Removal of Helper Method Frames (HMF)

[AaronRobinsonMSFT]

Background

Over the past several releases, there has been a slow, but intentional, effort to move more runtime code from C++ into C#. A consequence of this effort has meant the removal of Helper Method Frames (HMF). Removal of HMFs simplifies the runtime for efforts like cDAC, reduces opportunities for GC holes, and in some cases can result in improved performance when the JIT can see more of the code that is can optimize.

Prior work

Below is a list of various PRs that can be referenced that demonstrate that effort across multiple authors and releases.

#70000, #70055, #71873, #87166, #95038, #96926, #97590, #100116, #101353, #106793, #107058, #107218, #107648, #108167, #109135, #109996, #110064, #110211, #110377, #110481, #110627, #110766

Remaining work

List of JIT helpers that need the removal of explicit HMFs:

• JIT_GetFieldAddr_Framed
• JIT_New
• JIT_NewMaybeFrozen
• FramedAllocateString
• JIT_StrCns
• JIT_NewArr1
• JIT_NewArr1MaybeFrozen
• JIT_NewMDArr
• JIT_Box
• JIT_GetRuntimeFieldStub
• JIT_GetRuntimeMethodStub
• JIT_GetRuntimeType_Framed
• JIT_MonEnter_Helper
• JIT_MonTryEnter_Helper
• JIT_MonExit_Helper
• JIT_MonExit_Signal
• IL_Throw
• IL_Rethrow
• JIT_ThrowMethodAccessException
• JIT_ThrowFieldAccessException
• JIT_ThrowClassAccessException
• JIT_UserBreakpoint
• JIT_PollGC_Framed
• JIT_PInvokeEndRarePath
• JIT_RareDisableHelper/JIT_RareDisableHelperWorker
• JIT_StressGC
• JIT_Patchpoint_Framed
• JIT_PartialCompilationPatchpoint

List of JIT helpers with implicit HMFs through a use of FCThrow.

• JIT_Div, JIT_Mod, JIT_UDiv, JIT_UMod, JIT_LDiv, JIT_LMod, JIT_ULDiv, JIT_ULMod - Move arithmetic helpers to managed code #109087

The following clean-up can occur after all the above is complete:

• Removal of fcall.h and fcall.cpp
• Update BoTR to remove description of FCalls.

[dotnet-policy-service]

Tagging subscribers to this area: @mangod9
See info in area-owners.md if you want to be subscribed.

[huoyaoyuan]

Also #95695

[AaronRobinsonMSFT]

Also #95695

Thanks, I added it. I wasn't trying to be exhaustive. Wanted to provide a bunch of examples people could reference.

[AaronRobinsonMSFT]

/cc @davidwrighton @elinor-fung @am11 @MichalStrehovsky @janvorli

[am11]

JIT_MonEnter_Helper

The way @VSadov consolidated lock primitives with NativeAOT, if we continue on that path, maybe it is possible to share the entire implementation with AOT (which has this HCall written in C#).

JIT_Patchpoint_Framed

There is also a related JIT_Patchpoint HCall, which ends up using VirtualUnwindToFirstManagedCallFrame, not sure if there is a better structured way to tackle it without the libunwind dependency but would be nice to switch to one if it exists.

[huoyaoyuan]

Also #95695

Thanks, I added it. I wasn't trying to be exhaustive. Wanted to provide a bunch of examples people could reference.

It was tracking issue instead of PR. There is also more context about HMF.

[AaronRobinsonMSFT]

Also #95695

Thanks, I added it. I wasn't trying to be exhaustive. Wanted to provide a bunch of examples people could reference.

It was tracking issue instead of PR. There is also more context about HMF.

Ah, I see now. Hmmm. @jkotas Should I close this one?

[am11]

Removal of HMF has come a long way:

Release	FCThrow	HELPER_METHOD_FRAME
v5.0.0	187	353
v6.0.0	179	326
v7.0.0	164	298
v8.0.0	158	308
v9.0.0	97	152
main (+ #110766 and #109087)	0	37

[AaronRobinsonMSFT]

Closing this one in lieu of #95695.