Bug [IN-43583] Possible memory corruption/overwrite with High Quality Line Rendering

I've been having an issue with builds of my project hard crashing at random points (2023.1b14, and happening in 2023.1b20). It would happen seconds or minutes into running the build at seemingly random points.

From debugging the stack traces look to be at different origins. However, all of them have in common that it happens somewhere in either allocation or deallocation of memory. From this pattern I assume this is due to memory corruption from overwrite and/or double deallocation.

Here are some typical examples:

Code (CSharp):

1. 0x00007FFE0C33F222 (UnityPlayer) block_remove
2. 0x00007FFE0C33FA3A (UnityPlayer) tlsf_free
3. 0x00007FFE0B4378C3 (UnityPlayer) DynamicHeapAllocator::Deallocate
4. 0x00007FFE0B43EC9B (UnityPlayer) DualThreadAllocator<DynamicHeapAllocator>::TryDeallocate
5. 0x00007FFE0B43B360 (UnityPlayer) MemoryManager::Deallocate
6. 0x00007FFE0B6BB13D (UnityPlayer) profiler_start_new_frame
7. 0x00007FFE0B6A366E (UnityPlayer) `InitPlayerLoopCallbacks'::`2'::InitializationProfilerStartFrameRegistrator::Forward
8. 0x00007FFE0B69AAD7 (UnityPlayer) ExecutePlayerLoop
9. 0x00007FFE0B69AC84 (UnityPlayer) ExecutePlayerLoop
10. 0x00007FFE0B69B034 (UnityPlayer) PlayerLoop
11. 0x00007FFE0B8E36BB (UnityPlayer) PerformMainLoop
12. 0x00007FFE0B8E5A1B (UnityPlayer) MainMessageLoop
13. 0x00007FFE0B8E8D80 (UnityPlayer) UnityMainImpl
14. 0x00007FFE0B8E8F8B (UnityPlayer) UnityMain

Code (CSharp):

1.  
2. 0x00007FFEBD6A7D32 (UnityPlayer) block_remove
3. 0x00007FFEBD6A854A (UnityPlayer) tlsf_free
4. 0x00007FFEBC7A7983 (UnityPlayer) DynamicHeapAllocator::Deallocate
5. 0x00007FFEBC7ACEE2 (UnityPlayer) DelayedPointerDeletionManager::CleanupPendingMainThreadPointersInternal
6. 0x00007FFEBC7AE8A9 (UnityPlayer) DualThreadAllocator<DynamicHeapAllocator>::Allocate
7. 0x00007FFEBC7AAE88 (UnityPlayer) MemoryManager::Allocate
8. 0x00007FFEBCADA630 (UnityPlayer) dynamic_array_detail::dynamic_array_data::resize_buffer
9. 0x00007FFEBC6D6BBC (UnityPlayer) dynamic_array<double,0>::resize_buffer_nocheck
10. 0x00007FFEBC9453EF (UnityPlayer) RenderingCommandBuffer::PPtrResolver<Material>::Resolve
11. 0x00007FFEBC93EAA8 (UnityPlayer) RenderingCommandBuffer::PrepareState
12. 0x00007FFEBC999445 (UnityPlayer) ScriptableRenderContext::ExecuteScriptableRenderLoop
13. 0x00007FFEA2F3BCE2 (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\Unity.RenderPipelines.HighDefinition.Runtime__2.cpp:70710] HDRenderPipeline_Render_mD5A5DFADAEB2A8DB36E4261A8154AB6206D56557
14. 0x00007FFEA32056CD (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\UnityEngine.CoreModule__4.cpp:27502] RenderPipelineManager_DoRenderLoop_Internal_mB646C8738F4A9859101F3BE94809E2E10BBDB1FB
15. 0x00007FFE9FE70D16 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b14\Editor\Data\il2cpp\libil2cpp\vm\Runtime.cpp:638] il2cpp::vm::Runtime::InvokeWithThrow
16. 0x00007FFE9FE707B9 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b14\Editor\Data\il2cpp\libil2cpp\vm\Runtime.cpp:623] il2cpp::vm::Runtime::Invoke
17. 0x00007FFEBC99B1CD (UnityPlayer) ScriptableRenderContext::ExtractAndExecuteRenderPipeline
18. 0x00007FFEBC83A4FC (UnityPlayer) RenderManager::RenderCamerasWithScriptableRenderLoop
19. 0x00007FFEBC83AC5C (UnityPlayer) RenderManager::RenderCameras
20. 0x00007FFEBCA0F45F (UnityPlayer) PlayerRender
21. 0x00007FFEBCA06C37 (UnityPlayer) ExecutePlayerLoop
22. 0x00007FFEBCA06DE4 (UnityPlayer) ExecutePlayerLoop
23. 0x00007FFEBCA07194 (UnityPlayer) PlayerLoop
24. 0x00007FFEBCC4E86B (UnityPlayer) PerformMainLoop
25. 0x00007FFEBCC50BCB (UnityPlayer) MainMessageLoop
26. 0x00007FFEBCC53F30 (UnityPlayer) UnityMainImpl
27. 0x00007FFEBCC5413B (UnityPlayer) UnityMain
28.  

