OSDN Git Service
Christopher Wiley [Tue, 8 Mar 2016 18:49:51 +0000 (10:49 -0800)]
libbinder: Make null handling explicit for strong binders
Bug:
27156495
Test: Compiles
Change-Id: I62cade1761377a9ec2921a758307582552e4c2ac
Dan Stoza [Thu, 3 Mar 2016 15:36:33 +0000 (15:36 +0000)]
Merge ""#warning" conflict with -Werror"
Zhiquan Liu [Thu, 3 Mar 2016 07:24:41 +0000 (15:24 +0800)]
"#warning" conflict with -Werror
if code run #warning, it will have compile error.
Change-Id: Ic7f05cd722c14f3c5da89e2c3792615dd15cb798
Signed-off-by: Zhiquan Liu <zhiquan.liu@intel.com>
Christopher Wiley [Wed, 17 Feb 2016 17:32:15 +0000 (17:32 +0000)]
Merge "Allocate buffers locally in systems without SurfaceFlinger"
Christopher Wiley [Wed, 10 Feb 2016 23:33:14 +0000 (15:33 -0800)]
Allocate buffers locally in systems without SurfaceFlinger
In Brillo systems, we have no applications and configure
SELinux policy statically. In this model, we do not need
a special service to allocate buffers, since any process
that needs buffers can be appropriately marked at build time.
Bug:
26936651
Test: A test program on brilloemulator can allocate buffers for
the camera with this patch.
Change-Id: Ifcdb7a4b878f9a26f4899c7b75a12447dcdb0ac9
Samuel Tan [Tue, 16 Feb 2016 20:47:42 +0000 (20:47 +0000)]
Merge "libbinder: add getters for PersistableBundle keys"
Aurimas Liutikas [Tue, 16 Feb 2016 19:30:38 +0000 (19:30 +0000)]
Merge "Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp"
Aurimas Liutikas [Tue, 16 Feb 2016 18:42:52 +0000 (10:42 -0800)]
Fix -Wgnu-designator warnigns in cmds/flatland/Main.cpp
Bug: http://b/
27068438
Change-Id: I3064950f761411fe5271c6259867ad437c71e66d
Christopher Wiley [Tue, 16 Feb 2016 16:50:01 +0000 (16:50 +0000)]
Merge "Replace AppOpsManager locking with std::lock_guard"
Dimitry Ivanov [Sat, 13 Feb 2016 01:43:29 +0000 (01:43 +0000)]
Merge "Add missing liblog dependency"
Dimitry Ivanov [Sat, 13 Feb 2016 01:13:10 +0000 (17:13 -0800)]
Add missing liblog dependency
Bug: http://b/
27171986
Change-Id: Ifbe01c6b325e7198a95ea865f557b0c9acb65809
Christopher Wiley [Fri, 12 Feb 2016 23:41:08 +0000 (15:41 -0800)]
Replace AppOpsManager locking with std::lock_guard
Change-Id: Ice4c101659837cea594194cfb335f9f3f6837f38
Christopher Wiley [Fri, 12 Feb 2016 23:43:07 +0000 (23:43 +0000)]
Merge "Don't rely on AppOpsManager in systems without applications"
Christopher Wiley [Fri, 12 Feb 2016 23:43:00 +0000 (23:43 +0000)]
Merge "Fix race in AppOpsManager client library"
Samuel Tan [Fri, 12 Feb 2016 21:56:17 +0000 (13:56 -0800)]
libbinder: add getters for PersistableBundle keys
Add getters for the keys in each map stored in
PersistableBundle. This will allow users of
PersistableBundle to iterate over entries of
a specific type in PersistableBundle.
Also, add "using" statements to PersistableBundle
for std containers to reduce verbosity.
BUG:
27171235
Change-Id: Idd4b3bf713071ca8d8d023f1f83bdaf3ccbdd46f
TEST: libbinder compiles successfully.
