OSDN Git Service
Andreas Gampe [Sun, 24 Mar 2019 20:06:03 +0000 (13:06 -0700)]
Merge changes from topic "agampe:a-b-ota:vendor-product-logical-partitions"
am:
2c0b5118c3
Change-Id: I877a2eb1f00ba520287f6add55eb942d56893eb6
Andreas Gampe [Sun, 24 Mar 2019 19:45:36 +0000 (19:45 +0000)]
Merge changes from topic "agampe:a-b-ota:vendor-product-logical-partitions"
* changes:
OtaPreopt: Add support for logical partitions
OtaPreopt: Refactor code
Andreas Gampe [Fri, 22 Mar 2019 18:21:34 +0000 (11:21 -0700)]
OtaPreopt: Add support for logical partitions
Use logical partitions prepared by update_engine if possible.
Bug:
128867786
Test: DexOptOtaTests
Change-Id: I6979487b91d24b7309c876f2bdc26a827e2fcd1e
Andreas Gampe [Thu, 21 Mar 2019 18:34:19 +0000 (11:34 -0700)]
OtaPreopt: Refactor code
Create a common function for mounting vendor and product. This is in
preparation for dynamic partition support.
Bug:
128867786
Test: DexOptOtaTests
Change-Id: I020a35bcdd4b2a122c967d3a5173ea289f284689
David Brazdil [Fri, 22 Mar 2019 14:00:26 +0000 (07:00 -0700)]
Merge "dexopt: Open class loader context dex files for dex2oat"
am:
77ec238fda
Change-Id: Iaf42b8186c2fea4878256c45cc48d74d24ae1345
David Brazdil [Fri, 22 Mar 2019 13:48:58 +0000 (13:48 +0000)]
Merge "dexopt: Open class loader context dex files for dex2oat"
Martijn Coenen [Fri, 22 Mar 2019 07:42:43 +0000 (00:42 -0700)]
Merge "Update otapreopt_chroot with new libapexd dependencies."
am:
2d8e815a26
Change-Id: I9cda409bd13a5c885782dcd691b52d038550011f
Martijn Coenen [Fri, 22 Mar 2019 07:33:25 +0000 (07:33 +0000)]
Merge "Update otapreopt_chroot with new libapexd dependencies."
Andreas Gampe [Thu, 21 Mar 2019 22:03:57 +0000 (15:03 -0700)]
Merge "Fix installd_dexopt_test."
am:
c684ad61dc
Change-Id: Ia6d03d6a880647c0ba2e6caf4aee5a7ce83b3eb1
Andreas Gampe [Thu, 21 Mar 2019 21:57:40 +0000 (21:57 +0000)]
Merge "Fix installd_dexopt_test."
Martijn Coenen [Wed, 13 Mar 2019 11:59:57 +0000 (12:59 +0100)]
Update otapreopt_chroot with new libapexd dependencies.
libapexd has added dependencies to libvold_binder (static) and
libbinder/libutils (shared).
Bug:
126740531
Test: builds
Change-Id: Ia1c1a94efc98fb15aa4c7d3cd22c6229c1931b28
Martijn Coenen [Thu, 21 Mar 2019 12:14:59 +0000 (13:14 +0100)]
Fix installd_dexopt_test.
The way Binder status error messages are printed was changed in
I911c3a667ddbe39245e5c3bbbd09154b2902aad8. Fix tests to deal
with it.
Bug:
128555505
Test: atest installd_dexopt_test
Change-Id: I9372eccf5bc9e97fc043112fa4b463560a3ff2a5
David Brazdil [Tue, 19 Mar 2019 11:44:21 +0000 (11:44 +0000)]
dexopt: Open class loader context dex files for dex2oat
When compiling secondary dex files, the class loader context may contain
files with 'app_data_file' selabel which dex2oat/dexoptanalyzer are not
allowed to open. Extend installd/dexopt to open these files for them and
pass in their file descriptors.
Dexopt of secondary dex files now starts by invoking dexoptanalyzer with
'--flatten-class-loader-context'. In this mode, dexoptanalyzer will only
parse the given class loader context, flatten it and print a
colon-separated list of dex files to standard output. Dexopt captures
the output and splits the string to obtain a list of context dex files.
It then continues by opening these files for dexoptanalyzer under
dropped privileges, passing the FDs using command line arguments, and
later opening the files again for dex2oat.
