Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRDMLSH (scalar, H)

Test 1: uops

Code:

  sqrdmlsh h0, h1, h2
  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
100430372266125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000116830003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303723012825482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqrdmlsh h0, h1, h2
  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)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)9fa8acc5cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000012006129548251010010010000100100005004277313030022300373003728265328745101002001000020030000300373003711102011009910010010000100100710132422296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038
1020430037225000027006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100030710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121623296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000710121622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030a0b191e3f4e51inst 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
10024300372250009612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216322963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640416222963010000103003830038300383003830038
10024300372240000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731303001830037300372828732876710010201000022314583013230085411002110910101000010000640216222963010000103003830038300383003830038
10024300372250006612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400030612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731313001830037302252828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqrdmlsh h0, h0, h1
  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)0308091e3f4e51inst 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)9fa0a9cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071021611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037225007261295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0307080b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000000001032954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100100000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
10024300372250000025500612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000822954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqrdmlsh h0, h1, h0
  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)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)5e60696d6edispatch 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)9fa1a6a7a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000000612954825101001001000010010000500427731301300183003730037282653287451010020010000200300003003730037111020110099100100100001000000111071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731301300183003730037282653287451010020010000200300003003730037111020110099100100100001000000108071011611296340100001003003830038300383003830038
10204300372240000000612954825101001001000010010000500427731301300183003730037282653287451010020010000200300003003730037111020110099100100100001000000120071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731301300653003730037282653287451010020010000200300003003730037111020110099100100100001000000102071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731301300183008430085282653287451010020010000200300003003730037111020110099100100100001000000117071011611296340100001003003830038300383003830038
1020430037225000000061295482510100100100001001000050042773130130018300373003728265328745101002001000020030000300373003711102011009910010010000100000030071011611296340100001003003830038300383003830038
1020430037225000000061295482510100100100001001000050042773130130018300373003728265328745101002001000020030000300373003711102011009910010010000100000099071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731301300183003730037282653287451010020010000200300003003730037111020110099100100100001000000114071011611296340100001003003830038300383003830038
10204300372240000000612954825101001001000010010000500427731301300183003730037282653287451010020010000200300003003730037111020110099100100100001000000117071011611296340100001003003830038300383003830038
102043003722500000008229548251010010010000100100005004277313013001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0309181e1f3f4e51inst 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)9fa0a6a7a8acc2cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100001274600640216222963010000103003830038300383003830038
100243003723800150612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
10024300372240060612954825100101010000101000050427731330162300373003728287328767100102010000203000030037300371110021109101010000100001294300640216222963010000103003830038300383003830038
100243003722500906129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000062300661216222969810000103003830038300383003830038
100243003722500002102954825100101210000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
1002430037225001504302954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830038300383003830038
100243003722400120612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100001020640224322963010000103003830038300383013130038
1002430037225020104822954825100101010000101000050427731330018302253003728287328767100102010000203000030037300371110021109101010000100000000640216222963010000103003830133300383003830038
100243003724400001032954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000300640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqrdmlsh h0, h8, h9
  movi v1.16b, 0
  sqrdmlsh h1, h8, h9
  movi v2.16b, 0
  sqrdmlsh h2, h8, h9
  movi v3.16b, 0
  sqrdmlsh h3, h8, h9
  movi v4.16b, 0
  sqrdmlsh h4, h8, h9
  movi v5.16b, 0
  sqrdmlsh h5, h8, h9
  movi v6.16b, 0
  sqrdmlsh h6, h8, h9
  movi v7.16b, 0
  sqrdmlsh h7, h8, h9
  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)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)696d6escheduler 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
16020420065150003925801001008000010080000500640000200452006420064322801002008000020024000020064200641116020110099100100160000100000430010111416112006101600001002006520065200652006520065
1602042006415500392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000610111116112006101600001002006520065201342006520065
1602042006415000392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
160204200641502250392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
160204200641502610392580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
1602042006415025507042580100100800001008000050064000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065203072006520065

1000 unrolls and 10 iterations

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

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)5f6061696d6e74scheduler 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)9fa0a1a8a9accdcfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024200761500000572780012128000012800006264000011520032200512005103228001220800002024000020062200621116002110910101600001000000010053114230363322827200592412160000102006320063200632006320063
160024200621502201632980012128000013800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010054115325363222729200592412160000102006320063200632006320063
160024200621502201572980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010055146226363222728200592412160000102006320063200632006320063
160024200621502200782980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010057116228363222628200592412160000102006320063200632006320063
1600242006215033011172980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010059115227363221527200592412160000102006320063200632006320063
160024200621503200632980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010054115323363222728200592412160000102006320063200632006320063
160024200621502200632980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010056146229363223014200592412160000102006320063200632006320063
160024200621502200632980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010057115228363222827200592412160000102006320063200632006320063
160024200621502301692980012128000012800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010055146329363222227200592412160000102006320063200632006320063
1600242006215122001202980012128000011800006264000001520043200622006203228001220800002024000020062200621116002110910101600001000000010043115217453222716200592412160000102006320063200632006320063

