OSDN Git Service
Tao Bao [Fri, 4 Sep 2015 04:17:37 +0000 (21:17 -0700)]
releasetools: Fix the bug when deleting files.
For file-based OTAs, we used to remove unneeded files in ascending
order, which failed to delete non-empty directories. Reverse the order
to fix the issue.
For example, now we have the following in our generated script:
delete("/system/app/Calculator/arm/Calculator.odex",
"/system/app/Calculator/arm/",
"/system/app/Calculator/Calculator.apk",
"/system/app/Calculator/");
Bug:
22960996
Change-Id: I0d36d29b7862fb53bf55bf5685a990180f9c0b3b
Tao Bao [Thu, 3 Sep 2015 01:32:08 +0000 (01:32 +0000)]
Merge "releasetools: Fix the bug in symlink deletion."
Tao Bao [Wed, 2 Sep 2015 17:28:08 +0000 (10:28 -0700)]
releasetools: Fix the bug in symlink deletion.
For file-based OTAs, symlinks in the source build but not in the target
build will be deleted. However, if a symlink is replaced by a regular
file in the target build, the file will be accidentally deleted when
applying (resuming) the same package again.
Verify the checksum of a symlink that will be unpacked or renamed to.
Delete the file only if it doesn't have the target checksum.
Bug:
23646151
Change-Id: Ie322abb6022b6fa812c6b36a3127872d9614fc3b
Ying Wang [Wed, 2 Sep 2015 20:48:50 +0000 (20:48 +0000)]
Merge "Clean up Javac a little bit."
Evgenii Stepanov [Wed, 2 Sep 2015 20:13:37 +0000 (20:13 +0000)]
Merge "Add AddressSanitizer linker flags for x86."
Ivan Krasin [Wed, 2 Sep 2015 18:56:40 +0000 (11:56 -0700)]
Add AddressSanitizer linker flags for x86.
This is the 1/4 change required to get AddressSanitizer
for Android x86 to work.
Change-Id: I848805af34fdd32a313c991504d6014f9b97ab88
Ying Wang [Wed, 2 Sep 2015 02:12:05 +0000 (19:12 -0700)]
Clean up Javac a little bit.
- Added GLOBAL_JAVAC_DEBUG_FLAGS and merge it to
PRIVATE_JAVACFLAGS/PRIVATE_JACK_FLAGS to get rid of
PRIVATE_JAVAC_DEBUG_FLAGS/PRIVATE_JACK_DEBUG_FLAGS.
- With Java rules out of base_rules.mk we can get rid
of java_alternative_checked_module now.
Change-Id: I1a14716c785e3d49330a75044107662ce96a9307
Shinichiro Hamaji [Tue, 1 Sep 2015 23:53:41 +0000 (23:53 +0000)]
Merge "Set --ninja_suffix based on make/mm/mmm targets"
Shinichiro Hamaji [Tue, 1 Sep 2015 05:24:18 +0000 (14:24 +0900)]
Set --ninja_suffix based on make/mm/mmm targets
This fixes two issues:
- Switching make target didn't work well when the latter
make target is not a subset of the first make target.
- Switching between m and mm/mmm forced build.ninja to be
regenerated.
Bug:
23685536
Change-Id: I3c3550d6cca3007d166a13367d6bc22cc3feede2
Shinichiro Hamaji [Tue, 1 Sep 2015 02:26:15 +0000 (02:26 +0000)]
Merge "Show error message when resource limit too low for USE_GOMA=true"
Dan Willemsen [Tue, 1 Sep 2015 00:44:21 +0000 (00:44 +0000)]
Merge "Blacklist libc++ from windows host builds"
Richard Uhler [Mon, 31 Aug 2015 16:53:55 +0000 (16:53 +0000)]
Merge "Add ahat to runtime_libart product."
Ying Wang [Sat, 29 Aug 2015 18:09:20 +0000 (18:09 +0000)]
Merge "Cleaned the outdated incrementaljavac."
Ying Wang [Fri, 8 May 2015 20:27:25 +0000 (13:27 -0700)]
Cleaned the outdated incrementaljavac.
Nobody is using this feature and now we switched to jack.
