Apple M1 Microarchitecture Research by Dougall Johnson

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

SMLAL (vector, 8H)

Test 1: uops

Code:

  smlal v0.8h, v1.8b, v2.8b
  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)03091e1f3f4e51inst 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
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110000073216112630100030383038303830383038
10043037220906125482510001000100039831303018303730372415328951000100030003037307311100110000073116112630100030383038303830383038
100430372302106125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
100430372200886125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303722019806125482510001000100039831303018303730372415328951000100030003037303711100110001073116112630100030383038303830383038
100430372304206125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030833038303830383038
10043037220006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  smlal v0.8h, v1.8b, v2.8b
  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)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)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)9fa8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250008229548251010010010000100100005004277313130126300373003728265032874510100200100002003000030037300371110201100991001001000010021671003162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001009071002162329634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001007071002162229634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265032874510100200100002003000030037300371110201100991001001000010033371002162229634100001003003830038300383003830038
10204302282250006129548251010010010000100100005004277313130018300843013328265032874510100200100002003000030037300371110201100991001001000010003971002162229634100001003003830038300383003830038
1020430037224000612954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001006071012162229634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001006071212162329634100001003003830038300383003830038
10204300372250014922954825101001001000010010000500427731303001830037300372826503287451010020010000200300003003730037111020110099100100100001007071012162229634100001003003830038300383003830038
1020430037225000942954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001007071012162229634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265032874510100200100002003000030037300371110201100991001001000010029371012162229634100001003003830038300383003830038

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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500003300612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000002006402162229630010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000612954844100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372240000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000600612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372240000000612954825100101010000101000050427731303001830227300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372240000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
100243003722500003900612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  smlal v0.8h, v0.8b, v1.8b
  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)030818191e1f3a3f4e51inst 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)9fa0a1a6a8a9acc2c5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000092029548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071021611296340100001003003830038300383003830038
1020430037225000420072629548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372240000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000005030071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100006134277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
102043003722500000048529548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730180111020110099100100100001000000060071011611296340100001003003830038300383003830038
102043003722500000032529548251010010510000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000020000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
1020430037225000000124329548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308181e1f3f4e51inst 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)9fa0a1a6a8acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000061295482510010101000010100006142773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
1002430037225000061295482510010101000010100006142773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
1002430037225000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229632010000103003830038300383003830038
1002430037225000061295482510010101000011100005042773130300183003730037282873287671001020100002030000300373003751100211091010100001000000006402162229630010000103003830038300383003830038
1002430037225000061295482510010101000010105965042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103008530038300383003830038
10024300372250000189295482510010101000010100005042813841300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
10024300372250012061295482510010101000010100005042773131300183003730037282873288431001020100002030000300373003711100211091010100001000003006402162229630010000103003830038300383022830038
10024300372250051061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
1002430037225000061295302510010101000010100005042773130300183022530037282873287671001020100002030000300373003711100211091010100001000000006402402229630010000103003830038300383003830038
100243003722500120156295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000013007252162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  smlal v0.8h, v1.8b, v0.8b
  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)031e1f3f4e51inst 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)9fa8cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037224002122954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722500612954825101001041000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
1020430037225001262954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
1020430037225002332954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100207102162229634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
1020430037225003462954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722590612954844101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731330018300373003728265328766101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038

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)9fa8cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250189295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
10024300372250764295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001006402162729668010000103003830038300383003830038
1002430037225084295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
10024300372250124295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
10024300372250726295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
10024300372250747295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
1002430037224082295482510010101000010100005042773131300183003730084282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
10024300372250149295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
10024300372250172295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038
1002430037225061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001006402162229630010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  smlal v0.8h, v8.8b, v9.8b
  movi v1.16b, 0
  smlal v1.8h, v8.8b, v9.8b
  movi v2.16b, 0
  smlal v2.8h, v8.8b, v9.8b
  movi v3.16b, 0
  smlal v3.8h, v8.8b, v9.8b
  movi v4.16b, 0
  smlal v4.8h, v8.8b, v9.8b
  movi v5.16b, 0
  smlal v5.8h, v8.8b, v9.8b
  movi v6.16b, 0
  smlal v6.8h, v8.8b, v9.8b
  movi v7.16b, 0
  smlal v7.8h, v8.8b, v9.8b
  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)6061696d6e74scheduler 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)9fa8cdcfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200651500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064532280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
1602042006415002892580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611201271600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115011611200611600001002006520065200652006520065
160204200641500812580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010000101115111611200611600001002006520065200652006520065
160204200641500392580100100800001008000050064000015200452006420064032280100200800002002400002006420064111602011009910010016000010001101115111613200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