Bug:
126674985
Test: atest installd_dexopt_test
Change-Id: Ib9017a67c514ed711e1a03cfac313df796af6af9
Alec Mouri [Thu, 21 Mar 2019 06:48:20 +0000 (23:48 -0700)]
Merge changes from topic "RenderArea"
am:
3b14b7db6f
Change-Id: I6eea9c398eae9b8f67e20eb115e644bb0f7dd481
Treehugger Robot [Thu, 21 Mar 2019 06:33:46 +0000 (06:33 +0000)]
Merge changes from topic "RenderArea"
* changes:
Respect source crop when capturing layers.
libgui: add docs to geometry states and captureScreen
surfaceflinger: fix captureScreen for landscape LCM
surfaceflinger: improve RenderArea needsFiltering
surfaceflinger: respect install orientation in DisplayRenderArea
surfaceflinger: add install orientation to DisplayDevice
surfaceflinger: make mPrimaryDisplayOrientation static
surfaceflinger: clean up captureScreen
surfaceflinger: silence some RenderArea errors
surfaceflinger: fix race conditions in captureScreen
surfaceflinger: remove ISurfaceComposer.h from RenderArea
surfaceflinger: reorder width and height in RenderArea ctor
surfaceflinger: documents RenderArea
Alec Mouri [Fri, 26 Oct 2018 22:49:45 +0000 (15:49 -0700)]
Respect source crop when capturing layers.
Source crop we receive from window manager appears to be correct, no
need for this workaround.
Bug:
117892959
Test: adb shell screencap, rotation animation, power cycle display
Change-Id: I54ea3fb561ce4f2ea6f48079ad51efdd9984d186
Merged-In: I54ea3fb561ce4f2ea6f48079ad51efdd9984d186
Chia-I Wu [Wed, 22 Aug 2018 22:27:32 +0000 (15:27 -0700)]
libgui: add docs to geometry states and captureScreen
I think captureScreen is a bit fishy. We want it to capture the
logical display, but neither of the framework or surfaceflinger is
ready.
Bug:
113041375
Test: builds
Change-Id: Id5c76a4551e5260e83e64f0ca3ed2afaa34eecb5
Merged-In: Id5c76a4551e5260e83e64f0ca3ed2afaa34eecb5
Chia-I Wu [Wed, 29 Aug 2018 16:25:59 +0000 (09:25 -0700)]
surfaceflinger: fix captureScreen for landscape LCM
Make DisplayRenderArea::getSourceCrop return the display scissor
when the source crop is empty. Force the source crop to be empty in
captureScreen. This makes sure the install orientation is applied
on the source crop once, not twice.
Bug:
113041375
Test: force primaryDisplayOrientation, rotate screen, and switch
apps
Change-Id: I15006f867ff2d4a92ebccb1334ce59ab32abe69a
Merged-In: I15006f867ff2d4a92ebccb1334ce59ab32abe69a
Chia-I Wu [Tue, 28 Aug 2018 18:01:44 +0000 (11:01 -0700)]
surfaceflinger: improve RenderArea needsFiltering
Compare source crop (instead of the logical render area) against
physical render area to decide whether filtering is required. This
allows us to get rid of Layer::setFiltering.
As a result, captureLayers for Recents no longer enables filtering.
Screenshots under landscape mode no longer enables filtering.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: Ida95fdfec3a0dde7a19adf35c91bf3d570bab6bb
Merged-In: Ida95fdfec3a0dde7a19adf35c91bf3d570bab6bb
Chia-I Wu [Mon, 27 Aug 2018 21:54:37 +0000 (14:54 -0700)]
surfaceflinger: respect install orientation in DisplayRenderArea
DisplayRenderArea getTransform takes install orientation into
consideration. Fix getSourceCrop and getRotationFlags to respect
install orientation as well.