Change-Id: I749b486eb347cbf3ee7b107565fc800eeb238c44
(cherry-picked from commit
90598cb70c3a6e65223f02c4104e823024caf693)
Ying Wang [Sat, 29 Aug 2015 16:40:30 +0000 (16:40 +0000)]
Merge "Move Java stuff out of base_rules.mk"
Shinichiro Hamaji [Sat, 29 Aug 2015 02:58:47 +0000 (11:58 +0900)]
Show error message when resource limit too low for USE_GOMA=true
This should be particularly helpful for Mac users, because the
default limits of Mac are too low.
Change-Id: I03f0f76a0707b4ce5ef9cddf5df6e3857a215996
Ying Wang [Fri, 28 Aug 2015 18:47:03 +0000 (11:47 -0700)]
Move Java stuff out of base_rules.mk
- Moved the target-only .aidl and .logtag rules to java.mk.
- Moved the .proto rules, Java library dependency calculation,
Java resource processing and Java-related
target-specific and global variables setup to a new java_common.java,
which is shared by both target and host Java modules.
- Minor cleanup for the moved code.
Bug:
23597357
Change-Id: Ic3eb72e26bb1a663e5739abe0a935f96b88bc352
Richard Uhler [Fri, 31 Jul 2015 16:39:26 +0000 (09:39 -0700)]
Add ahat to runtime_libart product.
Change-Id: Iae5837f884c926a9f0e57df56bd21b6260ee8b2d
Elliott Hughes [Fri, 28 Aug 2015 19:47:36 +0000 (19:47 +0000)]
Merge "Remove AndroidConfig.h."
Tao Bao [Fri, 28 Aug 2015 18:33:25 +0000 (18:33 +0000)]
Merge "Use the basename of file_contexts from the info dict."
Tao Bao [Fri, 28 Aug 2015 17:52:03 +0000 (10:52 -0700)]
Use the basename of file_contexts from the info dict.
Changes in [1] break the backward compatibility of the OTA scripts. Fix
the issue by extracting the file basename from the info dict in
META/misc_info.txt.
[1] commit
d3a803e6680e86f7b4960d51ab0b620728b793ca
Bug:
23620049
Change-Id: I40a5625bbe0beb295f34b2328906a4a4283fe59b
Elliott Hughes [Fri, 28 Aug 2015 17:31:28 +0000 (10:31 -0700)]
Remove AndroidConfig.h.
This change leaves soong references.
Change-Id: Iee32b6b97ad9711f69e999c9d01f01d71562b63f
Dan Willemsen [Fri, 28 Aug 2015 03:39:34 +0000 (03:39 +0000)]
Merge "Add support for LOCAL_SRC_FILES_(host)_(arch)"
Dan Willemsen [Wed, 19 Aug 2015 00:47:06 +0000 (17:47 -0700)]
Add support for LOCAL_SRC_FILES_(host)_(arch)
For boringssl, which has architecture specific host assembly files
Bug:
23566667
Change-Id: Ia4602d28e93a1b3d1400cf95417b5e446c5fb84b
Dan Willemsen [Fri, 28 Aug 2015 03:37:30 +0000 (03:37 +0000)]
Merge "Add _(darwin|linux|windows)"
Dan Willemsen [Fri, 14 Aug 2015 00:51:40 +0000 (17:51 -0700)]
Add _(darwin|linux|windows)
So that we can support building both linux and windows binaries at the
same time on a linux host. This replaces the ifeq($(HOST_OS),...) checks
in Android.mk files.
Bug:
23566667
Change-Id: I693e11984e36d55bb6f09fa0d49bc485463e16fb
Tao Bao [Thu, 27 Aug 2015 19:16:06 +0000 (19:16 +0000)]
Merge "rangelib.py: Fix the bug in monotonic check."
Tao Bao [Thu, 27 Aug 2015 00:07:14 +0000 (17:07 -0700)]
rangelib.py: Fix the bug in monotonic check.
RangeSet("2-10").monotonic gives a wrong result of "False". Fix the bug
and add more tests into test_rangelib.py.
