I have a wild guess at a potential fix in the meantime:
diff --cc llvm/lib/Transforms/Utils/SimplifyCFG.cpp
index 3add561c66d5,3add561c66d5..bedbca9dd4b7
--- a/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
+++ b/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
@@@ -3242,9 -3242,9 +3242,9 @@@ bool llvm::FoldBranchToCommonDest(Branc
// as "bonus instructions", and only allow this transformation when the
// number of the bonus instructions we'll need to create when cloning
into
// each predecessor does not exceed a certain threshold.
-- unsigned NumBonusInsts = 0;
bool SawVectorOp = false;
const unsigned PredCount = Preds.size();
++ InstructionCost IC;
for (Instruction &I : *BB) {
// Don't check the branch condition comparison itself.
if (&I == Cond)
@@@ -3258,16 -3258,16 +3258,13 @@@
SawVectorOp |= isVectorOp(I);
// Account for the cost of duplicating this instruction into each
-- // predecessor. Ignore free instructions.
-- if (!TTI ||
-- TTI->getUserCost(&I, CostKind) != TargetTransformInfo::TCC_Free) {
-- NumBonusInsts += PredCount;
--
-- // Early exits once we reach the limit.
-- if (NumBonusInsts >
-- BonusInstThreshold * BranchFoldToCommonDestVectorMultiplier)
-- return false;
-- }
++ // predecessor.
++ IC += (TTI ? TTI->getUserCost(&I, CostKind)
++ : TargetTransformInfo::TCC_Basic) *
++ PredCount;
++ // Early exits once we reach the limit.
++ if (IC > BonusInstThreshold * BranchFoldToCommonDestVectorMultiplier)
++ return false;
auto IsBCSSAUse = [BB, &I](Use &U) {
auto *UI = cast<Instruction>(U.getUser());
@@@ -3280,9 -3280,9 +3277,8 @@@
if (!all_of(I.uses(), IsBCSSAUse))
return false;
}
-- if (NumBonusInsts >
-- BonusInstThreshold *
-- (SawVectorOp ? BranchFoldToCommonDestVectorMultiplier : 1))
++ if (IC > BonusInstThreshold *
++ (SawVectorOp ? BranchFoldToCommonDestVectorMultiplier : 1))
return false;
// Ok, we have the budget. Perform the transformation.
On Mon, Sep 27, 2021 at 10:05 AM Arthur Eubanks <aeuba...@google.com> wrote:
> Could I get the source file with S_regmatch()?
>
> On Mon, Sep 27, 2021 at 6:07 AM Maxim Kuvyrkov <maxim.kuvyr...@linaro.org>
> wrote:
>
>> Hi Arthur,
>>
>> Your patch seems to be slowing down 400.perlbench by 6% — due to slow
>> down of its hot function S_regmatch() by 14%.
>>
>> Could you take a look if this is easily fixable, please?
>>
>> Regards,
>>
>> --
>> Maxim Kuvyrkov
>> https://www.linaro.org
>>
>> > On 24 Sep 2021, at 15:07, ci_not...@linaro.org wrote:
>> >
>> > After llvm commit e7249e4acf3cf9438d6d9e02edecebd5b622a4dc
>> > Author: Arthur Eubanks <aeuba...@google.com>
>> >
>> > [SimplifyCFG] Ignore free instructions when computing cost for
>> folding branch to common dest
>> >
>> > the following benchmarks slowed down by more than 2%:
>> > - 400.perlbench slowed down by 6% from 9730 to 10312 perf samples
>> > - 400.perlbench:[.] S_regmatch slowed down by 14% from 3660 to 4188
>> perf samples
>> >
>> > Below reproducer instructions can be used to re-build both "first_bad"
>> and "last_good" cross-toolchains used in this bisection. Naturally, the
>> scripts will fail when triggerring benchmarking jobs if you don't have
>> access to Linaro TCWG CI.
>> >
>> > For your convenience, we have uploaded tarballs with pre-processed
>> source and assembly files at:
>> > - First_bad save-temps:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/build-e7249e4acf3cf9438d6d9e02edecebd5b622a4dc/save-temps/
>> > - Last_good save-temps:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/build-32a50078657dd8beead327a3478ede4e9d730432/save-temps/
>> > - Baseline save-temps:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/build-baseline/save-temps/
>> >
>> > Configuration:
>> > - Benchmark: SPEC CPU2006
>> > - Toolchain: Clang + Glibc + LLVM Linker
>> > - Version: all components were built from their tip of trunk
>> > - Target: aarch64-linux-gnu
>> > - Compiler flags: -O3
>> > - Hardware: NVidia TX1 4x Cortex-A57
>> >
>> > This benchmarking CI is work-in-progress, and we welcome feedback and
>> suggestions at linaro-toolchain@lists.linaro.org . In our improvement
>> plans is to add support for SPEC CPU2017 benchmarks and provide "perf
>> report/annotate" data behind these reports.
