Remove the VP8 and VP9 decoder and the subsequently unused functions.
This drops the size to libmozavcodec to around 1MB down from 4MB.
MozReview-Commit-ID: Ge57fauG35L
--HG--
extra : rebase_source : 3f667d7bf89036e9b059727d846af2504ce488b3
For now this is a C decoder only and FFmpeg isn't optimised to only decode flac (both vp8 and vp9 C decoders code is included, however they aren't enabled).
MozReview-Commit-ID: 7ulwvJDJqVg
--HG--
extra : rebase_source : 7090eee32ba394224ef8c3c1c31abf1543b69975
This also introduces C{XX}_LDFLAGS variables which contain cflags that
are meant to be passed to the linker, and adds them to various linker
command lines in place of CFLAGS.
MozReview-Commit-ID: GyKlD9nMqrt
This commit establishes a separate variable to add PROFILE_GEN and PROFILE_USE
CFLAGS to compile and link command lines. Currently the make backend
orchestrates the pgo build steps and is the only thing aware of whether
we're in the profile generate or profile use stage. The flags are separated
here to allow other flags to be moved to mozbuild, but this will not yet
sufficient to perform a PGO build independent of the make backend.
MozReview-Commit-ID: IX30l2MvvNc
When we move to a shell aware split for AC_SUBST_LIST it will become an error
to emit an unquoted make variable reference. Currently this happens to only
occur during artifact builds when setting cflags related variables that aren't
needed there anyway, so here we skip settting those variables when a compile
environment isn't present.
MozReview-Commit-ID: EnHu48yyZ1C
The Rust optimization logic is tied to --enable-optimize/MOZ_OPTIMIZE
and --enable-debug/MOZ_DEBUG. In order to more easily implement more
customization, let's move --enable-optimize/MOZ_OPTIMIZE to
moz.configure so its value can be consulted there.
The logic here is a bit wonky. The option behaves like a boolean
or a string. If a string, MOZ_OPTIMIZE is set to 2. Otherwise it
is 1 or unset depending on the boolean value.
The custom compiler flags string is passed to old-configure, where it
overwrites whatever old-configure derived as the default value.
We stop short of moving all references to MOZ_OPTIMIZE_FLAGS to
moz.configure because there are a handful of them and I don't want
to scope bloat.
MozReview-Commit-ID: 6iNDu2HwLGr
--HG--
extra : rebase_source : a64f1236012d13913f21253df1b9b5ff0ae8ea6e
It is now closer in the file to where other default values are computed.
MozReview-Commit-ID: BffCEb6FAUP
--HG--
extra : rebase_source : 81b17131f9330d89818a36ffff625b672c19c01e
Bug 1365460 introduced code paths behind MOZ_DEBUG #ifdefs, but
MOZ_DEBUG is never defined, while it is available in CONFIG in
moz.builds. This is kind of a confusing situation, but the fact that
we've been able to avoid those problems for so long would tend to
put the blame on mozjemalloc, and fixes should go there.
Except that bug 1261161 explains that the only existing alternative
(the DEBUG #define), as used in MFBT, is not working for spidermonkey,
so it actually makes sense to converge to MOZ_DEBUG rather than DEBUG.
So start defining MOZ_DEBUG globally, fixing the mozjemalloc issues of
not having the debug code enabled. Bug 1261161 can then take care of
changing the DEBUG #ifdefs.
--HG--
extra : rebase_source : 37e3d03ac8350c62c8059d4ca01d1ecfdf5f421a
Currently mozconfig.cache overrides a few build options for sccache. This
patch moves them into toolchain.configure so that the build system will
set them properly when sccache is in use. Additionally, {CC,CXX}_WRAPPER
are set in config.mk, so just avoid setting them when sccache is in use.
MozReview-Commit-ID: FYlVKRI8OiN
--HG--
extra : rebase_source : 00715beb5fbd2c11311dec43809bd1febab56a11
extra : intermediate-source : 0f2b1b75b83737378d882a3c3e0d8dfb4efecd1f
extra : source : a8032ae9cb2ad1c4574c6ac6f5c2778863cd71e0
This adds a section in both old-configure.in's for hardening flags.
In old-configure.in we add $HARDENING_CFLAGS (which are turned on by
--enable-hardening) and are defined in toolchain.configure (and
which does compiler detection there.) We then add non-optional
hardening flags, performing compiler detection here.
In js/src/old-configure.in we follow the same pattern, but omit
$HARDENING_CFLAGS because we don't apply the current lone flag
to js (doing so is Bug 1359905).
