Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMLSL (by element, S)

Test 1: uops

Code:

  sqdmlsl s0, h1, v2.h[1]
  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)03080b1e3f4e51inst 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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110000373116112630100030383038303830383038
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110000673116112630100030383038303830383038
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100030003037303711100110000092116112630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037231038225482510001000100039831313018303730372415328951000100030003037303711100110001073116112660100030383038308630383038
100430372300061254825100010001000398313130183037303724153289510001000300030373037111001100006073116112630100030383038303830383038
100430372300186125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110009073116112630100030383038303830383038
100430372300010325482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqdmlsl s0, h1, v2.h[1]
  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)03080b18191e1f3a3f4e51inst 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)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)9fa0a1a6a7a8a9acc2c5branch mispredict (cb)cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000000061295482510100100100001001000050042773133001830086300372827272874110100200100082003002430037300371110201100991001001000010000000000000710021622296340100001003003830038300383009030038
1020430037225010013200612954825101001041000810010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000000240000752122422296340100001003003830038300383003830038
102043003723301120001602954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000000140280000710121622296340100001003003830038300383003830038
1020430037225100000061295482510100100100001001000050042773133001830037300372826532874510100200103322283000030037300371110201100991001001000010000000000000710121632296340100001003003830038300383003830038
1020430037225000000061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010020004000000710121623296340100001003003830038300383003830038
1020430085225000000061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710121622296340100001003003830038300383003830038
1020430037225000000061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710121622296340100001003003830038300383003830038
10204300372250000000103295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710121622296340100001003003830038300383003830038
1020430037225000000061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710131622296340100001003003830086300383003830038
1020430037225000000061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010002201000000710121622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b181e3a3f4e51inst 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)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)9fa0a1a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640316222963010000103003830038300383003830038
10024300372250000028662954825100101010000101000050427731330018300373003728287328787100102010000203000030132300371110021109101010000100000300640216222963010000103003830038300383003830038
100243003722500000662954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
10024300372250001320612954825100101010000101000050427731330018300373003728287328767100102010000203000030085300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018300373003728287328786100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018300373003728287328767100102010000203000030084300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722400000612954825100101010000101000050427731330018300843008528287328767101592010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqdmlsl s0, h0, v1.h[1]
  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)03080b1e3f4e51inst 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)9fa1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037224000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500061295481041010010010000100100005004277313130054300373003728265328745101002001000020030000300373003711102011009910010010000100023120071011611296340100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100001000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000800071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830086300383003830038
1020430037225000612954825101001001000012210000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000100071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090a0b18191e1f3a3f4e51inst 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)60696b6d6edispatch 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)9fa0a1a6a7a8a9abacc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722501000200061295482510010101000014100005042773130300180300373003728287328767100102010000203000030037300371110021109101010000100021020006403242429702210000103003830038300383003830038
10024300372240000001200103295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300851110021109101010000100000000306402162229630010000103003830038300383003830038
100243003722400001015880168295482510019101000810104475042786701300180300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000061295482510010101000010100005042773131300180300373003728287728767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000061295482510010101000010100005042773130300180300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
10024300372250000000006129548251001010100001010000504277313030018030037300372828732876710010201000020300003003730037111002110910101000010002112401400606402162229630010000103003830038300383003830038
10024300832250000002700147295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000061295482510010101000010100005042773130300180300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103008630038300863003830038
100243003722500000090061295482510010101000010100005042773130300180300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqdmlsl s0, h1, v0.h[1]
  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)030809191e1f3f4e51inst 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)9fa6a8a9acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037226000002512954825101001001000010010000500427731303001830037300372827262874010100200100082003002430037300371110201100991001001000010000031117180160029647100001003003830038300383003830038