Casey Dahlin [Thu, 11 Feb 2016 23:24:41 +0000 (23:24 +0000)]
Merge "libbinder: Allow readByteVector to take uint8_t"
Mark Salyzyn [Wed, 10 Feb 2016 16:12:50 +0000 (16:12 +0000)]
Merge "dumpstate: Continued build breakage"
Mark Salyzyn [Wed, 10 Feb 2016 15:41:12 +0000 (07:41 -0800)]
dumpstate: Continued build breakage
- drop assignment length in scanf
- drop -std=gnu99 as invalid for C++
Change-Id: I924d2561069b3c289cb9deb475f732c46bb5a67e
Casey Dahlin [Tue, 9 Feb 2016 19:08:35 +0000 (11:08 -0800)]
libbinder: Allow readByteVector to take uint8_t
Change-Id: I2f0d1f3b4c04f04e7a866e9a4cf76c899b4b4980
Test: AIDL integration tests pass
Bug:
27078230
Mark Salyzyn [Tue, 9 Feb 2016 17:22:50 +0000 (17:22 +0000)]
Merge changes I6b499f53,If821ed21
* changes:
dumpstate: report process times
Add option to #define _DUMPSTATE_DRY_RUN_ to skip the actual dumps.
Mark Salyzyn [Fri, 5 Feb 2016 23:33:17 +0000 (15:33 -0800)]
dumpstate: report process times
- Fix per-pid helper to pick up comm if cmdline is empty
and assume kernel thread.
- Add per pid report command line user time if not kernel thread,
system time and I/O time, plus percentage if non zero.
Bug:
26966375
Change-Id: I6b499f535cc3687eb468b32ae39e7d151dffb11f
Felipe Leme [Wed, 11 Nov 2015 04:10:25 +0000 (20:10 -0800)]
Add option to #define _DUMPSTATE_DRY_RUN_ to skip the actual dumps.
(cherry picked from commit
93d705b6d742e8f00167ff2384e35c2205a6f9da)
Such option is useful when debugging dumpstate itself, since it will
finish much sooner and its output will be much smaller.
Change-Id: If821ed21715461bf82eea0b2be4b926239ad69da
John Reck [Sat, 6 Feb 2016 00:11:38 +0000 (00:11 +0000)]
Merge "Enable wildcard in the list of kernel functions"
Mark Salyzyn [Fri, 5 Feb 2016 22:38:34 +0000 (22:38 +0000)]
Merge "dumpstate: refine calculated eMMC performance values"
Mark Salyzyn [Thu, 4 Feb 2016 17:20:44 +0000 (09:20 -0800)]
dumpstate: refine calculated eMMC performance values
Correct the raw performance with the average queue depth. Add a
report for I/O per second, and the average queue depth. Round
values up.
Bug:
26966375
Change-Id: I2a8d233872d5e5ff79d433abba3c34d5945701f4
Christopher Wiley [Fri, 5 Feb 2016 17:08:23 +0000 (09:08 -0800)]
Don't rely on AppOpsManager in systems without applications
Brillo has no applications, and doesn't run AppOpsManager. Instead,
services are granted statically configured SELinux permissions at
build time. Rely on that configuration rather than Android's
dynamically configurable permission model.
Bug:
26936651
Test: Test code on Brillo system is able to connect to the camera.
Change-Id: I84b72a762c2f534c2e1cc6f99ef2003388fb1265
Mark Salyzyn [Fri, 5 Feb 2016 18:05:23 +0000 (18:05 +0000)]
Merge "system_server BINDER_TYPE_FD driver ashmem accessors"
Christopher Wiley [Fri, 5 Feb 2016 17:06:30 +0000 (09:06 -0800)]
Fix race in AppOpsManager client library
If the service stays down for more than 10 seconds, client threads
will time out attempting to obtain a reference, and never release
the mutex. This causes deadlock on the next attempt to speak
with the AppOpsManager.
Test: System without AppOpsManager service no longer hangs in camera
code.
Change-Id: I5a8ac2f0f35345f72c3d2a45b9d1fc5a78d28a51
Thomas Buhot [Wed, 27 Jan 2016 08:44:31 +0000 (09:44 +0100)]
Enable wildcard in the list of kernel functions
The handling of the -k option is too restrictive
as it stricly checks the names of the kernel functions
after it wrote them in /d/tracing/set_ftrace_filter.
However, a common usage of that function filtering with ftrace
is to use a wildcard character (*) that the ftrace tracer
automatically expands to all the matching kernel functions.
This enables the support of the wildcard (*) character
in the -k option to trace kernel functions.