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)5f60696d6escheduler 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)9fa0a8a9accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420085150011629800121280000128000062640000012004120060200603228001220800002024000020060200601116002110910101600001000001003131183421146202422201160000102005220052200522005220052
16002420051150026127800121280000128000062640000112004120051200513228001220800002024000020051200511116002110910101600001000001003231152521185202322201160000102005220052200522005220052
16002420051150065927800121280000128000062640000112003220051200603228001220800002024000020051200511116002110910101600001000001002831162521185202232201160000102005220052200522005220052
1600242005115004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001001001003131162521158202442201160000102005220052200522005220052
16002420051150034527800121280000128000062640000112003220051200513228011820800002024000020051200511116002110910101600001000001002731182521166202212201160000102005220052200522005220052
1600242005115004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000001003331142521166202212201160000102005220052200522005220052
1600242005115004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000001002831142521185202112201160000102005220052200522005220052
1600242005115004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000001002731142521166202162201160000102005220052200522005220061
1600242005115004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000001002931162521166202192201160000102005220052200522005220052
16002420051150010829800121280000128000062640000112003220051200513228001220800002024000020060200511116002110910101600001000001002931162521164202192201160000102005220052200522005220052

Test 6: throughput

Count: 16

Code:

  smlal v0.8h, v16.8b, v17.8b
  smlal v1.8h, v16.8b, v17.8b
  smlal v2.8h, v16.8b, v17.8b
  smlal v3.8h, v16.8b, v17.8b
  smlal v4.8h, v16.8b, v17.8b
  smlal v5.8h, v16.8b, v17.8b
  smlal v6.8h, v16.8b, v17.8b
  smlal v7.8h, v16.8b, v17.8b
  smlal v8.8h, v16.8b, v17.8b
  smlal v9.8h, v16.8b, v17.8b
  smlal v10.8h, v16.8b, v17.8b
  smlal v11.8h, v16.8b, v17.8b
  smlal v12.8h, v16.8b, v17.8b
  smlal v13.8h, v16.8b, v17.8b
  smlal v14.8h, v16.8b, v17.8b
  smlal v15.8h, v16.8b, v17.8b
  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)03080b181e373f51inst 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)6061696d6edispatch 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)9fa6a8acc5cfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602044005930000000832516010010016000010016000050012800001040020400394003919973319997160100200160000200480000400484003911160201100991001001600001000100101100011611400361600001004004040040400404004940040
1602044003930000000502516010010016000010016000050012800000040020400394003919973320006160100200160000200480000400484003911160201100991001001600001000000101100011611400451600001004004940040400404004040049
1602044004829901000412516010010016000010016000050012800001040020400394003919973320006160100200160000200480000400394004811160201100991001001600001000000101100011611400361600001004004940040400494004040040
1602044003930000000412516011710016001710016000050012800000040029400484003919973319997160100200160000200480000400394003911160201100991001001600001000000101100011611400361600001004004940040400494004040040
1602044004830000000412516010010016000010016000050012800001040020400394003919973319997160100200160000200480000400394003911160201100991001001600001000000101100011611400361600001004004040040400404004040040
16020440039300000007152516010010016001710016000050012800001040029400394003919973319997160100200160000200480000400484003911160201100991001001600001000000101100011611400451600001004004040049400404005040040
1602044003930000000692516010010016000010016000050023989990040020400394003919973320023160100200160000200480000400394004811160201100991001001600001000000101100014611400361600001004004940040400494004040040
1602044003930000000412516010010016000010016000050012800001040020400394003919973319997160100200160000200480000400394003911160201100991001001600001000000101100011611400361600001004004040049400404004940040
16020440039300000570412516010010016001710016000050012800001040020400394004819973319997160100200160000200480000400394004811160201100991001001600001000000101100011611400361600001004004040040400404004940049
1602044004829900000412516011710016000010016000050012800000040020400394003919973319997160100200160000200480000400494003911160201100991001001600001000000101100011611400451600001004004040049400404004040040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03373f51inst 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)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)9facc5branch mispredict (cb)cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400483000552516001010160000101600005024388651154002040039400391999632002816001020160000204800004003940039111600211091010160000100001002213418162114240045206160000104004940040400404004140040
16002440039300175525160027101600181016011250239908211104002040039400391999632001916001020160000204800004003940049111600211091010160000100001002213616162114240037209160000104004940040400404004040041
1600244003930014725160028101600001016000050243886511104002140040400491999632002916001020160000204800004003940040111600211091010160000100001002213515162113440046206160000104004140040400504004040040
16002440048300184625160028101600011016000050128000011104002040048400401999632002816001020160000204800004004040049111600211091010160000100001002213514162114240036206160000104004140050400494004040040
160024400493001771125160027101600171016000050131999811104003040048400401999632001916001020160000204800004003940039111600211091010160000100001002213614162113440046209160000104004040049400404004940040
1600244003930004625160027101600001016000050128000011104002040039400391999632002816001020160000204800004003940040111600211091010160000100001002213616162114440037206160000104004040041400494004940040
16002440040300184625160028101600171016000050243886511104002040048400391999632002816001020160000204800004003940039111600211091010160000100001002213618162112440037207160000104004040049400414005040040
1600244003929904625160010101600171016000050243886511104002040049400391999632001916001020160000204800004003940048111600211091010160000100001002213614162114240045206160000104004140040400414005040050
16002440048300175525160010101600001016000050128000011104002040039400481999632001916001020160000204800004003940048111600211091010160000100001002213616164114340037206160000104004140050400504004040040
160024400392991772025160010101600001016000050239899911104002040039400491999632001916001020160000204800004004840040111600211091010160000100001002213615162334440036206160000104004940050400414005040049