>> >
>> > THIS IS THE END OF INTERESTING STUFF. BELOW ARE LINKS TO BUILDS,
>> REPRODUCTION INSTRUCTIONS, AND THE RAW COMMIT.
>> >
>> > This commit has regressed these CI configurations:
>> > - tcwg_bmk_llvm_tx1/llvm-master-aarch64-spec2k6-O3
>> >
>> > First_bad build:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/build-e7249e4acf3cf9438d6d9e02edecebd5b622a4dc/
>> > Last_good build:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/build-32a50078657dd8beead327a3478ede4e9d730432/
>> > Baseline build:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/build-baseline/
>> > Even more details:
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/
>> >
>> > Reproduce builds:
>> > <cut>
>> > mkdir investigate-llvm-e7249e4acf3cf9438d6d9e02edecebd5b622a4dc
>> > cd investigate-llvm-e7249e4acf3cf9438d6d9e02edecebd5b622a4dc
>> >
>> > # Fetch scripts
>> > git clone https://git.linaro.org/toolchain/jenkins-scripts
>> >
>> > # Fetch manifests and test.sh script
>> > mkdir -p artifacts/manifests
>> > curl -o artifacts/manifests/build-baseline.sh
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/manifests/build-baseline.sh
>> --fail
>> > curl -o artifacts/manifests/build-parameters.sh
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/manifests/build-parameters.sh
>> --fail
>> > curl -o artifacts/test.sh
>> https://ci.linaro.org/job/tcwg_bmk_ci_llvm-bisect-tcwg_bmk_tx1-llvm-master-aarch64-spec2k6-O3/23/artifact/artifacts/test.sh
>> --fail
>> > chmod +x artifacts/test.sh
>> >
>> > # Reproduce the baseline build (build all pre-requisites)
>> > ./jenkins-scripts/tcwg_bmk-build.sh @@
>> artifacts/manifests/build-baseline.sh
>> >
>> > # Save baseline build state (which is then restored in
>> artifacts/test.sh)
>> > mkdir -p ./bisect
>> > rsync -a --del --delete-excluded --exclude /bisect/ --exclude
>> /artifacts/ --exclude /llvm/ ./ ./bisect/baseline/
>> >
>> > cd llvm
>> >
>> > # Reproduce first_bad build
>> > git checkout --detach e7249e4acf3cf9438d6d9e02edecebd5b622a4dc
>> > ../artifacts/test.sh
>> >
>> > # Reproduce last_good build
>> > git checkout --detach 32a50078657dd8beead327a3478ede4e9d730432
>> > ../artifacts/test.sh
>> >
>> > cd ..
>> > </cut>
>> >
>> > Full commit (up to 1000 lines):
>> > <cut>
>> > commit e7249e4acf3cf9438d6d9e02edecebd5b622a4dc
>> > Author: Arthur Eubanks <aeuba...@google.com>
>> > Date: Fri Aug 27 12:32:59 2021 -0700
>> >
>> > [SimplifyCFG] Ignore free instructions when computing cost for
>> folding branch to common dest
>> >
>> > When determining whether to fold branches to a common destination by
>> > merging two blocks, SimplifyCFG will count the number of
>> instructions to
>> > be moved into the first basic block. However, there's no reason to
>> count
>> > free instructions like bitcasts and other similar instructions.
>> >
>> > This resolves missed branch foldings with -fstrict-vtable-pointers in
>> > llvm-test-suite's lambda benchmark.
>> >
>> > Reviewed By: spatel
>> >
>> > Differential Revision: https://reviews.llvm.org/D108837
>> > ---
>> > llvm/lib/Transforms/Utils/SimplifyCFG.cpp | 17 ++++++-----
>> > llvm/test/CodeGen/AArch64/csr-split.ll | 34
>> +++++++++++-----------
>> > .../fold-branch-to-common-dest-free-cost.ll | 5 ++--
>> > 3 files changed, 29 insertions(+), 27 deletions(-)
>> >
>> > diff --git a/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
>> b/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
>> > index 2ff98b238de0..a3bd89e72af9 100644
>> > --- a/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
>> > +++ b/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
>> > @@ -3258,13 +3258,16 @@ bool llvm::FoldBranchToCommonDest(BranchInst
>> *BI, DomTreeUpdater *DTU,
>> > SawVectorOp |= isVectorOp(I);
>> >
>> > // Account for the cost of duplicating this instruction into each
>> > - // predecessor.