Change-Id: Ifffae975c20e1c253157a3a6b44a14b4f342b9d0
Signed-off-by: Thomas Buhot <thomas.buhot@intel.com>
Signed-off-by: Zhiquan Liu <zhiquan.liu@intel.com>
Eino-Ville Talvala [Fri, 5 Feb 2016 00:52:11 +0000 (00:52 +0000)]
Merge "Binder::Status: Add operator<< to help with gtest logging"
Eino-Ville Talvala [Wed, 3 Feb 2016 21:19:49 +0000 (13:19 -0800)]
Binder::Status: Add operator<< to help with gtest logging
Change-Id: I22fcf7c6314451701ba947af2d69d4e45b22345a
Mark Salyzyn [Tue, 2 Feb 2016 18:27:03 +0000 (10:27 -0800)]
system_server BINDER_TYPE_FD driver ashmem accessors
check if device matches the ashmem rdev, before calling
ashmem_get_size_region. This eliminates making this call
when associated with other driver file descriptors.
Bug:
26374183
Bug:
26918423
Bug:
26871259
Change-Id: I1f88c2c93ea35a73c8e14125f3d1a6c67fa4f15b
Dan Albert [Wed, 3 Feb 2016 01:16:40 +0000 (01:16 +0000)]
Merge "Fix typo from last commit."
Dan Albert [Wed, 3 Feb 2016 01:08:05 +0000 (17:08 -0800)]
Fix typo from last commit.
Bad commit: https://android-review.googlesource.com/#/c/200449/1
Change-Id: I412e2740951f15cf843bd36a50476070c0cb1ebd
Dan Albert [Wed, 3 Feb 2016 00:40:26 +0000 (00:40 +0000)]
Merge "Clean up warnings in inputflinger."
Dan Albert [Wed, 3 Feb 2016 00:40:03 +0000 (00:40 +0000)]
Merge "Clean up warnings in EGL."
Dan Albert [Tue, 2 Feb 2016 23:11:57 +0000 (15:11 -0800)]
Clean up warnings in inputflinger.
Bug: http://b/
26936282
Change-Id: Ie1fb281425c3b8c4b3e6e88dc5722d51b77ee093
Dan Albert [Tue, 2 Feb 2016 23:08:34 +0000 (15:08 -0800)]
Clean up warnings in EGL.
Bug: http://b/
26936282
Change-Id: I059628332d638f080a5236eed62cf0afa1d4d6ba
Andreas Gampe [Mon, 1 Feb 2016 22:23:33 +0000 (22:23 +0000)]
Merge "Binder: Add log output in IPCThreadState::self()"
Christopher Wiley [Mon, 1 Feb 2016 21:40:33 +0000 (21:40 +0000)]
Merge "Add support for decoding UTF16 to UTF8"
Andreas Gampe [Mon, 1 Feb 2016 21:21:56 +0000 (13:21 -0800)]
Binder: Add log output in IPCThreadState::self()
In the failure case, and on shutdown, add log output. It's highly
likely the client will crash when a null value is returned. Displaying
a possible root cause helps diagnosing issues.
Bug:
26865304
Change-Id: Ie5255ac50d6dcdf9f918dcef785788ea795a4791
Christopher Wiley [Fri, 29 Jan 2016 00:56:53 +0000 (16:56 -0800)]
Add support for decoding UTF16 to UTF8
This is helpful because many native service developers use std::string
internally, and String16 doesn't work very well with many C++
primitives.
Bug:
26729450
Test: Integration test appears to show this is correct.
Change-Id: Ib75068881da28fd8515685bcbcbe864ba3b7ec23
Mathieu Chartier [Thu, 28 Jan 2016 19:13:02 +0000 (19:13 +0000)]
Merge "Application image support for installd DO NOT MERGE"
Mathieu Chartier [Wed, 21 Oct 2015 21:05:28 +0000 (14:05 -0700)]
Application image support for installd DO NOT MERGE
DO NOT MERGE
sort of
(cherry picked from commit
edc8bc48fbf143e35578bf6cccf980dfab076196)
Bug:
22858531
Change-Id: I5a9dfcf67d2232e8755886085e9ba1bb37494d39
Mark Salyzyn [Wed, 27 Jan 2016 19:04:53 +0000 (19:04 +0000)]
Merge "system_server BINDER_TYPE_FD sockets using ashmem accessors"
Mark Salyzyn [Wed, 27 Jan 2016 16:02:48 +0000 (08:02 -0800)]
system_server BINDER_TYPE_FD sockets using ashmem accessors
check if device is a character device, before calling
ashmem_get_size_region. We do not check if the st_rdev
matches /dev/ashmem. So this at least eliminates making
this call when associated with a socket.
