Skip to content

LLVM optimization on PR 157983 [bot] #67

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.

Sign up for GitHub

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

Open
wants to merge 1 commit into
base: pr-157983-before-1758677361
Choose a base branch
from
Open

LLVM optimization on PR 157983 [bot] #67

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
677 changes: 335 additions & 342 deletions ...Benchmarks/LCALS/SubsetALambdaLoops/CMakeFiles/lcalsALambda.dir/LambdaSubsetAbenchmarks.s
View file
Open in desktop

Large diffs are not rendered by default.

677 changes: 335 additions & 342 deletions ...lts/MicroBenchmarks/LCALS/SubsetARawLoops/CMakeFiles/lcalsARaw.dir/RawSubsetAbenchmarks.s
View file
Open in desktop

Large diffs are not rendered by default.

24 changes: 12 additions & 12 deletions ...Benchmarks/LoopVectorization/CMakeFiles/LoopInterleavingBenchmarks.dir/LoopInterleaving.s
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1608,15 +1608,15 @@ _ZL27loopWithReductionWithVW1IC4i: # @_ZL27loopWithReductionWithVW1IC4i
addi.d $a1, $a1, 16
bnez $a3, .LBB21_5
# %bb.6: # %middle.block
vpickve2gr.w $a1, $vr0, 0
vpickve2gr.w $a3, $vr0, 1
add.d $a1, $a3, $a1
vpickve2gr.w $a3, $vr0, 2
add.d $a1, $a3, $a1
vreplvei.w $vr1, $vr0, 1
vadd.w $vr1, $vr1, $vr0
vreplvei.w $vr2, $vr0, 2
vadd.w $vr1, $vr2, $vr1
vpickve2gr.w $a1, $vr1, 0
vpickve2gr.w $a3, $vr0, 3
add.w $a1, $a3, $a1
beq $a2, $a0, .LBB21_9
.LBB21_7: # %.lr.ph.preheader14
.LBB21_7: # %.lr.ph.preheader18
pcalau12i $a3, %pc_hi20(A)
addi.d $a3, $a3, %pc_lo12(A)
alsl.d $a3, $a2, $a3, 2
Expand Down Expand Up @@ -3709,15 +3709,15 @@ _ZL30bigLoopWithReductionWithVW1IC4i: # @_ZL30bigLoopWithReductionWithVW1IC4i
addi.d $a1, $a1, 16
bnez $a7, .LBB42_5
# %bb.6: # %middle.block
vpickve2gr.w $a1, $vr0, 0
vpickve2gr.w $a4, $vr0, 1
add.d $a1, $a4, $a1
vpickve2gr.w $a4, $vr0, 2
add.d $a1, $a4, $a1
vreplvei.w $vr1, $vr0, 1
vadd.w $vr1, $vr1, $vr0
vreplvei.w $vr2, $vr0, 2
vadd.w $vr1, $vr2, $vr1
vpickve2gr.w $a1, $vr1, 0
vpickve2gr.w $a4, $vr0, 3
add.w $a1, $a4, $a1
beq $a3, $a0, .LBB42_9
.LBB42_7: # %.lr.ph.preheader17
.LBB42_7: # %.lr.ph.preheader21
pcalau12i $a4, %pc_hi20(A)
addi.d $a4, $a4, %pc_lo12(A)
alsl.d $a4, $a3, $a4, 2
Expand Down
86 changes: 37 additions & 49 deletions ...s/MicroBenchmarks/SLPVectorization/CMakeFiles/SLPVectorizationBenchmarks.dir/Versioning.s
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1042,14 +1042,11 @@ _Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj2EdEvRN9benchmark5S
.p2align 4, , 16
.LBB14_3: # %.lr.ph
# =>This Inner Loop Header: Depth=1
fld.d $fa0, $sp, 32
fld.d $fa1, $sp, 16
fld.d $fa2, $sp, 24
fld.d $fa3, $sp, 40
fmadd.d $fa1, $fa0, $fa1, $fa0
fst.d $fa1, $sp, 32
fmadd.d $fa0, $fa0, $fa2, $fa3
fst.d $fa0, $sp, 40
vld $vr0, $sp, 32
vld $vr1, $sp, 16
vreplvei.d $vr2, $vr0, 0