Test 6: throughput

Count: 16

Code:

  sqrdmlsh h0, h16, h17
  sqrdmlsh h1, h16, h17
  sqrdmlsh h2, h16, h17
  sqrdmlsh h3, h16, h17
  sqrdmlsh h4, h16, h17
  sqrdmlsh h5, h16, h17
  sqrdmlsh h6, h16, h17
  sqrdmlsh h7, h16, h17
  sqrdmlsh h8, h16, h17
  sqrdmlsh h9, h16, h17
  sqrdmlsh h10, h16, h17
  sqrdmlsh h11, h16, h17
  sqrdmlsh h12, h16, h17
  sqrdmlsh h13, h16, h17
  sqrdmlsh h14, h16, h17
  sqrdmlsh h15, h16, h17
  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)030b18191e1f373f51inst 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)9fa1a7a8a9acc5branch mispredict (cb)cdcfd5d6d9daddinst fetch restart (de)e0? int output thing (e9)eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602044003932200000681602251601001001600001001600005001280000140020400394003919973319997160100200160033200480096400484004811160201100991001001600001000000011110118116000040045001600001004004040041400414004140040
16020440040299000001730251601081001600081001600205001280132140020400394003919977619990160120200160032200480096400394003911160201100991001001600001000000011110118016000040036001600001004004040041400404004940040
16020440048310000000516251601001001600001001600005001280000140020400394012719973320006160100200160000200480000400484003911160201100991001001600001000000000110110116021140036001600001004004040040400404004940040
160204400483000000017872516011710016000010016000050023989991400294004840048199733199971601002001600002004800004003940039111602011009910010016000010000000000101101170011400451601600001004004940049400494004940041
1602044010032200000041251601001001600171001600005001280000140020400394003919973320006160100200160000200480000400394003911160201100991001001600001000000000010110116201140036001600001004034840040400404004940040
16020440039300000001741251601001001600171001600005001280000140029400394003919973319997160100200160000200480000400394004811160201100991001001600001000000000010110116201140045001600001004004040040400404004040049
1602044003930000000041251601001001600171001600005002399027140020400394003919973319997160100200160000200480000400394003911160201100991001001600001000000000110110116001140036001600001004004040040400404004040040
1602044003930000000050251601171001600001001600005002399027140020400484003919973319997160100200160000200480000400394003911160201100991001001600001000000000010110116201140036001600001004004940040400404004940049
16020440048300000120192226028916183612316167412416211560514280731405184054240843201976320419161933200161438204486090409434108511116020110099100100160000100401271980001040121752021408142501600001004083940460409604082640964
1602044083330411414172513204127732911617511241609611271618376312122257140029400484003919973319997160100200160000200480000400394004811160201100991001001600001000000000010110116001140036001600001004005040049400494004040040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030a181e373a3f51inst 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)9fa8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600244005630000010269251600271016001710160000501319998110400204004840039199960320019160010201600002048000040048400491116002110910101600001000100228311016211129400450155160000104004040040400494004940040
160024400393000001705225160010101600001016000050128000001540029400494004019996032002816001020160000204800004003940048111600211091010160000100010022311111621112124004503010160000104004940041400414004040040
16002440039300000005625160027101600171016000050239899910540029400484003919996032002816001020160000204800004003940039111600211091010160000100010022831111621111114004903010160000104004140040400504005040040
16002440039300000005225160010101600001016000050239899910540021400394003919996032001916001020160000204800004003940040111600211091010160000100010022831916211109400360305160000104004040041400404004040049
160024400403000000046251600101016001710160000501280000115400204004840040199960320019160010201600002048000040039400391116002110910101600001000100408319162119114003703111160000104004940040400494004040041
16002440039299000170552516002710160017101600005012800001154002040039400401999603200201600102016000020480000400404004911160021109101016000010001002411421116322109400360155160000104004040040400494004040049
16002440048300000170522516001010160000101600005012800000154002040039400401999603200191600102016000020480000400394004811160021109101016000010001002283191621112124004503010160000104004940040400494004040040
16002440039300000170462516001010160000101600005013199991154002940049400401999603200191600102016000020480000400404004911160021109101016000010201002283111162111010400370155160000104004040050400404004040040
16002440039300000105525160010101600011016000050128000011540020400394004819996032002816001020160000204800004003940048111600211091010160000100010022831916211864003603010160000104004140049400404004140041
16002440049300000170552516001010160017101600005023989991154002040039400481999603200281600102016000020480000400394009111160021109101016000010001002411429163221111400360155160000104004040041400404004140049