Bug:
26374183
Change-Id: I68ed9d1c2cd4c47228ed065e3e18eb4151f038f4
Mark Salyzyn [Wed, 27 Jan 2016 17:59:58 +0000 (17:59 +0000)]
Merge changes I4ebccfd0,I3507bdfd,Iaf02f55c
* changes:
Parcel: file descriptor leak
Parcel: include Android coding standard
Parcel: LOG_TAG cleanup
Mark Salyzyn [Wed, 27 Jan 2016 16:02:48 +0000 (08:02 -0800)]
Parcel: file descriptor leak
Resolve a file descriptor leak when a request for
ashmem size adjustment is not filed.
Change-Id: I4ebccfd096ec5313725fd99dc3e025f9561d061f
Mark Salyzyn [Wed, 27 Jan 2016 16:02:48 +0000 (08:02 -0800)]
Parcel: include Android coding standard
Change-Id: I3507bdfd04a06db85b46539338c787395ec652ff
Mark Salyzyn [Wed, 27 Jan 2016 16:02:48 +0000 (08:02 -0800)]
Parcel: LOG_TAG cleanup
NB: several ALOG messages behind comments, recommend removal
Change-Id: Iaf02f55cbf7ef0ee7c4a6ee67c7f8b57cf97cd26
Mark Salyzyn [Tue, 26 Jan 2016 20:41:51 +0000 (20:41 +0000)]
Merge "Revert "dumpstate: Add security log dump""
Mark Salyzyn [Tue, 26 Jan 2016 15:24:08 +0000 (07:24 -0800)]
Revert "dumpstate: Add security log dump"
This reverts commit
c7ad8cb82f035576c28707a8cbeb16b7ead2a4e7.
Bug:
26029733
Change-Id: I0c61f4d2c0aa9e07c6cc7b190d803e98c3a98d40
Martijn Coenen [Mon, 25 Jan 2016 09:47:27 +0000 (09:47 +0000)]
Merge "Implementation of HCE for NFC-F."
Yoshinobu Ito [Mon, 18 Jan 2016 12:02:47 +0000 (21:02 +0900)]
Implementation of HCE for NFC-F.
This impementation includes HCE(Host-based Card Emulation) for NFC-F
which emulates NFC Forum Type 3 Tag on a smartphone.
Signed-off-by: Daisuke Ito <DaisukeE.Ito@jp.sony.com>
Yoshinobu Ito <Yoshinobu.Ito@jp.sony.com>
Change-Id: I1618a7fba801e0705ff3cc078206a1446b3dd56d
Samuel Tan [Tue, 12 Jan 2016 23:09:44 +0000 (23:09 +0000)]
Merge "Add PersistableBundle.aidl into binder includes"
Samuel Tan [Mon, 21 Dec 2015 20:11:00 +0000 (12:11 -0800)]
Add PersistableBundle.aidl into binder includes
Add the PersistableBundle.aidl into the new /aidl/binder/
subdirectory (along with the necessary 'android' and 'os'
subdirectories) so that the aidl file will live alongside
its native implementation in libbinder. This change will
allow native-only Android checkouts (e.g. Brillo) to use
PersistableBundle.
The /aidl/ subdirectory should henceforth contain any
aidl files that live with native code.
BUG:
26292234
Change-Id: I07b49dad8afd679f33e41fa9044dffc02ec25e3e
Colin Cross [Tue, 5 Jan 2016 21:46:28 +0000 (21:46 +0000)]
Merge "libinstalld: add dependency on libselinux"
Nick Kralevich [Thu, 24 Dec 2015 02:59:16 +0000 (02:59 +0000)]
Merge "More O_CLOEXEC"
Nick Kralevich [Thu, 24 Dec 2015 02:58:05 +0000 (18:58 -0800)]
More O_CLOEXEC
Change-Id: I31991ed089f395174857e86e509b8190ad98f92a
Colin Cross [Tue, 22 Dec 2015 00:21:45 +0000 (16:21 -0800)]
libinstalld: add dependency on libselinux
libinstalld uses headers from libselinux, add a LOCAL_SHARED_LIBRARIES
dependency on libselinux.
