Apple M1 Microarchitecture Research by Dougall Johnson

Firestorm: Overview | Base Instructions | SIMD and FP Instructions
Icestorm:  Overview | Base Instructions | SIMD and FP Instructions

SQRDMLAH (scalar, S)

Test 1: uops

Code:

  sqrdmlah s0, s1, s2
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

retire (01)cycle (02)03191e3f4e51inst issue (52)~issue fp/simd (54)~dispatch fp/simd (57)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op fp/simd (7e)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst neon or fp (9a)a6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037230061254825100010001000398313030183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230061254825100010001000398313030183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037220061254825100010001000398313130183037303724153289510001000300030373037111001100020073216222630100030383038303830383038
100430372203061254825100010001000398313130183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037220061254825100010001000398313130183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037231061254825100010001000398313130183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230961254825100010001000398313130183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037220061254825100010001000398313130183037303724153289510001000300030373037111001100000073216222702100030383038303830383038

Test 2: Latency 1->1

Code:

  sqrdmlah s0, s1, s2
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e3a3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa6a7a8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000006971012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000006671212162229634100001003003830038300383003830038
10204300372250161295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000001871212162229634100001003003830038300383003830038
102043003722500191029548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000025371012162229634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000011771012162229634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000017471012162229634100001003003830038300383003830038
102043003722500141029548251010010010000100100005554277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308181e3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)72scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037224001358322954825100101010000101000050427731303005430037300372829003287671001020100002030000300373003711100211091010100001013640425432963010000103008530038300383008530038
10024300372251006129548251001010100001010000504277313130018300373003728287032876710010201000020300003003730037111002110910101000010072640216222963010000103003830038300383003830038
10024300372250002512954825100101010000101000050427731303001830037300372828703287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250001032954825100101010000101000050427731313001830037300372828703287671001020100002030000300373003711100211091010100001003640216222963010000103003830038300383003830038
1002430037225120612954825100101010000101000050427867013009030037300372828708287671015920100002030486300843008411100211091010100001013640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828703287671001020100002030000300373003711100211091010100001006640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828703287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828703287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313001830037300372828707287671001020100002030000300373008511100211091010100001000640216222963010000103003830038300383003830038
10024300372250009642954825100101010000101000050427731303001830037300372828703287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqrdmlah s0, s0, s1
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03081e3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)696d6edispatch stall (70)72scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa1a8a9c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500822954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722400612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300843003728265032874510100200100002003000030037300371110201100991001001000010000107101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722400612954825101001001000010010000500427731330018300373003728265032874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265032874510100200100002043000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e1f3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa0a1a6a7a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000183082295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000001820006407162229630010000103003830038300383003830038
1002430037225000000082429548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300791110021109101010000100000100006402162229630010000103003830038300383003830038
1002430037225000000012429548251001010100001010000504277313030018300373003728287328767100102010000203000030037300372110021109101010000100000000006402162229630010000103003830038300383003830038
1002430037225000000012429548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
1002430037224000000012429548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100200000006402162229630010000103003830038300383003830038
1002430037225000000014529548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
1002430037225000000017229548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
1002430037225000000012429548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
1002430037225000000012429548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqrdmlah s0, s1, s0
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e1f3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa1a6a7a8a9acc2cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000000111295482510100100100001001000050042773131300183003730037282653287451010020010000208300003003730037111020110099100100100001000000000710011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730182282653287451010020010000200300003003730037111020110099100100100001000000000710011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000004000710011611296340100001003003830038300383003830038
10204300372250000120103295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000732011611296340100001003003830038300383003830038
102043003722400000061295482510100100100001261000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000710011611296340100001003003830038300383003830038
1020430037225000000103295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000710011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000710012511296340100001003003830038300383003830038
1020430037225000000124295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000710011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000710011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000030710011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03040b1e3a3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722501021082954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000644111611102963010000103003830038300383003830038
10024300372240002108295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000064481610112963010000103003830038300383003830038
100243003722500122662954825100101010000101000050427731313001830037300372828732878610010201000020300003003730037111002110910101000010000644121610112963010000103003830038300383003830038
10024300372250002108295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000064420167102963010000103003830038300383003830038
1002430037224000266295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000064461611112963010000103003830038300383003830038
10024300372250002662954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000644121610102963010000103003830038300383003830038
10024300372250002662954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000644101610102963010000103003830038300383003830038
1002430037225000266295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000064451610102963010000103003830038300383003830038
10024300372250002662954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000644101610112963010000103003830038300383003830038
100243003722500542662954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010100644101611112963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqrdmlah s0, s8, s9
  movi v1.16b, 0
  sqrdmlah s1, s8, s9
  movi v2.16b, 0
  sqrdmlah s2, s8, s9
  movi v3.16b, 0
  sqrdmlah s3, s8, s9
  movi v4.16b, 0
  sqrdmlah s4, s8, s9
  movi v5.16b, 0
  sqrdmlah s5, s8, s9
  movi v6.16b, 0
  sqrdmlah s6, s8, s9
  movi v7.16b, 0
  sqrdmlah s7, s8, s9
  movi v8.16b, 9
  movi v9.16b, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2508

