-
Notifications
You must be signed in to change notification settings - Fork 14.3k
[RISCV] Add scheduler model for sifive-p450. #77989
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
Merged
Changes from all commits
Commits
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,354 @@ | ||
//==- RISCVSchedSiFiveP400.td - SiFiveP400 Scheduling Defs ---*- tablegen -*-=// | ||
// | ||
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. | ||
// See https://llvm.org/LICENSE.txt for license information. | ||
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception | ||
// | ||
//===----------------------------------------------------------------------===// | ||
|
||
//===----------------------------------------------------------------------===// | ||
|
||
def SiFiveP400Model : SchedMachineModel { | ||
let IssueWidth = 3; // 3 micro-ops are dispatched per cycle. | ||
let MicroOpBufferSize = 56; // Max micro-ops that can be buffered. | ||
let LoadLatency = 4; // Cycles for loads to access the cache. | ||
let MispredictPenalty = 9; // Extra cycles for a mispredicted branch. | ||
let PostRAScheduler = true; | ||
let UnsupportedFeatures = [HasStdExtZbkb, HasStdExtZbkc, HasStdExtZbkx, | ||
HasStdExtZcmt, HasStdExtZknd, HasStdExtZkne, | ||
HasStdExtZknh, HasStdExtZksed, HasStdExtZksh, | ||
HasStdExtZkr]; | ||
let CompleteModel = false; | ||
} | ||
|
||
// The SiFiveP400 microarchitecure has 6 pipelines: | ||
// Three pipelines for integer operations. | ||
// One pipeline for FPU operations. | ||
// One pipeline for Load operations. | ||
// One pipeline for Store operations. | ||
let SchedModel = SiFiveP400Model in { | ||
|
||
def SiFiveP400IEXQ0 : ProcResource<1>; | ||
def SiFiveP400IEXQ1 : ProcResource<1>; | ||
def SiFiveP400IEXQ2 : ProcResource<1>; | ||
def SiFiveP400FEXQ0 : ProcResource<1>; | ||
def SiFiveP400Load : ProcResource<1>; | ||
def SiFiveP400Store : ProcResource<1>; | ||
|
||
def SiFiveP400IntArith : ProcResGroup<[SiFiveP400IEXQ0, SiFiveP400IEXQ1, SiFiveP400IEXQ2]>; | ||
defvar SiFiveP400Branch = SiFiveP400IEXQ0; | ||
defvar SiFiveP400SYS = SiFiveP400IEXQ1; | ||
defvar SiFiveP400MulDiv = SiFiveP400IEXQ2; | ||
defvar SiFiveP400I2F = SiFiveP400IEXQ2; | ||
def SiFiveP400Div : ProcResource<1>; | ||
|
||
defvar SiFiveP400FloatArith = SiFiveP400FEXQ0; | ||
defvar SiFiveP400F2I = SiFiveP400FEXQ0; | ||
def SiFiveP400FloatDiv : ProcResource<1>; | ||
|
||
let Latency = 1 in { | ||
// Integer arithmetic and logic | ||
def : WriteRes<WriteIALU, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteIALU32, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteShiftImm, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteShiftImm32, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteShiftReg, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteShiftReg32, [SiFiveP400IntArith]>; | ||
// Branching | ||
def : WriteRes<WriteJmp, [SiFiveP400Branch]>; | ||
def : WriteRes<WriteJal, [SiFiveP400Branch]>; | ||
def : WriteRes<WriteJalr, [SiFiveP400Branch]>; | ||
} | ||
|
||
// CMOV | ||
def P400WriteCMOV : SchedWriteRes<[SiFiveP400Branch, SiFiveP400IEXQ1]> { | ||
let Latency = 2; | ||
let NumMicroOps = 2; | ||
} | ||
def : InstRW<[P400WriteCMOV], (instrs PseudoCCMOVGPRNoX0)>; | ||