This results in some changes
- source crop is no longer rotated by install orientation for
LayerRenderArea
- rotation flags is no longer rotated by install orientation for
LayerRenderArea
- source crop is no longer set to getBounds by default for
LayerRenderArea (it is already done in captureLayers)
- source crop is always rotated by install orientation for
DisplayRenderArea
AFAICT, all the changes are actually fixes. Quickly verified by
forcing primaryDisplayOrientation to DisplayState::eOrientation90.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: If19df222ae52f6b276f9b0572e7b9bec872e3ae4
Merged-In: If19df222ae52f6b276f9b0572e7b9bec872e3ae4
Chia-I Wu [Mon, 27 Aug 2018 21:38:23 +0000 (14:38 -0700)]
surfaceflinger: add install orientation to DisplayDevice
Rather than querying it from SurfaceFlinger, initialize
DisplayDevice with the install orientation.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: Ibffe47033276e938388af775749c56f170fe8c77
Merged-In: Ibffe47033276e938388af775749c56f170fe8c77
Chia-I Wu [Mon, 27 Aug 2018 21:38:14 +0000 (14:38 -0700)]
surfaceflinger: make mPrimaryDisplayOrientation static
The convention is to store configstore values in static variables.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: I085178803bc897e3e9201fd10bd8731cc5b617c1
Merged-In: I085178803bc897e3e9201fd10bd8731cc5b617c1
Chia-I Wu [Tue, 28 Aug 2018 19:57:23 +0000 (12:57 -0700)]
surfaceflinger: clean up captureScreen
dispScissor is never empty. Remove the check and update the
comments so that the intention is clear.
It also becomes obvious that reqWidth/reqHeight is never zero when
we reach "// get screen geometry". All code following the comment
is actually dead, which means we probably broke it since commit
06a58e22.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: Ida1430383ce62271365d9ef64ad9b055638e1eac
Merged-In: Ida1430383ce62271365d9ef64ad9b055638e1eac
Chia-I Wu [Mon, 27 Aug 2018 22:40:25 +0000 (15:40 -0700)]
surfaceflinger: silence some RenderArea errors
Allow physical size and source crop size to be larger than the
logical size.
Bug:
113041375
Test: builds
Change-Id: I5a0e2158d114a6e06c8241b48d90bbb6172216cc
Merged-In: I5a0e2158d114a6e06c8241b48d90bbb6172216cc
Chia-I Wu [Thu, 23 Aug 2018 19:55:44 +0000 (12:55 -0700)]
surfaceflinger: fix race conditions in captureScreen
The display was looked up without holding the state lock. It was
also used (including indirectly from updateDimensions) without
holding the lock either.
Inline updateDimensions in captureScreen with the state lock held.
Stop calling updateDimensions in captureLayers.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: I8b361847c44373ce08930d906ec4a995efd1c21b
Merged-In: I8b361847c44373ce08930d906ec4a995efd1c21b
Chia-I Wu [Fri, 24 Aug 2018 22:34:02 +0000 (15:34 -0700)]
surfaceflinger: remove ISurfaceComposer.h from RenderArea
RenderArea can be made more generic by removing ISurfaceComposer.h
dependency. The caller also prefers to work with
ui::Transform::orientation_flags than ISurfaceComposer::Rotation (we
want to move updateDimensions to the caller).
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: I16e1392d5c92c2f423f98307e867918415404d26
Merged-In: I16e1392d5c92c2f423f98307e867918415404d26
Chia-I Wu [Thu, 23 Aug 2018 20:44:43 +0000 (13:44 -0700)]
surfaceflinger: reorder width and height in RenderArea ctor
Height is before width only in a crazy world.
Bug:
113041375
Test: take screenshot, rotate screen, screencap
Change-Id: Ia10b26dbba9a6a91abb5dae9fbe20bf17cd3e78f
Merged-In: Ia10b26dbba9a6a91abb5dae9fbe20bf17cd3e78f
Chia-I Wu [Thu, 23 Aug 2018 19:05:27 +0000 (12:05 -0700)]
surfaceflinger: documents RenderArea
Bug:
113041375
Test: builds
Change-Id: I7cf73b862aac38a782c4d0b154ed2357e1ef8dfe
Merged-In: I7cf73b862aac38a782c4d0b154ed2357e1ef8dfe
Nandana Dutt [Wed, 20 Mar 2019 18:32:45 +0000 (11:32 -0700)]
Merge "dumpstate: remove obsolete TODO"
am:
81c1bf02a8
Change-Id: I0c98b92ce87bbde4d8814621a53e3b417c6cdc21
Nandana Dutt [Wed, 20 Mar 2019 18:11:18 +0000 (18:11 +0000)]
Merge "dumpstate: remove obsolete TODO"
Nandana Dutt [Wed, 20 Mar 2019 15:36:18 +0000 (15:36 +0000)]
dumpstate: remove obsolete TODO
The Bugreporting API only supports the case where the calling
and consuming app are the same. The TODO is obsolete.
