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, 4H)

Test 1: uops

Code:

  mla v0.4h, v1.4h, v2.4h
  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)030b1e3f4e51inst 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
1004303722006125482510001000100039831313018303730372415328951000100030003037303711100110000073216112630100030383038303830383038
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303722006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230661254825100010001000398313130183037303724153289510001000300030373037111001100002473116112700100030383038303830383038
1004303723006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110000373116112630100030383038303831213038
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110001073116112630100030383038303830383038
1004303722006125482510001000100039831303018303730372415328951000100030003037303711100110000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  mla v0.4h, v1.4h, v2.4h
  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)0307080a0b18191e1f3a3f4e51inst 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)9fa0a1a6a8a9acc2c5cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225100000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000123000710021622296340100001003003830038300383003830038
10204300372250000000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000200710021623296340100001003003830038300383003830038
1020430037225000000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000111000710121622296340100001003003830038300383003830038
1020430037225000000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000126000710121633296340100001003003830038300383003830038
1020430037224000000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000129000710121622296340100001003003830038300383003830038
1020430037225000000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000114000710121622296340100001003003830038300383003830038
10204300372250000000006129548251010010010000100100005004281384130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000010710121622296340100001003003830038300383003830038
1020430037225000000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000111000710121622296340100001003003830038300383003830038
102043003722500000027006129539251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000003000710121624296340100001003003830038300383003830038
102043003722500000000010329548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100020120000710121622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030a0b1e3f4e51inst 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)9fa8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010260640416442963010000103003830038300383003830038
100243003722400053629548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000102190640616452963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828732876710310201000020300003003730037111002110910101000010030640616442963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000102860640416452963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000101030640417452963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018300373003728287262876710010201000020300003003730037111002110910101000010130640516542963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010030640416442963010000103003830038300383003830038
1002430037225000822954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010060640416742963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010030640416442963010000103003830038300383003830038
1002430037224000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010130640416442963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  mla v0.4h, v0.4h, v1.4h
  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)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)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
102043003722509432954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001001971011611296340100001003003830038300383003830038
10204300372250612954825101001001000010010000626427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001000971011611296340100001003003830038300383003830038
10204300372250612954825101001001000012510000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001001371011611296340100001003003830038300383003830038
10204300372250612954825101251251000013110000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001000671011611296340100001003003830038300383003830038
1020430037225038252954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001001371011611296340100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001004671011611296340100001003003830038300383003830038
102043003722506129548251010010010000100100005004277313130018300373003728265032874510100200100002003000030037300371110201100991001001000010009371021611296340100001003003830038300383003830038
10204300372240612954825101001001000012510000500427731303001830037300372826503287451010020010000200300003003730037111020110099100100100001001671011611296340100001003003830038300383003830082
102043003722506129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010006710116112963425100001003003830038300383003830038
1020430037225028662954825101001001000010010000500427731313001830037300372826503287451010020010000200300003003730037111020110099100100100001001371011611296340100001003003830038300383003830038

1000 unrolls and 10 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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001020640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037225001562954825100101010000101000050427731313001830037300372828725287561001020100002030000300373003711100211091010100001010640316222963010000103003830038300383003830038
100243003722500726295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
100243003722500612954810310010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001023640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001010640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001010640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001010640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001013640216222963010000103003830038300383003830038
100243003722500842954810310010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001020640216222963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  mla v0.4h, v1.4h, v0.4h
  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)03191e3f4e51inst 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)9fa8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007102161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003008430084211020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000017101161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001001907101161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
102043003722500251295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225006129548651010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100561207101161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001001307101161129634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730085282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038

1000 unrolls and 10 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)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)9fa8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500612954825100101010000101000050427731303001830037300372830532876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010008101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010100640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427779703001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  mla v0.4h, v8.4h, v9.4h
  movi v1.16b, 0
  mla v1.4h, v8.4h, v9.4h
  movi v2.16b, 0
  mla v2.4h, v8.4h, v9.4h
  movi v3.16b, 0
  mla v3.4h, v8.4h, v9.4h
  movi v4.16b, 0
  mla v4.4h, v8.4h, v9.4h
  movi v5.16b, 0
  mla v5.4h, v8.4h, v9.4h
  movi v6.16b, 0
  mla v6.4h, v8.4h, v9.4h
  movi v7.16b, 0
  mla v7.4h, v8.4h, v9.4h
  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)033f51inst 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020420089150392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010111216112006101600001002006520065200652006520065
16020420064150392580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001002310111116112006101600001002006520065200652006520065
16020420064151392580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000010111116112006101600001002006520065200652006520065
160204200641503925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010030310111116112006101600001002006520065200652006520065
16020420064150392580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000010111116112006101600001002006520065200652006520065
16020420064150392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010111116112006101600001002006520065200652006520065
16020420064150392580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000010111116112006101600001002006520065200652006520065
160204200641503925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010015310111116112006101600001002006520065200652006520065
160204200641503925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010051010111116112006101600001002006520065200652006520065
16020420064150392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010111116112006101600001002006520065200652006520065

