[PATCH] Review for machine combiner pass
Gerolf Hoflehner
ghoflehner at apple.com
Tue Jul 15 18:42:58 PDT 2014
+llvm-commits
Reminder/Background:
I implemented a pass for a late machine instruction combiner that may replace an instruction sequence by combined instruction(s) when it is beneficial to do so. It provides the infrastructure to evaluate instruction combining patterns like mul+add->madd based on machine trace information. Currently the DAG Combiner greedily generates combined instructions, which usually is a win for code size, but unfortunately can cause performance losses. To remedy this the new pass changes the logic from always generate combined instruction(s) to only do so when beneficial.
The design choice was driven by the desire to make it simple to a) add new pattern and b) add support for machine combining in a target. Consequently the combiner pass comes in 3 patches: First, the target independent driver that walks all instructions of a basic block, asks the target for possible combiner pattern, evaluates each pattern by having the target generate the instruction sequence represented by the pattern and finally replaces the old code when the new sequence is more efficient. The pattern and the new code sequence are opaque to the driver. Second, the target dependent code which currently supports only AArch64: for a given instruction it records the possible combiner pattern and on demand generates the instruction sequence it represents. Third, optional dumps the critical path length for tuning support.
http://reviews.llvm.org/D4367
* Motivation + Example
The opportunity for this optimization is across the llvm test suite and benchmarks.
Specific example: SingleSource/Benchmarks/Shootout/matrix (compiled with O3 flto for AArch64 gives a >20% gain):
Current assembly snippet:
0000000100007d24 mul w6, w23, w6 // Chain of madds
0000000100007d28 madd w5, w7, w5, w6 // All multiplies on critical path!
0000000100007d2c ldp w6, w7, [x4, #8]
0000000100007d30 ldr w23, [x11, x2]
0000000100007d34 madd w5, w23, w6, w5
0000000100007d38 ldr w6, [x12, x2]
0000000100007d3c madd w5, w6, w7, w5
0000000100007d40 ldr w6, [x13, x2]
0000000100007d44 ldp w7, w23, [x4, #16]
0000000100007d48 madd w5, w6, w7, w5
0000000100007d4c ldr w6, [x14, x2]
0000000100007d50 madd w5, w6, w23, w5
0000000100007d54 ldr w6, [x15, x2]
0000000100007d58 ldp w7, w23, [x4, #24]
0000000100007d5c madd w5, w6, w7, w5
0000000100007d60 ldr w6, [x16, x2]
0000000100007d64 madd w5, w6, w23, w5
0000000100007d68 ldr w6, [x17, x2]
0000000100007d6c ldp w7, w23, [x4, #32]
0000000100007d70 ldr w24, [x0, x2]
0000000100007d74 madd w5, w6, w7, w5
…
With machine combiner the multiplies can execute in parallel shortening the critical path (>20% gain):
0000000100007cf4 mul w5, w7, w5 // Multiplies can execute in parallel
0000000100007cf8 ldp w7, w23, [x4, #8] // off critical path
0000000100007cfc ldr w24, [x10, x2]
0000000100007d00 mul w6, w24, w6
0000000100007d04 ldr w24, [x11, x2]
0000000100007d08 mul w7, w24, w7
0000000100007d0c ldr w24, [x12, x2]
0000000100007d10 mul w23, w24, w23
0000000100007d14 ldr w24, [x13, x2]
0000000100007d18 add w5, w6, w5
0000000100007d1c ldp w6, w25, [x4, #16]
0000000100007d20 mul w6, w24, w6
0000000100007d24 ldr w24, [x14, x2]
0000000100007d28 mul w24, w24, w25
0000000100007d2c ldr w25, [x15, x2]
0000000100007d30 add w5, w7, w5
0000000100007d34 ldp w7, w26, [x4, #24]
0000000100007d38 mul w7, w25, w7
0000000100007d3c ldr w25, [x16, x2]
0000000100007d40 mul w25, w25, w26
0000000100007d44 ldr w26, [x17, x2]
0000000100007d48 add w5, w23, w5
0000000100007d4c ldp w23, w27, [x4, #32]
0000000100007d50 mul w23, w26, w23
0000000100007d54 ldr w26, [x0, x2]
0000000100007d58 mul w26, w26, w27
0000000100007d5c add w5, w6, w5
….
On Jul 15, 2014, at 6:26 PM, Gerolf Hoflehner <ghoflehner at apple.com> wrote:
>
> Changes:
> 1) Added bool alwaysCombine() (Target/TargetInstrInfo.h) so targets
> can decide to always replace a given pattern. This should be equivalent to
> the current code in DAGCombine when a given pattern is disabled.
> 2) InstrDepth is now a small vector (MachineCombiner.cpp)
> 3) Added helper function instr2instrSC (MachineCombiner.cpp)
> 4) Improved comments as suggested by reviewers
>
> http://reviews.llvm.org/D4367
>
> Files:
> include/llvm/CodeGen/MachineCombinerPattern.h
> include/llvm/CodeGen/MachineTraceMetrics.h
> include/llvm/CodeGen/Passes.h
> include/llvm/CodeGen/TargetSchedule.h
> include/llvm/InitializePasses.h
> include/llvm/Target/TargetInstrInfo.h
> lib/CodeGen/CMakeLists.txt
> lib/CodeGen/CodeGen.cpp
> lib/CodeGen/MachineCombiner.cpp
> lib/CodeGen/MachineScheduler.cpp
> lib/CodeGen/MachineTraceMetrics.cpp
> lib/CodeGen/TargetSchedule.cpp
> lib/Target/AArch64/AArch64InstrFormats.td
> lib/Target/AArch64/AArch64InstrInfo.cpp
> lib/Target/AArch64/AArch64InstrInfo.h
> lib/Target/AArch64/AArch64TargetMachine.cpp
> test/CodeGen/AArch64/aarch64-neon-mul-div.ll
> test/CodeGen/AArch64/early-ifcvt.ll
> <D4367.11483.patch>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140715/71d688b4/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MC_TIv2.patch
Type: application/octet-stream
Size: 29311 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140715/71d688b4/attachment.obj>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140715/71d688b4/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MC_AArch64v2.patch
Type: application/octet-stream
Size: 67889 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140715/71d688b4/attachment-0001.obj>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140715/71d688b4/attachment-0002.html>
More information about the llvm-commits
mailing list