OSDN Git Service
Dan Willemsen [Tue, 7 Jun 2016 21:25:14 +0000 (14:25 -0700)]
Check that NDK-built modules only link to NDK-built modules
Modules built against the NDK should only link against modules also
built against the NDK (or link to the NDK prebuilts). This patch
attempts to catch these cases, and prints a large warning when this is
violated. Once the tree is cleaned up, this will change to an error.
Change-Id: Ib6ffcc38d9161abdbe45a58af26ba429fb6f1876
Dan Willemsen [Thu, 16 Jun 2016 02:20:57 +0000 (19:20 -0700)]
Add macros for printing pretty warnings/errors in rules
Change-Id: Ia58e6bc1328c84e5f4ba1f6a2fd2d650e94e127e
Treehugger Robot [Wed, 15 Jun 2016 22:41:32 +0000 (22:41 +0000)]
Merge "Fix kati --no_ignore_dirty usage"
Dan Willemsen [Wed, 15 Jun 2016 22:38:30 +0000 (22:38 +0000)]
Merge "Forbid libstdc++ on Linux and Darwin"
Dan Willemsen [Wed, 15 Jun 2016 21:55:58 +0000 (14:55 -0700)]
Fix kati --no_ignore_dirty usage
Kati only supports a single use of --no_ignore_dirty, so we were
ignoring the SOONG_ANDROID_MK file, and only detecting changes to
SOONG_MAKEVARS_MK. Fix this by using a pattern that should apply to both
of those makefiles.
Change-Id: I46390f9887f95f1db0efe4e93339667f35ebc67c
Dan Willemsen [Wed, 15 Jun 2016 20:41:45 +0000 (13:41 -0700)]
Forbid libstdc++ on Linux and Darwin
The last user has been removed, forbid any modules from selecting this.
Change-Id: Idd9ef6ca4b6c6754935d59c4a09c7d213ae481c1
Alex Klyubin [Wed, 15 Jun 2016 20:32:34 +0000 (20:32 +0000)]
Merge "Don't depend on Bouncy Castle."
Treehugger Robot [Tue, 14 Jun 2016 22:37:10 +0000 (22:37 +0000)]
Merge "Faster and cleaner way to obtain UTF-8 encoded form."
Treehugger Robot [Tue, 14 Jun 2016 22:14:06 +0000 (22:14 +0000)]
Merge "Use more prebuilt build-tools"
Alex Klyubin [Tue, 14 Jun 2016 21:18:21 +0000 (14:18 -0700)]
Faster and cleaner way to obtain UTF-8 encoded form.
Instead of specifying character encoding by name, the faster, cleaner,
and safer way is to use StandardCharsets.UTF_8.
Bug:
27461702
Change-Id: I897284d3ceeb44a21cc74de09a9b25f6aec8c205
Alex Klyubin [Tue, 14 Jun 2016 20:56:27 +0000 (20:56 +0000)]
Merge "APK JAR signature verifier."
Dan Willemsen [Mon, 6 Jun 2016 21:44:00 +0000 (14:44 -0700)]
Use more prebuilt build-tools
For acp, we've been using an old prebuilt in prebuilts/sdk, but it's not
part of the SDK. Instead, we'll use a prebuilt in the build-tools
repository.
For ijar, we've been using the host libstdc++ to workaround the lack of
libc++ on some unbundled branches. Instead, use a prebuilt that can use
libc++.
For ziptime, we've been disabling it on unbundled branches, due to the
lack of libc++. Instead, use a prebuilt version of ziptime that can use
the prebuilt libc++.
Change-Id: If80f845ea06f76e3fe6765964e77c864eaf303d0
Treehugger Robot [Tue, 14 Jun 2016 18:34:38 +0000 (18:34 +0000)]
Merge "Add a libc++ version of ijar for prebuilt use"
Alex Klyubin [Tue, 14 Jun 2016 17:26:41 +0000 (10:26 -0700)]
Don't depend on Bouncy Castle.
This switches PKCS#7 SignedData generation code from Bouncy Castle to
OpenJDK's proprietary internal API. This is to avoid depending on a
huge library that's not really needed. In the longer term, it's best
to add our own implementation of PKCS#7 SignedData building, parsing,
and verification. This will give the code more power to mimic what the
Android platform does.