102043003722500000612954825101001001000010010000500427731303001830037300372827262874010100204100082003002430037300371110201100991001001000010000031117170160029647100001003003830038300383003830038
1020430037225000907262954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
102043003722500000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
102043003722500000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
1020430037225000450612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010022030007101161129634100001003003830038300383003830038
102043003722500000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
102043003722500000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
102043003722500239688612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010001000007101161129634100001003003830038300383003830038
102043003722611000612954825101001001000010010000500427731303001830037300372826532876310100200100002003000030037300371110201100991001001000010000200007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03191e3a3f4e51inst 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)9fa0a1a8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500061295482510010101000010100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000100070064021622296300010000103003830038300383003830038
100243003722500061295482510010101000010100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000100000164021622296300010000103003830038300383003830038
1002430037224000536295482510010101000010100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000100000064031622296300010000103003830038300383003830038
1002430037225000388295482510010101000010100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000100020064021622296300010000103003830038300383003830038
1002430037225000620295304310010101000010100005042773131300183003730037282870328767100102010000203000030037300371110021109101010000100450064021622296300010000103003830038300383003830038
1002430037225000726295482510010101000010100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000100010064021632296300010000103003830038300383003830038
100243003722509061295482510010101000010100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000102012497806401290911299543010000103003830038300383032230038
100243013122500061295482510010101000710100005042773130300183003730037282870328767100102010000203000030037300371110021109101010000100000064021622296300010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828703287671001020100002033561304633046710110021109101010000100010064021622296302010000103003830086300383003830038
100243003722500061295484410010101000010100005042773131300183003730037282870328767100102010000203000030037300371110021109101010000100000064021622296300010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlsl s0, h8, v9.h[1]
  movi v1.16b, 0
  sqdmlsl s1, h8, v9.h[1]
  movi v2.16b, 0
  sqdmlsl s2, h8, v9.h[1]
  movi v3.16b, 0
  sqdmlsl s3, h8, v9.h[1]
  movi v4.16b, 0
  sqdmlsl s4, h8, v9.h[1]
  movi v5.16b, 0
  sqdmlsl s5, h8, v9.h[1]
  movi v6.16b, 0
  sqdmlsl s6, h8, v9.h[1]
  movi v7.16b, 0
  sqdmlsl s7, h8, v9.h[1]
  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)03080a0b18191e1f3a3f51inst 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)60696d6e74scheduler 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)9fa0a1a6a8a9acc2c5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020420078151000000003925801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
16020420064150000000006225801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
160204200641500000000020925801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000002010111116112006101600001002006520065200652006520065
1602042006415000000000134825801001008000010080000500640000120045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
16020420064151000000006225801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
160204200641500000000047925801001008000010080000500640000120045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
16020420064150000000008125801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
16020420064150000000006225801001008000010080000500640000120045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
160204200641500000000018825801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065
16020420064151000000003925801001008000010080000500640000020045200642006403228010020080000200240000200642006411160201100991001001600001000000000010111116112006101600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)031e1f3f51inst 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)9fa8a9acc5branch mispredict (cb)cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024200661500888729800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100000010043115217252111616200482201160000102006120052200522005220052
1600242005115015902362980012128020812800006264251201520173203702029111228011620801052024000020051201321116002110910101600001000000100408511625211165200482201160000102005220052201432029620061
160024200511500047927800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100000010040851625211166200482201160000102005220052200522005220052
1600242005115000703278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000000100408511625211177200482401160000102005220052200522005220052
16002420051151004442278001212800001280000626400001152003220051200513228001220800002024000020131200511116002110910101600001000000100308511634211166200482201160000102005220052200522005220052
160024200511510068278001212800001280000626400001152003220060200603228001220800002024000020051200511116002110910101600001000000100408511625422166200482401160000102005220052200522005220052
160024200511500016927800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100000010040851625211616200482201160000102005220052200522005220052
16002420051150402039927800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100000010040851625211616200482201160000102005220052200522005220052
160024200511500045278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000000100408511625211616200482201160000102005220052200522005220052
16002420051150001732780012128000012800006264000011520032200512005132280012208000020240000200512005111160021109101016000010000001004085116252111616200482201160000102005220052200522005220052