Change-Id: I04780571b45ecafd34040f405756b9745a9e21a5
Tao Bao [Wed, 26 Aug 2015 23:26:54 +0000 (23:26 +0000)]
Merge "Add unit tests for rangelib.py."
Tao Bao [Wed, 26 Aug 2015 23:13:18 +0000 (16:13 -0700)]
Add unit tests for rangelib.py.
Separate the test cases in rangelib.py comments into a python unittest
file.
To invoke:
$ python -m unittest test_rangelib
Change-Id: I59e4ce0aaccc83bcbb42fee15775683dcd559d99
Tao Bao [Wed, 26 Aug 2015 19:39:31 +0000 (19:39 +0000)]
Merge "Split large files for BBOTA v3."
Tao Bao [Tue, 25 Aug 2015 22:10:10 +0000 (15:10 -0700)]
Split large files for BBOTA v3.
For BBOTA v3, we need to stash source blocks to support resumable
feature. However, with the growth of file size and the shrink of the
cache size, source blocks that represent a file are too large to be
stashed as a whole. CL in [1] solves the issue by replacing the diff
command with a "new" command. However, it may increase the generated
package size substantially (e.g. from ~100MB to ~400MB).
With this CL, if a file spans too many blocks, we split it into smaller
pieces by generating multiple commands. For the same case above, it
reduces the package size to ~150MB.
One potential downside is that after splitting, files like .jar,
.apk and .zip can no longer use imgdiff. We may lose the potential
benefit of using imgdiff for patch size reduction.
[1] commit
82c47981bd0602a1c7b50dfabf9a6a2412993bae
Bug:
22430577
Change-Id: I5684fab0ac41c8d94fe18f52a34cef77e06f741c
Nick Kralevich [Tue, 25 Aug 2015 22:41:57 +0000 (22:41 +0000)]
Merge "Only allow toolbox exec where /system exec was already allowed."
Ying Wang [Tue, 25 Aug 2015 17:12:36 +0000 (17:12 +0000)]
Merge "Dist the otatools package."
Stephen Smalley [Tue, 25 Aug 2015 15:42:17 +0000 (11:42 -0400)]
Only allow toolbox exec where /system exec was already allowed.
When the toolbox domain was introduced, we allowed all domains to exec it
to avoid breakage. However, only domains that were previously allowed the
ability to exec /system files would have been able to do this prior to the
introduction of the toolbox domain. Remove the rule from domain.te and add
rules to all domains that are already allowed execute_no_trans to system_file.
Requires coordination with device-specific policy changes with the same Change-Id.
Change-Id: Ie46209f0412f9914857dc3d7c6b0917b7031aae5
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Ying Wang [Tue, 25 Aug 2015 00:13:53 +0000 (17:13 -0700)]
Dist the otatools package.
Bug:
23495952
Change-Id: I1f61946580dd8ea29c3d5e1f4e859cc6d1f8554a
Tom Cherry [Mon, 24 Aug 2015 23:23:36 +0000 (23:23 +0000)]
Merge "remove TARGET_PROVIDES_INIT_RC"
Tom Cherry [Mon, 24 Aug 2015 20:55:02 +0000 (13:55 -0700)]
remove TARGET_PROVIDES_INIT_RC
Change-Id: I765b80d48333f54fb995e4ccfd77d38683c6cbb9
Christopher Wiley [Fri, 21 Aug 2015 20:46:21 +0000 (20:46 +0000)]
Merge "Change .dbus.xml extension to .dbus-xml"
Shinichiro Hamaji [Fri, 21 Aug 2015 03:24:27 +0000 (03:24 +0000)]
Merge "Use goma when USE_GOMA=true is set"
Elliott Hughes [Thu, 20 Aug 2015 23:56:07 +0000 (23:56 +0000)]
Merge "Fix Linux host checkbuilds."
Elliott Hughes [Thu, 20 Aug 2015 23:52:47 +0000 (16:52 -0700)]
Fix Linux host checkbuilds.
Turns out that thanks to
9d59f417767991246848c3e101cb27d2dfea5988 we actually
build *all* the host libraries both 32- and 64-bit. Rather than fix the
libraries (or fixing things so we don't build stuff we don't need) right now,
let's just put _FILE_OFFSET_BITS and _LARGEFILE_SOURCE back.
