| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When speaker angles are queried, the size of the array for the
returned data is 3x the number of channels (where really it
should be max(2, nbChannels)). The code assumed it was at
least 3x2 (where 2 is the number of virtual speakers this
effect supports) and would thus crash when called for a mono
channel mask.
Test: see repro steps in bug
Bug: 32591350
AOSP-Change-Id: I33d4bff6b2e19a9fc4284a85a446804878d3a410
CVE-2017-0545
Change-Id: Ie4480d9abcfafcd53fca15ab2fd8ef7ecb6fd48d
(cherry picked from commit e5a54485e08400a976092cd5b1c6d909d0e1a4ab)
|
|
|
|
|
|
|
|
|
| |
Bug: 32705438
Bug: 32703959
Test: cts security test
Change-Id: I8900c92fa55b56c4c2c9d721efdbabe6bfc8a4a4
(cherry picked from commit e275907e576601a3579747c3a842790bacf111e2)
(cherry picked from commit b0bcddb44d992e74140a3f5eedc7177977ea8e34)
|
|
|
|
|
|
|
| |
Bug: 31781965
Change-Id: I1c439a0d0f6aa0057b3c651499f28426e1e1f5e4
(cherry picked from commit 9a2732ba0a8d609ab040d2c1ddee28577ead9772)
(cherry picked from commit 557bd7bfe6c4895faee09e46fc9b5304a956c8b7)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug: 32448258
Bug: 32095626
Test: Use POC bug or cts security test
Change-Id: I69f24eac5866f8d9090fc4c0ebe58c2c297b63df
(cherry picked from commit 01183402d757f0c28bfd5e3b127b3809dfd67459)
(cherry picked from commit 321ea5257e37c8edb26e66fe4ee78cca4cd915fe)
Fix security vulnerability: Equalizer command might allow negative indexes
Bug: 32247948
Bug: 32438598
Bug: 32436341
Test: use POC on bug or cts security test
Change-Id: I91bd6aadb6c7410163e03101f365db767f4cd2a3
(cherry picked from commit 0872b65cff9129633471945431b9a5a28418049c)
(cherry picked from commit e981cca9fff3608af22bdf8fc1acef5470e25663)
(cherry picked from commit c66c43ad571ed2590dcd55a762c73c90d9744bac)
|
|
|
|
|
|
|
|
|
|
|
| |
CYNGNOS-3312
Bug: 30229821
Test: fixing CL. Existing unit tests still pass.
Change-Id: I6e4abd759d5d2abc3b391e92e2e18f060cab7af0
(cherry picked from commit 874f9e0b8eb0cbe508d15c8c03796c863851f21f)
(cherry picked from commit 244e7fd2a45b4e7d70d2c2e550181220371b7edf)
|
|\
| |
| |
| |
| |
| |
| |
| | |
CYNGNOS-3303
Android 6.0.1 release 74
Change-Id: I0a14578751f4ecb8d13def26b9ffe5dcba4afd72
|
| |
| |
| |
| |
| |
| |
| | |
Bug: 30229821
Change-Id: Idd3c1563dc9d3261e6e168e945005bf133ab2cdb
(cherry picked from commit 099ab280775946e7c36c73fde47f2ee5a2579f53)
(cherry picked from commit 46dc714d523a41a4f886eecbe5b9947a4c900510)
|
|\ \
| |/
| |
| |
| | |
Android 6.0.1 release 17
Ticket: CYNGNOS-1854
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Add checks on parameter size field in effect command handlers
to avoid overflow leading to invalid comparison with min allowed
size for command and reply buffers.
Bug: 26347509.
Change-Id: I20e6a9b6de8e5172b957caa1ac9410b9752efa4d
(cherry picked from commit ad1bd92a49d78df6bc6e75bee68c517c1326f3cf)
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
* In case a device has a /vendor partition which cannot be
modified, this allows us to short-circuit the audio_effects.conf
file which might be placed there with one of our own which
lives on /system.
Change-Id: Ief87bd4cfba2c3188b0dff122d91f773b7f3d92d
|
| |
| |
| |
| |
| |
| |
| | |
Warn allocation failures explicitly rather than crash
trying to access unallocated memory
Change-Id: Ie86c3ac130917e1f4030eb8207ac8350cba7711d
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Validate channel mask in downmixer configure, this returns
early warning to calling module which can fallback to another
downmixing module or return error instead of playing improper
data.