MozReview-Commit-ID: EFE0Pc7yZHa
--HG--
extra : rebase_source : f793e665cabe3d33bdc73fc8a68208d69b55137a
The goal is to use a newer Android-Gradle build plugin version (2.3.3
is latest stable). That requires a modern Gradle (anything 3.3+, but
3.4.1 is the default from my Android Studio), and also a newer
build-tools (25.0.3 is latest stable).
The locations of lint output changed, and we want to use the standard
output location because it's difficult to accommodate variant details
in custom names. We change the location of findbugs output to follow
suit.
This requires either:
- fixing lint errors
- adding to the lint whitelist
- using the new lint baseline
It's best to use the new lint baseline, which will happen in the next commit.
MozReview-Commit-ID: D19FzIDCJrE
--HG--
extra : rebase_source : 12d132c0c3e0dbe2b8873b31360ea96d612de44c
Nothing in js/ uses TARGET_XPCOM_ABI, so the code in js's configure is
dead. Since TARGET_XPCOM_ABI is a Gecko-only thing, the reasonable
place for it is toolkit/moz.configure.
TARGET_XPCOM depends on TARGET_COMPILER_ABI, and the value of
TARGET_COMPILER_ABI is easily derivable from the current target and
compiler. The only notable change we make in the conversion to
moz.configure is that the ARM old-ABI is no longer supported by any of
our current targets: Android and Linux for ARM targets are both
exclusively EABI nowadays, so we have one less case to worry about.
TARGET_COMPILER_ABI's sole purpose was to provide information for
TARGET_XPCOM_ABI, and since we are deriving TARGET_XPCOM_ABI exclusively
inside moz.configure, we can remove TARGET_COMPILER_ABI.
Use of this define was removed in bug 866988, but is still listed in
non_global_defines.
MozReview-Commit-ID: BIiwlxL6i7i
--HG--
extra : rebase_source : 76488041e4144eccc63854a9267bd42adf3b4a85
Some gaia-related code was also found and removed as part of the
cleanup.
MozReview-Commit-ID: DEjVSljzzu1
--HG--
extra : rebase_source : 58c4c34df44a258d90029853f29ea01338bd142b
This option was added ~decade ago; AFAICT from bug archaeology, the
option was added to prevent our servers from being overwhelmed.
Somewhere over the years, however, we obtained more capable servers and
the option disappeared from mozconfigs. It seems moderately unlikely
that we'll have a need for this option again, and we could reintroduce
this patch very easily in any event. Let's go ahead and remove it.
This is primarily for local builds that use the latest Android SDK.
MozReview-Commit-ID: 5uIjC6EhCho
--HG--
extra : rebase_source : 338caa12cf72a711cf3dbc2fd67df0afef1ef373
This patch does two things:
- add a Gradle-only ANDROID_COMPILE_SDK_VERSION substitution;
- uses it while uniformizing all of the Gradle Android SDK version
configurations.
The approach is fairly standard (and we were using it already); see,
for example
https://medium.com/@ali.muzaffar/gradle-configure-variables-for-all-android-project-modules-in-one-place-5a6e56cd384e
This will make bumping the Gradle configuration versions forward
easier.
MozReview-Commit-ID: 1j5siCvR5qt
--HG--
extra : rebase_source : 07afb00de0e4a72af4026eb19ff4f2530c119336
MOZ_REPLACE_MALLOC_LINKAGE was added back when there were problems with
getting weak references working properly for replace-malloc.
Versions of OSX < 10.6 needed flat namespace, but aren't supported
anymore.
Versions of Xcode < 4.5 required flat namespace + a dummy library in
order to produce proper weak references. There is virtually nobody still
building with such an ancient toolchain.
Keeping those around doesn't /really/ hurt, except recent versions of
Xcode don't expose dyldinfo in /usr/bin, used for the configure test.
Consequently, MOZ_REPLACE_MALLOC_LINKAGE ended up being set to use the
dummy library setup, which, by using flat namespace, now causes harm in
bug 1356701, causing bug 1378332.
--HG--
extra : rebase_source : e3edc1f2cf905943c33fafeb631f2f88fc87167e
Add the dependency between "MOZ_ENABLE_SKIA_PDF" and "MOZ_TREE_FREETYPE" on Windows:
- let |tree_freetype| returns true if |skia_pdf| returns true on Windows, and
- avoid defining "MOZ_ENABLE_CAIRO_FT" on Windows ("cairo-ft-font.c" includes <dlfcn.h>, which only exists on posix platforms)
MozReview-Commit-ID: 6CWVwzIHL1Q
The linker uses zlib. The linker is in mozglue, zlib is in libxul by
default. As a consequence, we made --with-system-zlib a requirement for
builds enabling the linker.