Code (CSharp):

1.  
2. 0x00007FFE0C33F222 (UnityPlayer) block_remove
3. 0x00007FFE0C33FA3A (UnityPlayer) tlsf_free
4. 0x00007FFE0B4378C3 (UnityPlayer) DynamicHeapAllocator::Deallocate
5. 0x00007FFE0B43EC9B (UnityPlayer) DualThreadAllocator<DynamicHeapAllocator>::TryDeallocate
6. 0x00007FFE0B43B360 (UnityPlayer) MemoryManager::Deallocate
7. 0x00007FFE0B4380D4 (UnityPlayer) operator delete
8. 0x00007FFE0B6ED36F (UnityPlayer) std::_Tree_val<std::_Tree_simple_types<HWND__ * __ptr64> >::_Erase_tree<std::allocator<std::_Tree_node<HWND__ * __ptr64,void * __ptr64> > >
9. 0x00007FFE0B6EA70F (UnityPlayer) ContainerClear<std::map<DeprecatedFastPropertyNameSerialization,float,std::less<DeprecatedFastPropertyNameSerialization>,std::allocator<std::pair<DeprecatedFastPropertyNameSerialization const ,float> > > >
10. 0x00007FFE0B6DC028 (UnityPlayer) UnityPropertySheet::operator=
11. 0x00007FFE0B6DF777 (UnityPlayer) Material::CopyPropertiesFromMaterial
12. 0x00007FFE0B25FDC4 (UnityPlayer) Material_CUSTOM_CopyPropertiesFromMaterial_Injected
13. 0x00007FFE001B6049 (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\Unity.DemoTeam.Hair.Runtime.cpp:16394] HairInstance_UpdateRendererState_mDF3D80D342772090F4792EBA1923809AFDDF3EA9
14. 0x00007FFE001B6A69 (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\Unity.DemoTeam.Hair.Runtime.cpp:16193] HairInstance_UpdateRenderingState_m960FD4D03064646506BBED4A8CED51AA88A46A23
15. 0x00007FFE001B30E4 (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\Unity.DemoTeam.Hair.Runtime.cpp:14428] HairInstance_HandlePrerequisiteCompleted_m0CE6D3A6CA3FF2827B6961860CA7A58C7B92D6EB
16. 0x00007FFE001B422D (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\Unity.DemoTeam.Hair.Runtime.cpp:14480] HairInstance_LateUpdate_mD2B573892A690C28007C6C6858C0E1D174E855E1
17.  

To debug this, I first began to deconstruct my scene, which was not very complex (just one animated character). Turning off the hair got rid of the crashes.