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)9fa8accfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600242006315022001232580012128000012800006264000010200270200482004832280012208000020240000200482004811160021109101016000010001004231202420211211620043216160000102004720047200472004920047
1600242004815110001902580012128000012800006264000010200270200462004632280012208000020240000200462004611160021109101016000010001004131101920211192120043215160000102004920049200472004720047
160024200481501100722580012128000012800006264000011200290200482004632280012208000020240000200462004611160021109101016000010101003831101920111202220045215160000102004720047200472004720049
16002420046150110012172580012128000012800006264000011200290200462004632280012208000020240000200482004811160021109101016000010001004131101824111211520043215160000102004920049200472004720047
16002420046150130069625800121280314128000062640000102002702004620046322800122080000202400002004820052111600211091010160000103310042311020117211192420045216160000102029020049200492004720049
16002420048150002408242580012128000012800006264000010200270200462004634380012208000020240000200482004811160021109101016000010001004231102022211202020043215160000102004720047200492004920047
160024200461501200512580012128000012800006264000010200270200482004632280012208000020240000200482004611160021109101016000010001004531102022211221920043215160000102004720047200492005320047
160024200461501160722580012128000012800006264000011200290200502004632280012208000020240000200502004611160021109101016000010001004232111820221161620043216160000102004720047200472004720051
1600242004815011308082580012128000012800006264000011200270200462004632280012208000020240000200462004611160021109101016000010101004331102020211162020043215160000102004720047200472004920047
160024200461511100452580012128000012800006264000011200290200482004632280012208000020240000200462004611160021109101016000010001004331201920211202420049215160000102004920053200492004720047

Test 6: throughput

Count: 16

Code:

  mla v0.4h, v16.4h, v17.4h
  mla v1.4h, v16.4h, v17.4h
  mla v2.4h, v16.4h, v17.4h
  mla v3.4h, v16.4h, v17.4h
  mla v4.4h, v16.4h, v17.4h
  mla v5.4h, v16.4h, v17.4h
  mla v6.4h, v16.4h, v17.4h
  mla v7.4h, v16.4h, v17.4h
  mla v8.4h, v16.4h, v17.4h
  mla v9.4h, v16.4h, v17.4h
  mla v10.4h, v16.4h, v17.4h
  mla v11.4h, v16.4h, v17.4h
  mla v12.4h, v16.4h, v17.4h
  mla v13.4h, v16.4h, v17.4h
  mla v14.4h, v16.4h, v17.4h
  mla v15.4h, v16.4h, v17.4h
  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)03070a191e1f373a3f51inst 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)9fa8acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440061300110000470225160108100160017100160020500128013214002040039400391997761999016012020016003220048009640039400391116020110099100100160000100001111012213261311400361600001004004040040400494004040040
1602044003930011000172372516011710016001710016002050012801320400204003940048199776199901601202001600322004800964003940039111602011009910010016000010000000101141316613400361600001004004040040400494004040040
1602044003930011000024825160100100160000100160000500239899904002040049400391997331999716010020016000020048000040039400391116020110099100100160000100000001011415161311400361600001004004040040400494004040040
1602044003930011000172482516010010016000010016000050012800000400294003940039199733199971601002001600002004800004003940039111602011009910010016000010000000101141116713400451600001004004140040400494004040040
1602044003930011000024825160100100160000100160000500128000004002040039400391997331999716010020016000020048000040048400391116020110099100100160000100000001011411161212400361600001004004940040400404004040049
160204400483001100002532516010010016000010016000050012800000400204003940039199733199971601002001600002004800004003940048111602011009910010016000010000000101171416615400361600001004004040049400414004040049
1602044004830011000025725160117100160017100160000500128000014002040039400481997332000616010020016000020048000040039400481116020110099100100160000100000001011414161214400361600001004004940040400404004940040
160204400393001100002482516010010016000010016000050012800001400294003940039199733199971601002001600002004800004003940039111602011009910010016000010000000101141316136400361600001004004040040400404004040049
16020440048300110000252325160218100160000100160000500128000014002040039400391997331999716010020016000020048000040039400391116020110099100100160000100060001011415161214400361600001004004040040400404004040040
1602044003930011090024825160100100160000100160000500239899904002040039400391997331999716010020016000020048000040039400391116020110099100100160000100000001011414161313400361600001004004040040400404004940041

1000 unrolls and 10 iterations

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

retire (01)cycle (02)031e1f373f51inst 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)9fa8accdcfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002440060300900362251600101016001810160000501280000115400204003940039199963200281600102016000020480000400394003911160021109101016000010000100241131171621187400361560160000104004040041400404004040040
16002440039300120189425160010101600171016000050128000011540020400394003919996320019160010201600002048000040039400391116002110910101600001010010024831111622246400361550160000104004040040400404004040040
1600244004030000046251600271016000010160000501280000015400204003940039199963200191600102016000020480000400394003911160021109101016000010000100248411316412138400461550160000104005040040400404004940050
1600254003930000046251600271016001710160000502398999115400294004940039199963200191600102016000020480000400404003911160021109101016000010000100228316162111374003615150160000104004040040400404004040040
16002440039300000892516001010160018101600005012800001154002040039400391999632001916001020160000204800004008940039111600211091010160000100931100228311216211654003615517160000104005040050400404004140040
160024400493000004625160010101600001016000050128000011540020400494003919996320019160010201600002048000040089400391116002110910101600001000010022831416211611400461550160000104004040040400494004940040
16002440039300000462516001010160000101600005012800001154002040039400391999632001916001020160000204800004003940039111600211091010160000100001004283110162111112400361550160000104004040040400404004040040
160024400393000108046251600101016000010160000501280000115400204003940039199963200191600102016000020480000400394003911160021109101016000010000100228211116211139400361550160000104004040040400404004040040
16002440048300001856251600101016000010160000501280000115400204003940039199963200191600102016000020480000400394003911160021109101016000010000100221131716211764003615100160000104004040040400404004040040
1600244003929900185625160010101600181016000050128000011540030400494003919996320029160010201600002048000040039400391116002110910101600001000010022841816221145400361550160000104004040040400404004040040