Oculus 2 apk does not work correctly

viartsnet · November 10, 2021, 1:49pm

Hello everyone!
Has anyone managed to run the cloudXR app on the latest firmware (v34) in oculus quest 2 and steamVR version 1.21.1?

My application often crashes and is not usable.
The cloudXR server is running locally on windows 10, gtx 1650 video card, using Wi-Fi 5Gz. Even if you do not run any application, but simply is in steamVR, the application crashes after a while, also periodically steamVR itself crashes with a critical error.

There is a suspicion that the ovr-sample.apk application has not been updated for a long time and conflicts with the new firmware quest 2 (((

viartsnet · November 11, 2021, 1:38pm

UPD: If you specify the -ird parameter on the client, then everything works. Without it, the application crashes and the server also crashes.
Need help from developers, what could be the problem?

alex112 · November 14, 2021, 8:42pm

hey there-- I’m not getting crashes like you are, but I am getting a black screen on the client. adding -ird to CloudXRLaunchOptions.txt doesn’t seem to have any effect on my end. would love to solve this!

viartsnet · November 15, 2021, 2:08pm

with the -ird parameter, the application runs for 10-12 minutes and then goes into a deep black color ((
The case is repeated from time to time.

turn on the cloudXR server
run the cloudXR client
a second or two black screen, then the SteamVR environment is loaded
I run the game in steamVR for example Half Life Alyx through the menu in HMD
after loading the game a minute later - two stably for some time the screen ripples with different colors (like a chessboard, there are only many different colors). Then it lets go and the game streams very well without re-rippling.
After 9-10 minutes, the client goes to the black screen and does not leave it anymore, and the sound from the game is broadcast without interruption, clean. At the same time, SteamVR starts displaying 39-47ms of 72fps rendering instead of the ~ 13ms standard. Once the client is disconnected, the render time goes back to ~ 13ms in the SteamVR window.

An error is observed in Android Studio debugging:

2021-11-15 16:40:17.776 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=72/72,Prd=34ms,Tear=0,Early=0,Stale=0,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,1478/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=1353MHz,Free=2422MB,PLS=0,Temp=39.7C/0.0C,TW=4.33ms,App=2.48ms,GD=0.20ms,CPU&GPU=8.13ms,LCnt=1,GPU%=0.48,CPU%=0.66(W0.86),DSF=1.00
2021-11-15 16:40:17.920 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:17.920 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.096 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:40:18.099 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.100 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.104 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.104 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.109 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.109 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.113 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.113 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.543 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:40:18.547 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.547 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:40:18.550 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.550 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:40:18.553 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.553 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:40:18.557 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.557 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.751 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:40:18.751 30823-30860/com.nvidia.cloudxr.ovr I/chatty: uid=10095(com.nvidia.cloudxr.ovr) main identical 1 line
2021-11-15 16:40:18.751 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:40:18.755 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.755 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.758 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.758 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.761 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.761 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.766 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.766 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: SetTextureSwapChainSamplerState: No change in sampler state.
2021-11-15 16:40:18.775 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=63/72,Prd=34ms,Tear=0,Early=0,Stale=44,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,1478/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=2092MHz,Free=2421MB,PLS=0,Temp=39.7C/0.0C,TW=4.24ms,App=2.01ms,GD=0.20ms,CPU&GPU=10.68ms,LCnt=1,GPU%=0.49,CPU%=0.66(W0.75),DSF=1.00
2021-11-15 16:40:18.787 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.787 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.791 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.791 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.794 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.794 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)
2021-11-15 16:40:18.798 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:40:18.798 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (10560 KB); Flags: (0x100600)