>> > - NumBonusInsts += PredCount;
>> > -
>> > - // Early exits once we reach the limit.
>> > - if (NumBonusInsts >
>> > - BonusInstThreshold * BranchFoldToCommonDestVectorMultiplier)
>> > - return false;
>> > + // predecessor. Ignore free instructions.
>> > + if (!TTI ||
>> > + TTI->getUserCost(&I, CostKind) !=
>> TargetTransformInfo::TCC_Free) {
>> > + NumBonusInsts += PredCount;
>> > +
>> > + // Early exits once we reach the limit.
>> > + if (NumBonusInsts >
>> > + BonusInstThreshold * BranchFoldToCommonDestVectorMultiplier)
>> > + return false;
>> > + }
>> >
>> > auto IsBCSSAUse = [BB, &I](Use &U) {
>> > auto *UI = cast<Instruction>(U.getUser());
>> > diff --git a/llvm/test/CodeGen/AArch64/csr-split.ll
>> b/llvm/test/CodeGen/AArch64/csr-split.ll
>> > index 1bee7f05acec..de85b4313433 100644
>> > --- a/llvm/test/CodeGen/AArch64/csr-split.ll
>> > +++ b/llvm/test/CodeGen/AArch64/csr-split.ll
>> > @@ -82,22 +82,22 @@ define dso_local signext i32 @test2(i32* %p1)
>> local_unnamed_addr {
>> > ; CHECK-NEXT: .cfi_def_cfa_offset 16
>> > ; CHECK-NEXT: .cfi_offset w19, -8
>> > ; CHECK-NEXT: .cfi_offset w30, -16
>> > -; CHECK-NEXT: cbz x0, .LBB1_2
>> > -; CHECK-NEXT: // %bb.1: // %if.end
>> > +; CHECK-NEXT: cbz x0, .LBB1_3
>> > +; CHECK-NEXT: // %bb.1: // %entry
>> > ; CHECK-NEXT: adrp x8, a
>> > ; CHECK-NEXT: ldrsw x8, [x8, :lo12:a]
>> > ; CHECK-NEXT: mov x19, x0
>> > ; CHECK-NEXT: cmp x8, x0
>> > -; CHECK-NEXT: b.eq .LBB1_3
>> > -; CHECK-NEXT: .LBB1_2: // %return
>> > -; CHECK-NEXT: mov w0, wzr
>> > -; CHECK-NEXT: ldp x30, x19, [sp], #16 // 16-byte Folded Reload
>> > -; CHECK-NEXT: ret
>> > -; CHECK-NEXT: .LBB1_3: // %if.then2
>> > +; CHECK-NEXT: b.ne .LBB1_3
>> > +; CHECK-NEXT: // %bb.2: // %if.then2
>> > ; CHECK-NEXT: bl callVoid
>> > ; CHECK-NEXT: mov x0, x19
>> > ; CHECK-NEXT: ldp x30, x19, [sp], #16 // 16-byte Folded Reload
>> > ; CHECK-NEXT: b callNonVoid
>> > +; CHECK-NEXT: .LBB1_3: // %return
>> > +; CHECK-NEXT: mov w0, wzr
>> > +; CHECK-NEXT: ldp x30, x19, [sp], #16 // 16-byte Folded Reload
>> > +; CHECK-NEXT: ret
>> > ;
>> > ; CHECK-APPLE-LABEL: test2:
>> > ; CHECK-APPLE: ; %bb.0: ; %entry
>> > @@ -108,26 +108,26 @@ define dso_local signext i32 @test2(i32* %p1)
>> local_unnamed_addr {
>> > ; CHECK-APPLE-NEXT: .cfi_offset w29, -16
>> > ; CHECK-APPLE-NEXT: .cfi_offset w19, -24
>> > ; CHECK-APPLE-NEXT: .cfi_offset w20, -32
>> > -; CHECK-APPLE-NEXT: cbz x0, LBB1_2
>> > -; CHECK-APPLE-NEXT: ; %bb.1: ; %if.end
>> > +; CHECK-APPLE-NEXT: cbz x0, LBB1_3
>> > +; CHECK-APPLE-NEXT: ; %bb.1: ; %entry
>> > ; CHECK-APPLE-NEXT: Lloh2:
>> > ; CHECK-APPLE-NEXT: adrp x8, _a@PAGE
>> > ; CHECK-APPLE-NEXT: Lloh3:
>> > ; CHECK-APPLE-NEXT: ldrsw x8, [x8, _a@PAGEOFF]
>> > ; CHECK-APPLE-NEXT: mov x19, x0
>> > ; CHECK-APPLE-NEXT: cmp x8, x0
>> > -; CHECK-APPLE-NEXT: b.eq LBB1_3
>> > -; CHECK-APPLE-NEXT: LBB1_2: ; %return
>> > -; CHECK-APPLE-NEXT: ldp x29, x30, [sp, #16] ; 16-byte Folded Reload
>> > -; CHECK-APPLE-NEXT: mov w0, wzr
>> > -; CHECK-APPLE-NEXT: ldp x20, x19, [sp], #32 ; 16-byte Folded Reload
>> > -; CHECK-APPLE-NEXT: ret
>> > -; CHECK-APPLE-NEXT: LBB1_3: ; %if.then2
>> > +; CHECK-APPLE-NEXT: b.ne LBB1_3
>> > +; CHECK-APPLE-NEXT: ; %bb.2: ; %if.then2
>> > ; CHECK-APPLE-NEXT: bl _callVoid
>> > ; CHECK-APPLE-NEXT: ldp x29, x30, [sp, #16] ; 16-byte Folded Reload
>> > ; CHECK-APPLE-NEXT: mov x0, x19
>> > ; CHECK-APPLE-NEXT: ldp x20, x19, [sp], #32 ; 16-byte Folded Reload
>> > ; CHECK-APPLE-NEXT: b _callNonVoid
>> > +; CHECK-APPLE-NEXT: LBB1_3: ; %return
>> > +; CHECK-APPLE-NEXT: ldp x29, x30, [sp, #16] ; 16-byte Folded Reload
>> > +; CHECK-APPLE-NEXT: mov w0, wzr
>> > +; CHECK-APPLE-NEXT: ldp x20, x19, [sp], #32 ; 16-byte Folded Reload
>> > +; CHECK-APPLE-NEXT: ret
>> > ; CHECK-APPLE-NEXT: .loh AdrpLdr Lloh2, Lloh3
>> > entry:
>> > %tobool = icmp eq i32* %p1, null
>> > diff --git
>> a/llvm/test/Transforms/SimplifyCFG/fold-branch-to-common-dest-free-cost.ll
>> b/llvm/test/Transforms/SimplifyCFG/fold-branch-to-common-dest-free-cost.ll
>> > index ace2a5ed35ca..27df5ec44582 100644
>> > ---
>> a/llvm/test/Transforms/SimplifyCFG/fold-branch-to-common-dest-free-cost.ll
>> > +++
>> b/llvm/test/Transforms/SimplifyCFG/fold-branch-to-common-dest-free-cost.ll
>> > @@ -8,12 +8,11 @@ declare void @g2()
>> >
>> > define void @f(i8* %a, i8* %b, i1 %c, i1 %d, i1 %e) {
>> > ; CHECK-LABEL: @f(
>> > -; CHECK-NEXT: br i1 [[C:%.*]], label [[L1:%.*]], label [[L3:%.*]]
>> > -; CHECK: l1:
>> > ; CHECK-NEXT: [[A1:%.*]] = call i8*
>> @llvm.strip.invariant.group.p0i8(i8* [[A:%.*]])
>> > ; CHECK-NEXT: [[B1:%.*]] = call i8*
>> @llvm.strip.invariant.group.p0i8(i8* [[B:%.*]])
>> > ; CHECK-NEXT: [[I:%.*]] = icmp eq i8* [[A1]], [[B1]]
>> > -; CHECK-NEXT: br i1 [[I]], label [[L2:%.*]], label [[L3]]
>> > +; CHECK-NEXT: [[OR_COND:%.*]] = select i1 [[C:%.*]], i1 [[I]], i1
>> false
>> > +; CHECK-NEXT: br i1 [[OR_COND]], label [[L2:%.*]], label [[L3:%.*]]
>> > ; CHECK: l2:
>> > ; CHECK-NEXT: call void @g1()
>> > ; CHECK-NEXT: br label [[RET:%.*]]
>> > </cut>
>>
>>
_______________________________________________
linaro-toolchain mailing list
linaro-toolchain@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/linaro-toolchain