CRs-Fixed: 862105
Change-Id: Ic19c4913608468d5c5197f65bb5d6a640c9d2d74
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When switching from tunnel to arm effects, effect chain will be moved
from offload thread to mixer thread. During this procedure effect instance
is not destroyed, instead it's preserved and mounted onto new chain.
The migrant sequence is 1) disable instance -> 2) move to new effect
chain -> 3) enable instance again. However, the accumulated effect
enabled count is not decreased when effect instance is disabled.
As a result, LvmBundle_process() is only called once when Effect_process()
called twice, which causes input buffer get processed alternately and
glitch sound appears.
Change-Id: Idd3b8753a73694cc36617ed55b40fda93155abed
CRs-Fixed: 636353
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Use boolean property ro.audio.ignore_effects to load or ignore the
audio effects.
Bug 21906334
Change-Id: I778f8b10a1caf25f7679705b1c83775223df011d
|
|/
|
|
|
|
|
|
|
|
|
| |
Check consistency of effect command reply sizes before
copying to reply address.
Also add null pointer check on reply size.
Also remove unused parameter warning.
Bug: 21953516.
Change-Id: I4cf00c12eaed696af28f3b7613f7e36f47a160c4
|
|
|
|
|
|
|
|
|
| |
Fixed automatic level compensation for bundled effects. Fixed algorithm
now works even in corner cases previously reported.
Cross band coefficients and cross interaction of BassBoost fixed.
bug: 18769691
Change-Id: I51570836bdbff37e7c4b26f2b4282e4ded7fa56d
|
|\
| |
| |
| |
| |
| |
| | |
compensation" into lmp-mr1-dev
* commit 'd11d1be13a6391290fb1dbc572292973dae59909':
Fix for Bundled Effects level compensation
|
| |\
| | |
| | |
| | |
| | |
| | |
| | | |
into lmp-mr1-dev
* commit '7ebc167072be27f42b7c58bd7f9378c8311533ab':
Fix for Bundled Effects level compensation
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
From the bundled effects, only the EQ had some sort of level compensation
based on the settings of the effect. This patch takes into account all
the status and levels of EQ, BassBoost and Virtualizer bundled effects
and makes an informed decision on the overall level compensation.
bug: 18213779
Change-Id: I23c7b94fe7da5482f0bbef0cbe715ea464631420
|
|\ \ \
| |/ /
| | |
| | |
| | |
| | |
| | | |
TARGET_SIMULATOR checks"
* commit 'e10615e7f8749de887c8d25579398ad9e4094077':
Remove obsolete TARGET_SIMULATOR checks
|
| |\ \
| | | |
| | | |
| | | |
| | | | |
* commit '6b3e1a6d59b304a2dde7f03ba184feac0dec4ee7':
Remove obsolete TARGET_SIMULATOR checks
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I9d499190e49fed16aac686ae5da79ab870605ff4
|
|\ \ \ \
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | | |
virtualization mode" into lmp-mr1-dev
* commit '59ca7038364cd6fa3e7e1b52e0e90fa0960af038':
Fix for AOSP force virtualization mode
|
| |\ \ \
| | |/ /
| |/| /
| | |/
| | |
| | |
| | | |
into lmp-mr1-dev
* commit '4df79c2c13f6163f7d99360a7fb34bf3bd2a5115':
Fix for AOSP force virtualization mode
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Forcing virtualization mode with AOSP code might erroneously return
a failed code, whilst it was correct. Fixing the condition for
Automatic mode.
bug: 18538537
Change-Id: I5837ff40cfed04c99889813cd56ebe3d90793e74
|
|\ \ \
| |/ /
| | |
| | |
| | | |
* commit '575174d62b665900203d70a4d410329b8918c2af':
Move libldnhncr to libc++.
|
| | |
| | |
| | |
| | |
| | | |
Bug: 15193147
Change-Id: If59c4a53a497f501881d26d58b80b11371906527
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
uuids need to be unique, and things don't work properly when they're
not.
Also fix/enhance/extend the dumpEffectDescriptor() method, and
include a list of effects in audioflinger dumpsys.
