Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMLAL (vector, 2D)

Test 1: uops

Code:

  sqdmlal 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)03080b1e1f3f4e51inst 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)a1a6a7a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004303722000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303723000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303722000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303722000061254825100010001000398313130183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303722000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303722000061254825100010001000398313130183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
10043037230000891254825100010001000398313030183037303724153289510001000300030373037111001100000010073116112630100030383038303830383038
10043037230000116254825100010001000398313130183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303722000061254825100010001000398313130183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
1004303723000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqdmlal 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)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)9fa8cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100171013162229634100001003003830038300383003830038
102043003722400000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012163229634100001003003830038300383003830038
102043003722500000016129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162229634100001003003830038300383003830038
102043003722400000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162229634100001003003830038300383003830038
1020430037224000000010329548251010010010000100101495004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162329634100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162229634100001003003830038300383003830038
1020430037224000000010329548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012163229778100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162229634100001003003830038300383003830038
102043003722500000006129548251010010010000104100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162229634100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b1e1f3a3f4e51inst 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
100243003722500008682954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010012640816892963010000103003830038300383003830038
1002430037225030061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640816872963010000103003830038300383003830038
10024300372250000783295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640716882963010000103003830038300383003830038
1002430037225000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640716682963010000103003830038300383003830038
10024300372250000684295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640816872963010000103003830038300383003830038
1002430037225000082295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640816862963010000103003830038300383003830038
10024300372240000247295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640716782963010000103003830038300383003830038
100243003722500001644295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640916792963010000103003830038300383003830038
10024300372250000839295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640816872963010000103003830038300383003830038
10024300372250000872295482510019101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640716772963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqdmlal 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)030b1e3a3f4e51inst 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)9fa1a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007102161129634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129644100001003003830038300383003830038
1020430037224000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000607101161129634100001003003830038300383003830038
10204300372250009642954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037225001612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250120612954825101001201000810010000500427731313001830037300372826532874510100204100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001830037300372826532874510100204100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037225000822954825101001001000010010148500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000047101161129634100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03191e1f3f4e51inst 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)9fa1a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103008530038300383003830133
100243003723303010329548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001021000640216222963010000103003830038300383003830038
10024300372240006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222977410000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqdmlal 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)03191e1f3a3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)696b6d6edispatch 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)9fa0a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722400008229548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225012006129548251010010010000100100005004277313300183300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500006129548251010010410000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010148500427731330065030037300372826532874510100200100002003000030037300371110201100991001001000010000159171011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100006073411611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b191e1f3f4e51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000002512954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003723800000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313030018300373003728287252876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500012887262954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216422963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlal v0.2d, v8.2s, v9.2s
  movi v1.16b, 0
  sqdmlal v1.2d, v8.2s, v9.2s
  movi v2.16b, 0
  sqdmlal v2.2d, v8.2s, v9.2s
  movi v3.16b, 0
  sqdmlal v3.2d, v8.2s, v9.2s
  movi v4.16b, 0
  sqdmlal v4.2d, v8.2s, v9.2s
  movi v5.16b, 0
  sqdmlal v5.2d, v8.2s, v9.2s
  movi v6.16b, 0
  sqdmlal v6.2d, v8.2s, v9.2s
  movi v7.16b, 0
  sqdmlal 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)030809191e3a3f51inst 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)9fa1a7a8acbranch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200901500000051425801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
1602042006415100000394280101100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000002700010111116112006101600001002006520065200652006520065
16020420064151000003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
16020420064150000003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
16020420064150000003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
160204200641500000051825801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065
16020420064150000303925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000130010111116112006101600001002006520065200652006520065
16020420064150000003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112019201600001002006520065200652006520065
160204200641500000039258010010080000100800005006400000200452006420156322801002008000020024000020064200641116020110099100100160000100011300101111161120061121600001002006520065200652006520065
16020420146150000003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000000010111116112006101600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307090b1e3f51inst 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)9fa8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420087151000305127800121280000128000062640000115200322006020060322800122080000202400002005120060111600211091010160000100010030822783441147200572402160000102006120174200612006120052
1600242006015000004527800121280000128000062640000015200322006020060322800122080000202400002005120060111600211091010160000100010029822693421137200572402160000102005220061200522006120052
1600242005115000005127800121280000128000062640000105200412005120060322800122080000202400002005120051111600211091010160000100010026812532521156200482201160000102005220052200522005220052
1600242005115000004527800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100010028842552521157200482201160000102005220052200522005220052
1600242005115000004527800121280000128000062640000110200322005120051322800122080000202400002005120051111600211091010160000100010026842552521155200482201160000102005220052200522005220052
16002420051150000045278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000100268421525211511200482201160000102005220052200522005220052
1600242005115000004527800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100010028842052521294200482201160000102005220052200522005220052
1600242005115000004527800121280000128000062640000115200322005120051322800122080000202400002006220051111600211091010160000100010028842053422164200482201160000102005220061200522005220052
16002420051150000045298001212800001280000626400001152003220051200603228001220800002024000020051200601116002110910101600001000100261141932522156200482201160000102005220052200522005220052
1600242005115000004527800121280000128000062640000115200322005120051322800122080000202400002005120060111600211091010160000100010028841952521194200482201160000102005220052200522005220052