Test: builds
Change-Id: I7a408752d2b71d9e4b0ff77f0ae3f5f3a606239c
Jeongik Cha [Tue, 19 Mar 2019 21:27:34 +0000 (14:27 -0700)]
Merge "Add exceptionToString method into binder::Status"
am:
7cee1b4803
Change-Id: I4ef45660f1c57aa3b6f52530905c5a81dd4c1849
Treehugger Robot [Tue, 19 Mar 2019 21:15:41 +0000 (21:15 +0000)]
Merge "Add exceptionToString method into binder::Status"
Steven Moreland [Tue, 19 Mar 2019 19:50:29 +0000 (12:50 -0700)]
Merge "libbinder_ndk: unlinkToDeath dead proxy"
am:
a6697ec140
Change-Id: I7c78eaca55cfffe39669bfc204a5ff2d283a8b3f
Steven Moreland [Tue, 19 Mar 2019 19:37:39 +0000 (19:37 +0000)]
Merge "libbinder_ndk: unlinkToDeath dead proxy"
Jeongik Cha [Mon, 18 Mar 2019 05:38:33 +0000 (14:38 +0900)]
Add exceptionToString method into binder::Status
String exceptionToString(int32_t exceptionCode)
Add this method for logging purpose
Test: m -j
Bug:
127187108
Change-Id: I911c3a667ddbe39245e5c3bbbd09154b2902aad8
Steven Moreland [Wed, 13 Mar 2019 16:25:44 +0000 (09:25 -0700)]
libbinder_ndk: unlinkToDeath dead proxy
This CL automatically unlinks to death when a proxy dies (to clean
up additional data) and documents the behavior.
Bug: N/A
Test: ./runtests.sh
Change-Id: Ib092b25273a51f5afb12904bfe491927ad404f8d
Steven Moreland [Fri, 15 Mar 2019 16:47:08 +0000 (09:47 -0700)]
Merge "libbinder_ndk: fix UB memory access for race"
am:
91e66f9ba1
Change-Id: I638cc9c0e4c2b33492284aec6a4d9846b898cfde
Steven Moreland [Fri, 15 Mar 2019 16:32:37 +0000 (16:32 +0000)]
Merge "libbinder_ndk: fix UB memory access for race"
Nandana Dutt [Fri, 15 Mar 2019 11:27:42 +0000 (04:27 -0700)]
Merge "Check for user consent denial when dumping traces."
am:
cd3d535d38
Change-Id: I911de39b0eec73bcd761c584d0f931b633dae28e
Nandana Dutt [Fri, 15 Mar 2019 11:16:51 +0000 (11:16 +0000)]
Merge "Check for user consent denial when dumping traces."
Nandana Dutt [Fri, 15 Mar 2019 09:22:50 +0000 (02:22 -0700)]
Merge "Handle user consent denial sooner"
am:
62b3ec5981
Change-Id: I9403e618bc647742b11c4b4428e4bebef78728d9
Jiyong Park [Fri, 15 Mar 2019 09:16:32 +0000 (02:16 -0700)]
Merge "Revert "Have `otapreopt_chroot` bind-mount Bionic artifacts from the Runtime APEX.""
am:
0472824d3e
Change-Id: Ifc4865ae42f3cae802197bc4655e204e37258e62
Nandana Dutt [Thu, 14 Mar 2019 10:40:17 +0000 (10:40 +0000)]
Check for user consent denial when dumping traces.
VM traces is one of the longer running operations. Check consent
periodically and return early if necessary.
BUG:
128270426
Test: manual; by canceling when traces was running.
Test: verified "VM TRACES JUST NOW" shows up as expected
after refactoring in the normal flow as well.
Change-Id: Ic5282b948929857850730d6b06ec0cae8c39a99f
Nandana Dutt [Fri, 15 Mar 2019 09:04:16 +0000 (09:04 +0000)]
Merge "Handle user consent denial sooner"
Treehugger Robot [Fri, 15 Mar 2019 09:02:18 +0000 (09:02 +0000)]
Merge "Revert "Have `otapreopt_chroot` bind-mount Bionic artifacts from the Runtime APEX.""
Roland Levillain [Thu, 14 Mar 2019 19:35:29 +0000 (12:35 -0700)]
Merge "Fix a typo in an error message of dexopt."
am:
5f05746eb1
Change-Id: I37cbbf793465ca541ad0245e2c6a22d8bd69cc8e
Roland Levillain [Thu, 14 Mar 2019 19:30:29 +0000 (19:30 +0000)]
Merge "Fix a typo in an error message of dexopt."