I investigated more, and turning off High Quality Line Rendering either on the hair or via HDRP settings also removed the crashes.

Further testing revealed that it was the composition mode the Line Rendering was set to.

Setting the HQ Line Rendering Composition mode to "After Temporal Antialiasing" in the HDRP settings is causing the issue (I had been using it since a custom pass was messing with the depth buffer creating artifacts on the hair. Turning off custom passes did not get rid of the crashes, however.)
(edit: see posts down below)

Here is some more information that might be useful:

• Happens on both il2cpp and mono
• Happens with incremental garbage collection and also without
• Happens on Direct3D_11, Direct3D_12, but seemingly not Vulcan (which ran significantly slower)
• Does not happen in editor (both edit and play mode). It only happens on a build
• It didn't seem to happen when building development builds. With release builds it would happen quickly (30 sec - 2 min).
• Tested on Windows 10, Intel i9-12900KF, NVIDIA 3090 with latest drivers
• Tested Unity versions: 2023.1b14 and 2023.1b20

I've ran the build with the '-debugallocator' cmd line argument, but I couldn't see any difference in the stack traces.


I'm not sure if I can quickly throw together and test a stripped down project for reproducing as I'm currently very busy, but I will try should it be necessary.

 

The crashes still seem to happen on other composition modes, though less frequently. I will try some things to see if it really is caused by line rendering, or if that just makes it easier for it to trigger.

EDIT:
It seems like the crashes are happening on all composition modes, but they happen a lot more frequently on "After Temporal Antialiasing". I have gotten the crash on all modes, also while turning custom pass off.

It doesn't seem to happen when turning off High Quality Line Renderer, still.

The crashes do not happen in a development build. I had it running for 30+ minutes, while without development build it crashes within 30 sec.

 

So... it doesn't crash in development build.

But it does crash when I run the development build with the "-debugallocator" argument.

Code (CSharp):

1. ========== OUTPUTTING STACK TRACE ==================
2. 0x00007FFE99189A4A (UnityPlayer) profiling::ProfilerManager::DisposeProfilerRecorder
3. 0x00007FFE987FCE0B (UnityPlayer) ProfilerRecorder_CUSTOM_Control_Injected
4. 0x00007FFE380BFD84 (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\UnityEngine.CoreModule.cpp:8952] ProfilerRecorder_Control_m523BF56A25C70F4A89D3D1FF79B2596E8F3A7873
5. 0x00007FFE380C020B (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\UnityEngine.CoreModule.cpp:9103] ProfilerRecorder_Dispose_m6154715CD36B2CDE7B9DB390E0FEC335D75C96EB
6. 0x00007FFE38208D0D (GameAssembly) [\Library\Bee\artifacts\WinPlayerBuildProgram\il2cppOutput\cpp\UnityEngine.CoreModule__3.cpp:32020] Recorder_Finalize_mBF7E97C65CF9B2082B073AC844783635EF623330
7. 0x00007FFE38FC8D26 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\libil2cpp\vm\Runtime.cpp:638] il2cpp::vm::Runtime::InvokeWithThrow
8. 0x00007FFE38FC87C9 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\libil2cpp\vm\Runtime.cpp:623] il2cpp::vm::Runtime::Invoke
9. 0x00007FFE38FFAAED (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\libil2cpp\gc\GarbageCollector.cpp:200] il2cpp::gc::GarbageCollector::RunFinalizer
10. 0x00007FFE3904FA08 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\external\bdwgc\finalize.c:1318] GC_invoke_finalizers
11. 0x00007FFE38FF98F0 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\libil2cpp\gc\GarbageCollector.cpp:121] il2cpp::gc::FinalizerThread
12. 0x00007FFE3903DA69 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\libil2cpp\os\Thread.cpp:217] il2cpp::os::Thread::RunWrapper
13. 0x00007FFE390286B3 (GameAssembly) [C:\Program Files\Unity\Hub\Editor\2023.1.0b20\Editor\Data\il2cpp\libil2cpp\os\Win32\ThreadImpl.cpp:29] il2cpp::os::ThreadStartWrapper
14. 0x00007FFF66107614 (KERNEL32) BaseThreadInitThunk
15. 0x00007FFF666A26A1 (ntdll) RtlUserThreadStart
16. ========== END OF STACKTRACE ===========

 

Probably best to do a bug report , else chances are small to get this fixed. I already have submitted 5 critical bug reports for 2023.1, one of these is also about high quality lines. It seems to take a few weeks to month to get them fixed.