Change-Id: Ic12c55547526638810bcc1ac952d205f2f163fd3
Samuel Tan [Thu, 17 Dec 2015 01:19:10 +0000 (01:19 +0000)]
Merge "Fix format specifiers for size_t types"
Samuel Tan [Thu, 17 Dec 2015 01:13:29 +0000 (17:13 -0800)]
Fix format specifiers for size_t types
Use %zu instead of %u for logging size_t types, which
are of the unsigned long type, not the unsigned int type.
BUG: None
Change-Id: I5af6c2ea544a5f11b283bc9aebe8b5f0cfc605ec
Samuel Tan [Thu, 17 Dec 2015 00:35:28 +0000 (00:35 +0000)]
Merge "Add support for PersistableBundle in C++"
Nick Kralevich [Wed, 16 Dec 2015 21:31:46 +0000 (21:31 +0000)]
Merge "atrace.rc: tracing should be disabled by default."
Nick Kralevich [Wed, 16 Dec 2015 20:37:25 +0000 (12:37 -0800)]
atrace.rc: tracing should be disabled by default.
Bug:
26217098
Change-Id: I6066c433d19c2d1c220669a9a78f5085af8c1d4b
Casey Dahlin [Tue, 15 Dec 2015 21:04:16 +0000 (21:04 +0000)]
Merge "Allow null values for Parcel readers and writers"
Casey Dahlin [Wed, 25 Nov 2015 23:09:45 +0000 (15:09 -0800)]
Allow null values for Parcel readers and writers
Test: AIDL integration tests pass
Bug:
25969194
Change-Id: Ib7023c8e02af49e08c9921f1d097e5967c1d378e
Signed-off-by: Casey Dahlin <sadmac@google.com>
Casey Dahlin [Tue, 15 Dec 2015 01:09:50 +0000 (01:09 +0000)]
Merge "Don't redefine B_PACK_CHARS"
Casey Dahlin [Tue, 15 Dec 2015 00:09:06 +0000 (16:09 -0800)]
Don't redefine B_PACK_CHARS
Change-Id: I5dfaca405f0f866e697d5ae546ead8a8eb6de7ca
Test: make checkbuild succeeds
Bug:
25868260
Signed-off-by: Casey Dahlin <sadmac@google.com>
Felipe Leme [Mon, 14 Dec 2015 23:25:48 +0000 (23:25 +0000)]
Merge "Migrated dumpstate to C++."
Felipe Leme [Mon, 14 Dec 2015 23:09:56 +0000 (15:09 -0800)]
Migrated dumpstate to C++.
Changes required:
- Explicity casting function pointers.
- Moving variables initialization before 'goto' statements.
- Changing string references from 'char *' to 'const char *'.
- Using 'extern "C"' so device-dependent implementation of
'dumpstate_board()' can be written in C or C++.
BUG:
25563823
Change-Id: Ibeb70fd23e64e9b530736b554a372cb232fe1a44
Samuel Tan [Tue, 24 Nov 2015 02:22:12 +0000 (18:22 -0800)]
Add support for PersistableBundle in C++
Add support for PersistableBundle, a mapping from
String values to various types, in C++.
BUG:
25815410
Change-Id: If609b294a1709314bb4220afc4f2269b556babb8
Greg Hackmann [Mon, 14 Dec 2015 21:12:52 +0000 (21:12 +0000)]
Merge "atrace: add database tag"
Mark Salyzyn [Mon, 14 Dec 2015 16:33:00 +0000 (16:33 +0000)]
Merge "dumpstate: Add security log dump"
Mark Salyzyn [Fri, 11 Dec 2015 21:04:02 +0000 (13:04 -0800)]
dumpstate: Add security log dump
Bug:
26029733
Change-Id: I2e5b44d642bbac7f62f63388dd8a3123f2abeafe
Mark Salyzyn [Fri, 11 Dec 2015 18:43:21 +0000 (18:43 +0000)]
Merge "dumpstate: build breakage @
2482293"
Mark Salyzyn [Fri, 11 Dec 2015 18:41:52 +0000 (10:41 -0800)]
dumpstate: build breakage @
2482293
aosp is in C, internal is C++, struct must be explicit
Bug:
24200279
Change-Id: I1c9d8156a86db6de73ee5b70063e63b38170b052
Mark Salyzyn [Fri, 11 Dec 2015 15:17:26 +0000 (15:17 +0000)]
Merge "dumpstate: Dump eMMC 5.0 stats"
Mark Salyzyn [Wed, 9 Dec 2015 19:21:28 +0000 (11:21 -0800)]
dumpstate: Dump eMMC 5.0 stats
Extended csd EXT_CSD_REV, PRE_EOL_INFO, DEVICE_LIFE_TIME_EST_TYP_A,
and DEVICE_LIFE_TIME_EST_TYP_B fields printed if available. Allow
dumpstate access to /sys/kernel/debug/mmc0/mmc0:0001/ext_csd.
