Found when attempting an enum conversion.
Not sure what effect this will have, it was introduced with
commit a0c7b2bc45
Author: Oliver Bolte <obo@openoffice.org>
Date: Tue Nov 16 10:24:21 2004 +0000
INTEGRATION: CWS eforms2 (1.48.52); FILE MERGED
which pushed a whole bunch of stuff together.
Change-Id: I571fc228c3c0e164aba261e49233ef68ba87e77d
Reviewed-on: https://gerrit.libreoffice.org/31491
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>
With Fedora 25 GCC 6.2.1 and my usual config (except without
--with-package-format since that is slowest now) on commit
086631af59 i get with CCACHE_DISABLE=1:
With -O0:
"make check"
44490.70user 3127.01system 1:15:23elapsed 1052%CPU (0avgtext+0avgdata 2541312maxresident)k
"make check", incremental
7368.19user 153.98system 15:30.06elapsed 808%CPU (0avgtext+0avgdata 510200maxresident)k
"make CppunitTest_sc_functions_test"
272.13user 0.65system 4:33.47elapsed 99%CPU (0avgtext+0avgdata 258884maxresident)k
With -Og:
"make check"
46076.01user 3170.91system 1:16:19elapsed 1075%CPU (0avgtext+0avgdata 2613716maxresident)k
"make check", incremental
5478.33user 157.07system 11:50.42elapsed 793%CPU (0avgtext+0avgdata 454980maxresident)k
"make CppunitTest_sc_functions_test"
188.48user 0.62system 3:09.59elapsed 99%CPU (0avgtext+0avgdata 259096maxresident)k
So we now have so many tests that the time taken by GCC doing optimization
is brought back via the tests finishing faster, particularly if we
assume a non-negligible ccache hit rate for developers in practice.
Add this to gb_DEBUG_CFLAGS instead of gb_COMPILERNOOPTFLAGS because
presumably the bridges code that uses gb_COMPILERNOOPTFLAGS really wants
-O0; the gb_DEBUG_CFLAGS is added later so -Og overrides -O0.
It is an open question how well debugging in gdb actually works with
-Og, some experimentation is needed; "man gcc" claims:
If you are not using some other optimization option, consider using -Og
with -g. With no -O option at all, some compiler passes that collect
information useful for debugging do not run at all, so that -Og may
result in a better debugging experience.
Change-Id: I103499f398b69397cf5aa9993a242680966ce999
Reviewed-on: https://gerrit.libreoffice.org/31334
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
gcc-4.8 introduced a new optimization level -Og which enables
optimizations that do not interfere with debugging.
When configured with --enable-debug, use -Og rather than -O0
if available.
Change-Id: Iff3f98d736681ae34e49b96510228a14ce456b34
Reviewed-on: https://gerrit.libreoffice.org/31333
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
And replace gb_AllLangResTarget_ALLTARGETS with
gb_AllLangResTarget_REGISTERED which should have
the same content and is already used.
This actually helped to find a problem, fixed in
52d409f0d6
Change-Id: Iae551d7be221c5655dee1bc9ad273c8822d45178
If we skip to the first NL, then we start tokenizing some XML as PDF
data and soon error out due to an unexpected keyword.
Change-Id: I86b540a014e5a92ea4376ed765385a2ee568a3c1
Reviewed-on: https://gerrit.libreoffice.org/31472
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Assume T0 calls ThreadPool::pushTask twice, then ThreadPool::waitUntilDone:
T0 calls ThreadTaskTag::onTaskPushed:
mnTasksWorking = 1, maTasksComplete.reset
T1 runs the 1st task to completion and calls ThreadTaskTag::onTaskWorkerDone:
mnTasksWorking = 0; suspended...
T0 calls ThreadTaskTag::onTaskPushed:
mnTaskWorking = 1, maTasksComplete.reset
T1 continues in the call to ThreadTaskTag::onTaskWorkerDone:
..., maTasksComplete.set
T0 calls ThreadTaskTag::waitUntilDone and immediately returns
T2 only now starts to run the 2nd task
Change-Id: Ic29101a4791fca2a1a4d54b559f10ff706e8a20d
...seen JunitTest_sc_unoapi_4 fail once in
sc.ScHeaderFooterTextCursor::com::sun:⭐:text::XTextRange with an empty
RuntimeException when calling some remote getString.
Change-Id: Id631feffce810b40825fe0fa49d8f1846f045033
Otherwise, the 'newline' is passed as a second parameter to
gb_Module_add_moduledir which breaks build. :/
Change-Id: I02eba5ddef9f7a6dccbb7d135a7e8c68629da9c4
We do not require to read in gif extension data when it is ignored
This is used to skip the Adobe XMP gif extension
Change-Id: Id64f373a4eae764227c0b7ac6687697666991ec7
Reviewed-on: https://gerrit.libreoffice.org/31410
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
