caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
be1294de9dd93ae02002bd1aa4b4a8d62e562c21
clang-p2996/llvm/test/CodeGen/AArch64/vldn_shuffle.ll
David Green adec922361 [AArch64] Make -mcpu=generic schedule for an in-order core 
We would like to start pushing -mcpu=generic towards enabling the set of
features that improves performance for some CPUs, without hurting any
others. A blend of the performance options hopefully beneficial to all
CPUs. The largest part of that is enabling in-order scheduling using the
Cortex-A55 schedule model. This is similar to the Arm backend change
from eecb353d0e which made -mcpu=generic perform in-order scheduling
using the cortex-a8 schedule model.

The idea is that in-order cpu's require the most help in instruction
scheduling, whereas out-of-order cpus can for the most part out-of-order
schedule around different codegen. Our benchmarking suggests that
hypothesis holds. When running on an in-order core this improved
performance by 3.8% geomean on a set of DSP workloads, 2% geomean on
some other embedded benchmark and between 1% and 1.8% on a set of
singlecore and multicore workloads, all running on a Cortex-A55 cluster.

On an out-of-order cpu the results are a lot more noisy but show flat
performance or an improvement. On the set of DSP and embedded
benchmarks, run on a Cortex-A78 there was a very noisy 1% speed
improvement. Using the most detailed results I could find, SPEC2006 runs
on a Neoverse N1 show a small increase in instruction count (+0.127%),
but a decrease in cycle counts (-0.155%, on average). The instruction
count is very low noise, the cycle count is more noisy with a 0.15%
decrease not being significant. SPEC2k17 shows a small decrease (-0.2%)
in instruction count leading to a -0.296% decrease in cycle count. These
results are within noise margins but tend to show a small improvement in
general.

When specifying an Apple target, clang will set "-target-cpu apple-a7"
on the command line, so should not be affected by this change when
running from clang. This also doesn't enable more runtime unrolling like
-mcpu=cortex-a55 does, only changing the schedule used.

A lot of existing tests have updated. This is a summary of the important
differences:
 - Most changes are the same instructions in a different order.
 - Sometimes this leads to very minor inefficiencies, such as requiring
   an extra mov to move variables into r0/v0 for the return value of a test
   function.
 - misched-fusion.ll was no longer fusing the pairs of instructions it
   should, as per D110561. I've changed the schedule used in the test
   for now.
 - neon-mla-mls.ll now uses "mul; sub" as opposed to "neg; mla" due to
   the different latencies. This seems fine to me.
 - Some SVE tests do not always remove movprfx where they did before due
   to different register allocation giving different destructive forms.
 - The tests argument-blocks-array-of-struct.ll and arm64-windows-calls.ll
   produce two LDR where they previously produced an LDP due to
   store-pair-suppress kicking in.
 - arm64-ldp.ll and arm64-neon-copy.ll are missing pre/postinc on LPD.
 - Some tests such as arm64-neon-mul-div.ll and
   ragreedy-local-interval-cost.ll have more, less or just different
   spilling.
 - In aarch64_generated_funcs.ll.generated.expected one part of the
   function is no longer outlined. Interestingly if I switch this to use
   any other scheduled even less is outlined.

Some of these are expected to happen, such as differences in outlining
or register spilling. There will be places where these result in worse
codegen, places where they are better, with the SPEC instruction counts
suggesting it is not a decrease overall, on average.

Differential Revision: https://reviews.llvm.org/D110830
2021-10-09 15:58:31 +01:00

