Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMLSL (vector, 2D)

Test 1: uops

Code:

  sqdmlsl v0.2d, 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)031e3f4e51inst 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
1004303723012425482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000373116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303723186125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000116230003037303711100110000073116112630100030383038303830383038
1004303722546125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqdmlsl v0.2d, 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)03080b1e3a3f4e51inst 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)9fa8acc2cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722410006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100060710121622296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300841110201100991001001000010001950710121622296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100060710121622296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100030710121622296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100002710121623296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100030710121622296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010001590710121622296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100060710121622296340100001003003830038300383003830038
1020430037225000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000900710121622296340100001003003830038300383003830038
102043003722400006129548251010010010000100100005004277313130018300373003728265328745101002041000020030000300373003711102011009910010010000100000710121622296680100001003003830038300383003830038

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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640416222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037225661295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037224061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001003640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001003640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001003640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001003640216232963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001003640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqdmlsl v0.2d, 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)03070a1e3f4e51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730084111020110099100100100001000137102162229634100001003003830038300383003830038
1020430037225000612954825101001001000012610000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000097102162229634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313030018030037300372826532874510100200100002003000030037300371110201100991001001000010001667102162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001001137102162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000167102162229634100001003003830038300383003830038
10204300372250054612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000137102162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000207102162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000137102162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000137102162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000297102162229634100001003003830038300383003830038

1000 unrolls and 10 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)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)9fa0a1a6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000009661216222963010000103003830038300383003830038
1002430037224000030061295482510010101000010100005042773130300183003730037282873287671015920100002030000300373003711100211091010100001000000105640216222963010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000195640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000003640216222963010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000111640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000054640216222963010000103013330038300383003830038
10024300372240000000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
100243003722500000001032954825100101010000101000050427731303001830037301792828732876710010201000020300003003730037111002110910101000010000006640216222963010000103041830419304143041530359

Test 4: Latency 1->3

Code:

  sqdmlsl v0.2d, 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)030818191e1f3f4e51inst 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)9fa0a1a6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225100006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000301813003711102011009910010010000100000008471011611296340100001003003830038300383003830038
1020430037225000120612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300374110201100991001001000010000010071011611296340100001003003830038300383003830038
10204300372250000061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000012371011611296340100001003003830038300383003830038
1020430037225000390612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000971011611296340100001003003830038300383003830038
102043003722500000673295484210120100100001001000050042773130301623003730037282653287451010020010000200300003003730037111020110099100100100001000000012671015011296340100001003003830038300383003830038
1020430037225000005272954825101001001000010010000500427731303001830037300372826532874510100200100002003000030229300371110201100991001001000010000000971011611296340100001003003830038300383003830038
1020430037225000001032954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000371011611296340100001003003830038300383003830038
102043003722500000231295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000001018071011611296340100001003003830038300383003830038
10204300372250001201032954825101001001000010010000578427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000673511611296340100001003003830038300383003830038
102043003722500000612954825101001001000010010000500427731303001830037300372827732874510100200100002003000030037300371110201100991001001000010000000371011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0307090a1e3a3f4e51inst 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)9fa8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722510102682954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000644101611112963010000103003830038300383003830038
10024300372251010268295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001040364481610112963010000103003830038300383003830038
100243003722410102682954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006644111610102963010000103003830038300383003830038
1002430037224101026829548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000104500644101611112963010000103003830038300383003830038
100243003722510102682954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010103644111611102963010000103003830038300383003830038
1002430037225101026829548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000102800644111611112963010000103003830038300383003830038
100243003722410102682954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010103644111610102963010000103003830038300383003830038
10024300372251010268295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000664410161162963010000103003830038300383003830038
100243003722410102682954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010203644111610112963010000103003830038300383003830038
100243003722510102682954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010100644101611102963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlsl v0.2d, v8.2s, v9.2s
  movi v1.16b, 0
  sqdmlsl v1.2d, v8.2s, v9.2s
  movi v2.16b, 0
  sqdmlsl v2.2d, v8.2s, v9.2s
  movi v3.16b, 0
  sqdmlsl v3.2d, v8.2s, v9.2s
  movi v4.16b, 0
  sqdmlsl v4.2d, v8.2s, v9.2s
  movi v5.16b, 0
  sqdmlsl v5.2d, v8.2s, v9.2s
  movi v6.16b, 0
  sqdmlsl v6.2d, v8.2s, v9.2s
  movi v7.16b, 0
  sqdmlsl v7.2d, 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)03091e3f51inst 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)60696d6escheduler 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
1602042009015010104258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100001011331633200611600001002006520065200652006520065
160204200641500039258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011331633200611600001002006520065200652006520065
160204200641500062258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011231633200611600001002006520065200652006520065
160204200641500060258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011331633200611600001002006520065200652006520065
160204200641500039258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100401011331633200611600001002006520065200652006520065
160204200641500039258010010080318100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011331633200611600001002006520065200652006520065
160204200641500039258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011331632200611600001002006520065200652006520065
1602042006415000534258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100061011331633200611600001002006520065200652006520065
1602042006415000467258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011321633200611600001002006520065200652006520065
160204200641510062258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100001011321633200611600001002006520065200652006520065

