Set big core to group_cfs_skip_mask for sys group when screen is off,
so that sys group tasks will run less in big core.
Bug: 328210236
Test: suspend/resume test
Change-Id: I4868ba0303e981250105b1ff69a6db6458df844a
This reverts commit 2cc7bc7c3d.
Reason for revert: may get geekbench5/6 regressed
Bug: 338140604
Bug: 338141118
Change-Id: I76b2b98ca233fccc1f9a15c99bbb622bd7869392
Add the "UclampMax_EfficientBase" value to the ADPF config in the
powerhint.json to specify efficient core scheduling in the static
case, and "UclampMax_EfficientOffset" for the dynamic case, for this
device.
Bug: 277285195
Test: manual
Change-Id: If377f6384a1a7863abf80f0a4a0cf0fe3074ec21
This reverts previous JankCheckTimeFactor from 1.43 to 1.2.
Since the current reported duration doesn't include the GPU
part, the JankCheckTimeFactor needs to be smaller for matching
jank frame detection.
Bug: 204204746
Test: adb shell dumpsys android.hardware.power.Ipower/default;
check adpf sessions' traces.
Change-Id: I36ab5e699187650011a91a2c0814c33e62e4e125
Also reset the mask for launch and other camera streaming modes that
do not use it.
Bug: 248621289
Test: build pass
Change-Id: I752bcf27ac6c9064a464af54200a506fd6db03ff
1298MHz is no longer in the list of available frequencies of the big
core. This CL replaces this value with the next available one which is
1396MHz.
Bug: 304536786
Test: device boot
Change-Id: I2f58c18c41494e0a0ea4b006f641f55405b72cfd
Update it to align with the HWUI target duration scale factor,
so the adjusted jank check time duration is same as expected frame
timeline's duration.
HWUI target duration scale factor is 70%, and its inverse value is
approximately 1.43.
Bug: 204204746
Test: adb shell dumpsys android.hardware.power.IPower/default;
Check the device adpf sessions' traces.
Change-Id: Id835b30e7acd50b34840131669eaf3894ea5e629
This reverts commit 8116ee60ba.
Reason for revert: need more perf evaluation
Bug: 328210236
Test: build pass
Change-Id: Id389c7b78c8c168a3015cd207b6d7b1a00c5d3a2
Set cpuset system group to little and mid cores only when screen
off.
Bug: 328210236
Test: build pass
Change-Id: I3c116998e3bb5f347b8a0d401b7a22c458b19657
since exynos driver take a longer time during access MEM/INT/DSU/BCI file nodes, change the order of powerhint and do other boost first
Bug: 329209796
Change-Id: I63a709d0607f5999cd30c8fe12de0699934b4c53
Signed-off-by: Chungkai Mei <chungkai@google.com>
In auto prefer idle mode, a task in top-app will be granted prefer
idle if it satisfied the conditions. So turn auto prefer idle off
if top-app prefer_idle is explictly disabled. Also turn auto prefer
off when other group enables prefer_idle explictly.
Bug: 236868627
Test: build pass
Change-Id: I6e5694dfd9b5ed30bfad31509415027d5bac71cb
We already have the CPU scheduler optimization for game mode to sustain
the power and perf, so keep PMU_POLL being disabled to avoid the interfence on CPU scheduler.
Bug: 329697728
Test: run genshin
Change-Id: Ifb6838be9767d67cc1cf6a8109a33c5bfd402727
This CL enables CDPreferIdle for both 30 and 60fps use-cases. It also
limits the CDCpuset to 0-6 to avoid using the BigCore. For 60fps
use-case, it additionally bumps up the MIF frequency to 1.539GHz which
can help reduce the power regression caused by enabling CDPreferIdle
(limiting it to around 10mW regression in CPU power).
Bug: 325510979
Test: https://buganizer.corp.google.com/issues/325510979#comment23
Change-Id: Ib5312882f5de80a95c6061f636d7f1f9642b6299
Bug: 328685643
Test: Power test show ~20mW CPU power increase. but cpu profiling show CPU operation frequency are lower than before, which will limit the impact when system level CPU throttle start.
- Project11 power hint reuse "streaming_standard" with extra MIF target tuning
Change-Id: If62e93032f9f4d69caa2f74e8a4cf4b2cf68be86
We already have the CPU scheduler optimization for game mode to sustain
the power and perf, so remove the CPU thermal policy at lower
temperature to avoid the interfence on CPU scheduler.
Bug: 329697728
Test: run genshin
Change-Id: Ib0d47b3de7b3550a3af4db37ef062f8c5d2ea262