In the meanwhile, we added an option that makes zlib built in mozglue
for different needs, which, in fact, also allows to do that when the
linker is enabled.
So, allow to build without system zlib when the linker is enabled.
--HG--
extra : rebase_source : 873a87b17b306fc392018049e01cf794b63a6206
For parts of configuring Stylo, we need information about the library
extensions on all of our platforms, and this change is a reasonable way
to get at that information without duplicating it in two places. Plus
moving more things to moz.configure is more better.
Before, MSVS was set in old-configure and could only be unset or
"2015." We move the definition of the variable to moz.configure
and support defining its value as "2017" when VS2017 is being used.
As part of this, I discovered that GYP barfs with a "2017" value.
This is likely a limitation of the legacy version of GYP we have
vendored. Rather than go down the rabbit hole of upgrading GYP,
I added code to convert the value to "2015." This preserves existing
behavior and unblocks us from setting MSVS_VERSION properly. A
warning is emitted to remind us to remove this hack once GYP is
upgraded.
After this commit, we now generate native VS2017 solutions and
projects when building with VS2017.
MozReview-Commit-ID: BvNJX3F8qCn
--HG--
extra : rebase_source : 13a495856a83d9ca7afbb4770ebab1cc7f651cae
The test in old-configure.in for MOZ_REPLACE_MALLOC_LINKAGE runs
dyldinfo, which is a native tool on OSX. We have it in the cross
toolchain as well, but as x86_64-apple-darwin11-dyldinfo. We should use
the TOOLCHAIN_PREFIX here to make sure we get the same result as native
builds.
MozReview-Commit-ID: 3jyzpaM8ZGy
--HG--
extra : rebase_source : fa0d2c1470f8d61b3f31c9a46609d8911ae2ff91
js/src/builtin/Intl.js will spew three C4819 warnings every time non-Western
developers build the tree locally on Windows because of non-ASCII characters
in the file. Usually we have compilers use UTF-8 to prevent this, but it did
not work for Intl.js because embedjs.py did not pass around $CPPFLAGS.
--enable-warnings-as-errors did not catch this because embedjs.py did not
pass around warnings-as-errors flags either.
This patch will fix both issues.
MozReview-Commit-ID: 5D1TCGnIX1T
--HG--
extra : rebase_source : 9f353d46d0a5320f9a0b9be2aea11fb66d6a88b4
-Os generates smaller code, at the expense of performance. On desktop,
this tradeoff is not necessarily the best, especially when considering
the vast performance difference. Most downstream redistributors also
don't do PGO and don't override defaults, so they would tend to ship
slower builds as a consequence.
We however keep -Os as default for debug builds for now, because -O2
triggers -Werror=strict-overflow failures somehow.
--HG--
extra : rebase_source : ad2f4fedb7934b6b23b84412c86b30d29a8fb5f8
-Os generates smaller code, at the expense of performance. On desktop,
this tradeoff is not necessarily the best, especially when considering
the vast performance difference. Most downstream redistributors also
don't do PGO and don't override defaults, so they would tend to ship
slower builds as a consequence.
--HG--
extra : rebase_source : 7d66393d03b9424fdcc46874a09239ffe46cfb33
This intentionally allows to set MOZ_INSTALL_TRACKING without
reference to the milestone being release or beta. That is, we
separate the default value (which depends on release or beta) from the
value specified, making life easier for developers.
MozReview-Commit-ID: 3vPF7KO7fEX
--HG--
extra : rebase_source : 8d5764104b5322a32e4a048bfd3222f62fed73bb
We now have code that unconditionally requires the rust
compiler and are committed to adding more. Remove this
last vestige of conditional support.
MozReview-Commit-ID: EK6FBnAbR
--HG--
extra : rebase_source : 6efda10a74f9ca0482304c2b1ffe6941e42138f8
The first attempt at this patch in Bug 1314979 turned off the entire
ANGLE section for MINGW builds on Linux for Windows. This was
incorrect, as it also turned off includes of very important files
like d3d11.h
This changes it so we go through the ANGLE section, but if we don't find
a Windows SDK we only consider that an error when we're compiling with
MinGW for Windows _on_ Windows. If the Host and Target OSes don't match
(meaning we're compiling on Linux for Windows) we ignore the lack of a
Windows SDK.
