
in OpenGL's glext.h, which
<https://github.com/anholt/libepoxy/commit/0cfb0a044bc43245409c020e2ca34091c96283d4>
"Fix most GLhandleARB warnings on OS X with a big comment in our code" had tried
to address by adding "a cast to uintptr_t to shut up the compiler" in its
src/gen_dispatch.py (but without stating what compiler diagnostics exactly were
supposed to be silenced by that cast that now started to cause the above issues
with Clang 16 trunk).
It turns out that at least my macOS build with Clang 16 trunk builds fine
without that cast, so just drop it for good. (And patching epoxy's
src/gen_dispatch.py means that UnpackedTarball_epoxy needs to switch from
gb_UnpackedTarball_set_pre_action to gb_UnpackedTarball_set_post_action.)
Change-Id: I2c87bca2cc5510d17098028d4532989f48e349a9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/138407
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
removed epoxy.pyresource-leak.patch in favor of upstream
https://github.com/anholt/libepoxy/pull/243/commits/d0b319e27b874a13f6b41f4759c272b08d74cb8a
Change-Id: I9d3545b1dcf9dbdd5e28a54e13477a027636c444
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/123882
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I3f7bd1befffb45b12bc178f9addf1445a6c33921
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/113944
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
As in, really disable, so that they do not even show. This moreover
avoids tons of D9025 warnings from MSVC about overriding -W4 with -w.
Change-Id: Ia2e72fd72d883d91bdd89e467ee42f259e2ae033
Reviewed-on: https://gerrit.libreoffice.org/72899
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
https://github.com/anholt/libepoxy/issues/180 sounds very similar and
1.5.3 apparently fixes that
Change-Id: I009f5bc82f9e8326a7028ed29d86733cce649d15
Reviewed-on: https://gerrit.libreoffice.org/71733
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
This gets rid of the horrible hack in gbuild.mk to accomodate the
case-incorrect iOS platform makefiles that cannot be renamed without
upsetting git on file systems that sadly lack the case sensitivity
feature.
Keep the macro defined to IOS though.
Change-Id: I1022bfef4900da00e75fc1ccce786b20f8673234
Reviewed-on: https://gerrit.libreoffice.org/62705
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
|
|
Change-Id: Ic3093d3c12c33cbcc09903409daec6277428ecd6
Reviewed-on: https://gerrit.libreoffice.org/59664
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
chmod -x for .patch, .pptm, and .vb
Change-Id: I98e1221e48df22e8b58aaf305898cbe301f187ce
Reviewed-on: https://gerrit.libreoffice.org/52568
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
Change-Id: I931f0d95cad0a63d344c84d89ba7c80f14be8ec9
|
|
…as building with debug on Android for example fails when linking due to
unsatisfied symbol glXGetProcAddressARB
Change-Id: Ibea56d182fcabff182b1622599950e3e7ebf3aaa
Reviewed-on: https://gerrit.libreoffice.org/42335
Reviewed-by: Christian Lohmaier <lohmaier+LibreOffice@googlemail.com>
Tested-by: Jenkins <ci@libreoffice.org>
|
|
Looks like neither the __GNUC__ specific __attribute__((constructor)) definition
of CONSTRUCT nor the _MSC_VER specific one (if the #ifs were reordered so that
clang-cl would pick the latter) with __declspec(allocate(".CRT$XCU")) are
supported by clang-cl, and both are rather silently ignored. That means that
library_init is not called, library_initialized remains false, and the first
call to get_dlopen_handle aborts.
But this whole "verify that get_dlopen_handle isn't called too early" business
is somewhat pointless here (and that's the only use of the CONSTRUCT macro, and
DESTRUCT isn't used at all), so just short-circuit it for clang-cl for now.
Change-Id: I5d50df3574d350f9591e807ef0fb6a1b02dc34ec
|
|
Change-Id: Ib1d4d3f88c1cc11ba440b1d8f7208575b5bbb42d
|
|
Change-Id: I44b935e2093a5b630f8009af939ce8e83413c35b
|
|
to build on the festering baseline on which gtk3 is disabled
Change-Id: I6c074382590fc60ae8572fbde225679e4309494e
|
|
because that works under wayland out of the box and gtk3 uses it already
Change-Id: Iefaac31e325534a81a5389f752804af917c1baef
Reviewed-on: https://gerrit.libreoffice.org/31213
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|