 

I managed to reproduce the bug in a simple project with just one hair asset. It takes a bit longer for the bug to happen (probably due to less memory access), but it does happen after ~15 minutes for me.

I sent in the project with my bug report. I hope it gets addressed soon, as it pretty much makes the high quality hair unusable.

Incident number: IN-43583

 

Could you please post here in case QA rejects the report (which often is the case, as they might not be able to do a build etc.) in that case I would also try to reproduce and submit a case, as this feature is very important for us.

 

Did you hear back from QA about the issue?

 

Nothing yet, unfortunately. Only the initial email saying that they received the report. it's marked as "open" on atlassian.

Did you get a similar issue?

 

The bug has been rejected initially, since the reviewer wasn't able to reproduce it. I expected as much since the build would sometimes only crash after 20 minutes, and judging from the emails between "under review" and the answer, only 30 minutes passed, so subtracting download/setup time, the build might not have been running long enough on their side.

I've reopened it with an improved version, in which I just copied the hair a couple of times, which as I expected did the trick.Non-development builds from this project will crash in under a minute for me.

I uploaded the project here: https://drive.google.com/file/d/18G8XO2o65B_cTzsr2CrRdVASdxzZ-owM/view?usp=sharing (there is also an already built version in there for reference)

so anyone interested can try and reproduce the issue. I used Unity 2023.1.0b20 for this.

Let me know if you can reproduce the issue, since this is pretty critical for anyone using the high quality lines.

 

Hey, I am testing the build right now. So far the build is running for like 10 minutes without a crash on a NVIDIA 3080 (not latest driver, this I'll try next). I guess the build inside is a non dev build?

 

stopped it after running for nearly 30 minutes, no crash. I saw a slight increase in GPU memory, but this is usually not a reliable indicator (was like 200MB increase during the 30 minutes of running).

 

after updating drivers to latest, still running fine (did a 10 minutes run in fullscreen, around 24 fps, full GPU usage, around 2.2GB of GPU mem used, not really increasing over time).
Maybe it's something with your GPU? I had issues with my first 3080, it crashed after running something full load for a while, after exchanging issue was gone.

 

last thing I tried was do a new build, this also running fine since a few minutes, no crash....

BTW, where did you get the hair asset from? I am always on the search for new hair assets, difficult to find good ones.

 

I will try to update my drivers and try again. It would be good if it was just that. I shouldn't have ruled out the possibility of drivers being the issue. I guess I did because it was only happen on builds and not in editor, though I realize now that memory management for the hardware might be different as well.

I had the same issue as you with the hair assets. I eventually learned making hair in blender.

 

Unfortunately, updating the drivers didn't help. Still getting crashes within like 30 sec.

Code (CSharp):