Change-Id: I6c2d6a6919d1518f775e0d6c050e2c774994c5bd
Elliott Hughes [Thu, 20 Aug 2015 23:10:05 +0000 (23:10 +0000)]
Merge "Kill host Linux AndroidConfig.h."
Elliott Hughes [Thu, 20 Aug 2015 21:23:59 +0000 (21:23 +0000)]
Merge "Revert "Fix Windows SDK build.""
Elliott Hughes [Thu, 20 Aug 2015 21:22:19 +0000 (21:22 +0000)]
Revert "Fix Windows SDK build."
This reverts commit
b8dd176441f6ba9527d24a5065d8225437fe39d5.
This wasn't the problem, and mingw doesn't even use _LARGEFILE_SOURCE. The real problem (using WIN32 instead of _WIN32 in external/llvm) is fixed now.
Change-Id: I2b708a006b530cb18d03b1425cd65edda35ee17e
Elliott Hughes [Thu, 20 Aug 2015 21:21:01 +0000 (21:21 +0000)]
Merge "Stop defining WIN32."
Elliott Hughes [Thu, 20 Aug 2015 21:04:18 +0000 (14:04 -0700)]
Stop defining WIN32.
Change-Id: Iac0b4745579e6de55bea02571617ae561308e1a9
Ying Wang [Thu, 20 Aug 2015 20:47:53 +0000 (20:47 +0000)]
Merge "Support LOCAL_INIT_RC."
Ying Wang [Thu, 20 Aug 2015 19:02:10 +0000 (12:02 -0700)]
Support LOCAL_INIT_RC.
A module can install a companion init.rc file,
by specifying:
LOCAL_INIT_RC := <init.rc-file-path-replative-to-LOCAL_PATH>
You can also use the variant with _32 or _64 suffix.
Bug:
23186545
Change-Id: I00a96509f5707ae39361a0c5555fa59d46c90322
Elliott Hughes [Thu, 20 Aug 2015 17:16:40 +0000 (10:16 -0700)]
Kill host Linux AndroidConfig.h.
We only build 64-bit now anyway.
Change-Id: I4bc9487522d215234509510bee66e4f210d86765
Elliott Hughes [Thu, 20 Aug 2015 18:11:18 +0000 (18:11 +0000)]
Merge "Also -DWIN32 to keep llvm building."
Elliott Hughes [Thu, 20 Aug 2015 18:10:22 +0000 (11:10 -0700)]
Also -DWIN32 to keep llvm building.
Change-Id: Ic8e70f35dda23226d60010a38cf5633840772310
Elliott Hughes [Thu, 20 Aug 2015 17:20:45 +0000 (17:20 +0000)]
Merge "Fix Windows SDK build."
Elliott Hughes [Thu, 20 Aug 2015 17:19:26 +0000 (10:19 -0700)]
Fix Windows SDK build.
I forgot that removing a file doesn't cause a rebuild in our lame build
system, so I hadn't actually tested my previous change.
Change-Id: Ida19d4a8fd40e37029031eac4b4ca5f0cc5c895b
Elliott Hughes [Thu, 20 Aug 2015 16:59:03 +0000 (16:59 +0000)]
Merge "Remove Windows AndroidConfig.h."
Elliott Hughes [Thu, 20 Aug 2015 00:45:17 +0000 (17:45 -0700)]
Remove Windows AndroidConfig.h.
The only things left that we're using are the Windows target version
and _FILE_OFFSET_BITS=64, and they can go in the combo .mk.
Also fix the unused Windows 64 .mk.
Change-Id: I5f2458d67c0a8201196a339573f861bbf18b7eb8
Christopher Wiley [Thu, 20 Aug 2015 16:08:17 +0000 (09:08 -0700)]
Change .dbus.xml extension to .dbus-xml
Work around gyp's inability to handle compound extensions by expecting
a similar looking simple extension for XML files definine DBus
interfaces. We'll need to rename these sources in the places we're
using them already.