Change-Id: I3dfbc5ed0f7272c7809e337f2929212ece047ee4
|
|\ \ \
| |/ /
| | |
| | |
| | | |
* commit 'fed1594d707b4623015e538b2f69faf76bc38d5b':
Clean up some makefiles.
|
| |\ \
| | |/
| |/|
| | |
| | | |
* commit '9916dcb6ef041752fda5f13192ee6a36e51e73a5':
Clean up some makefiles.
|
| | |
| | |
| | |
| | |
| | |
| | | |
The build system takes care of these things for you.
Change-Id: Ife9605f15786e11d6a3578f56ed83dd9ae7fd799
|
| | |
| | |
| | |
| | |
| | | |
Change-Id: I88106ba25f9b3f8fd7616307ab6b7fbe7afb6e91
(cherry picked from commit bfd3b9764841971be2d2f04a3fdbd41a2144beea)
|
| | |
| | |
| | |
| | | |
Change-Id: I19252ea8d7dd8b9eee2532059c5a3a98d53c3b28
|
|/ /
| |
| |
| | |
Change-Id: I88106ba25f9b3f8fd7616307ab6b7fbe7afb6e91
|
| |
| |
| |
| |
| |
| |
| | |
No unused parameter warning in EffectBundle.
Remove unused variable declaration in Virtualizer_getParameter
Change-Id: Ibe1c051bc21ac53a6c770b4aa28e31ac5d559d9b
|
| |
| |
| |
| | |
Change-Id: I210129f5742b046f7ceef48194f039352eff596d
|
| |
| |
| |
| |
| | |
Change-Id: Ia658ab4b6320d19fdb50f123c930918724ff0ef3
Signed-off-by: Andy Hung <hunga@google.com>
|
|\ \
| |/
| |
| | |
Change-Id: Ic4e75541193f733f477cfa4a2a87cce4c9e64f45
|
| |
| |
| |
| |
| |
| | |
and use same suffixes
Change-Id: I90b74ede171ba2550db4a220cfd1ad2e3caefe2d
|
| |
| |
| |
| |
| |
| | |
and corresponding downmix effect
Change-Id: I1576aa373ca7acbb84d7742bacbd9c2da04a7a4c
|
|\ \
| |/
| |
| |
| | |
* commit 'e1ff1051ffee8fb650741ad133f0f28b73eb7a73':
libeffects: 64 bit compile issues
|
| |
| |
| |
| | |
Change-Id: I32dd13d307572fee91150f8e113df78924ca0067
|
|\ \
| |/
| |
| |
| | |
* commit '3a46b975070b503bc71b53b70f9a8f5bd03bba1c':
LP64 fixes for media/libeffects
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Changes include:
[x] In get parameter series of functions, replaced size_t*
formal parameter type with uint32_t* where actual parameter
passed was uint32_t*.
[x] In set parameter series of functions, changed size_t
formal parameter to uint32_t where actual parameter was
uint32_t.
[x] Changed the definition of LVM_UINT32 from unsigned
long to uint32_t as unsigned long is 64-bit in LP64.
[x] Used other stdint.h types for other LVM_types for
consistency.
[x] Use of uintptr_t for the pNextMember of the INST_ALLOC
structure, rather than LVM_UINT32, for portablility.
[x] Use of uintptr_t where pointers are used in arithmetic.
[x] Replaced the use of 0xFFFFFFFC with ~3 in places where
it was used to clear last two bits.
[x] Removed int casts where cmdSize and *replySize, both
uint32_t, were being compared with sizeof().
Change-Id: Ibec0b4d8e9b855f44b1cd853be6df84d13cf4186
Signed-off-by: Marcus Oakland <marcus.oakland@arm.com>
Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
|
|\ \ \
| |/ /
| | |
| | |
| | |
| | |
| | | |
capture command"
* commit 'e276aef8779a7763b318ebd3e8c09f35ff37a95c':
Fix preventing from heap overwrite in capture command
|
| |\ \ |
|
| | |/
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
There was heap usage error in case of stop of audio framework while
capturePoint was negative. Pointer to reply data was moved but
final silence write was done using original buffer size.
Now silence set is done at the beginning under the condition
that framework has stopped.
Change-Id: I7dab1e922f1813e5fbfd4a64c8b0d15d9465520c
|
|/ /
| |
| |
| |
| | |
bug:13225538
Change-Id: I4fccc414923f7e62cd46d691c67cb44b9692c225
|