1. ========== OUTPUTTING STACK TRACE ==================
2.  
3. 0x00007FFD40BD2965 (UnityPlayer) block_remove
4. 0x00007FFD40BD312A (UnityPlayer) tlsf_free
5. 0x00007FFD3FCBC193 (UnityPlayer) DynamicHeapAllocator::Deallocate
6. 0x00007FFD3FCC356B (UnityPlayer) DualThreadAllocator<DynamicHeapAllocator>::TryDeallocate
7. 0x00007FFD3FCBFC30 (UnityPlayer) MemoryManager::Deallocate
8. 0x00007FFD3FF40E5D (UnityPlayer) profiler_start_new_frame
9. 0x00007FFD3FF28C2E (UnityPlayer) `InitPlayerLoopCallbacks'::`2'::InitializationProfilerStartFrameRegistrator::Forward
10. 0x00007FFD3FF1EEDB (UnityPlayer) ExecutePlayerLoop
11. 0x00007FFD3FF1F078 (UnityPlayer) ExecutePlayerLoop
12. 0x00007FFD3FF1F413 (UnityPlayer) PlayerLoop
13. 0x00007FFD401721BB (UnityPlayer) PerformMainLoop
14. 0x00007FFD4017451B (UnityPlayer) MainMessageLoop
15. 0x00007FFD40177950 (UnityPlayer) UnityMainImpl
16. 0x00007FFD40177B6B (UnityPlayer) UnityMain
17. 0x00007FF647CE11F2 (TestHairHDRP) __scrt_common_main_seh
18. 0x00007FFE014B7614 (KERNEL32) BaseThreadInitThunk
19. 0x00007FFE02A426F1 (ntdll) RtlUserThreadStart
20.  
21. ========== END OF STACKTRACE ===========
22.  

 

I just tested and it also crashed on a completely different second machine (with ATI instead of intel, and a 2080 instead of a 3090).

Did you run the build that was already in the build folder @Qleenie , or did you build it yourself. I wonder if there is something with my unity install that maybe creates builds with this bug. I'll install and try building the project from the other machine.

 

I ran with the build unchanged in Build folder for like 30 minutes, then did a new build, which I again ran for like 10 minutes.
So it cannot be your Unity installation.

 

I'll test your build on a few different machines with different GPU / CPU / Windows versions and see if any is giving same issue, which might help in narrowing down the reason.

 

I tried to run it on the other machine again for like 10 minutes. Just when I thought the first crash was a fluke, it did crash again, in the same spot, though with a slightly different stack trace.

Not sure how to proceed from here. I have two machines that reproduce the crash.

One more question if you don't mind. Are you running windows 10 or 11, since I'm still on 10. (grasping to straws right now)

 

I tested with windows 10, next test will be on laptop with windows 11 in a second.

 

OK, it's crashing on the Laptop (Win 11, 2080 super GPU), stacktrace:

Code (CSharp):

1. ========== OUTPUTTING STACK TRACE ==================
2.  
3. 0x00007FFDC4F96C72 (UnityPlayer) PrepareComputeBufferParameters<OverridingParameterPreparer>
4. 0x00007FFDC4F914A0 (UnityPlayer) GpuProgramParameters::PrepareOverridingValues
5. 0x00007FFDC519BB7B (UnityPlayer) GfxDeviceD3D11Base::SetShaderPropertiesCopied
6. 0x00007FFDC5962A03 (UnityPlayer) GfxDeviceWorker::RunCommand
7. 0x00007FFDC59613CB (UnityPlayer) GfxDeviceWorker::RunExt
8. 0x00007FFDC5961317 (UnityPlayer) GfxDeviceWorker::RunGfxDeviceWorker
9. 0x00007FFDC4F49877 (UnityPlayer) Thread::RunThreadWrapper
10. 0x00007FFE488626AD (KERNEL32) BaseThreadInitThunk
11. 0x00007FFE49C0A9F8 (ntdll) RtlUserThreadStart
12.  
13. ========== END OF STACKTRACE ===========

 

And another run gave the following stacktrace:

Code (CSharp):

1. An abnormal situation has occurred: the PlayerLoop internal function has been called recursively. Please contact Customer Support with a sample project so that we can reproduce the problem and troubleshoot it.
2. An abnormal situation has occurred: the PlayerLoop internal function has been called recursively. Please contact Customer Support with a sample project so that we can reproduce the problem and troubleshoot it.
3. GfxDeviceD3D11Base::WaitForLastPresentationAndGetTimestamp() was called multiple times in a row without calling GfxDeviceD3D11Base::PresentFrame(). This may result in a deadlock.