Bug:
27461702
Change-Id: I29f1ceea1293e35424fde69c0f2969d551345100
Dan Willemsen [Tue, 14 Jun 2016 17:30:06 +0000 (10:30 -0700)]
Add a libc++ version of ijar for prebuilt use
I'll remove the old version once we've moved over to the prebuilts.
Change-Id: Ifd8d396f8b653abbe958cd34e69831a6ca378649
Alex Klyubin [Wed, 8 Jun 2016 21:54:28 +0000 (14:54 -0700)]
APK JAR signature verifier.
This adds JAR signature verification to ApkVerifier.
Bug:
27461702
Change-Id: Id2b72bea7869be66268f6bc1387e1559ee02ff9d
Treehugger Robot [Tue, 14 Jun 2016 17:22:25 +0000 (17:22 +0000)]
Merge "Skip uses-library check for preopted apps."
Jeff Hao [Mon, 13 Jun 2016 23:53:11 +0000 (16:53 -0700)]
Skip uses-library check for preopted apps.
Bug:
26880306
(cherry-picked from commit
b00263f96a90c9f77cf7d8d90742a0884290bf60)
Change-Id: I22beccaf61ca30cf1487a25f80e1c3fd7bdf2c62
Yabin Cui [Tue, 14 Jun 2016 00:16:34 +0000 (00:16 +0000)]
Merge "Add option to link static lite protobuf library."
Alex Klyubin [Mon, 13 Jun 2016 19:49:01 +0000 (19:49 +0000)]
Merge "More general OutputStreamDataSink."
Alex Klyubin [Mon, 13 Jun 2016 18:14:18 +0000 (18:14 +0000)]
Merge "Fix inefficiency in APK entry data alignment."
Alex Klyubin [Mon, 13 Jun 2016 17:19:26 +0000 (10:19 -0700)]
More general OutputStreamDataSink.
This replaces the less general DataSink which outputs into a
ByteArrayOutputStream with a more general DataSink which outputs into
an OutputStream.
Bug:
27461702
Change-Id: I9467f38c41f586b71f35edb3602fd6e57153184f
David Zeuthen [Mon, 13 Jun 2016 15:54:36 +0000 (15:54 +0000)]
Merge "bvb: Update path to key since repo was moved."
David Zeuthen [Mon, 13 Jun 2016 15:26:26 +0000 (11:26 -0400)]
bvb: Update path to key since repo was moved.
We recently moved Brillo verified boot from system/bvb to external/bvb
so update the path accordingly.
TEST=Build with 'BOARD_BVB_ENABLE := true' succeeds.
BUG=
29099910
Change-Id: I66f3b90e23d6b2afc09f81af571ede7b573325e9
Yohann Roussel [Mon, 13 Jun 2016 14:33:47 +0000 (14:33 +0000)]
Merge "Do not force turn off of Jack warning"
Alex Klyubin [Sun, 12 Jun 2016 00:42:29 +0000 (17:42 -0700)]
Fix inefficiency in APK entry data alignment.
26f00cda4b979d7e74db6872990682335b36612b introduced a bug where an
APK entry's extra field is padded for alignment purposes when no
padding is necessary because the entry is aligned without any padding
bytes.
Bug:
27461702
Change-Id: Icb164dbaa26d9686412e2920318a9f40c5ce9751
Treehugger Robot [Sat, 11 Jun 2016 00:56:31 +0000 (00:56 +0000)]
Merge "Redirect subprocess stderr to stdout in verbose mode."
Alex Deymo [Fri, 10 Jun 2016 23:38:31 +0000 (16:38 -0700)]
Redirect subprocess stderr to stdout in verbose mode.
This patch uses subprocess.communicate instead of subprocess.wait to
prevent deadlock if any of the child processes outputs too much data,
and redirects the subprocess output to stdout when running in verbose
mode.
With this patch `ota_from_target_files -v` prints the delta_generator
output in stdout, and no output if '-v' is not passed.
Bug: None
TEST=ota_from_target_files -v ...
