All target devices use pre-R audio.bluetooth_qti.default.so and
pre-R audio hal, in which the new set_event_callback API is not
implemented and the audio service will crash due to mismatched struct.
jjpprrrr:
The audio flinger also warns about event callback when playing
audio through a USB audio adapter with DAC. Let's go back to v5
to avoid calling these unimplemented features at all.
...
D AudioFlinger: Failed to add event callback
...
Change-Id: Ia330ac05d84d3dfb3a639458d66a3dc29689e67c
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
Return a positive integer for perf lock acquire and release so
that Goodix/FPC fingerprint sensor blobs will not complain.
Goodix:
E [goodixHAL][gf_hal_milan_f_series]: goodix_perf_lock_acquire: Failed to acquire perf lock, err: 0
E [goodixHAL][gf_hal_milan_f_series]: goodix_perf_lock_release: Perf lock release error 0
FPC:
E fpc_tac : fpc_perf_lock_acquire: Incorrect params, Failed to acquire perf lock, err
E fpc_tac : fpc_perf_lock_release: Perf lock release error 0
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
Power hints tuned for our devices by Albert I.
Thanks-to: Albert I <kras@raphielgang.org>
Signed-off-by: Ivan Vecera <ivan@cera.cz>
Change-Id: I580a527b8621252b44e5a8bfcb906b8a98d2156b
from LA.QSSI.12.0.r1-05100-qssi.0
sdm710-lenovo: also drop WFD and update remaining libs
Note that ims factory jars are added for the new ImsRcsService.apk.
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
Basically any non-plain-text files in our proprietary file list
are ELF binaries. Granted that I can use llvm-readobj to check the
headers and build them through cc_prebuilt_binary or
cc_prebuilt_library_shared one by one, but the result is kind of
pointless.
The purpose of cc_prebuilt_binary is to track dependencies by
running check_elf_file.py on them. We have so many proprietary blobs
that it is nearly impossible (and sometimes impractical and undesirable)
to define dependencies.
Blurs on android12 are extremely janky, even with EXPENSIVE_RENDERING
raising GPU frequency.
W FrameTracker: Missed SF frame:1
W FrameTracker: Missed SF frame:4
W FrameTracker: Missed SF frame:68
W FrameTracker: Missed SF frame:68
W FrameTracker: Missed SF frame:65
...
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
Those commands have moved to init.rc.
Bug: 170507876
Bug: 171740453
Test: build pass
Change-Id: Ib7acc0fd83adca37e247cd9f3113c1f117e5196e
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
Quoting commit afc09932f6fa07c5f6beffbcdc2bb4691e527321 in
system/sepolicy [1]:
"Devices launching with R or eariler can this by setting following
under BoardConfig.mk"
It is impractical to replace every prop definition from qcom to
vendor_property_type for old platforms like SDM845, so relax the
enforcement to avoid triggering millions of neverallows.
ref:
[1] afc09932f6
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
The pinner now asks the runtime for these files.
Test: adb shell dumpsys pinner
Bug: 144773394
Change-Id: Ie633ea788dab3389a3407ec9a461b5837befca46
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
This change replaces the '/system/lib64/libsurfaceflinger.so' pin, as
the file was removed in ag/12524602. The updated pin relies on
go/aog/1552085, as system_service needs read access to SurfaceFlinger.
Bug: 176197656
Test: adb shell dumpsys pinner (coral)
- shows that /system/bin/surfaceflinger is successfully pinned
adb logcat | grep PinnerService (coral)
- no longer shows a file-not-found error in PinnerService
Change-Id: Iae2a96c5fe4b1f5698d21de9970140321d11c460
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
(nuke whole vendor configs here)
* Without CodeAurora's modifications in frameworks/av, the vendor suffixed
media codecs configs are not parsed at all. Move select vendor additions
into the main files in order to enumerate support for hardware-backed
HEVC encoder.
/dev/cpu_dma_latency accepts values in either s32 or hex format,
so convert 67 (dec) to 43 (hex).
sdm845-common comment:
I assume Google meant to set cpu dma latency to 44 in decimal
rather than hex because 44 is the lowest valid latency value for
C3 LPM state, according to qcom,latency-us in sdm845-pm.dtsi.
The pm qos request is only active when the fd is kept open, so
the dummy value of 100 does not really matter. Keep it as is.
Test: hexdump -C /dev/cpu_dma_latency
00000000 2c 00 00 00 |,...|
00000004
Test: kernel pr_err() in pm_qos_power_write() before
pm_qos_update_request()
Inspired by Artem's commit "qos: Workaround for faulty libperfmgr
config".
Signed-off-by: Chenyang Zhong <zhongcy95@gmail.com>
Enable 64 bit compilation for DRM plugins.
This will generate DRM plugins, HALs & oemcrypto_test in 64 bit.
Change-Id: I451db39edea3446d9816fcaba77caa8d3b097afb