Nandana Dutt [Thu, 14 Mar 2019 17:37:43 +0000 (10:37 -0700)]
Merge "dumpstate: Log duration only if significant"
am:
48159a81be
Change-Id: Ia273ed733289bcf80bb8f5c2c0543c372ad888ef
Treehugger Robot [Thu, 14 Mar 2019 17:15:24 +0000 (17:15 +0000)]
Merge "dumpstate: Log duration only if significant"
Roland Levillain [Thu, 14 Mar 2019 14:31:59 +0000 (14:31 +0000)]
Fix a typo in an error message of dexopt.
Test: m
Change-Id: I881f772afa41ea0389e99d3c972cacdc5e93ed1a
Gavin Corkery [Thu, 14 Mar 2019 14:22:04 +0000 (07:22 -0700)]
Merge "Rename data/pkg_staging to data/app-staging"
am:
4308cc478d
Change-Id: I3c1bfc034350839336395086504f513cd8a71002
Gavin Corkery [Thu, 14 Mar 2019 14:14:38 +0000 (14:14 +0000)]
Merge "Rename data/pkg_staging to data/app-staging"
Nandana Dutt [Thu, 14 Mar 2019 11:04:22 +0000 (11:04 +0000)]
dumpstate: Log duration only if significant
Ignore durations shorter than 500ms. Also adjust
the precision of the duration logged accordingly.
This is to avoid log spam.
Test: verified logcat on bugreport
Change-Id: Ibcef6b107de4327f769242c0c43d7f5317eee7f6
Nandana Dutt [Tue, 12 Mar 2019 10:52:56 +0000 (10:52 +0000)]
Handle user consent denial sooner
Currently dumpstate prompts the user for consent rightaway when called
via the API, but does not check for the result until the entire
bugreport is generated. This can make for a poor user experience - for
e.g. the client will continue to show a progress bar for a long time
after user denied consent.
Fix by checking for user consent intermittently.
BUG:
128270426
Test: Tested canceling during each long running operation in internal
Test: bugreport from power menu (ie non-api flow) still works as
expected
Test: adb shell
/data/nativetest64/dumpstate_smoke_test/dumpstate_smoke_test &&
adb shell /data/nativetest64/dumpstate_test/dumpstate_test
Change-Id: Icbed7c510ff9c9a882a7b49eac1a92fa17727635
Merged-In: Icbed7c510ff9c9a882a7b49eac1a92fa17727635
Gavin Corkery [Wed, 27 Feb 2019 12:24:19 +0000 (12:24 +0000)]
Rename data/pkg_staging to data/app-staging
Test: atest apex_e2e_tests
Bug:
126330086
(cherry picked from commit
950d5b4b8a55bda5874bfa75f3bcadf7160446a0)
Change-Id: Id98ecbad4f9452bc40a628ceb8c50ce20453d7ad
Merged-In: Id98ecbad4f9452bc40a628ceb8c50ce20453d7ad
Steven Moreland [Wed, 13 Mar 2019 16:23:18 +0000 (09:23 -0700)]
libbinder_ndk: fix UB memory access for race
AIBinder_DeathRecipient holds a strong pointer to its internal class
TransferDeathRecipient. If a user deletes a death recipient at the
same time a binder dies, the TDR still holds a reference to the
AIBinder_DeathRecipient object that may be invalidated.
Bug: N/A
Test: build, boot, and ./runtests.sh
Change-Id: Ic4cbc50c2d85ce52e36d4b157a50d0c75048e664
Martijn Coenen [Wed, 13 Mar 2019 12:09:33 +0000 (05:09 -0700)]
Merge "Revert "libbinder: Choose the binder driver at runtime based on system/vendor process.""
am:
1f69a63dda
Change-Id: Idcd26a40071e8a81d9d89edb275d61a159d205aa
Martijn Coenen [Wed, 13 Mar 2019 11:53:54 +0000 (11:53 +0000)]
Merge "Revert "libbinder: Choose the binder driver at runtime based on system/vendor process.""
Martijn Coenen [Wed, 13 Mar 2019 11:51:08 +0000 (11:51 +0000)]
Revert "libbinder: Choose the binder driver at runtime based on system/vendor process."
This reverts commit
46f95536bec334f8472ea92a7697c37602337398.