Change-Id: Id66e4f3360a6f91d61a3ce96d53afbccdaa19da5
David Turner [Fri, 10 Jun 2016 18:04:26 +0000 (18:04 +0000)]
Merge "goldfish_setup: grant /system/bin/sh exec access"
Nick Kralevich [Wed, 25 May 2016 00:36:23 +0000 (17:36 -0700)]
goldfish_setup: grant /system/bin/sh exec access
The goldfish_setup shell script needs the ability to execute
the shell script interpreter. Allow it.
Addresses the following denial:
avc: denied { getattr } for pid=1220 comm="init.goldfish.s"
path="/system/bin/sh" dev="vda" ino=442 scontext=u:r:goldfish_setup:s0
tcontext=u:object_r:shell_exec:s0 tclass=file permissive=0
(cherrypicked from commit
501c88c0290eb2308c110398f106411260863bb8)
Bug:
28941573
Change-Id: I22d26e90f107c8d801229354a5e0513c37e6c31d
Vikas Marwaha [Thu, 9 Jun 2016 02:30:57 +0000 (02:30 +0000)]
Merge "Add variable PRODUCT_SHIPPING_API_LEVEL and make files with the new read only product property ro.product.first_api_level."
Dan Willemsen [Wed, 8 Jun 2016 21:09:34 +0000 (21:09 +0000)]
Merge "Rewrite LDLIBS and SHARED_LIBRARIES"
Alex Klyubin [Wed, 8 Jun 2016 20:51:58 +0000 (20:51 +0000)]
Merge "APK Signature Scheme v2 APK verifier."
Alex Klyubin [Tue, 7 Jun 2016 20:18:57 +0000 (13:18 -0700)]
APK Signature Scheme v2 APK verifier.
This adds the ApkVerifier class which verifies APKs using APK
Signature Scheme v2 only. In a follow-up commit this class will be
extended to verify APKs using JAR signature scheme when necessary.
The APK verifier is designed to not just verify an APK, but also
report errors, warnings, and information about signers in a structured
way, to enable tools to surface this information to users in various
ways.
Bug:
27461702
Change-Id: I10c6ba436021d86b6dbf6d3cf44494652adacb66
Alex Klyubin [Wed, 8 Jun 2016 15:17:42 +0000 (15:17 +0000)]
Merge "Fix DSA APK signatures for API Level 8 and lower."
Yohann Roussel [Wed, 8 Jun 2016 15:00:09 +0000 (17:00 +0200)]
Do not force turn off of Jack warning
When ANDROID_JACK_EXTRA_ARGS is defined there is no reason to still turn
off those warnings.
Change-Id: Ic7ac02c72ed3b7b0eaca9394c9fc7d92d5dfa871
Dan Willemsen [Tue, 7 Jun 2016 23:55:57 +0000 (16:55 -0700)]
Rewrite LDLIBS and SHARED_LIBRARIES
LOCAL_LDLIBS was the only correct way to use NDK libraries, but few used
it correctly. It also often got confused with LOCAL_LDFLAGS, so move the
flags to the correct variable.
For binaries that weren't using the NDK (empty LOCAL_SDK_VERSION), it
was never valid to use LOCAL_LDLIBS, as dependencies would not be
properly set up, and could lead to random build failures. So convert any
-l linker flags to using LOCAL_SHARED_LIBRARIES automatically.
For binaries built using the NDK (LOCAL_SDK_VERSION set), they were
required to use LOCAL_LDLIBS for prebuilt NDK libraries, otherwise they
would get headers and dependencies to the platform versions. Any
non-prebuilt LOCAL_LDLIBS would miss dependencies. So move the NDK
prebuilt libraries to LDLIBS from SHARED_LIBRARIES, and move everything
else to SHARED_LIBRARIES.
So now, for device modules, LOCAL_SHARED_LIBRARIES should always be
used, and we'll do the right thing. LOCAL_LDLIBS should only be used for
host libraries from the system.
Change-Id: Ide34c7afdcfb6507a378d45a42471729e489a9e0
Yabin Cui [Wed, 8 Jun 2016 02:21:12 +0000 (19:21 -0700)]
Add option to link static lite protobuf library.
Bug:
28114205
Change-Id: Id2b7e42017a53138c60b92de8d196ca92bc38053
Dan Willemsen [Mon, 6 Jun 2016 23:59:23 +0000 (23:59 +0000)]
Merge "Add NATIVE_TESTS class, move host native tests"
Treehugger Robot [Mon, 6 Jun 2016 22:51:56 +0000 (22:51 +0000)]
Merge "Switch some build tools to Soong"
Alex Klyubin [Mon, 6 Jun 2016 21:58:23 +0000 (14:58 -0700)]
Fix DSA APK signatures for API Level 8 and lower.
