condense test into one line

pipeline is printed on a single line, so having CHECK lines on separate
lines can allow extra characters in between (and will not error out on
extra passes being in the pipeline)

Author: optimisan

llvm/test/CodeGen/AMDGPU/llc-pipeline-npm.ll

@@ -1,276 +1,19 @@
-; RUN: llc -enable-new-pm -disable-verify -mtriple=amdgcn--amdhsa -print-pipeline-passes < %s 2>&1 \
+; RUN: llc -enable-new-pm -mtriple=amdgcn--amdhsa -O0 -print-pipeline-passes < %s 2>&1 \
+; RUN:   | FileCheck -check-prefix=GCN-O0 %s
+
+; RUN: llc -enable-new-pm -mtriple=amdgcn--amdhsa -print-pipeline-passes < %s 2>&1 \
 ; RUN:   | FileCheck -check-prefix=GCN-O2 %s
 
-; RUN: llc -O3 -enable-new-pm -disable-verify -mtriple=amdgcn--amdhsa -print-pipeline-passes < %s 2>&1 \
+; RUN: llc -O3 -enable-new-pm -mtriple=amdgcn--amdhsa -print-pipeline-passes < %s 2>&1 \
 ; RUN:   | FileCheck -check-prefix=GCN-O3 %s
 
-; GCN-O2: require<MachineModuleAnalysis>
-; GCN-O2: require<profile-summary>
-; GCN-O2: require<collector-metadata>
-; GCN-O2: pre-isel-intrinsic-lowering
-; GCN-O2: function(expand-large-div-rem
-; GCN-O2: expand-fp)
-; GCN-O2: amdgpu-remove-incompatible-functions
-; GCN-O2: amdgpu-printf-runtime-binding
-; GCN-O2: amdgpu-lower-ctor-dtor
-; GCN-O2: function(amdgpu-image-intrinsic-opt)
-; GCN-O2: expand-variadics
-; GCN-O2: amdgpu-always-inline
-; GCN-O2: always-inline
-; GCN-O2: amdgpu-export-kernel-runtime-handles
-; GCN-O2: amdgpu-sw-lower-lds
-; GCN-O2: amdgpu-lower-module-lds
-; GCN-O2: function(infer-address-spaces
-; GCN-O2: amdgpu-atomic-optimizer
-; GCN-O2: atomic-expand
-; GCN-O2: amdgpu-promote-alloca
-; GCN-O2: separate-const-offset-from-gep<>
-; GCN-O2: slsr
-; GCN-O2: early-cse<>
-; GCN-O2: nary-reassociate
-; GCN-O2: early-cse<>
-; GCN-O2: amdgpu-codegenprepare
-; GCN-O2: loop-mssa(loop-reduce)
-; GCN-O2: mergeicmps
-; GCN-O2: expand-memcmp
-; GCN-O2: gc-lowering
-; GCN-O2: lower-constant-intrinsics
-; GCN-O2: UnreachableBlockElimPass
-; GCN-O2: consthoist
-; GCN-O2: ReplaceWithVeclib
-; GCN-O2: partially-inline-libcalls
-; GCN-O2: ee-instrument<post-inline>
-; GCN-O2: scalarize-masked-mem-intrin
-; GCN-O2: ExpandReductionsPass
-; GCN-O2: early-cse<>
-; GCN-O2: amdgpu-lower-kernel-arguments)
-; GCN-O2: amdgpu-lower-buffer-fat-pointers
-; GCN-O2: cgscc(function(codegenprepare
-; GCN-O2: load-store-vectorizer
-; GCN-O2: lower-switch
-; GCN-O2: lower-invoke
-; GCN-O2: UnreachableBlockElimPass
-; GCN-O2: flatten-cfg
-; GCN-O2: sink
-; GCN-O2: amdgpu-late-codegenprepare
-; GCN-O2: amdgpu-unify-divergent-exit-nodes
-; GCN-O2: fix-irreducible
-; GCN-O2: unify-loop-exits
-; GCN-O2: StructurizeCFGPass
-; GCN-O2: amdgpu-annotate-uniform
-; GCN-O2: si-annotate-control-flow
-; GCN-O2: amdgpu-rewrite-undef-for-phi
-; GCN-O2: lcssa))