MozReview-Commit-ID: KyggI6yJCEM
rustc generates .lib files for its libraries when compiling for Windows
(even using MinGW on Linux). But MinGW expects .a files. So we add in
rust-specific prefix and suffixes so MinGW builds can find the libs that
rustc generates. (And the RUST_LIB- variables default to the same vales
as the LIB_ variables otherwise.)
MozReview-Commit-ID: ClsA0YuJaxh
--HG--
extra : rebase_source : 7b46460c94ceb34b7a5a302ce91d3f1dca600041
This flag enables the stack-cookie exploit mitigation for all functions which
manipulate stack-based buffers, providing better protections than
-fstack-protector, at considerably lower performance overhead than
-fstack-protector-all.
r=froydnj
MozReview-Commit-ID: 7ZNAHHAf376
--HG--
extra : rebase_source : 6d5ccbb9537372912c3f5a73fe0fdc65bb68ac08
MINGW builds do not need any of the checks that are performed in the
windows.configure file. Nor do they the D3D compiler DLL that is
needed for ANGLE, so we can skip that entire section in
old-configure.in.
MozReview-Commit-ID: DqufbgGoGy4
--HG--
extra : rebase_source : d5f1ed371f79a8a16f888ccc5d058ac72a69f34f
Historically, we had support for some GNOME VFS protocols through the
gnomevfs library, and this was under extension. This may not have been
built by default when it was introduced, but GNOME upstream moved those
things into Gtk itself, and we then got support for the new Gio-based
protocol, similar to what we had through the gnomevfs library.
Time passes, and we switched off the gnomevfs library entirely, and
enabled the Gio-based protocol handlers by default. We then removed
everything related to the gnomevfs library.
Fast forward to now, and disabling Gio support in Firefox just doesn't
make sense, and leaving the gio protocol handler as an extension doesn't
make sense either.
As it is a protocol handler, its natural place is under
netwerk/protocol, which is where we're moving it here.
The netwerk/protocol subdirectories being handled automatically, we
don't need to add the moved directory in any DIRS variable.
--HG--
rename : extensions/gio/moz.build => netwerk/protocol/gio/moz.build
rename : extensions/gio/nsGIOProtocolHandler.cpp => netwerk/protocol/gio/nsGIOProtocolHandler.cpp
extra : rebase_source : 071a9cb1769f013717357458df24e2fd9570ccf4
Historically, we had support for some GNOME VFS protocols through the
gnomevfs library, and this was under extension. This may not have been
built by default when it was introduced, but GNOME upstream moved those
things into Gtk itself, and we then got support for the new Gio-based
protocol, similar to what we had through the gnomevfs library.
Time passes, and we switched off the gnomevfs library entirely, and
enabled the Gio-based protocol handlers by default. We then removed
everything related to the gnomevfs library.
Fast forward to now, and disabling Gio support in Firefox just doesn't
make sense, and leaving the gio protocol handler as an extension doesn't
make sense either.
As it is a protocol handler, its natural place is under
netwerk/protocol, which is where we're moving it here.
The netwerk/protocol subdirectories being handled automatically, we
don't need to add the moved directory in any DIRS variable.
--HG--
rename : extensions/gio/moz.build => netwerk/protocol/gio/moz.build
rename : extensions/gio/nsGIOProtocolHandler.cpp => netwerk/protocol/gio/nsGIOProtocolHandler.cpp
extra : rebase_source : fe3c9480cee468aa2a24fd34e569b58e4f2c9c9a
It looks like Google decided to split these jars out a bit, so we need to piece
them all back together.
We could probably just query the sdk version instead, but I'm not 100% sure
know when this setup changed - moreover we don't know when (if?) the paths
are likely to change again. SDK 26.0 still has lint 25.3.1, so the SDK and
lint versions don't appear to be tied.
It seems that only the lint* jars are needed to compile 'build/annotationProcessor',
however we need all the remaining jars in the classpath when running that code
in 'widget/android/bindings'.
MozReview-Commit-ID: GAKwMrVXW55
--HG--
extra : rebase_source : 4e790aaccae8ccc3f151c39bf1ef4404b2581d7a
Android/x86 runs xpcshell and a limited of mochitest. These tests are already passed on m-c.
MozReview-Commit-ID: 7Qkqi0dYh95
--HG--
extra : rebase_source : 4cc064aea022a8ba17d9a78a1f36a612ab877898
Now that we don't support XP it's reasonable for us to drop support for GDI
fonts and rely solely on DirectWrite. We should start this process by removing
support for building without DirectWrite.