This modifies JAR signing code to produce DSA signatures which are
accepted by all Android platforms rather than only API Level 9 and
higher.
The issue is that by default Bouncy Castle uses OID 1.2.840.10040.4.3
(dsaWithSha1) in PKCS #7 CMS SignerInfo whereas Android accepts that
only since API Level 9. However, OID 1.2.840.10040.4.1 (dsa) is
accepted by all Android platforms.
Bug:
27461702
Change-Id: I24256a255bcdc2108bdb447557af7568a2c096e3
Dan Willemsen [Mon, 6 Jun 2016 21:41:32 +0000 (14:41 -0700)]
Switch some build tools to Soong
In preparation for including them in prebuilts/build-tools.
acp: We use a prebuilt in prebuilts/sdk, but it's not part of the sdk.
ijar: We use the host libstdc++ to workaround the lack of libc++ on
some unbundled branches.
ziptime: We disable this on unbundled builds, due to the lack of libc++.
Change-Id: Ib9766b1dbddd151c38ff27c529865200ab37fce1
Treehugger Robot [Mon, 6 Jun 2016 21:35:47 +0000 (21:35 +0000)]
Merge "Remove dead rgb2565"
Treehugger Robot [Mon, 6 Jun 2016 21:22:57 +0000 (21:22 +0000)]
Merge "Drop CLANG and CLANG_CXX in the environment"
Alex Klyubin [Mon, 6 Jun 2016 21:07:52 +0000 (21:07 +0000)]
Merge "Improve efficiency of using the DataSource abstraction."
Dan Willemsen [Mon, 6 Jun 2016 20:32:11 +0000 (13:32 -0700)]
Remove dead rgb2565
Change-Id: I2397dc97768f24a7c416bff96dd983fc796cbb53
Dan Willemsen [Mon, 6 Jun 2016 19:21:06 +0000 (12:21 -0700)]
Drop CLANG and CLANG_CXX in the environment
We'll explicitly pass these variables to the static analyzer builds in
binary.mk
Bug:
29071982
Change-Id: Ifc7633d36f5785fe16b63f54b885396c2cab542a
Alex Klyubin [Mon, 6 Jun 2016 19:40:23 +0000 (19:40 +0000)]
Merge "Switch signapk to apksigner-core."
Alex Klyubin [Mon, 6 Jun 2016 18:12:50 +0000 (11:12 -0700)]
Improve efficiency of using the DataSource abstraction.
This adds getByteBuffer and copyTo methods to the DataSource
abstraction. These methods enable the client to avoid unnecessary
copying of the data source's data.
Bug:
27461702
Change-Id: If4e9f902ea75c1ca5c7be0e20c0e7218faf9c504
Dan Willemsen [Sat, 4 Jun 2016 01:03:20 +0000 (01:03 +0000)]
Merge "Error on external includes"
Vikas Marwaha [Fri, 3 Jun 2016 22:02:06 +0000 (15:02 -0700)]
Add variable PRODUCT_SHIPPING_API_LEVEL and make files with the
new read only product property ro.product.first_api_level.
PRODUCT_SHIPPING_API_LEVEL declares which API level a device first
shipped with. We use this variable to set a read-only property that
can be used in CTS/GTS tests.
Change-Id: If54202075d3b92ed380655cde8b3fc50c5ba73b3
Treehugger Robot [Thu, 2 Jun 2016 23:17:59 +0000 (23:17 +0000)]
Merge "Default to enable tidy warnings in header files."
Treehugger Robot [Thu, 2 Jun 2016 22:44:55 +0000 (22:44 +0000)]
Merge "Add support for TARGET_RECOVERY_BRICK."
Chih-Hung Hsieh [Thu, 2 Jun 2016 22:20:05 +0000 (15:20 -0700)]
Default to enable tidy warnings in header files.
Change-Id: I3831f03facbbd6d17117d82f9ccac49bc2645237
Tao Bao [Wed, 1 Jun 2016 06:56:23 +0000 (23:56 -0700)]
Add support for TARGET_RECOVERY_BRICK.