Bug:
24200279
Change-Id: I310bdcf546a9307cf927f109359ad5a6d24b70f8
Colin Cross [Thu, 10 Dec 2015 21:15:19 +0000 (21:15 +0000)]
Merge "Fix allocation count"
Christopher Wiley [Wed, 9 Dec 2015 22:34:58 +0000 (22:34 +0000)]
Merge "libbinder: Enable service specific error codes"
Christopher Wiley [Sat, 21 Nov 2015 01:48:27 +0000 (17:48 -0800)]
libbinder: Enable service specific error codes
Add another factory method that takes a message and service
specific error code.
Bug:
25800533
Test: system/tools/aidl integration tests pass
Change-Id: I592cb7def0538576965d14c200ab58548b3bef32
Colin Cross [Wed, 9 Dec 2015 01:15:50 +0000 (17:15 -0800)]
Fix allocation count
The realloc case in continueWrite should not increment the counter,
the pointer passed to realloc is guaranteed to be non-NULL so the total
number of allocations will not have changed.
When realloc is called in restartWrite mData has not been checked
against NULL, increment the counter if it was NULL.
Bug:
26086286
Change-Id: I4c8af450cca1868b91793c0c5f0d8c4b4b5badbe
Dan Stoza [Tue, 8 Dec 2015 17:08:13 +0000 (09:08 -0800)]
Merge "Fix the execution point of onFrameAvailable/onFrameReplaced callbacks"
am:
789ede379d
* commit '
789ede379d2f19ce284866c72bcaad7e8255b1cc':
Fix the execution point of onFrameAvailable/onFrameReplaced callbacks
Dan Stoza [Tue, 8 Dec 2015 17:05:25 +0000 (17:05 +0000)]
Merge "Fix the execution point of onFrameAvailable/onFrameReplaced callbacks"
Dimitry Ivanov [Mon, 7 Dec 2015 23:16:20 +0000 (23:16 +0000)]
Merge "Replace libGLESv3 symlink with shared library"
am:
18ff655766
* commit '
18ff6557667f3d130ad1f83e04f3b175f8b8cf6d':
Replace libGLESv3 symlink with shared library
Dimitry Ivanov [Mon, 7 Dec 2015 19:00:58 +0000 (19:00 +0000)]
Merge "Replace libGLESv3 symlink with shared library"
Christian Poetzsch [Mon, 7 Dec 2015 13:36:22 +0000 (13:36 +0000)]
Fix the execution point of onFrameAvailable/onFrameReplaced callbacks
In
a4650a5 the concept of a maximum frame number allowance for the consumer was
introduced. A call to acquireBuffers will only return buffers when their frame
number is less-than-or-equal-to this maximum frame number. When SurfaceFlinger
is the consumer, this maximum frame number is calculated in the
onFrameAvailable/onFrameReplaced callbacks. These callbacks are called when a
new buffer is dequeued by the application. The problem is that these callbacks
are called _after_ the fence wait which is used to throttle the frame
production of client apps. When the previous frame needs a long time to draw,
those waits can potentially be a long time. As a result SurfaceFlinger won't do
any composition with the new frame until the wait is over.
Normally this isn't a big problem because there is a queue of buffers for
SurfaceFlinger to work with. However, this changes massively when a client app
is using a swap interval of zero. In this case, a new frame will instantly
replace the previous queued frame. However, SurfaceFlinger doesn't know this
until the onFrameReplaced callback gets called - which is delayed by the fence
wait. If the timing is bad, SurfaceFlinger never gets a chance to pick up a new
frame to do the composition with.
