| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Following libs are explicitly marked as double_loadable since they are
one of the (indirect) dependencies of LLNDK libraries and at the same
time they themselves are marked as VNDK. Such lib can be double loaded
inside a vendor process.
* libgui and libbinder: due to indirect dependency from libmediandk via
libmediaomx. libmediandk is LLNDK)
* libui: due to dependency from libnativewindow, which is LLNDK.
Note: even without this change, the library is already capable of being
double loaded due to the dependency chain towards it. This change is to
make it explicit so that double loading of a library is carefully
tracked and signed-off by the owner of the lib.
Bug: 77155589
Test: m -j
Merged-In: Id4768162aeb72b71d63d7e4498980f276ef58e6b
Change-Id: Id4768162aeb72b71d63d7e4498980f276ef58e6b
(cherry picked from commit a75d3d6d9f2aef4b1855e6a58332859f3531143d)
|
|
These NDK docs weren't in any groups, so they don't show up in the
new site.
Test: https://irina-dot-devsite.googleplex.com/ndk/reference/group/native-activity
Test: https://irina-dot-devsite.googleplex.com/ndk/reference/group/networking
Test: https://irina-dot-devsite.googleplex.com/ndk/reference/group/tracing
Bug: http://b/77236573
Change-Id: Ic3154f77de1f13fe753b4c03537eab41b0a22848
|
|
Test: builds
Change-Id: I0008621f7c819744eae62db8ae51e3719ba84894
Merged-In: I0008621f7c819744eae62db8ae51e3719ba84894
|
|
If more than one pointer is present and identical coordinates are
encountered for any pointer, the lastResample value will not be
invalidated (by clearing its bits). As a result, the same lastResample
value will be used to subsequently rewriteMessage any time in the
future. This can cause significant jumps in the pointer coordinates, and
is a bug. To further clarify, this bug has only to do with resampling
and nothing to do with both pointers having the same X or Y value.
This CL makes this logic to be per-pointer. As soon as non-identical
value is encountered and the timing conditions are such that the
resampled value is not needed to be used, the lastResample bit for that
pointer will be cleared, meaning that the value in lastResample for this
pointer is stale. This value will no longer be used anywhere.
When performing resampling, allow the process to happen on a per-pointer
basis. If one of the pointers has encountered events with identical
coordinates, then use the previously resampled value (do not resample
again), if that value is still valid (see above). Otherwise, the normal
resampling path will be taken. On the other pointers that do not have
identical coordinates, go through the normal path as well.
Bug: 68840121
Test: recorded a repro event with inputstudio and replayed it while
observing the screen. Saw that the coordinates jump before the fix,
and do not jump with the fix.
Change-Id: If43c323759de8f0184b76221d1ae599a75349ce9
Merged-In: If43c323759de8f0184b76221d1ae599a75349ce9
|
|
InputPublisherAndConsumer_test currently segfaults due to
null pointer dereference. This is because 0 is passed instead
of a valid pointer to an allocated int32_t. Also, enforce strict
compiler warnings, include signed comparisons.
Bug: 62033391
Test: m -j libinput_tests_InputEvent_test
libinput_tests_InputChannel_test
libinput_tests_InputPublisherAndConsumer_test
libinput_tests_InputTransport_test libinput_tests_VelocityTracker_test
&& adb push out/target/product/taimen/data/nativetest/libinput_tests/
/data/nativetest/
then run the above tests individually, for example
/data/nativetest/libinput_tests # ./InputPublisherAndConsumer_test
and others in a similar manner
Change-Id: Ia030cdbfa22d2bf7bdf6274a337b059ca3f9a6c3
Merged-In: Ia030cdbfa22d2bf7bdf6274a337b059ca3f9a6c3
|
|
The implementation behind this cflag has already been removed.
Test: none
Change-Id: Ida61c64f86180c8a7d41b6af7036312ebe0dd3d4
|
|
|
|
This reverts commit ffb5375c37804f84750a2092779ced5b986850a3.
Clang 7.0 LTO shouldn't need arm instruction set,
if we revert one LLVM recent change.
Bug: 72619014
Change-Id: I513fc78b2fde7ee95cd1bf2153dd647d2fbd196e
Merged-In: Ic1ee146b823602ae419891f5e0956640109dd397
|
|
Test: builds
Change-Id: Idb7a51db4382e1012afe12df3c9fe807922b17fe
|
|
* Cast EXPECT_EQ arguments to the same signed/unsigned type.
Test: Build
Change-Id: I1055215045fb0904438d0a71bde034fab99111cc
|
|
* surfaceflinger.so uses LTO.
clang 7.0 LTO inlined more code and some references
are too far away to use thumb instructions.
Bug: 72619014
Test: build with Forrest
Merged-In: Ia09931e93a170ac58007f80da2d580bb9b9635a6
Change-Id: I2cbb652a9a8d908d9444a267e6d234b135c279b5
|
|
|
|
Prevent future cyclic dependency from libunwind changes.