TARGET_RECOVERY_BRICK specifies the file that lists all the partitions
to be bricked under recovery. The file, if present, will be copied to
'$(TARGET_RECOVERY_ROOT_OUT)/etc/recovery.brick'.
Bug:
27253717
Change-Id: Id84195b5ee1031a83bf7a077be070a3728416580
(cherry picked from commit
3051f4f11aef75af1d7cf623506ad6b4b648f3bd)
Treehugger Robot [Wed, 1 Jun 2016 23:28:10 +0000 (23:28 +0000)]
Merge "Add LOCAL_LOGTAGS_FILES that prebuilts can use"
Jeffrey Vander Stoep [Wed, 1 Jun 2016 23:18:45 +0000 (23:18 +0000)]
Merge "disable unpriv perf by default in user{,debug} builds"
Dan Willemsen [Wed, 1 Jun 2016 22:27:11 +0000 (15:27 -0700)]
Add LOCAL_LOGTAGS_FILES that prebuilts can use
Soong modules can define logtags files that need to be combined into
/system/etc/event-log-tags, so add a new LOCAL_LOGTAGS_FILES variable
that can be used to specify *.logtags file outside of LOCAL_SRC_FILES.
Bug:
28989759
Change-Id: I53c5d396dfb7c6006806758f351eb5cdde90fe74
Colin Cross [Wed, 1 Jun 2016 04:52:59 +0000 (04:52 +0000)]
Merge "Update default NINJA_STATUS to show finished edges"
Colin Cross [Wed, 1 Jun 2016 04:52:47 +0000 (04:52 +0000)]
Merge "Update Android.mk finder to work with Android.bp files"
Dan Willemsen [Wed, 1 Jun 2016 02:37:18 +0000 (02:37 +0000)]
Merge "Add Soong tools to PATH"
Dan Willemsen [Fri, 27 May 2016 22:15:47 +0000 (15:15 -0700)]
Add Soong tools to PATH
Change-Id: I8f8649106df4ba7cefbfa37a724ebe0635d90b76
Treehugger Robot [Tue, 31 May 2016 22:08:15 +0000 (22:08 +0000)]
Merge "Allow arm prebuilts for x86_arm builds"
Daniel Micay [Tue, 17 May 2016 04:40:52 +0000 (00:40 -0400)]
disable unpriv perf by default in user{,debug} builds
Bug:
29054680
Change-Id: If04bc63fc366006cde842a50b7c764f89d7449a6
Dan Willemsen [Tue, 31 May 2016 20:59:38 +0000 (13:59 -0700)]
Allow arm prebuilts for x86_arm builds
But only for static and shared libraries -- we disable building other
types of modules for arm, so don't support the prebuilts for them
either.
This is needed so that we can build these libraries under Soong, and
expose them to Make using BUILD_PREBUILT.
Bug:
27526885
Change-Id: I4f6ec922aed37591b6257b7022b81fb65d05b0fe
Colin Cross [Tue, 31 May 2016 19:38:04 +0000 (12:38 -0700)]
Update default NINJA_STATUS to show finished edges
Update default NINJA_STATUS that includes percentage to match
https://github.com/ninja-build/ninja/commit/
6c2c2eecbfd2e4ddeeacb0129944618a6ce3dc4f
Change-Id: I5694308abe26b71999d944c40962d59d50a73077
Colin Cross [Fri, 27 May 2016 22:23:13 +0000 (15:23 -0700)]
Update Android.mk finder to work with Android.bp files
The Android.mk finder should stop traversing when it finds an Android.bp
file, and any Android.soong.mk file at the same path should be included.
Update filter-soong-makefiles to accept a mixed list of Android.bp files
and Android.mk files, convert Android.bp files to Android.soong.mk files
if they exist, and ignore Android.mk files that have an Android.bp file.
Update first-makefiles-under to use the new --dir syntax for
findleaves.py and search for both Android.bp and Android.mk files.
Update all-makefiles-under and all-named-subdir-makefiles to search for
both Android.bp and Android.mk files.