retire (01)cycle (02)0318191e3f51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)5f6061696d6escheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8acbranch mispredict (cb)cfd0d2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602042006515100060258010010080000100800005006400000152004520064200643228010020080000200240000200642006411160201100991001001600001000001011100116112006101600001002006520065200652006520065
1602042006415100081258010010080000100800005006400000052004520064200643228010020080000200240000200642006411160201100991001001600001000001011100116112006101600001002006520065200652006520065
16020420064151000172258010010080000100800005006400000052004520064200643228010020080000200240000200642006411160201100991001001600001000001011150116112006101600001002006520065200652006520065
1602042006415000039258010010080000100800005006400000102004520064200643228010020080000200240000200642006411160201100991001001600001000001011100116112006101600001002006520065200652006520065
1602042006415000039258010010080000100800005006400000052004520064200643228010020080000200240000200642006411160201100991001001600001000001011150116112006101600001002006520065200652006520065
16020420064150000790258010010080000100800005006400000102004520064200643228010020080000200240000200642006411160201100991001001600001000001011151116112006101600001002006520065200652006520065
16020420064150000102258010010080000100800005006400001152004520064200643228010020080000200240000200642006411160201100991001001600001000001011100116112006101600001002006520065200652006520065
1602042006415000039258010010080000100800005006400000102004520064200643228010020080000200240000200642013211160201100991001001600001000001011100116112006101600001002006520065200652006520065
1602042006415000039258010010080000100800005006400000102004520064200643228010020080000200240000200642006411160201100991001001600001000001011100116112006101600001002006520065200652006520065
1602042006415000039258010010080000100800005006400000102004520064200643228010020080000200240000200642006411160201100991001001600001000001011151116112006101600001002006520065200652006520065

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2506

retire (01)cycle (02)03070a0b18191e3a3f51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)5f6061696d6escheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8acc2cfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024200721501100000120258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010042851301920211171220043215160000102004720047200472004720047
16002420046150120000051258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010043850301320211182020043215160000102004720047200472004720047
16002420046150020000045258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010038844412122211171820043215160000102004720047200472004720047
16002420046150010000051258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010042846301720211171720043215160000102004720047200472004720047
16002420046150110000074258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010067850301820211171720043215160000102004720047200472004720047
16002420046150110000051258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010042846301920211171720043215160000102004720047200472004720047
16002420046150110000051258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010044846301920211171720043215160000102004720047200472004720047
16002420046150010000057258001212800001280000626400001152002720046200463228001220800002024000020046200461116002110910101600001000010042844301920211202020043215160000102004720047200472004720047
160024200461501200000452580012128000012800006264000011520027200462004632280012208000020240000200502005011160021109101016000010000100391146301924422171820047230160000102005120051200512005120047
160024200501500000000710258001212800001280000626400001152002920046200463228001220800002024000020046200461116002110910101600001000010042848302024422181820047230160000102005120051200512005120051