 

This was the one I got when I first ran it on the 2080 machine. At the second time I got the allocator one.

 

One thing to mention is that it takes a lot longer for it to crash on the 2080 machine (a few minutes), while on my 3090 it happens almost instantly (within 30 sec)

 

on my 3080 it did not crash after 30 minutes, and on the 2080 it crashed two times after 1 minute

 

A bug like this is the worst kind of bug. I hope we can get someone from the Unity team to seriously take a look at it. Shipping HQLines in 2023.1 with this bug would be pretty bad.

 

agreed. can you add all the stacktraces to your internal ticket? you can also mention that someone else was able to reproduce, also link me for questions if needed.

 

Will do. I added a link to this thread as well. Thank you for your help.

 

it's on our best interest to get this resolved, too. there's another issue that this feature does not work in VR, but a fix seems to be i review for 2023.2 alpha at least and will hopefully be down ported.

 

on a third run i got the exact stack trace as you:

Code (CSharp):

1. ========== OUTPUTTING STACK TRACE ==================
2.  
3. 0x00007FFDC5B02965 (UnityPlayer) block_remove
4. 0x00007FFDC5B0312A (UnityPlayer) tlsf_free
5. 0x00007FFDC4BEC193 (UnityPlayer) DynamicHeapAllocator::Deallocate
6. 0x00007FFDC4BF356B (UnityPlayer) DualThreadAllocator<DynamicHeapAllocator>::TryDeallocate
7. 0x00007FFDC4BEFC30 (UnityPlayer) MemoryManager::Deallocate
8. 0x00007FFDC4E70E5D (UnityPlayer) profiler_start_new_frame
9. 0x00007FFDC4E58C2E (UnityPlayer) `InitPlayerLoopCallbacks'::`2'::InitializationProfilerStartFrameRegistrator::Forward
10. 0x00007FFDC4E4EEDB (UnityPlayer) ExecutePlayerLoop
11. 0x00007FFDC4E4F078 (UnityPlayer) ExecutePlayerLoop
12. 0x00007FFDC4E4F413 (UnityPlayer) PlayerLoop
13. 0x00007FFDC50A21BB (UnityPlayer) PerformMainLoop
14. 0x00007FFDC50A451B (UnityPlayer) MainMessageLoop
15. 0x00007FFDC50A7950 (UnityPlayer) UnityMainImpl
16. 0x00007FFDC50A7B6B (UnityPlayer) UnityMain
17. 0x00007FF7862511F2 (TestHairHDRP) __scrt_common_main_seh
18. 0x00007FFE488626AD (KERNEL32) BaseThreadInitThunk
19. 0x00007FFE49C0A9F8 (ntdll) RtlUserThreadStart
20.  
21. ========== END OF STACKTRACE ===========

 

@schema_unity I now also got a first crash on my main PC, it took around 10 minutes. The crash log looks very different this time, so this could be very interesting for QA, will put stack trace at bottom of this post!

I did test on two more PCs (one Desktop with 2080 GPU with i7 CPU, one with AMD GPU + AMP CPU). Both have same crash as post before. So far I tested on 4 machines, 3 did crash within one minute, one runs much longer before crash:

1) GPU 2080 + i7 CPU + Win 10 + RAM 16GB = Crash < 1 minute
2) GPU 2080 super mobile + i7 CPU + Win 11 + RAM 16GB = Crash < 1 minute
3) GPU RX 6700 XT + CPU Ryzen 7 + Win 10 + RAM 32GB = Crash < 1 minute
4) GPU 3080 + CPU Ryzen 7 + Win 10 + RAM 64GB = Crash >10 minutes, one time it ran for 30+ minutes

Here is the stack trace of 4):

Code (CSharp):