Change-Id: Iddc51bf84593d5de3add02f2b2c065da08962af4
Treehugger Robot [Sat, 28 May 2016 00:53:47 +0000 (00:53 +0000)]
Merge "Add --dir option to findleaves.py"
Colin Cross [Fri, 27 May 2016 22:20:31 +0000 (15:20 -0700)]
Add --dir option to findleaves.py
findleaves.py needs to be able to search multiple directories for
multiple filenames (Android.bp and Android.mk). Add support for a new
--dir= option to specify one or more directories to search, and then
treat any remaining options as filenames. If no directories are
specified, fall back to treating only the last option as a filename and
the rest as directories.
Change-Id: Ie879facd49192c6b4fafae070ad00eb21b1faa32
Alex Klyubin [Fri, 27 May 2016 21:36:05 +0000 (14:36 -0700)]
Switch signapk to apksigner-core.
This switches signapk's APK signing from its own signing logic to that
offered by apksigner-core library. OTA update package signing logic
remains inside signapk codebase.
Bug:
27461702
Change-Id: Ibf8435c555fe3f2b621d5189e7ae44f79082c810
Alex Klyubin [Fri, 27 May 2016 20:00:59 +0000 (20:00 +0000)]
Merge "Improve DataSource interface."
Alex Klyubin [Fri, 27 May 2016 19:57:11 +0000 (19:57 +0000)]
Merge "Store entry alignment information in APK."
Alex Klyubin [Fri, 27 May 2016 17:00:53 +0000 (10:00 -0700)]
Improve DataSource interface.
1. This adds way to obtain a DataSource which represents a region of
data contained in the DataSource.
2. This fixes a design bug in "feed" method where the size parameter
was an int instead of long.
3. This fixes a bug in ByteBufferDataSource where its mSize field was
a long instead of an int (ByteBuffer's length cannot be more than
2^31).
Bug:
27461702
Change-Id: Ib0812784beb581f19d2412e667b8bd018f0a3c78
Treehugger Robot [Fri, 27 May 2016 00:23:24 +0000 (00:23 +0000)]
Merge "Fix mm and mmm in Android.bp directories"
Colin Cross [Thu, 26 May 2016 22:32:15 +0000 (15:32 -0700)]
Fix mm and mmm in Android.bp directories
mm and mmm were not parsing out/soong/Android.mk, so they were ignoring
all modules defined in an Android.bp file. Always include
out/soong/Android.mk when ONE_SHOT_MAKEFILE is set, and convert the mm
and mmm functions to build MODULES-IN-$DIR instead of all_modules so
that only the subset of modules defined in out/soong/Android.mk for the
requested directory are built.
Bug:
28986194
Change-Id: I6e1ff8cacfc668a4154a0d5937450db840bb7a6b
Dan Willemsen [Thu, 26 May 2016 21:49:01 +0000 (21:49 +0000)]
Merge "Remove unnecessary variables"
Dan Willemsen [Tue, 17 May 2016 05:07:31 +0000 (22:07 -0700)]
Remove unnecessary variables
These are all either recently unused due to the removal of CFLAGS/etc,
or have been exported by Soong and are no longer necessary.
Change-Id: I5930d43fda21acc8202b3d8ea010fbefb6ae4cf1
Colin Cross [Thu, 26 May 2016 00:46:39 +0000 (00:46 +0000)]
Merge "Remove USE_SOONG=false option"
Colin Cross [Wed, 25 May 2016 19:55:43 +0000 (12:55 -0700)]
Remove USE_SOONG=false option
Soong is always required to build now.
Change-Id: I4e310e98e912b370ebf980bf609f04212a7e9a50
Dan Willemsen [Tue, 24 May 2016 20:06:07 +0000 (13:06 -0700)]
Error on external includes
To make the build hermetic, all include files should be located within
the source tree (or generated as output).
There was an Android.mk that was adding /usr/include to its include
path.
Change-Id: I4c40b908314239c8239c7835d592e409248a3150
Dan Willemsen [Wed, 2 Mar 2016 21:54:51 +0000 (13:54 -0800)]
Add NATIVE_TESTS class, move host native tests
Host native tests have been getting installed into
out/host/linux-x86/bin/..., but this pollutes the bin directory with a
lot of poorly named tests. Also, to support 32-bit and 64-bit tests, we
need to have different names with different suffixes. This causes
problems when tests expect to be named something specific (like gtest).