184 lines
7.9 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=aarch64-none-eabif | FileCheck %s
define void @vld2(float* nocapture readonly %pSrc, float* noalias nocapture %pDst, i32 %numSamples) {
; CHECK-LABEL: vld2:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: mov x8, xzr
; CHECK-NEXT: .LBB0_1: // %vector.body
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: ld2 { v0.4s, v1.4s }, [x0], #32
; CHECK-NEXT: fmul v2.4s, v0.4s, v0.4s
; CHECK-NEXT: fmla v2.4s, v1.4s, v1.4s
; CHECK-NEXT: str q2, [x1, x8]
; CHECK-NEXT: add x8, x8, #16
; CHECK-NEXT: cmp x8, #1, lsl #12 // =4096
; CHECK-NEXT: b.ne .LBB0_1
; CHECK-NEXT: // %bb.2: // %while.end
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%0 = shl i64 %index, 1
%next.gep = getelementptr float, float* %pSrc, i64 %0
%next.gep19 = getelementptr float, float* %pDst, i64 %index
%1 = bitcast float* %next.gep to <8 x float>*
%wide.vec = load <8 x float>, <8 x float>* %1, align 4
%2 = fmul fast <8 x float> %wide.vec, %wide.vec
%3 = shufflevector <8 x float> %2, <8 x float> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
%4 = fmul fast <8 x float> %wide.vec, %wide.vec
%5 = shufflevector <8 x float> %4, <8 x float> undef, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
%6 = fadd fast <4 x float> %5, %3
%7 = bitcast float* %next.gep19 to <4 x float>*
store <4 x float> %6, <4 x float>* %7, align 4
%index.next = add i64 %index, 4
%8 = icmp eq i64 %index.next, 1024
br i1 %8, label %while.end, label %vector.body
while.end: ; preds = %vector.body
ret void
}
define void @vld3(float* nocapture readonly %pSrc, float* noalias nocapture %pDst, i32 %numSamples) {
; CHECK-LABEL: vld3:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: mov x8, xzr
; CHECK-NEXT: .LBB1_1: // %vector.body
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: ld3 { v0.4s, v1.4s, v2.4s }, [x0], #48
; CHECK-NEXT: fmul v3.4s, v0.4s, v0.4s
; CHECK-NEXT: fmla v3.4s, v1.4s, v1.4s
; CHECK-NEXT: fmla v3.4s, v2.4s, v2.4s
; CHECK-NEXT: str q3, [x1, x8]
; CHECK-NEXT: add x8, x8, #16
; CHECK-NEXT: cmp x8, #1, lsl #12 // =4096
; CHECK-NEXT: b.ne .LBB1_1
; CHECK-NEXT: // %bb.2: // %while.end
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%0 = mul i64 %index, 3
%next.gep = getelementptr float, float* %pSrc, i64 %0
%next.gep23 = getelementptr float, float* %pDst, i64 %index
%1 = bitcast float* %next.gep to <12 x float>*
%wide.vec = load <12 x float>, <12 x float>* %1, align 4
%2 = fmul fast <12 x float> %wide.vec, %wide.vec
%3 = shufflevector <12 x float> %2, <12 x float> undef, <4 x i32> <i32 0, i32 3, i32 6, i32 9>
%4 = fmul fast <12 x float> %wide.vec, %wide.vec
%5 = shufflevector <12 x float> %4, <12 x float> undef, <4 x i32> <i32 1, i32 4, i32 7, i32 10>
%6 = fadd fast <4 x float> %5, %3
%7 = fmul fast <12 x float> %wide.vec, %wide.vec
%8 = shufflevector <12 x float> %7, <12 x float> undef, <4 x i32> <i32 2, i32 5, i32 8, i32 11>
%9 = fadd fast <4 x float> %6, %8
%10 = bitcast float* %next.gep23 to <4 x float>*
store <4 x float> %9, <4 x float>* %10, align 4
%index.next = add i64 %index, 4
%11 = icmp eq i64 %index.next, 1024
br i1 %11, label %while.end, label %vector.body
while.end: ; preds = %vector.body
ret void
}
define void @vld4(float* nocapture readonly %pSrc, float* noalias nocapture %pDst, i32 %numSamples) {
; CHECK-LABEL: vld4:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: mov x8, xzr
; CHECK-NEXT: .LBB2_1: // %vector.body
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: ld4 { v0.4s, v1.4s, v2.4s, v3.4s }, [x0], #64
; CHECK-NEXT: fmul v4.4s, v0.4s, v0.4s