Bug: 66919073
Test: make -j 50
Change-Id: I311d9a91f0efaf9806b46bf768ec2d9dc9065e3c
|
|
|
|
Discovered by the upcoming compiler update.
Test: m checkbuild
Change-Id: I8a4129904e25a28560d2481c4769146673fbc48d
|
|
Workaround for upstream Clang bug to fix Werror build.
Bug: 71356631
Test: Build
Change-Id: I4a3a2fa5aaee4fc2e981478c34fe7ac53eb8c928
|
|
Switches Layer::computeBounds to return a FloatRect instead of a Rect.
During the computation of the bounds, we apply the layer transformation
to its nominal dimensions, clip it against its bounds (which are either
its parents bounds or the screen bounds), and apply the inverse of the
layer transformation.
Previously, the intermediate position (after transformation/clip, but
before inverse transformation) was stored as Rect, which is to say that
it was truncated to integer coordinates. After applying the inverse
transformation, this loss of precision can cause glitches where a layer
that should be clipped against, e.g., the side of the screen no longer
creates a watertight seal against that side.
In order to fix this, we now store the intermediate value as a FloatRect
and propagate float precision back through computeBounds. The callers of
computeBounds tend to then immediately apply the transform again, at
which point it is safe to round back to integer.
Bug: 64070729
Bug: 66431327
Bug: 69935057
Test: Modified android.view.cts.SurfaceViewSyncTest#
testSurfaceViewBigScale no longer produces bogus display frames
Change-Id: If5987ca4ad76657f9670a5f59258f896180352e2
(cherry picked from commit 80d611613ef1a9c05a31cd3f47badfd4de9b1418)
Merged-In: If5987ca4ad76657f9670a5f59258f896180352e2
|
|
|
|
|
|
The integer overflow sanitizer is throwing unsigned integer overflow
errors in the list functions in ISensorService. This refactors the loops
to prevent the overflow on the last iteration.
Test: Compiles and device boots without sanitizer errors.
Bug: 30969751
Change-Id: I6a7993024fdc71702e8e4e8ae535cfaf999e9dab
Merged-In: I6a7993024fdc71702e8e4e8ae535cfaf999e9dab
|
|
Some devices don't have vndbinder yet.
Bug: 67975337
Test: media cas doesn't crash loop on bullhead
Change-Id: I381771ff971fadeae75a78b5d45d6d067adced44
|
|
|
|
|
|
A header file in private directory are not supposed to export with
llndk library. To avoid this, move the private directory to a new
include-private directory and export it for framework-only library.
Bug: 69235725
Test: build and boot any device.
Change-Id: I9f020411350cf5b5f1db01ef9add76d8c3b2c395
|
|
* EXPECT_EQ arguments must be both signed or both unsigned.
Merged-In: I18ed32862bbee207e2751139599eff4f6f6e8618
Test: normal build
Change-Id: I9e03a3457e09a23371227286eaf7744738721352
|
|
A recent change to the binder kernel driver delays the
delivery of BR_TRANSACTION_COMPLETE until a reply comes
in. The binderDriverInterfaceTest uses the non-blocking
interface, and assumes BR_TRANSACTION_COMPLETE comes in
on the first read. I don't think we depend on this
assumption in any of our real code, so change the test.
Bug: 69442864
Test: binderDriverInterfaceTest passes
Change-Id: Iafca061c1bb7badd6cb61b3e876d9c3bbfa18542
|
|
am: 7553b18b49
Change-Id: I60abdc7bbde7008a0b058495febd6f3d3bf75031
|
|
* Do not include <base/logging.h> to avoid redefinition warnings.
* Use unsigned 0U to avoid signed vs unsigned comparison.
* Fix typo of unused ret2.
Bug: 66996870
Test: build with WITH_TIDY=1
Change-Id: I5554ffaadd54d967580998672e1dd64428269566
|
|
Put AIDL files into a filegroup so they can be imported as sources
for framework.jar.
Bug: 69917341
Test: m checkbuild
Change-Id: I1aeb65b651f9097fa72212ca5cce9fb679b5a779
Merged-In: I1aeb65b651f9097fa72212ca5cce9fb679b5a779
(cherry picked from commit 21c210ec5ce370a2d73332be48372d16dd789a08)
|
|
Remove absent symbols - they were never exported. Explicitly add
private/ symbols to LIBNATIVEWINDOW_PLATFORM version.
Bug: http://b/69603741
Test: make && boot taimen
Change-Id: I6ff7c170a5ad5f0bcee5d65e27aef04310286a21
|
|
oc-mr1-dev-plus-aosp into stage-aosp-master.
Change-Id: I754fe8c1ec11f047e58694a2fdc8d9ab1b85cab5
|
|
* Suppress/fix warnings of unused variable, parameter,
private-field, and function.
* Fix warnings about typedef missing declarator.