It's also convenient to store test data next to the test itself.
So with this change, native tests will be installed in
out/host/linux-x86/nativetest[64]/$(LOCAL_MODULE)/$(LOCAL_MODULE_STEM)
just like target tests get installed into /data/nativetest[64].
Implement this using a new NATIVE_TESTS class, which is like
EXECUTABLES, but sets up the install path differently, and configures
the rpath to load shared libraries with the proper relative path.
LOCAL_MODULE_RELATIVE_PATH can be used to control the directory name, it
will default to $(LOCAL_MODULE). This way multiple related tests can be
grouped together.
Target native tests also use NATIVE_TESTS now, but nothing should change
other than LOCAL_MODULE_RELATIVE_PATH can be used.
Change-Id: I535e42b1a6b21c5b8d6a580aa2f944d2be35e27d
Alex Klyubin [Mon, 23 May 2016 21:19:51 +0000 (14:19 -0700)]
Store entry alignment information in APK.
Data of uncompressed APK entries is often aligned to a multiple of 4
or 4096 in the APK to make it easier to mmap the data. Unfortunately,
the current method for achieving alignment suffers from two issues:
(1) the way it uses the Local File Header extra field is not compliant
with ZIP format (for example, this prevents older versions of Python's
zipfile from reading APKs: https://bugs.python.org/issue14315), and
(2) it does not store information about the alignment multiple in the
APK, making it harder/impossible to preserve the intended alignment
when rearranging entries in the APK.
This change solves these issues by switching to a different method for
aligning data of uncompressed APK entries. Same as before, alignment
is achieved using Local File Header entry field. What's different is
that alignment is achieved by placing a well-formed extensible data
field/block into the extra field. The new field/block contains the
alignment multiple (e.g., 4 or 4096) as well as the necessary padding
(if any). Compared to the original alignment method, the new method
uses 6 more bytes for each uncompressed entry.
Bug:
27461702
Change-Id: I8cffbecc50bf634b28fca5bc39eb23f671961cf9
Vishwath Mohan [Mon, 23 May 2016 22:17:35 +0000 (22:17 +0000)]
Merge "Double quote some statements for zsh compatibility"
Colin Cross [Mon, 23 May 2016 22:05:27 +0000 (22:05 +0000)]
Merge "Error on USE_CLANG_PLATFORM_BUILD="
Treehugger Robot [Mon, 23 May 2016 21:18:57 +0000 (21:18 +0000)]
Merge "Add LOCAL_CTS_GTEST_LIST_EXECUTABLE"
Alex Klyubin [Mon, 23 May 2016 19:26:14 +0000 (19:26 +0000)]
Merge "Improve concurrency for JAR entry inspection requests."
Dan Willemsen [Mon, 23 May 2016 17:30:18 +0000 (17:30 +0000)]
Merge "Move to a single clang unknown flags list"
Dan Willemsen [Thu, 19 May 2016 21:39:25 +0000 (14:39 -0700)]
Move to a single clang unknown flags list
Soong is using a single list of unknown clang flags for all
architectures, instead of the per-architecture list here. Collapse this
down to a single list as well so that it can be verified against the
Soong list, and eventually replaced by the Soong version.
Change-Id: If43cd892105da5155907c05965a74b835920a369
Dan Willemsen [Thu, 19 May 2016 22:09:45 +0000 (22:09 +0000)]
Merge "Remove (C|CPP|CONLY|LD)FLAGS checked by Soong"
Colin Cross [Thu, 19 May 2016 19:59:54 +0000 (12:59 -0700)]
Error on USE_CLANG_PLATFORM_BUILD=
A blank USE_CLANG_PLATFORM_BUILD= on the command line for make causes
confusion between make and kati. make ignores writes to variables set
on the command line, so envsetup.mk cannot set it to true,
USE_CLANG_PLATFORM_BUILD is still empty in soong.mk, and soong.variables
gets DeviceUsesClang: false. kati gets USE_CLANG_PLATFORM_BUILD= in its
environment instead of on its command line, so envsetup.mk when parsed
by kati sets USE_CLANG_PLATFORM_BUILD=true.
Error out early if USE_CLANG_PLATFORM_BUILD is not true or false after
attempting to set it to true if it is empty.