; CHECK-NEXT: add x9, x1, x8
; CHECK-NEXT: add x8, x8, #32
; CHECK-NEXT: cmp x8, #2, lsl #12 // =8192
; CHECK-NEXT: fmla v4.4s, v1.4s, v1.4s
; CHECK-NEXT: fmul v5.4s, v2.4s, v2.4s
; CHECK-NEXT: fmla v5.4s, v3.4s, v3.4s
; CHECK-NEXT: st2 { v4.4s, v5.4s }, [x9]
; CHECK-NEXT: b.ne .LBB2_1
; CHECK-NEXT: // %bb.2: // %while.end
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%0 = shl i64 %index, 2
%next.gep = getelementptr float, float* %pSrc, i64 %0
%1 = shl i64 %index, 1
%2 = bitcast float* %next.gep to <16 x float>*
%wide.vec = load <16 x float>, <16 x float>* %2, align 4
%3 = fmul fast <16 x float> %wide.vec, %wide.vec
%4 = shufflevector <16 x float> %3, <16 x float> undef, <4 x i32> <i32 0, i32 4, i32 8, i32 12>
%5 = fmul fast <16 x float> %wide.vec, %wide.vec
%6 = shufflevector <16 x float> %5, <16 x float> undef, <4 x i32> <i32 1, i32 5, i32 9, i32 13>
%7 = fadd fast <4 x float> %6, %4
%8 = fmul fast <16 x float> %wide.vec, %wide.vec
%9 = shufflevector <16 x float> %8, <16 x float> undef, <4 x i32> <i32 2, i32 6, i32 10, i32 14>
%10 = fmul fast <16 x float> %wide.vec, %wide.vec
%11 = shufflevector <16 x float> %10, <16 x float> undef, <4 x i32> <i32 3, i32 7, i32 11, i32 15>
%12 = fadd fast <4 x float> %11, %9
%13 = getelementptr inbounds float, float* %pDst, i64 %1
%14 = bitcast float* %13 to <8 x float>*
%interleaved.vec = shufflevector <4 x float> %7, <4 x float> %12, <8 x i32> <i32 0, i32 4, i32 1, i32 5, i32 2, i32 6, i32 3, i32 7>
store <8 x float> %interleaved.vec, <8 x float>* %14, align 4
%index.next = add i64 %index, 4
%15 = icmp eq i64 %index.next, 1024
br i1 %15, label %while.end, label %vector.body
while.end: ; preds = %vector.body
ret void
}
define void @twosrc(float* nocapture readonly %pSrc, float* nocapture readonly %pSrc2, float* noalias nocapture %pDst, i32 %numSamples) {
; CHECK-LABEL: twosrc:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: mov x8, xzr
; CHECK-NEXT: .LBB3_1: // %vector.body
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: add x9, x0, x8
; CHECK-NEXT: ld2 { v0.4s, v1.4s }, [x9]
; CHECK-NEXT: add x9, x1, x8
; CHECK-NEXT: add x8, x8, #32
; CHECK-NEXT: cmp x8, #2, lsl #12 // =8192
; CHECK-NEXT: ld2 { v2.4s, v3.4s }, [x9]
; CHECK-NEXT: fmul v4.4s, v2.4s, v0.4s
; CHECK-NEXT: fmla v4.4s, v1.4s, v3.4s
; CHECK-NEXT: str q4, [x2], #16
; CHECK-NEXT: b.ne .LBB3_1
; CHECK-NEXT: // %bb.2: // %while.end
; CHECK-NEXT: ret
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
%index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
%0 = shl i64 %index, 1
%next.gep = getelementptr float, float* %pSrc, i64 %0
%1 = shl i64 %index, 1
%next.gep23 = getelementptr float, float* %pSrc2, i64 %1
%next.gep24 = getelementptr float, float* %pDst, i64 %index
%2 = bitcast float* %next.gep to <8 x float>*
%wide.vec = load <8 x float>, <8 x float>* %2, align 4
%3 = bitcast float* %next.gep23 to <8 x float>*
%wide.vec26 = load <8 x float>, <8 x float>* %3, align 4
%4 = fmul fast <8 x float> %wide.vec26, %wide.vec
%5 = shufflevector <8 x float> %4, <8 x float> undef, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
%6 = fmul fast <8 x float> %wide.vec26, %wide.vec
%7 = shufflevector <8 x float> %6, <8 x float> undef, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
%8 = fadd fast <4 x float> %7, %5
%9 = bitcast float* %next.gep24 to <4 x float>*
store <4 x float> %8, <4 x float>* %9, align 4
%index.next = add i64 %index, 4
%10 = icmp eq i64 %index.next, 1024
br i1 %10, label %while.end, label %vector.body
while.end: ; preds = %vector.body
ret void
}
Reference in New Issue View Git Blame Copy Permalink