-; GCN-O2: amdgpu-perf-hint
-; GCN-O2: cgscc(function(require<uniformity>
-; GCN-O2: callbr-prepare
-; GCN-O2: safe-stack
-; GCN-O2: stack-protector))
-; GCN-O2: cgscc(function(machine-function(amdgpu-isel
-; GCN-O2: si-fix-sgpr-copies
-; GCN-O2: si-i1-copies
-; GCN-O2: finalize-isel
-; GCN-O2: early-tailduplication
-; GCN-O2: opt-phis
-; GCN-O2: stack-coloring
-; GCN-O2: localstackalloc
-; GCN-O2: dead-mi-elimination
-; GCN-O2: early-machinelicm
-; GCN-O2: machine-cse
-; GCN-O2: machine-sink
-; GCN-O2: peephole-opt
-; GCN-O2: dead-mi-elimination
-; GCN-O2: si-fold-operands
-; GCN-O2: gcn-dpp-combine
-; GCN-O2: si-load-store-opt
-; GCN-O2: si-peephole-sdwa
-; GCN-O2: early-machinelicm
-; GCN-O2: machine-cse
-; GCN-O2: si-fold-operands
-; GCN-O2: dead-mi-elimination
-; GCN-O2: si-shrink-instructions
-; GCN-O2: detect-dead-lanes
-; GCN-O2: InitUndefPass
-; GCN-O2: ProcessImplicitDefsPass
-; GCN-O2: unreachable-mbb-elimination
-; GCN-O2: require<live-vars>
-; GCN-O2: require<machine-loops>
-; GCN-O2: phi-node-elimination
-; GCN-O2: two-address-instruction
-; GCN-O2: register-coalescer
-; GCN-O2: rename-independent-subregs
-; GCN-O2: machine-scheduler
-; GCN-O2: greedy<all>
-; GCN-O2: amdgpu-nsa-reassign
-; GCN-O2: VirtRegRewriterPass
-; GCN-O2: stack-slot-coloring
-; GCN-O2: machine-cp
-; GCN-O2: machinelicm
-; GCN-O2: si-fix-vgpr-copies
-; GCN-O2: si-optimize-exec-masking
-; GCN-O2: remove-redundant-debug-values
-; GCN-O2: fixup-statepoint-caller-saved
-; GCN-O2: PostRAMachineSinkingPass
-; GCN-O2: ShrinkWrapPass
-; GCN-O2: PrologEpilogInserterPass
-; GCN-O2: branch-folder
-; GCN-O2: tailduplication
-; GCN-O2: machine-latecleanup
-; GCN-O2: machine-cp
-; GCN-O2: post-ra-pseudos
-; GCN-O2: postmisched
-; GCN-O2: block-placement
-; GCN-O2: fentry-insert
-; GCN-O2: xray-instrumentation
-; GCN-O2: patchable-function
-; GCN-O2: gcn-create-vopd
-; GCN-O2: si-memory-legalizer
-; GCN-O2: si-insert-waitcnts
-; GCN-O2: si-late-branch-lowering
-; GCN-O2: si-pre-emit-peephole
-; GCN-O2: post-RA-hazard-rec
-; GCN-O2: AMDGPUWaitSGPRHazardsPass
-; GCN-O2: amdgpu-insert-delay-alu
-; GCN-O2: branch-relaxation
-; GCN-O2: remove-loads-into-fake-uses
-; GCN-O2: live-debug-values
-; GCN-O2: machine-sanmd
-; GCN-O2: stack-frame-layout)
-; GCN-O2: invalidate<machine-function-info>))
 
-; GCN-O3: require<MachineModuleAnalysis>
-; GCN-O3: require<profile-summary>
-; GCN-O3: require<collector-metadata>
-; GCN-O3: pre-isel-intrinsic-lowering
-; GCN-O3: function(expand-large-div-rem
-; GCN-O3: expand-fp)
-; GCN-O3: amdgpu-remove-incompatible-functions
-; GCN-O3: amdgpu-printf-runtime-binding
-; GCN-O3: amdgpu-lower-ctor-dtor
-; GCN-O3: function(amdgpu-image-intrinsic-opt)
-; GCN-O3: expand-variadics
-; GCN-O3: amdgpu-always-inline