Bug:
23380180
Change-Id: Ieb2050f3ef05456cd70de65c3e128d57a6a508f8
Shinichiro Hamaji [Mon, 17 Aug 2015 08:38:24 +0000 (17:38 +0900)]
Use goma when USE_GOMA=true is set
With USE_GOMA, the path to gomacc in $GOMA_DIR or $HOME/goma
will be appended to CC_WRAPPER and CXX_WRAPPER.
Note this works only with USE_NINJA. Unlike ninja, GNU make
cannot change the parallelism depending on targets. Specifying
-j500 to GNU make would mean you may run 500 local jobs in
parallel, but with -j32 goma will just slow down the build.
Change-Id: I0f571454fd2a5b525ee29b445f7ab8715927ca00
Shinichiro Hamaji [Thu, 20 Aug 2015 05:25:34 +0000 (05:25 +0000)]
Merge "Add two flags for the latest ckati"
Dan Albert [Thu, 20 Aug 2015 05:07:34 +0000 (05:07 +0000)]
Merge "Fix sanitizer choice for global vs module."
Dan Albert [Thu, 20 Aug 2015 03:13:33 +0000 (20:13 -0700)]
Fix sanitizer choice for global vs module.
The sanitizer chosen by the environment (either by SANITIZE_TARGET or
SANITIZE_HOST) should be chosen over the one specified by the module.
Bug: http://b/
23330588
Change-Id: I835b7d76e071fc0db2f859f98dfb9d7ff76af245
Elliott Hughes [Wed, 19 Aug 2015 22:09:32 +0000 (22:09 +0000)]
Merge "Stop defining HAVE_MS_C_RUNTIME."
Christopher Wiley [Wed, 19 Aug 2015 21:39:28 +0000 (21:39 +0000)]
Merge "Fix DBus proxies to use all definitions"
Elliott Hughes [Wed, 19 Aug 2015 21:22:59 +0000 (14:22 -0700)]
Stop defining HAVE_MS_C_RUNTIME.
Change-Id: Ie491547552ba64d60d669b74115da96a0fcced18
Christopher Wiley [Wed, 19 Aug 2015 20:48:53 +0000 (13:48 -0700)]
Fix DBus proxies to use all definitions
Bug:
22608897
Change-Id: Id17cf4b896c1c62be874599c7f346f3045e36819
Christopher Wiley [Wed, 19 Aug 2015 18:30:20 +0000 (18:30 +0000)]
Merge "Generate DBus proxies and adaptors separately"
Christopher Wiley [Fri, 14 Aug 2015 21:55:32 +0000 (14:55 -0700)]
Generate DBus proxies and adaptors separately
Enable daemons exposing an interface over DBus to easily
build client libraries. Now daemons can write rules like:
include $(CLEAR_VARS)
LOCAL_MODULE := libdbus-binding-example-client
LOCAL_DBUS_PROXY_PREFIX := dbus-example-example
LOCAL_SRC_FILES := \
dbus_bindings/org.chromium.Example.Manager.dbus.xml \
dbus_bindings/dbus-service-config.json
include $(BUILD_SHARED_LIBRARY)
to expose a client library.
While here, add support for generating independent adaptor header
files on a per interface basis.
Bug:
22608897
Change-Id: I011f9afc234811c31e445898321c2731c482fa77
Tao Bao [Wed, 19 Aug 2015 02:34:27 +0000 (02:34 +0000)]
Merge "build: Don't call make_recovery_patch if there's no recovery."
Dan Willemsen [Fri, 14 Aug 2015 00:41:56 +0000 (17:41 -0700)]
Blacklist libc++ from windows host builds
It's not ready, and instead of each Android.mk doing this, do it in the
core build.
Change-Id: Ie80b17ffcde64e72c98d310ca986d4cf775c099e
Tao Bao [Wed, 19 Aug 2015 01:03:56 +0000 (18:03 -0700)]
build: Don't call make_recovery_patch if there's no recovery.
Change-Id: Iea1bd529eca9cd607da3b420bc4287f95c18219d
Chih-hung Hsieh [Tue, 18 Aug 2015 22:11:03 +0000 (22:11 +0000)]
Merge "Remove special x86 flag that disables tail call optimization."