…

2021-11-15 16:41:08.382 30823-30873/com.nvidia.cloudxr.ovr I/Telemetry: App memory usage: PSS=429MB DalvikPSS=4 MB
2021-11-15 16:41:08.384 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=1/72,Prd=35ms,Tear=0,Early=0,Stale=72,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,1478/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=1353MHz,Free=2424MB,PLS=0,Temp=39.5C/0.0C,TW=1.97ms,App=1.03ms,GD=0.00ms,CPU&GPU=678.06ms,LCnt=1,GPU%=0.15,CPU%=0.69(W1.00),DSF=1.00
2021-11-15 16:41:08.506 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:08.508 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:08.778 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=1/72,Prd=39ms,Tear=0,Early=0,Stale=72,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,1478/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=1353MHz,Free=2424MB,PLS=0,Temp=39.5C/0.0C,TW=1.95ms,App=0.00ms,GD=0.00ms,CPU&GPU=710.06ms,LCnt=1,GPU%=0.14,CPU%=0.77(W0.82),DSF=1.00
2021-11-15 16:41:08.875 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:08.878 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:09.273 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:09.293 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:09.613 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:09.657 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:09.779 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=1/72,Prd=39ms,Tear=0,Early=0,Stale=72,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,710/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=451MHz,Free=2425MB,PLS=0,Temp=39.5C/0.0C,TW=1.95ms,App=0.00ms,GD=0.00ms,CPU&GPU=710.06ms,LCnt=1,GPU%=0.14,CPU%=0.77(W0.82),DSF=1.00
2021-11-15 16:41:09.989 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:10.014 30823-30860/com.nvidia.cloudxr.ovr W/main: LatchFrame failed, frame not ready for 2000 ms
2021-11-15 16:41:10.019 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.019 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.023 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.023 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.027 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.027 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.027 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:10.036 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.036 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.041 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.041 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.044 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.044 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.048 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.048 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.051 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.051 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.055 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.055 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.058 30823-30860/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0300945 (IOCTL_KGSL_GPUOBJ_ALLOC) failed: errno 12 Out of memory
2021-11-15 16:41:10.058 30823-30860/com.nvidia.cloudxr.ovr E/Adreno-GSL: <gsl_memory_alloc_pure:2875>: ERROR: kgsl_sharedmem_alloc() failed! Allocation size: (9472 KB); Flags: (0x100600)
2021-11-15 16:41:10.058 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: UpdateEvents: HMT was UNmounted
2021-11-15 16:41:10.059 30823-30860/com.nvidia.cloudxr.ovr W/OsSdk: Requested historical tracking state too far in the past; oldest: 25443.638594s, requested: 25441.898256s
2021-11-15 16:41:10.359 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:10.361 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:10.740 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:10.743 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:10.751 30823-30860/com.nvidia.cloudxr.ovr D/AAudio: AAudioStream_requestStop(s#1) called
2021-11-15 16:41:10.751 30823-30860/com.nvidia.cloudxr.ovr D/AudioTrack: stop(260): called with 33817924 frames delivered
2021-11-15 16:41:10.819 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=1/72,Prd=40ms,Tear=0,Early=0,Stale=72,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,710/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=200MHz,Free=2516MB,PLS=0,Temp=39.5C/0.0C,TW=1.87ms,App=0.68ms,GD=0.00ms,CPU&GPU=741.59ms,LCnt=1,GPU%=0.00,CPU%=0.00(W0.00),DSF=1.00