|
||
let Latency = 3 in { | ||
// Integer multiplication | ||
def : WriteRes<WriteIMul, [SiFiveP400MulDiv]>; | ||
def : WriteRes<WriteIMul32, [SiFiveP400MulDiv]>; | ||
// cpop[w] look exactly like multiply. | ||
def : WriteRes<WriteCPOP, [SiFiveP400MulDiv]>; | ||
def : WriteRes<WriteCPOP32, [SiFiveP400MulDiv]>; | ||
} | ||
|
||
// Integer division | ||
def : WriteRes<WriteIDiv, [SiFiveP400MulDiv, SiFiveP400Div]> { | ||
let Latency = 35; | ||
let ReleaseAtCycles = [1, 34]; | ||
} | ||
def : WriteRes<WriteIDiv32, [SiFiveP400MulDiv, SiFiveP400Div]> { | ||
let Latency = 20; | ||
let ReleaseAtCycles = [1, 19]; | ||
} | ||
|
||
let Latency = 1 in { | ||
// Bitmanip | ||
def : WriteRes<WriteRotateImm, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteRotateImm32, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteRotateReg, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteRotateReg32, [SiFiveP400IntArith]>; | ||
|
||
def : WriteRes<WriteCLZ, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteCLZ32, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteCTZ, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteCTZ32, [SiFiveP400IntArith]>; | ||
|
||
def : WriteRes<WriteORCB, [SiFiveP400IntArith]>; | ||
|
||
def : WriteRes<WriteREV8, [SiFiveP400IntArith]>; | ||
|
||
def : WriteRes<WriteSHXADD, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteSHXADD32, [SiFiveP400IntArith]>; | ||
|
||
def : WriteRes<WriteSingleBit, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteSingleBitImm, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteBEXT, [SiFiveP400IntArith]>; | ||
def : WriteRes<WriteBEXTI, [SiFiveP400IntArith]>; | ||
} | ||
|
||
// Memory | ||
let Latency = 1 in { | ||
def : WriteRes<WriteSTB, [SiFiveP400Store]>; | ||
def : WriteRes<WriteSTH, [SiFiveP400Store]>; | ||
def : WriteRes<WriteSTW, [SiFiveP400Store]>; | ||
def : WriteRes<WriteSTD, [SiFiveP400Store]>; | ||
def : WriteRes<WriteFST16, [SiFiveP400Store]>; | ||
def : WriteRes<WriteFST32, [SiFiveP400Store]>; | ||
def : WriteRes<WriteFST64, [SiFiveP400Store]>; | ||
} | ||
let Latency = 4 in { | ||
def : WriteRes<WriteLDB, [SiFiveP400Load]>; | ||
def : WriteRes<WriteLDH, [SiFiveP400Load]>; | ||
} | ||
let Latency = 4 in { | ||
def : WriteRes<WriteLDW, [SiFiveP400Load]>; | ||
def : WriteRes<WriteLDD, [SiFiveP400Load]>; | ||
} | ||
|
||
let Latency = 6 in { | ||
def : WriteRes<WriteFLD16, [SiFiveP400Load]>; | ||
def : WriteRes<WriteFLD32, [SiFiveP400Load]>; | ||
def : WriteRes<WriteFLD64, [SiFiveP400Load]>; | ||
} | ||
|
||
// Atomic memory | ||
let Latency = 3 in { | ||
def : WriteRes<WriteAtomicSTW, [SiFiveP400Store]>; | ||
def : WriteRes<WriteAtomicSTD, [SiFiveP400Store]>; | ||
def : WriteRes<WriteAtomicW, [SiFiveP400Load]>; | ||
def : WriteRes<WriteAtomicD, [SiFiveP400Load]>; | ||
def : WriteRes<WriteAtomicLDW, [SiFiveP400Load]>; | ||
def : WriteRes<WriteAtomicLDD, [SiFiveP400Load]>; | ||
} | ||
|
||
// Floating point | ||
let Latency = 4 in { | ||
def : WriteRes<WriteFAdd16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFAdd32, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFAdd64, [SiFiveP400FloatArith]>; | ||
|
||