-; GCN-O3: always-inline
-; GCN-O3: amdgpu-export-kernel-runtime-handles
-; GCN-O3: amdgpu-sw-lower-lds
-; GCN-O3: amdgpu-lower-module-lds
-; GCN-O3: function(infer-address-spaces
-; GCN-O3: amdgpu-atomic-optimizer
-; GCN-O3: atomic-expand
-; GCN-O3: amdgpu-promote-alloca
-; GCN-O3: separate-const-offset-from-gep<>
-; GCN-O3: slsr
-; GCN-O3: gvn<>
-; GCN-O3: nary-reassociate
-; GCN-O3: early-cse<>
-; GCN-O3: amdgpu-codegenprepare
-; GCN-O3: loop-mssa(loop-reduce)
-; GCN-O3: mergeicmps
-; GCN-O3: expand-memcmp
-; GCN-O3: gc-lowering
-; GCN-O3: lower-constant-intrinsics
-; GCN-O3: UnreachableBlockElimPass
-; GCN-O3: consthoist
-; GCN-O3: ReplaceWithVeclib
-; GCN-O3: partially-inline-libcalls
-; GCN-O3: ee-instrument<post-inline>
-; GCN-O3: scalarize-masked-mem-intrin
-; GCN-O3: ExpandReductionsPass
-; GCN-O3: gvn<>
-; GCN-O3: amdgpu-lower-kernel-arguments)
-; GCN-O3: amdgpu-lower-buffer-fat-pointers
-; GCN-O3: cgscc(function(codegenprepare
-; GCN-O3: load-store-vectorizer
-; GCN-O3: lower-switch
-; GCN-O3: lower-invoke
-; GCN-O3: UnreachableBlockElimPass
-; GCN-O3: flatten-cfg
-; GCN-O3: sink
-; GCN-O3: amdgpu-late-codegenprepare
-; GCN-O3: amdgpu-unify-divergent-exit-nodes
-; GCN-O3: fix-irreducible
-; GCN-O3: unify-loop-exits
-; GCN-O3: StructurizeCFGPass
-; GCN-O3: amdgpu-annotate-uniform
-; GCN-O3: si-annotate-control-flow
-; GCN-O3: amdgpu-rewrite-undef-for-phi
-; GCN-O3: lcssa))
-; GCN-O3: amdgpu-perf-hint
-; GCN-O3: cgscc(function(require<uniformity>
-; GCN-O3: callbr-prepare
-; GCN-O3: safe-stack
-; GCN-O3: stack-protector))
-; GCN-O3: cgscc(function(machine-function(amdgpu-isel
-; GCN-O3: si-fix-sgpr-copies
-; GCN-O3: si-i1-copies
-; GCN-O3: finalize-isel
-; GCN-O3: early-tailduplication
-; GCN-O3: opt-phis
-; GCN-O3: stack-coloring
-; GCN-O3: localstackalloc
-; GCN-O3: dead-mi-elimination
-; GCN-O3: early-machinelicm
-; GCN-O3: machine-cse
-; GCN-O3: machine-sink
-; GCN-O3: peephole-opt
-; GCN-O3: dead-mi-elimination
-; GCN-O3: si-fold-operands
-; GCN-O3: gcn-dpp-combine
-; GCN-O3: si-load-store-opt
-; GCN-O3: si-peephole-sdwa
-; GCN-O3: early-machinelicm
-; GCN-O3: machine-cse
-; GCN-O3: si-fold-operands
-; GCN-O3: dead-mi-elimination
-; GCN-O3: si-shrink-instructions
-; GCN-O3: detect-dead-lanes
-; GCN-O3: InitUndefPass
-; GCN-O3: ProcessImplicitDefsPass
-; GCN-O3: unreachable-mbb-elimination
-; GCN-O3: require<live-vars>
-; GCN-O3: require<machine-loops>
-; GCN-O3: phi-node-elimination
-; GCN-O3: two-address-instruction
-; GCN-O3: register-coalescer
-; GCN-O3: rename-independent-subregs
-; GCN-O3: machine-scheduler
-; GCN-O3: greedy<all>
-; GCN-O3: amdgpu-nsa-reassign
-; GCN-O3: VirtRegRewriterPass
-; GCN-O3: stack-slot-coloring
-; GCN-O3: machine-cp
-; GCN-O3: machinelicm
-; GCN-O3: si-fix-vgpr-copies
-; GCN-O3: si-optimize-exec-masking