* Use -Wno-error to keep existing warnings, to fix later.
Bug: 66996870
Test: build with WITH_TIDY=1
Change-Id: Iae96af3f6a17623befe73a9db2c81275fee44fa6
|
|
|
|
|
|
This makes the removal of HWC1 path in SurfaceFlinger possible.
Test: hikey960 still boots
Change-Id: I65a13db8c0abc45c16ee04718ddb104e437fafe7
|
|
|
|
This reverts commit 0494d6ab6f4b85eb28eeb71737398ba0b862a822.
Change-Id: I115dd7cd3e11bb573e17b3859133e0cad1afef49
|
|
* changes:
binder: send BC_REPLY and BC_FREE_BUFFER together
binder: remove unnecessary err check
|
|
In current BR_TRANSACTION handling logic, BC_REPLY and BC_FREE_BUFFER
are sent to kernel seperately which may introduce latency for freeing
buffer.
I think it's better to free buffer asap, this change fixes this
by putting BC_REPLY and BC_FREE_BUFFER in the same buffer. And then
call waitForResponse() to send them to kernel together.
After this, function sendReply() will not be called by anyone. So
sendReply() is also removed.
Below is the test result of "binderThroughputTest -w 100"
Env: android 7.1.2, 2G ram
---------------------
---> base:
iterations per sec: 31917.7
collecting results
average:2.83375ms worst:572.205ms best:0.054219ms
50%: 2.14844 90%: 5.27344 95%: 6.83594 99%: 12.3047
iterations per sec: 32142.3
collecting results
average:2.8186ms worst:550.884ms best:0.054948ms
50%: 2.53906 90%: 5.27344 95%: 6.83594 99%: 11.1328
iterations per sec: 31704.5
collecting results
average:2.85129ms worst:600.984ms best:0.053854ms
50%: 2.53906 90%: 5.27344 95%: 6.83594 99%: 12.3047
iterations per sec: 31932.3
collecting results
average:2.82549ms worst:563.098ms best:0.052396ms
50%: 2.53906 90%: 5.66406 95%: 6.83594 99%: 11.1328
---> patched:
iterations per sec: 32062.5
collecting results
average:2.85475ms worst:599.177ms best:0.054895ms
50%: 2.14844 90%: 5.27344 95%: 6.44531 99%: 12.6953
iterations per sec: 32241.7
collecting results
average:2.82851ms worst:634.984ms best:0.05651ms
50%: 2.53906 90%: 5.66406 95%: 6.83594 99%: 10.7422
iterations per sec: 31939
collecting results
average:2.86589ms worst:628.861ms best:0.058645ms
50%: 2.53906 90%: 5.66406 95%: 6.83594 99%: 11.1328
iterations per sec: 32149.4
collecting results
average:2.80785ms worst:658.889ms best:0.055573ms
50%: 2.53906 90%: 5.66406 95%: 6.83594 99%: 10.7422
It seems patched is better.
Change-Id: I0ef88864f4dbc63f3cdd2eba345b5dce6140b2d3
Suggested-by: Martijn Coenen <maco@google.com>
Signed-off-by: Ganesh Mahendran <opensource.ganesh@gmail.com>
|
|
* Fix/remove unused variables, parameters, functions.
* Add missing typedef name.
* Suppress warning of unused return value from release().
* Suppress warnings that are fixed in goog/master, to be exported to AOSP.
Bug: 66996870
Test: build with WITH_TIDY=1
Change-Id: If6b539db0a243b44fc6d51d90259f6564f306f6d
|
|
|
|
* Use const reference parameter type to avoid unnecessary copy.
* Use more efficient overloaded string methods.
* Use const reference type for loop index variables to avoid unnecessary copy.
(cherry picked from commit cb057c2e6bf23b43df49282c0db26b663e4535f7)
Bug: 30407689
Bug: 30411878
Bug: 30413223
Test: build with WITH_TIDY=1
Merged-In: I75ed62e2cb9939878b8932ed512ceb7e241edd6f
Change-Id: I75ed62e2cb9939878b8932ed512ceb7e241edd6f
|
|
|
|
because once the Android.bp finder is enabled, it will be found anyway,
and this will allow us to confirm separately that adding this Android.bp
is successful.
Bug: 64363847
Test: m -j
Change-Id: Id4593099cdb3d839054700cc198f6c808c3c6642
Merged-In: Id4593099cdb3d839054700cc198f6c808c3c6642
|
|
am: 3e90c7021a
Change-Id: I395a7229e5ba157c6472fdf718a9aef4743c351e
|
|
am: 451ff0d855
Change-Id: I1a462b24a9fddd57584a6d720c6e0f36932bfbad
|
|
|
|
3a566dd1e1
am: 14ac77cd1e
Change-Id: I529db9786afca459da51ec886f7e672743038bb9
|
|
am: c4ebf5bbab
Change-Id: Ia99830d3c2ffe68a05db21c56364f3fcd1965ac1
|
