PerformanceMetricsState.getForHierarchy API is a bit weird [233421985]

Fixed

Bug

Status Update

No update yet.

Description

py...@gmail.com

created issue #1

May 21, 2022 10:05PM

PerformanceMetricsState.getForHierarchy returns a MetricsStateHolder which is just a wrapper around PerformanceMetricsState?.

As I started using it, I did this:

// Named it "state" because I'm calling "get" on something a class which as a name ending in State.
val state = PerformanceMetricsState.getForHierarchy(view)
state.state?.let { state ->
 state.addState("foo", "bar")
}

So.. I'm getting a state, then getting a state from it or maybe not, and if I do have a state then I can add a state to it!

The other thing that's a bit weird about this is how PerformanceMetricsState is a separate class but it's directly coupled to JankState, i.e. MetricsStateHolder.state will always be set if I already called JankState.createAndTrack() for that window, but it'll be null if I didn't. This conditional nullability isn't super clear.

Suggestions:

Consider hiding MetricsStateHolder and having getForHierarchy directly return PerformanceMetricsState?
Consider a name changes so that we're not adding "state" (addState) to an object already named "state" (PerformanceMetricsState).
Consider either breaking the coupling entirely or making the coupling more explicit. Breaking the coupling would mean deleting the PerformanceMetricsState.create() function that JankStat is calling, and having MetricsStateHolder.getForHierarchy() automatically create it on first call. Making the coupling more explicit might mean making a PerformanceMetricsState instance directly available on a JankState instance.

Comments

py...@gmail.com <py...@gmail.com> #2May 21, 2022 10:15PM

A couple of questions:

1. Have you saw crash in real device or only in simulators?
2. Do you use dynamic feature for language ID?

ch...@google.com <ch...@google.com> #3Jul 11, 2022 10:15PM

Tested on Android 12 Emulator with custom executor, but cannot repro this issue.

ch...@google.com <ch...@google.com> #4Jul 11, 2022 10:56PM

Second crash in the description is from a real device. Experienced it myself on two different Xiaomi phones, plus lots of crashes from users in the Google Play console.
Dynamic features are not used in the application.

As a wild guess, I have downgraded build tools from 31.0.0 to 30.0.3, compileSdk from 31 to 30, and moved all work with Language ID to the service in a separate process (just to be sure that crash can kill secondary process instead of main). This combination is in beta for 2 days by now and I don't see any SIGSEGV crashes.

ch...@google.com <ch...@google.com> #5Jul 11, 2022 11:47PM

Hmm, I feel the crash might be something related to separate/secondary process.

I also changed compileSdk and targetSDK to 31 but still cannot repro this issue.

ap...@google.com <ap...@google.com> #6Jul 13, 2022 08:24PM

On the contrary, there was no separate process before, when crashes started.

In the new build (with the aforementioned changes) I can see SIGSEGV crash, but only one instead of dozens and it has a bit different backtrace:

signal 11 (SIGSEGV), code 1 (SEGV_MAPERR)
liblanguage_id_jni.so (offset 0x11e000)

backtrace:
  #00  pc 000000000003c7c0  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 000000000003b960  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 000000000003bb48  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 000000000003bafc  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 0000000000036c98  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 0000000000032714  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 0000000000031cac  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/split_config.arm64_v8a.apk!lib/arm64-v8a/liblanguage_id_jni.so (offset 0x11e000)
  #00  pc 0000000000057438  /data/app/azagroup.reedy-mF7zTu2bv_ELlbFArwNgqA==/oat/arm64/base.odex (offset 0x57000)

ch...@google.com <ch...@google.com> Jul 13, 2022 09:21PM

Marked as fixed.

py...@gmail.com <py...@gmail.com> #7Jul 14, 2022 08:44AM

FYI, ML Kit launched a new language ID SDK in the latest release, which uses a new language ID model. https://developers.google.com/ml-kit/release-notes#august_11_2021

Could you try the new SDK version(17.0.0) to check if you can still repro this native crash? Thanks!

ch...@google.com <ch...@google.com> #8Jul 14, 2022 03:59PM

Thank you, I'll try it and check.

Message last modified on Jul 14, 2022 04:00PM