vfmadd.d $vr0, $vr2, $vr1, $vr0
vst $vr0, $sp, 32
#APP
#NO_APP
#APP
Expand Down Expand Up @@ -1209,18 +1206,18 @@ _Z49benchmark_multiply_accumulate_runtime_checks_failILj3EdEvRN9benchmark5StateE
_Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj3EdEvRN9benchmark5StateE: # @_Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj3EdEvRN9benchmark5StateE
.cfi_startproc
# %bb.0: # %_ZN9benchmark5State13StateIteratorC2EPS0_.exit
addi.d $sp, $sp, -96
.cfi_def_cfa_offset 96
st.d $ra, $sp, 88 # 8-byte Folded Spill
st.d $fp, $sp, 80 # 8-byte Folded Spill
st.d $s0, $sp, 72 # 8-byte Folded Spill
st.d $s1, $sp, 64 # 8-byte Folded Spill
addi.d $sp, $sp, -112
.cfi_def_cfa_offset 112
st.d $ra, $sp, 104 # 8-byte Folded Spill
st.d $fp, $sp, 96 # 8-byte Folded Spill
st.d $s0, $sp, 88 # 8-byte Folded Spill
st.d $s1, $sp, 80 # 8-byte Folded Spill
.cfi_offset 1, -8
.cfi_offset 22, -16
.cfi_offset 23, -24
.cfi_offset 24, -32
move $fp, $a0
addi.d $a0, $sp, 40
addi.d $a0, $sp, 48
ori $a1, $zero, 3
pcaddu18i $ra, %call36(_ZL9init_dataIdEvPT_j)
jirl $ra, $ra, 0
Expand All @@ -1237,23 +1234,21 @@ _Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj3EdEvRN9benchmark5S
# %bb.1: # %_ZN9benchmark5State13StateIteratorC2EPS0_.exit
beqz $s0, .LBB17_4
# %bb.2:
addi.d $a0, $sp, 40
addi.d $a0, $sp, 48
addi.d $a1, $sp, 16
.p2align 4, , 16
.LBB17_3: # %.lr.ph
# =>This Inner Loop Header: Depth=1
fld.d $fa0, $sp, 40
fld.d $fa1, $sp, 16
fmadd.d $fa1, $fa0, $fa1, $fa0
fld.d $fa2, $sp, 24
fld.d $fa3, $sp, 48
fld.d $fa4, $sp, 32
fld.d $fa5, $sp, 56
fst.d $fa1, $sp, 40
fmadd.d $fa1, $fa0, $fa2, $fa3
fst.d $fa1, $sp, 48
fmadd.d $fa0, $fa0, $fa4, $fa5
fst.d $fa0, $sp, 56
vld $vr0, $sp, 16
vld $vr1, $sp, 48
fld.d $fa2, $sp, 48
fld.d $fa3, $sp, 32
fld.d $fa4, $sp, 64
vreplvei.d $vr5, $vr1, 0
vfmadd.d $vr0, $vr5, $vr0, $vr1
vst $vr0, $sp, 48
fmadd.d $fa0, $fa2, $fa3, $fa4
fst.d $fa0, $sp, 64
#APP
#NO_APP
#APP
Expand All @@ -1265,11 +1260,11 @@ _Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj3EdEvRN9benchmark5S
move $a0, $fp
pcaddu18i $ra, %call36(_ZN9benchmark5State17FinishKeepRunningEv)
jirl $ra, $ra, 0
ld.d $s1, $sp, 64 # 8-byte Folded Reload
ld.d $s0, $sp, 72 # 8-byte Folded Reload
ld.d $fp, $sp, 80 # 8-byte Folded Reload
ld.d $ra, $sp, 88 # 8-byte Folded Reload
addi.d $sp, $sp, 96
ld.d $s1, $sp, 80 # 8-byte Folded Reload
ld.d $s0, $sp, 88 # 8-byte Folded Reload
ld.d $fp, $sp, 96 # 8-byte Folded Reload
ld.d $ra, $sp, 104 # 8-byte Folded Reload
addi.d $sp, $sp, 112
ret
.Lfunc_end17:
.size _Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj3EdEvRN9benchmark5StateE, .Lfunc_end17-_Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj3EdEvRN9benchmark5StateE
Expand Down Expand Up @@ -1446,22 +1441,15 @@ _Z54benchmark_multiply_accumulate_no_runtime_checks_neededILj4EdEvRN9benchmark5S
.p2align 4, , 16
.LBB20_3: # %.lr.ph