Test 6: throughput

Count: 12

Code:

  sqdmlsl s0, h12, v13.h[1]
  sqdmlsl s1, h12, v13.h[1]
  sqdmlsl s2, h12, v13.h[1]
  sqdmlsl s3, h12, v13.h[1]
  sqdmlsl s4, h12, v13.h[1]
  sqdmlsl s5, h12, v13.h[1]
  sqdmlsl s6, h12, v13.h[1]
  sqdmlsl s7, h12, v13.h[1]
  sqdmlsl s8, h12, v13.h[1]
  sqdmlsl s9, h12, v13.h[1]
  sqdmlsl s10, h12, v13.h[1]
  sqdmlsl s11, h12, v13.h[1]
  movi v12.16b, 13
  movi v13.16b, 14

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)031e373f4e51inst 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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12020430039225004167132512010110012001810012000050096000013172930040300391497331499712010020012000020036000030040300391112020110099100100120000100000761041611300361200001003004130040317493004031749
12020431748225004102512011810012001810012000050096000003002031748300391497331499712010020012000020036000030039300401112020110099100100120000100000761011611300361200001003174930040317493004031749
120204300422250161025120100100120000100120000500439952403002030039300401665331499812010020012000020036000031748300391112020110099100100120000100000761011611300361200001003031930042300403004030041
120204317482256013667132512010110012001810012000050096000003002030039300401665331670612010020012000020036000031748300391112020110099100100120000100000761011611300361200001003004130040317633101131749
120204317482250186102512011810012001810012000050096000003002030039317481497331499712010020012000020036000030039317481112020110099100100120000100000761011611317451200001003174930040317493004030040
1202043003922412534167132512011810012000110012000050096000003172931748300391497331670612010020012000020036000031748300391112020110099100100120000100000761011611300361200001003174930040317493004031749
120204317482250184367132512011810012001810012000050096000003002030039300401497331499812010020012000020036000030039300401112020110099100100120000100000761011611317451200001003104430040300423004030040
120204300402250041025120100100120000100120000500439952413002030922300391497331499712010020012000020036000030039317481112020110099100100120000100000761011611300361200001003004130040317493004030041
120204300402252706267132512010110012001810012000050096000003172931748300391498931499712010020012000020036000030040300391112020110099100100120000100000761011611300361200001003174930040317493004030041
120204317482250041025120100100120000100120000500439952413002131748300391665331499812010020012000020036000030040300391112020110099100100120000100000761011611300361200001003004031749306393020030040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03090a0b181e1f373f4e51inst 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)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)9fa1a6a8accdcfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12002430048225000000041025120010101200001012000050990000013002030039300391667731501912001020120000203600003004030040111200211091010120000100000075200004160004430037120000103004130040300403004030040
120024317502250000000410251200101012008410120000509600000130020300393003916677315020120010201200002036000030039309511112002110910101200001000415075200006160004430039120000103004030923300413004130040
12002430039225000060140025120010101200001012000050990000013002030039300391499631673012001020120000203600003092230040111200211091010120000100000075200007160006430036120000103004130998300403004030923
120024300392250000001400251200111012003510120000504283400013002030039300391667731501912001020120000203600003003930039111200211091010120000100099075200006160004430036120000103004030041300413004130923
12002430039231000000040025120010101200001012000050960000013002130039300391499631501912001020120000203600003014230922111200211091010120000100000075200004160004331745120000103004030041317513004131751
12002431750238000060061025120011101200351012000050990000013002030039300391499631501912001020120000203600003004030039111200211091010120000100000075200004160004430036120000103004030923300413004030040
120024300402250000001741025120045101200011012000050960000013002030039300391499631501912001020120000203600003003930039111200211091010120000100000075200006160007630036120000103004130041300403004030040
12002430039231000000040025120011101200171012000050990000013090330039300391667731590212001020120000203600003003930039111200211091010120000100000075200006160006630036120000103092330040300403004030040
120024300402380000003540025120010101200001012000050960000013002030039300391499631501912001020120000203600003003930039111200211091010120000100020075200004160007430036120000103004030040300403004030040
1200243004022500000017705025120010101200001012000050990000013090330039300391499631501912001020120000203600003095130039111200211091010120000100000075380004160006630037120000103004030040300403004030040