-; GCN-O3: remove-redundant-debug-values
-; GCN-O3: fixup-statepoint-caller-saved
-; GCN-O3: PostRAMachineSinkingPass
-; GCN-O3: ShrinkWrapPass
-; GCN-O3: PrologEpilogInserterPass
-; GCN-O3: branch-folder
-; GCN-O3: tailduplication
-; GCN-O3: machine-latecleanup
-; GCN-O3: machine-cp
-; GCN-O3: post-ra-pseudos
-; GCN-O3: postmisched
-; GCN-O3: block-placement
-; GCN-O3: fentry-insert
-; GCN-O3: xray-instrumentation
-; GCN-O3: patchable-function
-; GCN-O3: gcn-create-vopd
-; GCN-O3: si-memory-legalizer
-; GCN-O3: si-insert-waitcnts
-; GCN-O3: si-late-branch-lowering
-; GCN-O3: si-pre-emit-peephole
-; GCN-O3: post-RA-hazard-rec
-; GCN-O3: AMDGPUWaitSGPRHazardsPass
-; GCN-O3: amdgpu-insert-delay-alu
-; GCN-O3: branch-relaxation
-; GCN-O3: remove-loads-into-fake-uses
-; GCN-O3: live-debug-values
-; GCN-O3: machine-sanmd
-; GCN-O3: stack-frame-layout)
-; GCN-O3: invalidate<machine-function-info>))
+; GCN-O0: require<MachineModuleAnalysis>,require<profile-summary>,require<collector-metadata>,pre-isel-intrinsic-lowering,function(expand-large-div-rem,expand-fp),amdgpu-remove-incompatible-functions,amdgpu-printf-runtime-binding,amdgpu-lower-ctor-dtor,expand-variadics,amdgpu-always-inline,always-inline,amdgpu-export-kernel-runtime-handles,amdgpu-sw-lower-lds,amdgpu-lower-module-lds,function(atomic-expand,verify,gc-lowering,lower-constant-intrinsics,UnreachableBlockElimPass,ee-instrument<post-inline>,scalarize-masked-mem-intrin,ExpandReductionsPass,amdgpu-lower-kernel-arguments),amdgpu-lower-buffer-fat-pointers,cgscc(function(lower-switch,lower-invoke,UnreachableBlockElimPass,amdgpu-unify-divergent-exit-nodes,fix-irreducible,unify-loop-exits,StructurizeCFGPass,amdgpu-annotate-uniform,si-annotate-control-flow,amdgpu-rewrite-undef-for-phi,lcssa,require<uniformity>,callbr-prepare,safe-stack,stack-protector,verify)),cgscc(function(machine-function(amdgpu-isel,si-fix-sgpr-copies,si-i1-copies,finalize-isel,localstackalloc,phi-node-elimination,two-address-instruction,regallocfast,si-fix-vgpr-copies,remove-redundant-debug-values,fixup-statepoint-caller-saved,prolog-epilog,post-ra-pseudos,fentry-insert,xray-instrumentation,patchable-function,si-memory-legalizer,si-insert-waitcnts,si-late-branch-lowering,post-RA-hazard-rec,AMDGPUWaitSGPRHazardsPass,branch-relaxation,remove-loads-into-fake-uses,live-debug-values,machine-sanmd,stack-frame-layout,verify),invalidate<machine-function-info>))
+
+