Until we have the ability to properly make these flags conditional in
moz.configure, do not perform these checks.
MozReview-Commit-ID: CexvgiadIw0
--HG--
extra : rebase_source : 3944a7c98a6570bf0f30dc01a9d895cc07cc0ff1
Although bug 1322703 is backed out, we can remove -FS options to build SIMPLE_PROGRAM.
MozReview-Commit-ID: 7uO3We5hc5n
--HG--
extra : rebase_source : 6f909c4f38ba5a5bdffed7f9ca5be0030b9c681b
Although I add -FS option to support VS2013+ by bug 915973, after landing bug 1322703, -FS option is unnecessary now.
MozReview-Commit-ID: 4SYmwGXHA9U
--HG--
extra : rebase_source : d6abc1d03f0fede48c11ee8a1c159d74d530169d
Currently mozconfig.cache overrides a few build options for sccache. This
patch moves them into toolchain.configure so that the build system will
set them properly when sccache is in use. Additionally, {CC,CXX}_WRAPPER
are set in config.mk, so just avoid setting them when sccache is in use.
MozReview-Commit-ID: FYlVKRI8OiN
--HG--
extra : rebase_source : cc7e4346869b98a52840c101824044abc236637f
This removes the UNIFY_DIST and UNIFIED_BUILD variables, as well as the
--unify flag from the packager and UnifiedBuildFinder from mozpack. As a
result the STAGEPATH variable is never defined anymore, so its uses can
be removed as well.
test_unify.py is currently the only mozbuild/mozpack test that fails
without running configure first, and there isn't much point in fixing
tests for things that we don't actually use anymore.
MozReview-Commit-ID: F5q1FPW3Did
--HG--
extra : rebase_source : cadbd237f51c23ea1983135294521d628d16f0df
Intl API (ECMA-402) already supports most web browsers such as Edge (including Windows 10 Mobile), Chrome (including Android), and upcoming Safari 10(https://developer.apple.com/library/prerelease/content/releasenotes/General/WhatsNewInSafari/Articles/Safari_10_0.html).
Also IDN2008 support, and number control by <input type="number"> require ICU. And, although gfx has own unicode table, we want to remove this if ICU is supported on all platform.
Also, new L20N framework (aka L20N) wants to use Intl API to localize UI.
So I would like to use ICU as default on Fennec Android.
Negative issue is file size. ICU has big table, file size of libxul.so will be following. This support requires additional 4.4MB.
before this ... 26287379 bytes
after this ... 30692696 bytes
Although android OS already has ICU, ICU version is different per OS version. And our ICU requirement is 50.1+ (Android 4.0 uses ICU 48) and exported fucntion name of ICU has ICU version suffix. So we cannot use it.
MozReview-Commit-ID: 5R5iNMzWNjS
--HG--
extra : rebase_source : fceb3ee320f885e7c03749f76cf0ad7699a21c2c
The constexpr warning no longer appears in VS2015 headers.
I spot-checked a few other warnings in the list, and we still need to keep them.
Notably, we still need -Wno-ignored-attributes, but now for a different reason!
MozReview-Commit-ID: LMqJX1KlAra
--HG--
extra : rebase_source : c1ad696f19975a1252a9bf2b01771530183c6c14
Jean-Yves Avenard
Chris Manchester
Sebastian Hengst
Daniel Holbert
Nick Alexander
Gregory Szorc
Mike Hommey
Ted Mielczarek
Tom Ritter
Jan Henning
Franziskus Kiefer
Glenn Randers-Pehrson
Nathan Froyd
Mike Shal
Ting-Yu Lin
Kai Engert
Liang Wei
Wes Kocher
Jon Coppeard
Ryan VanderMeulen
Nevin Chen
Eric Rahm
Carsten "Tomcat" Book
mozilla@mozilla-23773
Zibi Braniecki
Sebastian Kaspari
Marco Castelluccio
Makoto Kato
Bruce Sun
Mike Taylor
Stephen A Pohl
Masatoshi Kimura
Ben Hearsum
Petr Sumbera
Paolo Amadini
Jan Beich
Ralph Giles
Alex Gaynor
Iris Hsiao
Andrzej Hunt
Dão Gottwald
Jeff Muizelaar
David Anderson
Gian-Carlo Pascutto
Jeff Gilbert
Phil Ringnalda
Ted Campbell
David Major