Reason for revert: appears to break (some?) native tests, which start looking for binder services in /dev/vndbinder
Change-Id: Id27d90e484d23a5975869729fa274213a9b1e1e4
Xin Li [Wed, 13 Mar 2019 05:19:33 +0000 (22:19 -0700)]
[automerger skipped] Merge "DO NOT MERGE - Merge PPRL.190305.001 into master"
am:
873f499994 -s ours
am skip reason: subject contains skip directive
Change-Id: I5a5b3e970313b090e759c8d39f655855e8609200
Xin Li [Wed, 13 Mar 2019 04:46:50 +0000 (21:46 -0700)]
[automerger skipped] DO NOT MERGE - Merge PPRL.190305.001 into master
am:
b126899aa4 -s ours
am skip reason: subject contains skip directive
Change-Id: Ieeaac3cba2372d4865faa3338d22c194411880e0
Jayant Chowdhary [Wed, 13 Mar 2019 04:45:25 +0000 (21:45 -0700)]
Merge "libbinder: Choose the binder driver at runtime based on system/vendor process."
am:
81a91bd83e
Change-Id: Ied00c256d04718f6bfa85ed4ccdeddf90b9f8ec6
Xin Li [Wed, 13 Mar 2019 02:26:03 +0000 (02:26 +0000)]
Merge "DO NOT MERGE - Merge PPRL.190305.001 into master"
Jayant Chowdhary [Wed, 13 Mar 2019 00:07:05 +0000 (00:07 +0000)]
Merge "libbinder: Choose the binder driver at runtime based on system/vendor process."
Jayant Chowdhary [Fri, 8 Mar 2019 18:08:20 +0000 (10:08 -0800)]
libbinder: Choose the binder driver at runtime based on system/vendor process.
This is needed since llndk libraries may cause the system variant of libbinder (and therefore
compile time checks are not enough) to be loaded in a vendor process. In that case, it should not be
initing the driver with /dev/binder. It should use /dev/vndbinder instead.
Bug:
124128212
Test: Device boots
Test: play YouTube videos, use camera to take pictures / record video (sanity) on devices supporting
legacy vndk(sailfish) and also devices supporting current vndk.
Change-Id: Ia5581efa04c8d4adc6af39668b0aa98e84324a27
Signed-off-by: Jayant Chowdhary <jchowdhary@google.com>
Nandana Dutt [Tue, 12 Mar 2019 10:34:39 +0000 (03:34 -0700)]
Merge "Dumpstate: Refactor DumpHals"
am:
b0f2b48c0a
Change-Id: I8b6bcb887675d907e21bb818cf0408d54d975b14
Nandana Dutt [Tue, 12 Mar 2019 10:24:43 +0000 (10:24 +0000)]
Merge "Dumpstate: Refactor DumpHals"
Mohammad Samiul Islam [Mon, 11 Mar 2019 20:21:58 +0000 (13:21 -0700)]
Merge "Add new emails to OWNER file"
am:
4000dd6f03
Change-Id: I1c1b94062eb5cbfdfc6677fe5aa422288dda1d67
Treehugger Robot [Mon, 11 Mar 2019 19:05:50 +0000 (19:05 +0000)]
Merge "Add new emails to OWNER file"
Nandana Dutt [Mon, 11 Mar 2019 16:33:24 +0000 (16:33 +0000)]
Dumpstate: Refactor DumpHals
Prefactoring to handle user consent denial sooner.
BUG:
128270426
Test: "adb shell bugreport" for flat file bugreports works
Change-Id: Ifba9ec600ab7fe23b046414fd718220cd07a39d2
Mohammad Samiul Islam [Mon, 11 Mar 2019 15:11:58 +0000 (15:11 +0000)]
Add new emails to OWNER file
Test: builds successfully
Change-Id: I080d4cd20290f3662087ccb5034d09c43311ecfb
Xin Li [Mon, 11 Mar 2019 16:13:29 +0000 (09:13 -0700)]
DO NOT MERGE - Merge PPRL.190305.001 into master
Bug:
127812889
Change-Id: I1cb1c38609ee26bb7c3e7343a9ed4c85fb4fb0a6
Xin Li [Mon, 11 Mar 2019 16:09:43 +0000 (16:09 +0000)]
Merge "DO NOT MERGE - Merge PPRL.190305.001 into stage-aosp-master" into stage-aosp-master
Nandana Dutt [Mon, 11 Mar 2019 14:23:10 +0000 (07:23 -0700)]
Merge "Log duration to logcat always."
am:
ea7255933e
Change-Id: I4d4ddd7425beee3a5272c9bb2f652e1ddab042e4
Treehugger Robot [Mon, 11 Mar 2019 14:18:57 +0000 (14:18 +0000)]
Merge "Log duration to logcat always."