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2507

retire (01)cycle (02)031e3a3f51inst 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)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420088150004527800121280000128000062640000110200322005120051322800122080000202400002005120051111600211091010160000100001002581142521144200482402160000102006120061200612006120061
160024200601500051298001212800001280000626400000152004120060200603228001220800002024000020060200601116002110910101600001000010028113243442262200572402160000102006120052200612006120061
160024200601500051298001212800001280000626400000152004120060200603228001220800002024000020060200511116002110910101600001000010030113423442224200572402160000102006120061200612006120061
1600242006015000462298001212800001280000626400000152003220060200603228001220800002024000020060200511116002110910101600001020010030113243422224200572401160000102006120061200612006120061
160024200601500015629800121280000128000062640000015200412006020060322800122080000202400002006020051111600211091010160000100001002884243441242200572402160000102005220061200522006120061
1600242006015100137298001212800001280000626400000152004120060200513228001220800002024000020060200601116002110910101600001000010028114222542124200572402160000102006120061200612006120061
1600242006015100116278001212800001280000626400001152004120051200603228001220800002024000020060200601116002110910101600001000010028114243442242200572401160000102006120061200612006120061
160024200601500051298001212800001280000626400000152004120060200603228001220800002024000020060200601116002110910101600001000010030114242542242200572402160000102006120052200612005220061
160024200601510093298001212800001280000626400000152004120060200603228001220800002024000020060200601116002110910101600001000010028114263442264200572402160000102006120061200612006120061
16002420060150510856298001212800001280000626400000152004120060200603228001220800002024000020060200511116002110910101600001000010028114243422142200572402160000102006120061200612006120062

Test 6: throughput

Count: 16

Code:

  sqdmlsl v0.2d, v16.2s, v17.2s
  sqdmlsl v1.2d, v16.2s, v17.2s
  sqdmlsl v2.2d, v16.2s, v17.2s
  sqdmlsl v3.2d, v16.2s, v17.2s
  sqdmlsl v4.2d, v16.2s, v17.2s
  sqdmlsl v5.2d, v16.2s, v17.2s
  sqdmlsl v6.2d, v16.2s, v17.2s
  sqdmlsl v7.2d, v16.2s, v17.2s
  sqdmlsl v8.2d, v16.2s, v17.2s
  sqdmlsl v9.2d, v16.2s, v17.2s
  sqdmlsl v10.2d, v16.2s, v17.2s
  sqdmlsl v11.2d, v16.2s, v17.2s
  sqdmlsl v12.2d, v16.2s, v17.2s
  sqdmlsl v13.2d, v16.2s, v17.2s
  sqdmlsl v14.2d, v16.2s, v17.2s
  sqdmlsl v15.2d, 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)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)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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440059311001402516010010016001710016000050012800000400204004840039199730319998160100200160000200480000400394004811160201100991001001600001000010110116344003601600001004004940049400404004040049
1602044003930000502516010010016001710016000050023989990400294003940039199730320006160100200160000200480000400394003911160201100991001001600001000010110116134003601600001004004140040400404004940040
160204400403000171712516010010016000010016000050012800000401114003940039199730320006160100200160000200480000400394004811160201100991001001600001000010110116144003601600001004004040040400494004040049
1602044004830000412516011710016000010016000050012800000400204003940048199730320006160100200160000200480000400494003911160201100991001001600001000010110116134003601600001004004040050400404004040040
16020440048299017502516011710016000010016000050012800000400294003940039199730319997160100200160000200480000400484003911160201100991001001600001000310110116134004501600001004004040049400404004040040
1602044003929900502516010010016000010016000050012800000400204004840048199730319997160100200160000200480000400484003911160201100991001001600001000010110116114004501600001004004040049400404004040040
160204400393000177062516010010016001710016000050012800000400204003940039199730320006160100200160000200480000400484003911160201100991001001600001000010110116144003601600001004004040040400494004040040
16020440039300005002516010010016000010016000050023989990400204004940039199730319997160100200160000200480000400394003911160201100991001001600001000010110116124003601600001004004940040400404004140040
1602044003930000502516010010016000010016000050012800000400204003940048199730320006160100200160000200480000400394004011160201100991001001600001000010110116124003601600001004004040049400494004940049
1602044004830001412516010110016000010016000050023989990400294004840039199730320006160100200160000200480000400484004811160201100991001001600001001610110116164004501600001004004040049400404004040040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030b18191e373a3f51inst 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)5f6061696d6edispatch 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)9fa0a1a6a8a9acc2c5cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400493000000024862516001010160000101600005012800001154002040049400391999603200191600102016000020480000400394003911160021109101016000010000000001002282141641164400360155160000104004040040400404004040040
16002440039300000001972516001010160000101600005012800001154002040039400391999603200471601192016000020480000400394004911160021109101016000010000000001002282181621164400360155160000104005040050400404004040040
160024400392990000014625160027101600171016000050128000011540020400394003919996032002816001020160000204800004003940039111600211091010160000100000000010024113261642264400360155160000104004040040400404004040040
1600244004930000001821132516001010160017121600005024388651154002040040400481999603200191600102016000020480000400394003911160021109101016000010000000001002283241622269400360155160000104004040040400404004040040
1600244004030000000222225160010101600001016000050128000001540030400394003919996032002916001020160000204800004003940039111600211091010160000100000000010022821616211694003603012160000104004040049400494004940040
1600244003929900001816725160010101600001016000050128000011540020400394003919996032001916001020160000204800004003940039111600211091010160000100000000010022832416422864003603010160000104004040098400404005040040
1600244003930000000250425160010101600001016000050128000011540020400394003919996032001916001020160000204800004003940039111600211091010160000100000000010024821616411664004803010160000104004040040400494004940040
1600244003930001016252273425160010101600001016000050128000001540020400394004919996032001916001020160000204800004004840039111600211091010160000100000000010022821916221644003603010160000104004040040400404004040040
1600244003930000000111125160010101600171016000050239899901540029400394003919996032001916001020160000204800004003940039111600211091010160000100000000010022831616422854003603010160000104004040040400404004040040
160024400393000000024492516001010160000101600005012800001154002040039400391999603200191600102016000020480000400394004011160021109101016000010000003001002282241641188400450306160000104004040040400504005040040