The signature already existed before but once bug 1595994 landed, it got more frequent on Nightly 84.0a1 and affects the rdd and gpu processes (the latter ones had been affected before but less frequent).

Crash report: https://crash-stats.mozilla.org/report/index/63604d7b-c30e-4462-9463-9cd150201114

Reason: EXCEPTION_ACCESS_VIOLATION_READ

Top 10 frames of crashing thread:

0 atidxx64.dll AmdDxGsaFreeCompiledShader 
1 atidxx64.dll AmdDxGsaFreeCompiledShader 
2 atidxx64.dll AmdDxGsaFreeCompiledShader 
3 atidxx64.dll AmdDxGsaFreeCompiledShader 
4 atidxx64.dll AmdDxGsaFreeCompiledShader 
5 atidxx64.dll AmdDxGsaFreeCompiledShader 
6 atidxx64.dll XdxQueryTlsLookupTable 
7 d3d11.dll static void CContext::TID3D11DeviceContext_DrawIndexedInstanced_<1> 
8 libglesv2.dll rx::Renderer11::drawElements gfx/angle/checkout/src/libANGLE/renderer/d3d/d3d11/Renderer11.cpp:1595
9 libglesv2.dll rx::Context11::drawElementsInstanced gfx/angle/checkout/src/libANGLE/renderer/d3d/d3d11/Context11.cpp:323

this is a null deref, could be on mDeviceContext

I think this is the same underlying issue as bug 1673201, where most addresses within the amd driver binary are resolving as AmdDxGsaFreeCompiledShader.

That means a whole bunch of different driver crashes are being coalesced under this signature, rather than reporting the d3d11.dll symbol (CContext::TID3D11DeviceContext_DrawIndexedInstanced_<1> in the crash from comment 0).