2021-11-15 16:41:11.113 30823-30923/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:11.116 30823-30924/com.nvidia.cloudxr.ovr W/Adreno-GSL: <gsl_ldd_control:553>: ioctl fd 69 code 0xc0200948 (IOCTL_KGSL_GPUOBJ_IMPORT) failed: errno 12 Out of memory
2021-11-15 16:41:11.242 30823-30860/com.nvidia.cloudxr.ovr D/AAudio: AAudioStream_close(s#1) called ---------------
2021-11-15 16:41:11.246 30823-30860/com.nvidia.cloudxr.ovr D/AAudio: AAudioStream_close(s#1) returned 0 ---------
2021-11-15 16:41:11.587 30823-30928/com.nvidia.cloudxr.ovr D/SurfaceUtils: disconnecting from surface 0xc31b2808, reason disconnectFromSurface
2021-11-15 16:41:11.606 30823-30927/com.nvidia.cloudxr.ovr D/SurfaceUtils: disconnecting from surface 0xda183008, reason disconnectFromSurface
2021-11-15 16:41:11.614 30823-30860/com.nvidia.cloudxr.ovr I/VrApi: ---------- vrapi_LeaveVrMode [start] ----------
2021-11-15 16:41:11.617 30823-30860/com.nvidia.cloudxr.ovr D/VrApi: targetSDKVersion 29
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: DEVICE MODEL NUMBER = Quest
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: DEVICE BUILD NAME = user-20169900217500000
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: DEVICE BUILD TYPE = user
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: DEVICE OS VERSION = 10
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: API LOADER VERSION = 1.37.0
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: RUNTIME API VERSION = 1.51.0
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: RUNTIME VERSION = 34.0.0.91.366 Oct 18 2021 20:31:23 RELEASE
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: APP NAME = CloudXR Client
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: APP VERSION = 1.0 versionCode 1 internalVersionName
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: APP VR TYPE = vr_only
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: APP MODE TYPE =
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: APP PACKAGE NAME = com.nvidia.cloudxr.ovr
2021-11-15 16:41:11.619 30823-30860/com.nvidia.cloudxr.ovr I/VrRuntimeClient: APP ACTIVITY CLASS = com.nvidia.cloudxr.ovr.MainActivity
2021-11-15 16:41:11.620 30823-30860/com.nvidia.cloudxr.ovr D/BatteryReceiver: Unregistering battery receiver
2021-11-15 16:41:11.620 30823-30860/com.nvidia.cloudxr.ovr I/Sensor: OSSDKTRACKING Shutdown tracking
2021-11-15 16:41:11.623 30823-30873/com.nvidia.cloudxr.ovr I/VrRuntimeDeviceConfigClient: GetBoolean: oculus_mobile_disable_non_hands_app_hands_usage 0
2021-11-15 16:41:11.625 30823-30873/com.nvidia.cloudxr.ovr I/VrApi: FPS=1/72,Prd=40ms,Tear=0,Early=0,Stale=72,VSnc=1,Lat=0,Fov=0D,CPU4/GPU=4/0,710/305MHz,OC=FF,TA=0/0/0,SP=N/N/N,Mem=2092MHz,Free=2536MB,PLS=0,Temp=39.5C/0.0C,TW=1.87ms,App=0.68ms,GD=0.00ms,CPU&GPU=741.59ms,LCnt=1,GPU%=0.00,CPU%=0.00(W0.00),DSF=1.00
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr I/VrApi: OVR::Stats thread stopped
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr I/InputDeviceManager: ovrInputDeviceManager::~ovrInputDeviceManager
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr I/OSHandDataQuerier: OSSDKTRACKING shutting down HandsTracker
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr I/OSHandDataQuerier: OSSDKTRACKING shutting down HandsTracker
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr I/InputDevice_TrackedRemote: OSSDKTRACKING shutting down Haptics
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr I/InputDevice_TrackedRemote: OSSDKTRACKING shutting down Haptics
2021-11-15 16:41:11.627 30823-30860/com.nvidia.cloudxr.ovr D/OsSdk: TrackingServiceConnection dtor
2021-11-15 16:41:11.628 30823-30860/com.nvidia.cloudxr.ovr D/OsSdk: TrackingServiceConnection unregistered
2021-11-15 16:41:11.647 30823-30871/com.nvidia.cloudxr.ovr I/ClientInputThread_Internal: InputThreadRun ShutdownRequest received
2021-11-15 16:41:11.647 30823-30860/com.nvidia.cloudxr.ovr I/InputHooks_Internal: Disabling Native Input Hooks
2021-11-15 16:41:11.648 30823-30860/com.nvidia.cloudxr.ovr I/ClientInputTrackedObjectManager: TRACKEDKEYBOARD ClearTrackedObjectManager called on client
2021-11-15 16:41:11.650 30823-30876/com.nvidia.cloudxr.ovr I/FenceChecker: FenceChecker::MainLoop - ended
2021-11-15 16:41:11.651 30823-30860/com.nvidia.cloudxr.ovr I/VrApi: ---------- vrapi_LeaveVrMode [end] ----------
2021-11-15 16:41:11.653 30823-30823/com.nvidia.cloudxr.ovr I/Choreographer: Skipped 189 frames! The application may be doing too much work on its main thread.