Test 6: throughput

Count: 16

Code:

  sqrdmlah s0, s16, s17
  sqrdmlah s1, s16, s17
  sqrdmlah s2, s16, s17
  sqrdmlah s3, s16, s17
  sqrdmlah s4, s16, s17
  sqrdmlah s5, s16, s17
  sqrdmlah s6, s16, s17
  sqrdmlah s7, s16, s17
  sqrdmlah s8, s16, s17
  sqrdmlah s9, s16, s17
  sqrdmlah s10, s16, s17
  sqrdmlah s11, s16, s17
  sqrdmlah s12, s16, s17
  sqrdmlah s13, s16, s17
  sqrdmlah s14, s16, s17
  sqrdmlah s15, s16, s17
  movi v16.16b, 17
  movi v17.16b, 18

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2502

retire (01)cycle (02)0308090b181e1f373f51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)5f60696d6edispatch stall (70)72scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa0a1a8acbranch mispredict (cb)cdcfd5d6dbddinst fetch restart (de)e0ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204402623010000001865925160100100160017100160098500128000001400294004841430201770320007160100202163071200480000401014111751160201100991001001600001000003900101102160114004501600001004004940040400404004940040
160204400483000000000502516010010016000010016000050012800000140020400394003919973031999816010020016000020048000040039400481116020110099100100160000100000000101101160114003601600001004004040049400404004040040
160204400403000000000412516010010016000010016000050012800001140029400484003919973031999716010020016000020048000040039400481116020110099100100160000100000000101101160114003601600001004004040040400404004940040
160204400483000000391080502516010010016001710016000050023989990140020400484003919973031999716010020016000020048000040039400491116020110099100100160000100000000101101160114003601600001004004040040400414004040049
1602044003930000000005025160100100160000100160000500128000000400294004840039199730261999716010020016000020048000040039400491116020110099100100160000100000000101101160114003601600001004004140040400404004140040
160204400392990000000502516010010016000010016000050012800001140020400394003919973031999716010020016000020048000040040400391116020110099100100160000100000000101101160114004501600001004004040049400404004940040
1602044003930000000017412516010010016000010016000050023990270140020400404003919973031999716010020016000020048000040039400391116020110099100100160000100000000101101160114004501600001004004140040400414004040049
160204400483000000000412516010010016001710016000050012800000140029400484003919973032000616010020016000020048000040048400391116020110099100100160000100000000101101160114004501600001004004040049400404004940049
1602044004830000000007062516011710016001710016000050023989990140020400484003919973031999716010020016000020048000040039400491116020110099100100160000100000000101101160114004501600001004004040050400504004040040
160204400392990000000502516011710016001710016000050012800000140020400484003919973031999716010020016000020048000040039400481116020110099100100160000100000000101101160114003601600001004004040040400494004040049

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2503

retire (01)cycle (02)031e373f51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)5f6061696b6d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa0a6a8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400983000097251600101016001710160000501280000110400200400484003919996320019160010201600002048000040039400481116002110910101600001000001002461223162221820400363010160000104004940040400494004040040
160024400483000179025160010101600001016000050128000010040020040048400391999632001916001020160000204800004004940039111600211091010160000100000100223111716211201640036155160000104004940159401714004040040
1600244003930001753025160010101600001016000050239899900040029040039400481999632002816001020160000204800004003940039111600211091010160000100000100243111716211181640036155160000104004940040400494004040049
16002440048300004625160010101600171016000050239899901040020040048400391999632002816001020160000204800004004840048111600211091010160000100000100243211816211201940036155160000104004040049400404004940040
160024400393000174625160010101600001016000050239899911040029040048400391999632001916001020160000204800004003940048111600211091010160000100000100223111616211171440036305160000104004940040400404005040040
160024400483000175625160027101600171016000050128000010040029040048400391999632002816001020160000204800004003940039111600211091010160000100000100223111316211201440045305160000104004940040400494004040049
160024400483000046251600271016001710160000502398999110400290400394004819996320019160010201600002048000040039400481116002110910101600001000001002231119162122018402021510160000104004940040400494004040049
1600244004830001772025160010101600171016000050128000001040020040048400391999632001916001020160000204800004003940048111600211091010160000100000100246221816422191540046155160000104004940040400494004040049
16002440048300005525160010101600001016000050128000011040020040039400481999632002816001020160000204800004004840039111600211091010160000100000100223112016211161540045155160000104004040049400404004940040
16002440039300006925160010101600001016000050239899911040029040039400391999632001916001020160000204800004003940039111600211091010160000100000100223111716211201840036155160000104004040050400404004940040