# =>This Inner Loop Header: Depth=1
fld.d $fa0, $sp, 48
fld.d $fa1, $sp, 16
fld.d $fa2, $sp, 24
fld.d $fa3, $sp, 56
fmadd.d $fa1, $fa0, $fa1, $fa0
fst.d $fa1, $sp, 48
fmadd.d $fa1, $fa0, $fa2, $fa3
fld.d $fa2, $sp, 32
fld.d $fa3, $sp, 64
fld.d $fa4, $sp, 40
fld.d $fa5, $sp, 72
fst.d $fa1, $sp, 56
fmadd.d $fa1, $fa0, $fa2, $fa3
fst.d $fa1, $sp, 64
fmadd.d $fa0, $fa0, $fa4, $fa5
fst.d $fa0, $sp, 72
vld $vr0, $sp, 48
vld $vr1, $sp, 16
vld $vr2, $sp, 32
vld $vr3, $sp, 64
vreplvei.d $vr4, $vr0, 0
vfmadd.d $vr0, $vr4, $vr1, $vr0
vst $vr0, $sp, 48
vfmadd.d $vr0, $vr4, $vr2, $vr3
vst $vr0, $sp, 64
#APP
#NO_APP
#APP
Expand Down
24 changes: 12 additions & 12 deletions results/MultiSource/Applications/ClamAV/CMakeFiles/clamscan.dir/libclamav_special.s
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,19 +50,19 @@ cli_check_mydoom_log: # @cli_check_mydoom_log
vshuf4i.b $vr0, $vr0, 27
vreplgr2vr.w $vr1, $a0
vxor.v $vr0, $vr0, $vr1
vpickve2gr.w $a1, $vr0, 0
vpickve2gr.w $a2, $vr0, 1
add.d $a1, $a2, $a1
vpickve2gr.w $a2, $vr0, 2
ld.w $a3, $sp, 28
add.d $a1, $a2, $a1
vpickve2gr.w $a2, $vr0, 3
add.d $a1, $a2, $a1
revb.2w $a2, $a3
vreplvei.w $vr1, $vr0, 1
vadd.w $vr1, $vr1, $vr0
vreplvei.w $vr2, $vr0, 2
vadd.w $vr1, $vr2, $vr1
ld.w $a1, $sp, 28
vpickve2gr.w $a2, $vr1, 0
vpickve2gr.w $a3, $vr0, 3
add.d $a2, $a3, $a2
revb.2w $a1, $a1
ld.w $a3, $sp, 32
xor $a2, $a2, $a0
st.w $a2, $sp, 28
add.d $a1, $a2, $a1
xor $a1, $a1, $a0
st.w $a1, $sp, 28
add.d $a1, $a1, $a2
revb.2w $a2, $a3
xor $a2, $a2, $a0
ld.w $a3, $sp, 36
Expand Down
22 changes: 9 additions & 13 deletions results/MultiSource/Applications/JM/ldecod/CMakeFiles/ldecod.dir/erc_do_p.s
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1307,24 +1307,22 @@ conceal_lost_frames: # @conceal_lost_frames
move $s0, $a0
add.d $s4, $a0, $s5
st.w $s6, $s4, 76
st.w $s3, $s4, 8
stx.w $s3, $a0, $s5
st.w $zero, $s4, 32
st.d $s6, $s4, 24
ld.d $a0, $sp, 80 # 8-byte Folded Reload
ldx.w $a0, $fp, $a0
ld.d $a1, $sp, 72 # 8-byte Folded Reload
ldx.w $a1, $fp, $a1
st.w $s6, $s4, 272
st.w $zero, $s4, 216
st.w $s3, $s4, 8
stx.w $s3, $s0, $s5
stptr.w $s3, $fp, 5676
add.d $a1, $a1, $a0
st.w $a1, $s0, 8
vreplgr2vr.w $vr0, $a1
vst $vr0, $s0, 4
ld.d $a2, $sp, 64 # 8-byte Folded Reload
ld.w $a0, $a2, 28
st.w $a1, $s0, 12
st.w $a1, $s0, 16
st.w $a1, $s0, 4
st.w $zero, $s4, 32
st.w $s6, $s4, 272
st.w $zero, $s4, 216
addi.w $a0, $a0, -1
stptr.w $a1, $fp, 6056
bltz $a0, .LBB3_12
Expand Down Expand Up @@ -2263,11 +2261,9 @@ conceal_non_ref_pics: # @conceal_non_ref_pics
# %bb.10: # in Loop: Header=BB18_8 Depth=1
move $s1, $a0
stptr.w $s8, $a2, 6072
vreplgr2vr.w $vr0, $s8
vst $vr0, $a0, 4
ldptr.w $a0, $a2, 6068
st.w $s8, $s1, 8
st.w $s8, $s1, 12
st.w $s8, $s1, 16
st.w $s8, $s1, 4
ori $a4, $zero, 2
beq $a0, $a4, .LBB18_13
# %bb.11: # in Loop: Header=BB18_8 Depth=1
Expand Down
Loading