Ying Wang [Tue, 18 Aug 2015 21:35:42 +0000 (21:35 +0000)]
Merge "Rename libnanopb-c-2.8.0 to libprotobuf-c-nano."
Ying Wang [Tue, 18 Aug 2015 21:35:37 +0000 (21:35 +0000)]
Merge "Consolidate the nanopb-c and regular protobuf build rules."
Ying Wang [Wed, 8 Apr 2015 16:54:34 +0000 (09:54 -0700)]
Rename libnanopb-c-2.8.0 to libprotobuf-c-nano.
Bug:
20093047
Change-Id: I787925fde97d1a3965cb8a9bf012f3c5a45a6aed
Ying Wang [Tue, 7 Apr 2015 18:59:34 +0000 (11:59 -0700)]
Consolidate the nanopb-c and regular protobuf build rules.
Bug:
20093047
Change-Id: I38ebd748aacce6d215403da481ad04833810a995
Tao Bao [Tue, 18 Aug 2015 16:59:05 +0000 (16:59 +0000)]
Merge "Revise stash for BBOTAs when needed."
Dan Willemsen [Tue, 18 Aug 2015 04:31:07 +0000 (04:31 +0000)]
Merge "Remove legacy windows platform build support"
Dan Willemsen [Tue, 18 Aug 2015 03:30:11 +0000 (03:30 +0000)]
Merge "Use repeatable timestamps"
Ying Wang [Tue, 18 Aug 2015 00:06:52 +0000 (00:06 +0000)]
Merge "Set up dependency on ADDRESS_SANITIZER_LINKER"
Ying Wang [Mon, 17 Aug 2015 23:13:24 +0000 (16:13 -0700)]
Set up dependency on ADDRESS_SANITIZER_LINKER
Set up dependency on ADDRESS_SANITIZER_LINKER if address sanitizer is
enabled.
Bug:
22850550
Change-Id: I736fe1d4db9594edf9e82ae96e4631887881dfa5
Dan Albert [Mon, 17 Aug 2015 22:32:58 +0000 (22:32 +0000)]
Merge "Add BUILD(_HOST)?_FUZZ_TEST."
Dan Albert [Thu, 30 Jul 2015 23:56:30 +0000 (16:56 -0700)]
Add BUILD(_HOST)?_FUZZ_TEST.
These are tests that use LLVM's libFuzzer for testing with a
sanitizer.
Bug: http://b/
22850550
Change-Id: I6eeaff7877b7dd0c30f8910c6f27f5278b27a83d
Elliott Hughes [Mon, 17 Aug 2015 20:25:36 +0000 (20:25 +0000)]
Merge "Replace HAVE_MS_C_RUNTIME with _WIN32 in build."
Dan Albert [Mon, 17 Aug 2015 20:16:01 +0000 (20:16 +0000)]
Merge "Don't filter on a non-existent variable."
Elliott Hughes [Mon, 17 Aug 2015 19:37:37 +0000 (12:37 -0700)]
Replace HAVE_MS_C_RUNTIME with _WIN32 in build.
Change-Id: Iebca0c4eab975c4bebcd3658baefd85c51890db4
Dan Willemsen [Thu, 13 Aug 2015 21:31:36 +0000 (14:31 -0700)]
Remove legacy windows platform build support
We still support HOST_OS=windows for the SDK host tools cross-builds, but
that's only when USE_MINGW is set when running under linux.
Change-Id: I37da87dc9fbbd69ba10ce4d7f2668ab3f6482d92
Tao Bao [Mon, 17 Aug 2015 17:38:01 +0000 (17:38 +0000)]
Merge "Add an option to include a full bootloader in incremental OTA."
Dan Albert [Mon, 17 Aug 2015 17:36:09 +0000 (10:36 -0700)]
Don't filter on a non-existent variable.
This used to be used for aarch64
(https://android-review.googlesource.com/#/c/80126/), but has since
been removed and now does nothing. This was then reimplemented more
generally with my_illegal_flags
(https://android-review.googlesource.com/#/c/166405/). Drop
*GLOBAL_UNSUPPORTED_CFLAGS in favor of that.