1. ========== OUTPUTTING STACK TRACE ==================
2.  
3. 0x00007FFE50192A22 (UnityPlayer) block_remove
4. 0x00007FFE501931B4 (UnityPlayer) tlsf_free
5. 0x00007FFE4F27C263 (UnityPlayer) DynamicHeapAllocator::Deallocate
6. 0x00007FFE4F2819B2 (UnityPlayer) DelayedPointerDeletionManager::CleanupPendingMainThreadPointersInternal
7. 0x00007FFE4F283379 (UnityPlayer) DualThreadAllocator<DynamicHeapAllocator>::Allocate
8. 0x00007FFE4F27F708 (UnityPlayer) MemoryManager::Allocate
9. 0x00007FFE4F5B4D05 (UnityPlayer) dynamic_array_detail::dynamic_array_data::reserve
10. 0x00007FFE4F0A0A5F (UnityPlayer) Material_CUSTOM_GetShaderKeywords_Injected
11. 0x000001D37F530D6A (Mono JIT Code) (wrapper managed-to-native) UnityEngine.Material:GetShaderKeywords_Injected (intptr)
12. 0x000001D37F530C93 (Mono JIT Code) UnityEngine.Material:GetShaderKeywords ()
13. 0x000001D37F530BDB (Mono JIT Code) UnityEngine.Material:get_shaderKeywords ()
14. 0x000001D37F52E4F3 (Mono JIT Code) UnityEngine.Rendering.LineRendering/BindRendererToComputeKernel:.ctor (UnityEngine.Rendering.CommandBuffer,UnityEngine.Rendering.LineRendering/RendererData)
15. 0x000001D37F528163 (Mono JIT Code) UnityEngine.Rendering.LineRendering:ExecuteGeometryPass (UnityEngine.Rendering.CommandBuffer,UnityEngine.Rendering.LineRendering/GeometryPassData)
16. 0x000001D37F5261FB (Mono JIT Code) UnityEngine.Rendering.LineRendering/<>c__DisplayClass12_0:<DrawInternal>b__4 (UnityEngine.Rendering.LineRendering/GeometryPassData,UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraphContext)
17. 0x000001D3641E460B (Mono JIT Code) UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraphPass`1<PassData_REF>:Execute (UnityEngine.Experimental.Rendering.RenderGraphModule.InternalRenderGraphContext)
18. 0x000001D3641E3339 (Mono JIT Code) UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraph:ExecuteCompiledPass (UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraph/CompiledPassInfo&,int)
19. 0x000001D3641E311B (Mono JIT Code) UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraph:ExecuteRenderGraph ()
20. 0x000001D3641DA95B (Mono JIT Code) UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraph:Execute ()
21. 0x000001D3641DA80B (Mono JIT Code) UnityEngine.Experimental.Rendering.RenderGraphModule.RenderGraphExecution:Dispose ()
22. 0x000001D3585A4A4B (Mono JIT Code) UnityEngine.Rendering.HighDefinition.HDRenderPipeline:ExecuteWithRenderGraph (UnityEngine.Rendering.HighDefinition.HDRenderPipeline/RenderRequest,UnityEngine.Rendering.HighDefinition.AOVRequestData,System.Collections.Generic.List`1<UnityEngine.Rendering.RTHandle>,System.Collections.Generic.List`1<UnityEngine.Rendering.RTHandle>,UnityEngine.Rendering.ScriptableRenderContext,UnityEngine.Rendering.CommandBuffer)
23. 0x000001D354F41813 (Mono JIT Code) UnityEngine.Rendering.HighDefinition.HDRenderPipeline:ExecuteRenderRequest (UnityEngine.Rendering.HighDefinition.HDRenderPipeline/RenderRequest,UnityEngine.Rendering.ScriptableRenderContext,UnityEngine.Rendering.CommandBuffer,UnityEngine.Rendering.HighDefinition.AOVRequestData)
24. 0x000001D354B5F0D3 (Mono JIT Code) UnityEngine.Rendering.HighDefinition.HDRenderPipeline:Render (UnityEngine.Rendering.ScriptableRenderContext,System.Collections.Generic.List`1<UnityEngine.Camera>)
25. 0x000001D354B5D044 (Mono JIT Code) UnityEngine.Rendering.RenderPipeline:InternalRender (UnityEngine.Rendering.ScriptableRenderContext,System.Collections.Generic.List`1<UnityEngine.Camera>)
26. 0x000001D32EE1E213 (Mono JIT Code) UnityEngine.Rendering.RenderPipelineManager:DoRenderLoop_Internal (UnityEngine.Rendering.RenderPipelineAsset,intptr,UnityEngine.Object)
27. 0x000001D32EE1E493 (Mono JIT Code) (wrapper runtime-invoke) <Module>:runtime_invoke_void_object_intptr_object (object,intptr,intptr,intptr)
28. 0x00007FFE4E86D6A4 (mono-2.0-bdwgc) mono_jit_set_domain
29. 0x00007FFE4E7B85C4 (mono-2.0-bdwgc) mono_object_get_virtual_method
30. 0x00007FFE4E7B86D5 (mono-2.0-bdwgc) mono_runtime_invoke
31. 0x00007FFE4F472C8C (UnityPlayer) ScriptableRenderContext::ExtractAndExecuteRenderPipeline
32. 0x00007FFE4F30F20C (UnityPlayer) RenderManager::RenderCamerasWithScriptableRenderLoop
33. 0x00007FFE4F30F91C (UnityPlayer) RenderManager::RenderCameras
34. 0x00007FFE4F4E85DF (UnityPlayer) PlayerRender
35. 0x00007FFE4F4DEC1B (UnityPlayer) ExecutePlayerLoop
36. 0x00007FFE4F4DEDB8 (UnityPlayer) ExecutePlayerLoop
37. 0x00007FFE4F4DF153 (UnityPlayer) PlayerLoop
38. 0x00007FFE4F731B6B (UnityPlayer) PerformMainLoop
39. 0x00007FFE4F733ECB (UnityPlayer) MainMessageLoop
40. 0x00007FFE4F7372F0 (UnityPlayer) UnityMainImpl
41. 0x00007FFE4F73750B (UnityPlayer) UnityMain
42. 0x00007FF7E28211F2 (TestHairHDRP) __scrt_common_main_seh
43. 0x00007FFF2E977614 (KERNEL32) BaseThreadInitThunk
44. 0x00007FFF2F1C26F1 (ntdll) RtlUserThreadStart
45.  
46. ========== END OF STACKTRACE ===========