We see this behaviour on our TC development system (slow GPU) with legacy
on-screen benchmarks. Such apps are using a swap interval of zero and sometimes
frames don't get updated for several seconds. This behaviour can be also seen
on a Nexus5, although it isn't as obvious as on our TC.
The fix in this cl is to move the EGL throttling to the end of the queueBuffers
function. This ensures that if a frame gets replaced in the queue, all
consumers who installed the callbacks, get called in a timely fashion.
Change-Id: I36e9ecda162150f41e97d4fb7437963a3d86b371
Signed-off-by: Christian Poetzsch <christian.potzsch@imgtec.com>
Bart Sears [Sat, 5 Dec 2015 03:42:14 +0000 (03:42 +0000)]
Merge "Revert "libbinder: Don\'t redefine B_PACK_CHARS""
am:
3f46d82d7b
* commit '
3f46d82d7b8ee0ae6295723311e5c986b649d727':
Revert "libbinder: Don't redefine B_PACK_CHARS"
Bart Sears [Sat, 5 Dec 2015 03:38:59 +0000 (03:38 +0000)]
Merge "Revert "libbinder: Don't redefine B_PACK_CHARS""
Bart Sears [Sat, 5 Dec 2015 03:38:34 +0000 (03:38 +0000)]
Revert "libbinder: Don't redefine B_PACK_CHARS"
This reverts commit
3eab1b40acbcc2e82b3e90da26706129019eccf9.
Change-Id: Iedc99961c6014029722f814e621970c1732f9092
Elliott Hughes [Sat, 5 Dec 2015 01:39:36 +0000 (17:39 -0800)]
Merge "Track rename from base/ to android-base/."
am:
f31cf3ab0f
* commit '
f31cf3ab0f53394200c0f22a4591db741ba1ccb0':
Track rename from base/ to android-base/.
Elliott Hughes [Sat, 5 Dec 2015 01:33:52 +0000 (01:33 +0000)]
Merge "Track rename from base/ to android-base/."
Christopher Wiley [Sat, 5 Dec 2015 00:20:15 +0000 (16:20 -0800)]
Merge "libbinder: Don\'t redefine B_PACK_CHARS"
am:
7166e7823c
* commit '
7166e7823cedb817f0278f569940aa86f424aa3b':
libbinder: Don't redefine B_PACK_CHARS
Christopher Wiley [Sat, 5 Dec 2015 00:15:12 +0000 (00:15 +0000)]
Merge "libbinder: Don't redefine B_PACK_CHARS"
Christopher Wiley [Fri, 4 Dec 2015 22:15:27 +0000 (14:15 -0800)]
libbinder: Don't redefine B_PACK_CHARS
This macro is defined in linux/binder.h which is *sometimes* included
before binder/IBinder.h depending on codepath.
Bug:
25868260
Test: Compiles
Change-Id: Ied1e6658936668c16cbdacab3a3205107c907655
Elliott Hughes [Fri, 4 Dec 2015 23:39:32 +0000 (15:39 -0800)]
Track rename from base/ to android-base/.
Change-Id: I62c18891ffe319d20ff084f5ca2aa0c038c23cd3
Christopher Wiley [Thu, 3 Dec 2015 21:21:21 +0000 (21:21 +0000)]
Merge "libbinder: Handle transaction failures correctly"
am:
823fa55aec
* commit '
823fa55aec55684605e6f4c764ec20901066553c':
libbinder: Handle transaction failures correctly
Christopher Wiley [Thu, 3 Dec 2015 21:17:42 +0000 (21:17 +0000)]
Merge "libbinder: Handle transaction failures correctly"
Christopher Wiley [Mon, 23 Nov 2015 00:29:58 +0000 (16:29 -0800)]
libbinder: Handle transaction failures correctly
Java code expects status_t != OK to be caught at the JNI level in
android_util_Binder.cpp (see signalExceptionForError). We were
incorrectly mapping this kind of failure to a special exception type
and writing that exception type to parcels.
Instead, refuse to write EX_TRANSACTION_FAILED to a parcel and return
the status value instead.
While here, remove non-trivial constructors to push authors toward the
more explicit factory methods. Remove getException() and push authors
toward using the simpler getter methods. Fix minor camelCase issues.
Bug:
25615695
Test: system/tools/aidl integration tests still pass
Change-Id: I7cad3ac8ae8300b5ac0b466606f4934d01e503c5