Apple M1 Microarchitecture Research by Dougall Johnson

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

MLA (vector, 2S)

Test 1: uops

Code:

  mla v0.2s, v1.2s, v2.2s
  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)03081e1f3f4e51inst 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
10043037230906125482510001000100039831303018303730372415328951000100030003037303711100110000373116112630100030383038303830383038
10043037220006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230606125482510001000100039831313018303730372415328951000100030003037303711100110000673116112630100030383038303830383038
100430372203606125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110001073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110001073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  mla v0.2s, v1.2s, v2.2s
  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)03090e1e1f3a3f4e51inst 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)9fa0a1a8a9acbranch mispredict (cb)cdcfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000061295482510100100100001001000050042773131300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071015162229634100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773131300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071012160229634100001003003830038300383003830038
1020430037225000001036295482510100100100001001000050042773131300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071012163229634100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071012162229634100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773131300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071012162229634100001003003830038300383003830038
10204300372240000061295482510100100100001001000050042773131300183003730037282650328765101002001000020030000300373003711102011009910010010000100000000071013162229634100001003003830038300383003830038
10204300372240000061295482510100100100001001000050042773131300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071012162229634100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071212162329634100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000000071012162229634100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773131300183003730037282650328745101002001000020030000300373003711102011009910010010000100001000071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e3f4e51inst 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)9fa6a7a8a9acc2c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250116029548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100050000640316222963010000103003830038300383003830038
100243003722408429548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
10024300372252155229548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722539116129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
10024300372250101029548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216322963010000103003830038300383003830038
10024300372250107029548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000103000000640216222963010000103003830038300383003830038
1002430037225095229548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
1002430037225019929548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
1002430037225095229548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  mla v0.2s, v0.2s, v1.2s
  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)03191e1f3a3f4e51inst 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)9fa0a8accdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037233000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500108061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000171011611296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773130300183003730037282650328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e1f3a3f4e51inst 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722400061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383008530038
100243003722500066295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722500061295482510010101000010100005042773130300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722400161295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100006640216222963010000103003830038300383003830038
100243003722500061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722400061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722500061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722500061295484410010101000010100005042773130300180300373003728287328767100102010000203000030037300372110021109101010000100000640216222963010000103003830038300383003830038
100243003722500061295482510010101000010100005042773131300180300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  mla v0.2s, v1.2s, v0.2s
  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)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)9fa8a9cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225044461295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225046861295302510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037224046861295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
10204300372250375726295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037224046261295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037224039361295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225043561295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
10204300372250420631295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225040261295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296341100001003003830038300383003830038
10204300372240447726295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e3f4e51inst 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)9fa0a1a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372252796129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372251026129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372253996129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
100243003722539663129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372253996129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000100640216222963010000103003830038300383003830038
10024300372244026129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372263396129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372253936129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372253396129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  mla v0.2s, v8.2s, v9.2s
  movi v1.16b, 0
  mla v1.2s, v8.2s, v9.2s
  movi v2.16b, 0
  mla v2.2s, v8.2s, v9.2s
  movi v3.16b, 0
  mla v3.2s, v8.2s, v9.2s
  movi v4.16b, 0
  mla v4.2s, v8.2s, v9.2s
  movi v5.16b, 0
  mla v5.2s, v8.2s, v9.2s
  movi v6.16b, 0
  mla v6.2s, v8.2s, v9.2s
  movi v7.16b, 0
  mla v7.2s, v8.2s, v9.2s
  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)031e3f51inst 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)6061696b6d6escheduler 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)9fa0a8accfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020420089150603925801001008000010080000500640000152004502006420064322801002008000020024000020064200641116020110099100100160000100000101115111611200611600001002006520065200652006520065
160204200641501983925801001008000010080000500640000152004502006420064322801002008000020024000020064200641116020110099100100160000100000101595111611200611600001002006520065200652006520065
1602042006415003925801001008010510080103500640000152004502006420064322801002008000020024000020064200641116020110099100100160000100000101115111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000052004502006420064322801002008000020024000020064200641116020110099100100160000100000101115111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000052004502006420064322801002008000020024000020064200641116020110099100100160000100000101115111611200611600001002006520065200652006520065
16020420064150092125801001008000010080000500640000152004502006420064322801002008000020024000020064200641116020110099100100160000100000101115111611200611600001002006520065200652006520065
16020420064150153925801001008000010080000500640000152004502006420064322801002008000020024000020064200641116020110099100100160000100000101115111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640840152004532006420064322801002008000020024000020064200641116020110099100100160000100000101115011611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000152004502006420064322801002008000020024000020064200641116020110099100100160000100000101115011611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000102004502006420064322801002008000020024000020064200641116020110099100100160000100000101110111611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03070a1e3a3f51inst 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)5f60696b6d6escheduler 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)9fa8c3cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420064150229051258001212800001280000626400001120027020046202843228001220800002024000020046200461116002110910101600001000100383112120211121520043215160000102004720047200472004720047
16002420046150120045258001212800001280000626400001120027020046203273228001220800002024000020048200481116002110910101600001000100363211620211161820043215160000102004720047200472004720047
1600242004615031123057258001212800001280000626400001120027020046202403228001220800002024000020046200461116002110910101600001000100383111420211131620043215160000102004720047200472004720047
160024200461501169051258001212800001280000626400001120027020050203933228001220800002024000020046200461116002110910101600001000100363111520211141420043215160000102005120047200472004720047
16002420046150126045258001212800001280000626400001120027020046202463228001220800002024000020046200461116002110910101600001000100373111420211141420043215160000102004720047200472004720047
16002420046150110168258001212800001280000626400001120027020046201983228001220800002024000020046200461116002110910101600001000100373111320211151320043215160000102005120047200472004720047
1600242004615011605125800121280000128000062640000112002702004620242322800122080000202400002004620046111600211091010160000100010040311920211131320043215160000102004720047200472004720047
16002420046150126051258001212800001280000626400001120027020046202343228001220800002024000020046200461116002110910101600001000100363121625211131320043215160000102004720047200472004720047
16002420046150330057258001212800001280000626400001120027020046202913228001220801322024000020050200461116002110910101600001000100393121420211141520043215160000102005120047200472004720047
16002420046150210051258001212800001280000626400001120027020046202463228001220800002024000020046200461116002110910101600001000100383111620211131420043215160000102004720047200472004720047