def : WriteRes<WriteFMul16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFMul32, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFMul64, [SiFiveP400FloatArith]>; | ||
|
||
def : WriteRes<WriteFMA16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFMA32, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFMA64, [SiFiveP400FloatArith]>; | ||
} | ||
|
||
let Latency = 2 in { | ||
def : WriteRes<WriteFSGNJ16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFSGNJ32, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFSGNJ64, [SiFiveP400FloatArith]>; | ||
|
||
def : WriteRes<WriteFMinMax16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFMinMax32, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFMinMax64, [SiFiveP400FloatArith]>; | ||
} | ||
|
||
// Half precision. | ||
def : WriteRes<WriteFDiv16, [SiFiveP400FEXQ0, SiFiveP400FloatDiv]> { | ||
let Latency = 19; | ||
let ReleaseAtCycles = [1, 18]; | ||
} | ||
def : WriteRes<WriteFSqrt16, [SiFiveP400FEXQ0, SiFiveP400FloatDiv]> { | ||
let Latency = 18; | ||
let ReleaseAtCycles = [1, 17]; | ||
} | ||
|
||
// Single precision. | ||
def : WriteRes<WriteFDiv32, [SiFiveP400FEXQ0, SiFiveP400FloatDiv]> { | ||
let Latency = 19; | ||
let ReleaseAtCycles = [1, 18]; | ||
} | ||
def : WriteRes<WriteFSqrt32, [SiFiveP400FEXQ0, SiFiveP400FloatDiv]> { | ||
let Latency = 18; | ||
let ReleaseAtCycles = [1, 17]; | ||
} | ||
|
||
// Double precision | ||
def : WriteRes<WriteFDiv64, [SiFiveP400FEXQ0, SiFiveP400FloatDiv]> { | ||
let Latency = 33; | ||
let ReleaseAtCycles = [1, 32]; | ||
} | ||
def : WriteRes<WriteFSqrt64, [SiFiveP400FEXQ0, SiFiveP400FloatDiv]> { | ||
let Latency = 33; | ||
let ReleaseAtCycles = [1, 32]; | ||
} | ||
|
||
// Conversions | ||
let Latency = 2 in { | ||
def : WriteRes<WriteFCvtI32ToF16, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFCvtI32ToF32, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFCvtI32ToF64, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFCvtI64ToF16, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFCvtI64ToF32, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFCvtI64ToF64, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFCvtF16ToI32, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCvtF16ToI64, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCvtF16ToF32, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFCvtF16ToF64, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFCvtF32ToI32, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCvtF32ToI64, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCvtF32ToF16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFCvtF32ToF64, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFCvtF64ToI32, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCvtF64ToI64, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCvtF64ToF16, [SiFiveP400FloatArith]>; | ||
def : WriteRes<WriteFCvtF64ToF32, [SiFiveP400FloatArith]>; | ||
|
||
def : WriteRes<WriteFClass16, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFClass32, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFClass64, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCmp16, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCmp32, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFCmp64, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFMovI16ToF16, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFMovF16ToI16, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFMovI32ToF32, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFMovF32ToI32, [SiFiveP400F2I]>; | ||