Nandana Dutt [Mon, 11 Mar 2019 13:17:56 +0000 (06:17 -0700)]
Merge "dumpstate: move dump_traces to Dumpstate"
am:
6644784ead
Change-Id: I7d3609d6abd6a3021c525cc63244561384a840a6
Treehugger Robot [Mon, 11 Mar 2019 13:10:53 +0000 (13:10 +0000)]
Merge "dumpstate: move dump_traces to Dumpstate"
Nandana Dutt [Mon, 11 Mar 2019 11:00:58 +0000 (11:00 +0000)]
Log duration to logcat always.
Remove TODO and make the logging behavior intentional.
Rename field to make the intent clearer.
BUG:
128270426
Test: took bugreport and verified log lines: adb logcat | grep "Duration was"
Change-Id: I6a9379af9ee6a03587c40c35f3f242d84ddf3bc3
Nandana Dutt [Mon, 11 Mar 2019 09:23:09 +0000 (09:23 +0000)]
dumpstate: move dump_traces to Dumpstate
Prefactoring to handle user consent denial sooner.
BUG:
128270426
Test: adb shell
/data/nativetest64/dumpstate_smoke_test/dumpstate_smoke_test &&
adb shell /data/nativetest64/dumpstate_test/dumpstate_test
Change-Id: Ifbace4391d22901480e41d3125d8a89508ebd52d
The Android Open Source Project [Fri, 8 Mar 2019 23:56:59 +0000 (15:56 -0800)]
DO NOT MERGE - Merge PPRL.190305.001 into stage-aosp-master
Bug:
127812889
Change-Id: I6af64f4b2667eb4c0540b23cec899581620fc352
Steven Moreland [Fri, 8 Mar 2019 23:40:35 +0000 (15:40 -0800)]
Merge changes from topic "
b127336441"
am:
9b07b1ab41
Change-Id: I104b6cda22b6d1fc0de531c54cf895d254d4daa5
Steven Moreland [Fri, 8 Mar 2019 23:39:15 +0000 (15:39 -0800)]
libbinder_ndk: define userdata class in headers
am:
304f9d37b1
Change-Id: I76291bd2566509d1f39aaaa4e64c881be9985435
Steven Moreland [Fri, 8 Mar 2019 23:06:43 +0000 (23:06 +0000)]
Merge changes from topic "
b127336441"
* changes:
libbinder_ndk: Fix 'isRemote' value for Bn*.
libbinder_ndk: define userdata class in headers
android-build-team Robot [Fri, 8 Mar 2019 18:45:51 +0000 (18:45 +0000)]
Snap for
5240760 from
3ce2b66f92ea8c50005b38595bb1aed4c80135ed to pi-platform-release
Change-Id: I35689fd57b90f25332bd2e7219b3a72f3d386cab
Jiyong Park [Thu, 7 Mar 2019 15:04:17 +0000 (15:04 +0000)]
Revert "Have `otapreopt_chroot` bind-mount Bionic artifacts from the Runtime APEX."
This reverts commit
c3265aef20d4afca0eb62d0ce02fdbc8a254c421.
Reason for revert: No need to bind-mount bionic libraries
/system/lib/libc.so is a symlink directly to
/apex/com.android.runtime/lib/bootstrap/libc.so.
Bug:
125549215
Bug:
113373927
Bug:
120266448
Change-Id: I8a22d9e70ea017ff9a9f0b0f81f8a90fee3dabf2
Test: m
Test: asit/dexoptota/self_full
Steven Moreland [Wed, 6 Mar 2019 23:02:23 +0000 (15:02 -0800)]
libbinder_ndk: Fix 'isRemote' value for Bn*.
Bn* is not remote (it's local).
This wasn't noticed before since the AIDL generated method 'associate'
always wrapped a proxy, and since a wrapped Bn object will always have
a null remoteBinder, it would properly show that it was local even
though it was wrapped.
Bug:
127336441
Test: atest android.binder.cts
Change-Id: Ic4d96822a355e88bc8f3dbbc8fa602835653139e
Steven Moreland [Wed, 6 Mar 2019 21:55:49 +0000 (13:55 -0800)]
libbinder_ndk: define userdata class in headers
Rather than auto-generating it for each class.