Change-Id: I3347726dd92f9d3a16783e0d9d3b685a828400b4
leozwang [Fri, 14 Aug 2015 17:57:58 +0000 (10:57 -0700)]
Add an option to include a full bootloader in incremental OTA.
Bug:
23101469
Change-Id: Iac466b83f789f10d26e8d289dce766cf0d05148b
(cherry picked from commit
c1324751e79b23a63ea494f4bc784446c8f32435)
Tao Bao [Mon, 17 Aug 2015 16:45:13 +0000 (09:45 -0700)]
Revise stash for BBOTAs when needed.
When generating incremental BBOTAs (v2 and above), we need to ensure
that the needed runtime stash is below the given threshold. If it's
running out of space on /cache, we replace the command that uses a
stash with a "new" command instead.
This may increase the OTA package size, since it is carrying more full
blocks instead of patches. It gets even worse for large files that span
a number of blocks, because currently we will store all the blocks for
the file as "new" blocks if stashing cannot be satisfied. We may further
optimize by splitting them into smaller chunks so that most of them can
still be stashed.
Bug:
22430577
Change-Id: I5a49e361adc7d3d41de2e9c08ee9b08c1e6c091a
Shinichiro Hamaji [Mon, 17 Aug 2015 08:41:29 +0000 (17:41 +0900)]
Add two flags for the latest ckati
* --regen
Re-generate build.ninja only when necessary. If either
1. .mk file is updated,
2. environment variable is updated,
3. $(wildcard) result is changed, or
4. $(shell) result is changed,
ckati will regenerate ninja file. This check takes only ~1
second, so incremental build will become much faster even
without "fastincremental" target.
* --ignore_dirty=out/%:
Some .mk files in out/ (e.g.,
out/target/product/generic/previous_build_config.mk)
are updated while ckati is running. With this flag, ckati
does not regenerate build.ninja when they look modified.
This should be OK for ninja based build, as ninja handles
command line changes nicely.
Change-Id: I7a2fca0e327d999599d6b16f06358e8a5e657565
Stephen Hines [Sat, 15 Aug 2015 02:04:55 +0000 (02:04 +0000)]
Merge "Add -mfpu=neon-vfpv4 if cpu is krait"
Yang Ni [Fri, 14 Aug 2015 18:39:47 +0000 (18:39 +0000)]
Merge "Include target api version in build system for rscpp."
Dan Albert [Fri, 14 Aug 2015 00:59:52 +0000 (00:59 +0000)]
Merge "Link librt on the host when using sanitizers."
Dan Albert [Fri, 14 Aug 2015 00:25:10 +0000 (17:25 -0700)]
Link librt on the host when using sanitizers.
The sanitizers now use shared memory.
Change-Id: Ie8fa6690374eda08ee4138b68fd628ebbfa2d0c8
Noah Presler [Fri, 14 Aug 2015 00:04:10 +0000 (17:04 -0700)]
Include target api version in build system for rscpp.
b/
23161320
Change-Id: Ibc23309abed092e7648168fbf4b80891a44929a6
Tao Bao [Thu, 13 Aug 2015 22:22:16 +0000 (22:22 +0000)]
Merge "Free the stash after its use in BBOTA v2."
Tao Bao [Thu, 13 Aug 2015 21:44:15 +0000 (14:44 -0700)]
Free the stash after its use in BBOTA v2.
In BBOTA v2, the stash size computation is inaccurate. A stash will be
used only once, so we deduct its size from stashed_blocks right after
its use. However, the stash doesn't actually get freed until being
overwritten by another stash or destroyed at the end of an update. This
leads to OTA failures due to insufficient space on /cache. This CL
changes the BBOTA v2 script to generate an explicit free right after
its use.
Bug:
23119955
Change-Id: Icdd8352bb0ff778945937c870990a888c7a4d38b
Dan Albert [Thu, 13 Aug 2015 16:48:18 +0000 (16:48 +0000)]
Merge "Never allow -w."
Neil Fuller [Thu, 13 Aug 2015 10:40:39 +0000 (10:40 +0000)]
Merge "Include the okhttp.jar when compiling against internal classes"