def : WriteRes<WriteFMovI64ToF64, [SiFiveP400I2F]>; | ||
def : WriteRes<WriteFMovF64ToI64, [SiFiveP400F2I]>; | ||
} | ||
|
||
// Others | ||
def : WriteRes<WriteCSR, [SiFiveP400SYS]>; | ||
def : WriteRes<WriteNop, []>; | ||
|
||
// FIXME: This could be better modeled by looking at the regclasses of the operands. | ||
def : InstRW<[WriteIALU, ReadIALU], (instrs COPY)>; | ||
|
||
//===----------------------------------------------------------------------===// | ||
// Bypass and advance | ||
def : ReadAdvance<ReadJmp, 0>; | ||
def : ReadAdvance<ReadJalr, 0>; | ||
def : ReadAdvance<ReadCSR, 0>; | ||
def : ReadAdvance<ReadStoreData, 0>; | ||
def : ReadAdvance<ReadMemBase, 0>; | ||
def : ReadAdvance<ReadIALU, 0>; | ||
def : ReadAdvance<ReadIALU32, 0>; | ||
def : ReadAdvance<ReadShiftImm, 0>; | ||
def : ReadAdvance<ReadShiftImm32, 0>; | ||
def : ReadAdvance<ReadShiftReg, 0>; | ||
def : ReadAdvance<ReadShiftReg32, 0>; | ||
def : ReadAdvance<ReadIDiv, 0>; | ||
def : ReadAdvance<ReadIDiv32, 0>; | ||
def : ReadAdvance<ReadIMul, 0>; | ||
def : ReadAdvance<ReadIMul32, 0>; | ||
def : ReadAdvance<ReadAtomicWA, 0>; | ||
def : ReadAdvance<ReadAtomicWD, 0>; | ||
def : ReadAdvance<ReadAtomicDA, 0>; | ||
def : ReadAdvance<ReadAtomicDD, 0>; | ||
def : ReadAdvance<ReadAtomicLDW, 0>; | ||
def : ReadAdvance<ReadAtomicLDD, 0>; | ||
def : ReadAdvance<ReadAtomicSTW, 0>; | ||
def : ReadAdvance<ReadAtomicSTD, 0>; | ||
def : ReadAdvance<ReadFStoreData, 0>; | ||
def : ReadAdvance<ReadFMemBase, 0>; | ||
def : ReadAdvance<ReadFAdd16, 0>; | ||
def : ReadAdvance<ReadFAdd32, 0>; | ||
def : ReadAdvance<ReadFAdd64, 0>; | ||
def : ReadAdvance<ReadFMul16, 0>; | ||
def : ReadAdvance<ReadFMA16, 0>; | ||
def : ReadAdvance<ReadFMA16Addend, 0>; | ||
def : ReadAdvance<ReadFMul32, 0>; | ||
def : ReadAdvance<ReadFMA32, 0>; | ||
def : ReadAdvance<ReadFMA32Addend, 0>; | ||
def : ReadAdvance<ReadFMul64, 0>; | ||
def : ReadAdvance<ReadFMA64, 0>; | ||
def : ReadAdvance<ReadFMA64Addend, 0>; | ||
def : ReadAdvance<ReadFDiv16, 0>; | ||
def : ReadAdvance<ReadFDiv32, 0>; | ||
def : ReadAdvance<ReadFDiv64, 0>; | ||
def : ReadAdvance<ReadFSqrt16, 0>; | ||
def : ReadAdvance<ReadFSqrt32, 0>; | ||
def : ReadAdvance<ReadFSqrt64, 0>; | ||
def : ReadAdvance<ReadFCmp16, 0>; | ||
def : ReadAdvance<ReadFCmp32, 0>; | ||
def : ReadAdvance<ReadFCmp64, 0>; | ||
def : ReadAdvance<ReadFSGNJ16, 0>; | ||
def : ReadAdvance<ReadFSGNJ32, 0>; | ||
def : ReadAdvance<ReadFSGNJ64, 0>; | ||
def : ReadAdvance<ReadFMinMax16, 0>; | ||
def : ReadAdvance<ReadFMinMax32, 0>; | ||
def : ReadAdvance<ReadFMinMax64, 0>; | ||
def : ReadAdvance<ReadFCvtF16ToI32, 0>; | ||
def : ReadAdvance<ReadFCvtF16ToI64, 0>; | ||
def : ReadAdvance<ReadFCvtF32ToI32, 0>; | ||
def : ReadAdvance<ReadFCvtF32ToI64, 0>; | ||
def : ReadAdvance<ReadFCvtF64ToI32, 0>; | ||
def : ReadAdvance<ReadFCvtF64ToI64, 0>; | ||
def : ReadAdvance<ReadFCvtI32ToF16, 0>; | ||