From 'struct AIBinder' comment:
(process boundary)
|
MyInterface ---> AIBinder_Weak | ProxyForMyInterface
^ . | |
| . | |
| v | v
UserData <--- AIBinder <-|- AIBinder
ICInterfaceData represents 'UserData' here.
Bug:
127336441
Test: atest android.binder.cts
Change-Id: I2ffd7533dd81f06126547e4c82ec7e1e1cb0b0d0
Steven Moreland [Wed, 6 Mar 2019 00:53:57 +0000 (16:53 -0800)]
Merge "libbinder_ndk: support dump"
am:
36415570df
Change-Id: Ice93b194d4990fbd1315b41391bb8f8bcb7f4b55
Steven Moreland [Wed, 6 Mar 2019 00:22:37 +0000 (00:22 +0000)]
Merge "libbinder_ndk: support dump"
Steven Moreland [Tue, 5 Mar 2019 00:47:07 +0000 (16:47 -0800)]
libbinder_ndk: support dump
At the NDK layer:
AIBinder_Class_setOnDump (to implement dump)
AIBinder_dump (to call dump)
At the AIDL layer:
ICInterface::dump(...) to implement and call dump
Bug:
125733384
Test: atest android.binder.cts
Change-Id: I0873ad574bb7a8b9d645c7301fdfa0797e62c834
Jooyung Han [Tue, 5 Mar 2019 02:29:32 +0000 (18:29 -0800)]
Merge "Move PublicFormat from libandroid_runtime to libui"
am:
17f57c6ead
Change-Id: Ib3fe86fb8c1d2d6726bf4e439f8c8c4ca6ea8619
Treehugger Robot [Tue, 5 Mar 2019 02:16:42 +0000 (02:16 +0000)]
Merge "Move PublicFormat from libandroid_runtime to libui"
Greg Kaiser [Tue, 5 Mar 2019 01:00:24 +0000 (17:00 -0800)]
Merge "ip-up-vpn: Fix resource leaks in error cases"
am:
a5d85affa9
Change-Id: Id6a1442dd9226b50a392aca2d484afd357bcabd9
Treehugger Robot [Tue, 5 Mar 2019 00:35:12 +0000 (00:35 +0000)]
Merge "ip-up-vpn: Fix resource leaks in error cases"
Greg Kaiser [Mon, 4 Mar 2019 22:00:34 +0000 (14:00 -0800)]
ip-up-vpn: Fix resource leaks in error cases
These leaks aren't really consequential since this is a standalone
command, and these "leaks" are in main(). But by putting in simple
fixes for these, we reduce the noise when using static analysis to
find leaks in Android code.
Test: TreeHugger
Change-Id: I2cbb4585c6c8d59a0eb89a72f17dfc4fbc532249
Nandana Dutt [Mon, 4 Mar 2019 15:09:26 +0000 (07:09 -0800)]
Merge "dumpstate: handle errors gracefully"
am:
e953b04799
Change-Id: I6d55aa42a196ac53883057800596b0d065ce605c
Treehugger Robot [Mon, 4 Mar 2019 14:58:30 +0000 (14:58 +0000)]
Merge "dumpstate: handle errors gracefully"
Jooyung Han [Mon, 4 Mar 2019 07:28:48 +0000 (16:28 +0900)]
Move PublicFormat from libandroid_runtime to libui
PublicFormat is defined in android_view_Surface.
However this and related functions are also used in libmediandk
which is both NDK and LL-NDK. As a LL-NDK it should run without
libandroid_runtime.
By moving PublicFormat definition and its utility functions
into libui, LL-NDK can use it without cost of libandroid_runtime.
And also PublicFormat will be deprecated(b/
126594675).
This is a part cutting dependency from libmediandk to
libandroid_runtime.
Note: applied clang-format
Bug:
124268753
Test: m -j
Change-Id: I7215dde607b425f7a0d47a6541b83fbe3a0deaf4
Merged-In: I7215dde607b425f7a0d47a6541b83fbe3a0deaf4
Calin Juravle [Sat, 2 Mar 2019 02:49:00 +0000 (18:49 -0800)]
Merge "Ensure we select the right execution binary for all dexopt commands"
am:
3e167da916
Change-Id: I8b32fe8ad5a98e60ca353029cfa6933e59cee315