LatchFrame set the waiting time to 2000 ms, tried both 500 (standard) and 277, this does not affect the result. Stable after 10-12 minutes black screen and increased rendering times in SteamVR.

viartsnet · November 16, 2021, 12:24pm

UPD:
If, after the black screen started to appear (after 10-12 minutes), exit the cloudXR client application and try to reconnect, the black screen will still be displayed, sometimes a picture in poor resolution flickers. On the server, frame processing is increased by ~ 2.5 times.
But if you restart the helmet and start the client again, then you can play again for 10-12 minutes. The server continues to render the picture ~ 13 ms.

The above procedure is described with the cloudXR server always running (i.e. when we restart the helmet, the server continues to work).

There is an assumption that the cloudXR client for quest 2 has a memory leak somewhere, which ends in 10-12 and breaks the logic of work.

Nvidia please help!

viartsnet · November 19, 2021, 2:07pm

UPD:

After a lot of testing and debugging, I am inclined to believe that the problem is related to the video card ((
My home video card is gtx 1650 and it does not support HEVC b-frame, perhaps because of this, the client does not work for me without the -ird parameter.

I tried to run the application in the cloud with a Tesla T4 video card and the client worked. Without additional parameters and the same with the -ird parameter, and so it works. Of course, due to the remoteness of the cloud, lags are very noticeable and completely unsuitable for use. Therefore, it seems that you need to change the video card ((

Can anyone test the app on rtx 3060 - rtx 3090? you need to make sure that it will work on these cards too.

viartsnet · November 22, 2021, 11:20am

UPD:

I managed to try how the client would behave in the cloud with an rtx 3090 video card on board.

With the -ird parameter, the client also goes into a stupor after 7-8 minutes (most likely due to either the fact that there is a memory leak, or because this is an experimental parameter, without the source code of the cloudXR library it is difficult to say what this parameter is responsible for and How does it work). Those. this parameter does not depend on the video card (on gtx 1650, tesla t4 and rtx 3090, the result is the same). I have attached a piece of video showing what happens after 7-8 minutes, in the client I corrected the background color from black to gray, so that you can clearly see where the picture is being rendered and how it disappears and lags. at this moment OVR Metrics shows 2 FPS, processor load 80-100%, processor load level 4.

Well, good news, without the -ird parameter on the same client on the same rtx 3090 video card, the application worked for 1 hour, I did not test further, there was no time, but we can assume that the client will continue to work.

Therefore, we can conclude: you need to use rtx 3000 series video cards without the -ird parameter on the client side.

But I don’t understand why on gtx 1650 it is not possible to start the client without the -ird parameter, is it really because of the lack of HEVC b-frame support? in google search people said that this is a bug of nvidia drivers, the video card itself supports HEVC, but the programs cannot explicitly determine that it is supported. I am using the most recent video driver, but it didn’t help ((
Nvidia, @tegradave , @GJones-NVIDIA-XR-Team could you please provide some clarification?

P.S. is there a way to set the refresh rate to more than 72Hz? if you specify more in the client, it does not affect the server. ((
Is there any information when the next release of cloudXR is planned?

tegradave · November 23, 2021, 11:35pm

Release 3.1 came out yesterday, and did not have a fix for the memory leak. It’s actually due to steamvr thrashing the swap resolution, sometimes changing resolution 5-10x / s. There’s a bug in the way the sample code handles things. We’re going to look at whether we can post a fixed file (or details on how to fix).

And yes, looks like the original GTX 1650 didn’t have HEVC B-frame support in the encoder.

The -ird decoder is still in experimental state. The hope is it will eventually replace the default decoder and offer much better performance, but there is much work left to do and significant testing to put it through before it might become the default.

I should also note the original bug described here was not reproducible with release 3.1, so would ask you retry with the new release and see if it improves on the initial failure.

viartsnet · November 24, 2021, 10:49am

@tegradave I launched the client on version 3.1 and the result is the same, after 7-9 minutes of the session fps and red texture drop (I changed the color in the code on purpose to make it clearer). I started it the same way with the -ird parameter. without it, my video card does not seem to be able to work ((

I am attaching video and logs. Perhaps this will somehow help to identify the cause of the failures.

CloudXR Server - SteamVR Log 2021-11-24 13.19.13.txt (998.9 KB)
Streamer Server Log 2021-11-24 13.19.13.txt (232.6 KB)
CloudXR Client Log 2021-11-24 13.19.44.txt (610.3 KB)
Streamer Client Log 2021-11-24 13.19.44.txt (257.1 KB)
androidStudioLog.txt (1.0 MB)

tegradave · November 24, 2021, 5:02pm

Server log indicates constant errors and some serious resolution thrashing, the latter being the root cause of this class of issue with our samples and one we’re trying to improve in next release (noting that Alyx more than any app causes this).

One weird thing is both client and server streamer logs are showing ‘perfect’ QoS – which is unusual to say the least. Especially since the client is logging a lot of network dropped frames.

I’ve asked internally about hevc v h264, and issues related to h264 (since I have never tested it myself, always on hevc).

viartsnet · November 24, 2021, 6:27pm

@tegradave there is an assumption that this is not related due to the type of codec, since I wrote above that I tested it on the rtx 3090 video card with the -ird parameter and the result is the same as with gtx 1650, although HEVC is used. here the matter is apparently in the ird handler itself. Without this parameter, everything works well on the 3090, without these freezing problems.

tegradave · November 24, 2021, 6:45pm

Yes, the experimental decoder has more rough edges, doesn’t handle this issue at all, and some related issues and corner cases likely just aren’t handled yet. Again, hope to have it improved by next release.

Note that the current ovr sample does have similar ‘over time’ issues with Alyx, just much longer to show.