def : ReadAdvance<ReadFCvtI32ToF32, 0>; | ||
def : ReadAdvance<ReadFCvtI32ToF64, 0>; | ||
def : ReadAdvance<ReadFCvtI64ToF16, 0>; | ||
def : ReadAdvance<ReadFCvtI64ToF32, 0>; | ||
def : ReadAdvance<ReadFCvtI64ToF64, 0>; | ||
def : ReadAdvance<ReadFCvtF32ToF64, 0>; | ||
def : ReadAdvance<ReadFCvtF64ToF32, 0>; | ||
def : ReadAdvance<ReadFCvtF16ToF32, 0>; | ||
def : ReadAdvance<ReadFCvtF32ToF16, 0>; | ||
def : ReadAdvance<ReadFCvtF16ToF64, 0>; | ||
def : ReadAdvance<ReadFCvtF64ToF16, 0>; | ||
def : ReadAdvance<ReadFMovF16ToI16, 0>; | ||
def : ReadAdvance<ReadFMovI16ToF16, 0>; | ||
def : ReadAdvance<ReadFMovF32ToI32, 0>; | ||
def : ReadAdvance<ReadFMovI32ToF32, 0>; | ||
def : ReadAdvance<ReadFMovF64ToI64, 0>; | ||
def : ReadAdvance<ReadFMovI64ToF64, 0>; | ||
def : ReadAdvance<ReadFClass16, 0>; | ||
def : ReadAdvance<ReadFClass32, 0>; | ||
def : ReadAdvance<ReadFClass64, 0>; | ||
|
||
// Bitmanip | ||
def : ReadAdvance<ReadRotateImm, 0>; | ||
def : ReadAdvance<ReadRotateImm32, 0>; | ||
def : ReadAdvance<ReadRotateReg, 0>; | ||
def : ReadAdvance<ReadRotateReg32, 0>; | ||
def : ReadAdvance<ReadCLZ, 0>; | ||
def : ReadAdvance<ReadCLZ32, 0>; | ||
def : ReadAdvance<ReadCTZ, 0>; | ||
def : ReadAdvance<ReadCTZ32, 0>; | ||
def : ReadAdvance<ReadCPOP, 0>; | ||
def : ReadAdvance<ReadCPOP32, 0>; | ||
def : ReadAdvance<ReadORCB, 0>; | ||
def : ReadAdvance<ReadREV8, 0>; | ||
def : ReadAdvance<ReadSHXADD, 0>; | ||
def : ReadAdvance<ReadSHXADD32, 0>; | ||
def : ReadAdvance<ReadSingleBit, 0>; | ||
def : ReadAdvance<ReadSingleBitImm, 0>; | ||
|
||
//===----------------------------------------------------------------------===// | ||
// Unsupported extensions | ||
defm : UnsupportedSchedZbc; | ||
defm : UnsupportedSchedZbkb; | ||
defm : UnsupportedSchedZbkx; | ||
defm : UnsupportedSchedSFB; | ||
defm : UnsupportedSchedZfa; | ||
defm : UnsupportedSchedV; | ||
} |
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One question I should have asked before (not for this patch): Can we separate SiFive7 model into two files or two models (scalar scheduling and RVV scheduling)? I know the microarchitecture of X280 inherits from SiFive7, but there are some different designs I see from current scheduling model, which makes X280 and SiFive7 independent enough.
Maybe we can make scalar parts a multiclass and reuse it in X280 scheduling model.
cc @michaelmaitland
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not sure what different designs you're referring to here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, I may use the wrong words here. I meant that the CPU resources/pipeline are different, X280 has additional vector pipeline. Intuitively, they should be separated (I think). And, the outputs of
llvm-mca
contain vector resources even for scalar CPU likesifive-u74
.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am currently exploring some refactors of the scheduler model in our downstream for better code reusability. I will take a look to see if what you suggest is possible. I see your point about the llvm-mca reporting. But besides llvm-mca, the suggested change doesn't impact scheduling. In my experience, the relevant TblGen classes may make this change harder than it is worth.