Test 6: throughput

Count: 16

Code:

  sqdmlal v0.2d, v16.2s, v17.2s
  sqdmlal v1.2d, v16.2s, v17.2s
  sqdmlal v2.2d, v16.2s, v17.2s
  sqdmlal v3.2d, v16.2s, v17.2s
  sqdmlal v4.2d, v16.2s, v17.2s
  sqdmlal v5.2d, v16.2s, v17.2s
  sqdmlal v6.2d, v16.2s, v17.2s
  sqdmlal v7.2d, v16.2s, v17.2s
  sqdmlal v8.2d, v16.2s, v17.2s
  sqdmlal v9.2d, v16.2s, v17.2s
  sqdmlal v10.2d, v16.2s, v17.2s
  sqdmlal v11.2d, v16.2s, v17.2s
  sqdmlal v12.2d, v16.2s, v17.2s
  sqdmlal v13.2d, v16.2s, v17.2s
  sqdmlal v14.2d, v16.2s, v17.2s
  sqdmlal 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)03081e1f373f51inst 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)9fa8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204400593000000412516010010016000010016000050012800001400204004840039199730320006160100200160000200480000400394003911160201100991001001600001000001011011611400361600001004004040040400494004940049
160204400392990000412516010010016001710016000050012800001400294003940039199730319997160100200160000200480000400394004811160201100991001001600001000001011021611400361600001004004040040400494004040040
160204400482990000412516010010016000110016000050023989991400294003940039199730320007160100200160000200480000400484003911160201100991001001600001000001011021612400361600001004004140050400504005040040
160204400393000000622516011710016000010016000050012800001400204003940039199730319997160100200160000200480000400394003911160201100991001001600001000001011011633400361600001004004040040400404004940040
160204400403000000412516010010016000010016000050012800001400204004840039199730319997160100200160000200480000400394005211160201100991001001600001000001011021612400361600001004004040040400404004040040
1602044004830002131320412516010010016000010016000050012800001400204003940048199730320006160100200160000200480000400404004921160202100991001001600001002001011011622400451600001004004040040400404004040040
160204400393000000512516010010016000010016000050023989991400204003940040199730319997160100200160000200480000400404004911160201100991001001600001000001011011621400361600001004004040040400404004040040
160204400393000000412516010010016000010016000050012800001400204004940039199730319997160100200160000200480000400394003911160201100991001001600001000001011021622400461600001004004040040400404004040040
160204400903000001412516010010016000110016000050012800001400204003940039199730319997160100200160000200480000400394003911160201100991001001600001000001011021622400361600001004004040040400414004040049
160204400393000000422516010010016000010016000050012800001400204004940039199730319997160100200160000200480000400494004811160201100991001001600001000001011011612400361600001004004040040400404004040040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0308090a0b18191e1f373f51inst 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)9fa0a1a6a7a8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002440048300000000001464416001010160000101600005023989991104002040039400391999632002816001020160000204800004004040048111600211091010160000100001301008584171621157400360206160000104004040050400404004040040
16002440040299000000000462516001010160001101600005019033241154002940039400391999632001916001020160000204800004003940039111600211091010160000100400001008082151621177400450206160000104005040040400404004040040
16002440039300000000000462516002710160000101600005012800001154002040039400391999632002916001020160000204800004003940039111600211091010160000100000101009064181621178400360206160000104026040049400494004040040
1600244003930000000000046251600101016000010160000501280000115401264003940039199963200191600102016000020480000400394003911160021109101016000010000030100691152660222107400360406160000104004040040400404004040313
1600244003930000000000175625160010101600001016000050128000011540020400494026719996320019160010201600002048000040039400481116002110910101600001000000310064115191621177400450206160000104004940049400404004040040
1600244003929900000000071225160010101600001016000050128000011540020400394003919996320029160010201600002048000040039400391116002110910101600001000000010073841101621154400460209160000104004040040400414004040040
16002440049300000000000672516001010160000101600005012800001154002140039400391999632001916001020160000204800004004840039111600211091010160000100020001006484141621145400370207160000104004040040400404004040040
160024400393000000000004625160010101600001016000050128000011540020400394003919996320029160010201600002048000040048400481116002110910101600001000000010064115251621145400360207160000104004040041400504004940040
160024400392990000000004625160010101600001016000050128000011540020400394003919996320019160010201600002048000040039400391116002110910101600001000000010064115256122155400450206160000104004040040400504004040040
16002440049300000000900462516001010160000101600005012800001154002040039400391999632001916001020160000204800004003940039111600211091010160000100000001006884151621157400360206160000104004040040400404004040040