+; GCN-O2: require<MachineModuleAnalysis>,require<profile-summary>,require<collector-metadata>,pre-isel-intrinsic-lowering,function(expand-large-div-rem,expand-fp),amdgpu-remove-incompatible-functions,amdgpu-printf-runtime-binding,amdgpu-lower-ctor-dtor,function(amdgpu-image-intrinsic-opt),expand-variadics,amdgpu-always-inline,always-inline,amdgpu-export-kernel-runtime-handles,amdgpu-sw-lower-lds,amdgpu-lower-module-lds,function(infer-address-spaces,amdgpu-atomic-optimizer,atomic-expand,amdgpu-promote-alloca,separate-const-offset-from-gep<>,slsr,early-cse<>,nary-reassociate,early-cse<>,amdgpu-codegenprepare,verify,loop-mssa(loop-reduce),mergeicmps,expand-memcmp,gc-lowering,lower-constant-intrinsics,UnreachableBlockElimPass,consthoist,ReplaceWithVeclib,partially-inline-libcalls,ee-instrument<post-inline>,scalarize-masked-mem-intrin,ExpandReductionsPass,early-cse<>,amdgpu-lower-kernel-arguments),amdgpu-lower-buffer-fat-pointers,cgscc(function(codegenprepare,load-store-vectorizer,lower-switch,lower-invoke,UnreachableBlockElimPass,flatten-cfg,sink,amdgpu-late-codegenprepare,amdgpu-unify-divergent-exit-nodes,fix-irreducible,unify-loop-exits,StructurizeCFGPass,amdgpu-annotate-uniform,si-annotate-control-flow,amdgpu-rewrite-undef-for-phi,lcssa)),amdgpu-perf-hint,cgscc(function(require<uniformity>,callbr-prepare,safe-stack,stack-protector,verify)),cgscc(function(machine-function(amdgpu-isel,si-fix-sgpr-copies,si-i1-copies,finalize-isel,early-tailduplication,opt-phis,stack-coloring,localstackalloc,dead-mi-elimination,early-machinelicm,machine-cse,machine-sink,peephole-opt,dead-mi-elimination,si-fold-operands,gcn-dpp-combine,si-load-store-opt,si-peephole-sdwa,early-machinelicm,machine-cse,si-fold-operands,dead-mi-elimination,si-shrink-instructions,detect-dead-lanes,InitUndefPass,ProcessImplicitDefsPass,unreachable-mbb-elimination,require<live-vars>,require<machine-loops>,phi-node-elimination,two-address-instruction,register-coalescer,rename-independent-subregs,machine-scheduler,amdgpu-pre-ra-long-branch-reg,greedy<sgpr>,virt-reg-rewriter<no-clear-vregs>,stack-slot-coloring,si-lower-sgpr-spills,si-pre-allocate-wwm-regs,greedy<wwm>,si-lower-wwm-copies,virt-reg-rewriter<no-clear-vregs>,amdgpu-reserve-wwm-regs,greedy<vgpr>,amdgpu-nsa-reassign,virt-reg-rewriter,machine-cp,machinelicm,si-fix-vgpr-copies,si-optimize-exec-masking,remove-redundant-debug-values,fixup-statepoint-caller-saved,PostRAMachineSinkingPass,shrink-wrap,prolog-epilog,branch-folder,tailduplication,machine-latecleanup,machine-cp,post-ra-pseudos,postmisched,block-placement,fentry-insert,xray-instrumentation,patchable-function,gcn-create-vopd,si-memory-legalizer,si-insert-waitcnts,si-late-branch-lowering,si-pre-emit-peephole,post-RA-hazard-rec,AMDGPUWaitSGPRHazardsPass,amdgpu-insert-delay-alu,branch-relaxation,remove-loads-into-fake-uses,live-debug-values,machine-sanmd,stack-frame-layout,verify),invalidate<machine-function-info>))
+