Change-Id: I0e2731315711db9a6ac6ea2bf49c88bccacd35f7
Vishwath Mohan [Fri, 11 Mar 2016 18:00:40 +0000 (10:00 -0800)]
Double quote some statements for zsh compatibility
Two eval commands did not enclose their arguments in double quotes, and
this broke compatibility on zsh - preventing lunch from running to
completion. This CL adds the double quotes and fixes the issue.
Change-Id: I5ed884b455e2e61182f6c29fb807ed92372af039
Dan Willemsen [Thu, 19 May 2016 00:39:04 +0000 (17:39 -0700)]
Remove (C|CPP|CONLY|LD)FLAGS checked by Soong
At the beginning of every build, Soong has exported its version of these
variables, and has been ensuring that when sorted, the result is the
same. So these variables all have the same flags of the Soong ones, but
may be in different orders. We don't believe any of the remaining order
differences matter. As we remove the Make definitions, the exported
Soong variables will take over.
This only removes lines that change one of these variables:
[2ND_][CLANG_]HOST_CROSS_GLOBAL_CFLAGS
[2ND_][CLANG_]HOST_CROSS_GLOBAL_CONLYFLAGS
[2ND_][CLANG_]HOST_CROSS_GLOBAL_CPPFLAGS
[2ND_][CLANG_]HOST_CROSS_GLOBAL_LDFLAGS
[2ND_][CLANG_]HOST_GLOBAL_CFLAGS
[2ND_][CLANG_]HOST_GLOBAL_CONLYFLAGS
[2ND_][CLANG_]HOST_GLOBAL_CPPFLAGS
[2ND_][CLANG_]HOST_GLOBAL_LDFLAGS
[2ND_][CLANG_]TARGET_GLOBAL_CFLAGS
[2ND_][CLANG_]TARGET_GLOBAL_CONLYFLAGS
[2ND_][CLANG_]TARGET_GLOBAL_CPPFLAGS
[2ND_][CLANG_]TARGET_GLOBAL_LDFLAGS
Many other variables are unnecessary now too, but those will be removed
in another change. For those we can ensure the build.ninja file does not
change, but we expect it to change here due to the ordering differences.
Change-Id: I0bd0778706d02ee27b2784dd8dc6b2c71d37bd3a
Dan Willemsen [Thu, 19 May 2016 01:09:28 +0000 (01:09 +0000)]
Merge "Windows: Stop adding to GLOBAL_LD_DIRS"
Dan Willemsen [Wed, 18 May 2016 23:18:13 +0000 (16:18 -0700)]
Windows: Stop adding to GLOBAL_LD_DIRS
Soong does not have the idea of global linking directories, so move the
windows prebuilt library directory to GLOBAL_LDFLAGS instead.
Change-Id: Ie101146f4682cd85924463aa5acaa46869938e6c
Dan Willemsen [Wed, 18 May 2016 23:50:27 +0000 (23:50 +0000)]
Merge "Remove trailing newline from TARGET_C_INCLUDES"
Dan Willemsen [Wed, 18 May 2016 22:36:37 +0000 (15:36 -0700)]
Remove trailing newline from TARGET_C_INCLUDES
In preparation for comparing with the Soong variable.
Change-Id: I159507757ed7678a33a3c94e6d01d170485e251a
Alex Light [Wed, 18 May 2016 21:05:41 +0000 (21:05 +0000)]
Merge "Add fake libart"
Treehugger Robot [Wed, 18 May 2016 20:45:19 +0000 (20:45 +0000)]
Merge "Remove (TARGET|HOST)_NO_UNDEFINED_LDFLAGS"
Dan Willemsen [Wed, 18 May 2016 19:50:53 +0000 (12:50 -0700)]
Remove (TARGET|HOST)_NO_UNDEFINED_LDFLAGS
We never use it for host modules, and all the target configurations use
the same linker argument. This matches Soong.
Change-Id: Ibcba9a4ce340c7a12306d1fee620a04027c0e292
Alex Light [Fri, 15 Apr 2016 17:19:02 +0000 (10:19 -0700)]
Add fake libart
Bug:
27775991
Change-Id: I54cbdf2340dcfdd9b3973d9d10c8bd94492736fb
(cherry picked from commit
6da74d8f1f0a06366fe93b885dc26f53401ec6b4)