Test 6: throughput

Count: 16

Code:

  mla v0.2s, v16.2s, v17.2s
  mla v1.2s, v16.2s, v17.2s
  mla v2.2s, v16.2s, v17.2s
  mla v3.2s, v16.2s, v17.2s
  mla v4.2s, v16.2s, v17.2s
  mla v5.2s, v16.2s, v17.2s
  mla v6.2s, v16.2s, v17.2s
  mla v7.2s, v16.2s, v17.2s
  mla v8.2s, v16.2s, v17.2s
  mla v9.2s, v16.2s, v17.2s
  mla v10.2s, v16.2s, v17.2s
  mla v11.2s, v16.2s, v17.2s
  mla v12.2s, v16.2s, v17.2s
  mla v13.2s, v16.2s, v17.2s
  mla v14.2s, v16.2s, v17.2s
  mla v15.2s, v16.2s, v17.2s
  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)0318191e373f51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440074300001680412516010110016000110016000050023989994002040048400391997331999716010020016000020048000040039400401116020110099100100160000100001011021611400361600001004004040049400404004940040
16020440040300002705025160100100160000100160000500239899940020400484003919973319997160100200160000200480000400394004811160201100991001001600001005001011011611400451600001004004940041400404004940049
160204400483000000622516011710016000010016000050012800004002040048400401997331999716010020016000020048000040048400401116020110099100100160000100001011011611400451600001004004040049400404004940040
16020440039299002717502516011710016001710016000050023989994002940048400391997332000616010020016000020048000040039400481116020110099100100160000100001011011611400361600001004004940040400404004040049
1602044003930001180412516011710016001710016000050012800004002940039400391997332000616010020016000020048000040039400481116020110099100100160000100001011011611400451600001004004940040400404004940040
1602044004830000391767662516011710016000010016000050023989994002040039400391997331999716010020016000020048000040039400481116020110099100100160000100001011011611400361600001004004040049400404004040040
1602044003930000720502516011710016000010016000050023989994002940039400401997331999716010020016000020048000040039400401116020110099100100160000100001011011611400361600001004004040049400404004940049
16020440048299003017942516010110016001710016000050012800004002040039400491997331999716010020016000020048000040039400391116020110099100100160000100001011011611400451600001004004040050400404004940049
16020440048300004217502516011710016000010016000050023989994002140039400391997331999716010020016000020048000040040400391116020110099100100160000100001011011611400371600001004004940040400414004040040
1602044004929900617422516011710016001710016000050013199984003040039400391997331999816010020016000020048000040048400481116020110099100100160000100001011011611400451600001004004040041400404005340040

1000 unrolls and 10 iterations

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

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)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)9fa0a8acc5cdcfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400492990046251600101016000010160000502398999114003040040400391999603200281600102016000020480000400484003911160021109101016000010000001002431113162115940046155160000104004040049400404004940040
16002440039300005525160010101600001016000050128000011400214003940039199960320019160010201600002048000040039400391116002110910101600001000010100223116162117940036155160000104004940040400494004040040
160024400393000175525160010101600001016000050239899911400204003940039199960320029160010201600002048000040039400401116002110910101600001000000100223119164129540046155160000104004940040400494004040040
16002440039300004625160010101600001016000050128000011400214004840049199960320020160010201600002048000040048400391116002110910101600001000000100223115162115940036155160000104004940040400494004040041
16002440048300004625160027101600171016000050128000011400204004840039199960320019160122201600002048000040039400481116002110910101600001000000100223119162115940045155160000104004040040400404004940041
160024400393000056251600101016000110160000501319998114002140039400401999603200191600102016000020480000400494003911160021109101016000010000001002231161621191040036156160000104004040041400504004040050
160024400393000046251600271016001710160000502398999114002940048400391999603200191600102016000020480000400484003911160021109101016000010000001002231161621151040036155160000104004040040400404005040040
16002440048311004625160010101600001016000050239899921400294004840039199960320019160010201600002048000040039400391116002110910101600001000300100243115162116940046155160000104004040049400404004940040
160024400393000056251600281016001710160000501280000114002140039400481999603200281600102016000020480000400484003911160021109101016000010000001002231191621161040036155160000104004940040400494004040040
16002440039300017462516002710160000101600005012800000140020400394004819996032002816001020160000204800004004840039111600211091010160000100000010024622916422610400363010160000104004040040400494004040049