+; GCN-O3: require<MachineModuleAnalysis>,require<profile-summary>,require<collector-metadata>,pre-isel-intrinsic-lowering,function(expand-large-div-rem,expand-fp),amdgpu-remove-incompatible-functions,amdgpu-printf-runtime-binding,amdgpu-lower-ctor-dtor,function(amdgpu-image-intrinsic-opt),expand-variadics,amdgpu-always-inline,always-inline,amdgpu-export-kernel-runtime-handles,amdgpu-sw-lower-lds,amdgpu-lower-module-lds,function(infer-address-spaces,amdgpu-atomic-optimizer,atomic-expand,amdgpu-promote-alloca,separate-const-offset-from-gep<>,slsr,gvn<>,nary-reassociate,early-cse<>,amdgpu-codegenprepare,verify,loop-mssa(loop-reduce),mergeicmps,expand-memcmp,gc-lowering,lower-constant-intrinsics,UnreachableBlockElimPass,consthoist,ReplaceWithVeclib,partially-inline-libcalls,ee-instrument<post-inline>,scalarize-masked-mem-intrin,ExpandReductionsPass,gvn<>,amdgpu-lower-kernel-arguments),amdgpu-lower-buffer-fat-pointers,cgscc(function(codegenprepare,load-store-vectorizer,lower-switch,lower-invoke,UnreachableBlockElimPass,flatten-cfg,sink,amdgpu-late-codegenprepare,amdgpu-unify-divergent-exit-nodes,fix-irreducible,unify-loop-exits,StructurizeCFGPass,amdgpu-annotate-uniform,si-annotate-control-flow,amdgpu-rewrite-undef-for-phi,lcssa)),amdgpu-perf-hint,cgscc(function(require<uniformity>,callbr-prepare,safe-stack,stack-protector,verify)),cgscc(function(machine-function(amdgpu-isel,si-fix-sgpr-copies,si-i1-copies,finalize-isel,early-tailduplication,opt-phis,stack-coloring,localstackalloc,dead-mi-elimination,early-machinelicm,machine-cse,machine-sink,peephole-opt,dead-mi-elimination,si-fold-operands,gcn-dpp-combine,si-load-store-opt,si-peephole-sdwa,early-machinelicm,machine-cse,si-fold-operands,dead-mi-elimination,si-shrink-instructions,detect-dead-lanes,InitUndefPass,ProcessImplicitDefsPass,unreachable-mbb-elimination,require<live-vars>,require<machine-loops>,phi-node-elimination,two-address-instruction,register-coalescer,rename-independent-subregs,machine-scheduler,amdgpu-pre-ra-long-branch-reg,greedy<sgpr>,virt-reg-rewriter<no-clear-vregs>,stack-slot-coloring,si-lower-sgpr-spills,si-pre-allocate-wwm-regs,greedy<wwm>,si-lower-wwm-copies,virt-reg-rewriter<no-clear-vregs>,amdgpu-reserve-wwm-regs,greedy<vgpr>,amdgpu-nsa-reassign,virt-reg-rewriter,machine-cp,machinelicm,si-fix-vgpr-copies,si-optimize-exec-masking,remove-redundant-debug-values,fixup-statepoint-caller-saved,PostRAMachineSinkingPass,shrink-wrap,prolog-epilog,branch-folder,tailduplication,machine-latecleanup,machine-cp,post-ra-pseudos,postmisched,block-placement,fentry-insert,xray-instrumentation,patchable-function,gcn-create-vopd,si-memory-legalizer,si-insert-waitcnts,si-late-branch-lowering,si-pre-emit-peephole,post-RA-hazard-rec,AMDGPUWaitSGPRHazardsPass,amdgpu-insert-delay-alu,branch-relaxation,remove-loads-into-fake-uses,live-debug-values,machine-sanmd,stack-frame-layout,verify),invalidate<machine-function-info>))
 
 
 define void @empty() {