 

Thank you. I'll add that to the report

 

Any updates to this issue? If they don’t have enough info or still reject, I’d try to build a separate project and submit same bug again.

 

Yes, just tonight the bug was processed again and good news. It has been confirmed. Internal ID: UUM-41319

 

That's good news! I'd estimate a fixing time of 4 month, based on my former experience.

 

Oof, I can imagine they have a lot on their plate. As long as it is fixed in the official 2023.1 release I can't really complain. I'm still a few weeks from producing some first prototype test releases for my project, and for that I was planning to use dev builds anyway.

 

Is this the ticket?
https://issuetracker.unity3d.com/is...ommand-arg-release-when-it-runs-out-of-memory

Seems that QA found a pretty cryptic rephrasing for the issue... I also did not see increase of memory to be the issue for the crash (memory at least on GPU seemed to be stable).

 

Yeah, this should be the ticket. At least the internal ID matches, though the description is indeed a bit strange. I'm just speculating, but maybe only one stacktrace was looked at, and then assumed the method it was happen to crash on was just running out of memory.

I only hope this bug's severity isn't underestimated.

 

Maybe it's worth to post this in HDRP forum, as it's now officially released with 2023.1 (I already posted on the announcement thread yesterday). This thread here is in an archived forum. Would you like to do it or shall I?

 

Good idea. You can post it if you like.