Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMLSL (by element, 4S)

Test 1: uops

Code:

  sqdmlsl v0.4s, v1.4h, v2.h[1]
  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
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000373316332630100030383038303830383038
1004303723014925482510001000100039831313018303730372415328951000100030003037303711100110000073316332630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110000073316332630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110000073316332630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073316332630100030383038303830383038
100430372336125482510001000100039831303018303730372415328951000100030003037303711100110000073316432630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073316332630100030383038303830383038
10043037221210525482510001000100039831313018303730372415328951000100030003037303711100110003073316332630100030383038303830383038
100430372206125482510001000100039831303018303730372415328951000100030003037303711100110000073316332630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110000073316332630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqdmlsl v0.4s, v1.4h, v2.h[1]
  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)0308090b0f1e3a3f4e51inst 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)9fa0a8acbranch mispredict (cb)cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722510000025029548251014610010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
102043003722500000010529548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020431002300853003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
102043003722500000012329548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071212162229634100001003003830038300383003830038
102043003722500000012329548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745102532001000020030000300373003711102011009910010010000100000071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e3a3f4e51inst 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)9fa1accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372240078929548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640316222963010000103003830038300383003830038
1002430037225006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250010629548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216322963010000103003830038300383003830038
100243003722584025029548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830086
100243003722500200229548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
1002430037224006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216232963010000103003830038300383003830038
1002430037225006129548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216322963010000103003830038300383003830038
10024300372250072629548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216322963010000103003830038300383003830038
10024300372250016829548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216232963010000103003830038300383003830038
10024300372250045629548251001010100001010000504277313300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqdmlsl v0.4s, v0.4h, v1.h[1]
  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)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)9fa6a8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030084300374110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722400612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722400612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250519612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000007101162229634100001003003830038300383003830038
1020430037224012612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030084300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010020007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e1f3f4e51inst 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)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000061295482510010101000010100005042773130300183003730037282878287671016220100002030000300373003711100211091010100001000000006403162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830224300853003830038
100243003722501000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001004004006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722400000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006682162229630010000103003830038300383003830038
1002430037225000000156295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqdmlsl v0.4s, v1.4h, v0.h[1]
  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)03070b191e1f3a3f4e51inst 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)9fa6a8cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000007262954825101001001000010010000500427731313001830037300372826532874510125200100002003000030037300371110201100991001001000010000710011611296340100001003003830038300383003830038
1020430037225000000612954825101251001000012510000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000710011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110202100991001001000010000710011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100204100002023000030037300371110201100991001001000010010710011633296340100001003003830038300383003830038
1020430037225000900612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000710011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874410100200100002003000030037300371110201100991001001000010000710011633296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000710011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000710011651296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510125200100002003000030037300371110201100991001001000010000710011611296340100001003003830038300383003830038
10204300372250007500612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000710011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e1f3f4e51inst 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640316222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372240000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731330018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlsl v0.4s, v8.4h, v9.h[1]
  movi v1.16b, 0
  sqdmlsl v1.4s, v8.4h, v9.h[1]
  movi v2.16b, 0
  sqdmlsl v2.4s, v8.4h, v9.h[1]
  movi v3.16b, 0
  sqdmlsl v3.4s, v8.4h, v9.h[1]
  movi v4.16b, 0
  sqdmlsl v4.4s, v8.4h, v9.h[1]
  movi v5.16b, 0
  sqdmlsl v5.4s, v8.4h, v9.h[1]
  movi v6.16b, 0
  sqdmlsl v6.4s, v8.4h, v9.h[1]
  movi v7.16b, 0
  sqdmlsl v7.4s, v8.4h, v9.h[1]
  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)0318191e1f3f51inst 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602042008915000408039258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100000010111316112006101600001002006520065200652006520065
16020420064150000039258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
1602042006415000270514258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
16020420064150000039258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
1602042006415000002001258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
1602042006415000105039258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
160204200641510015039258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
160204200641500063039258010010080000100800005006400000200452006420064322801002008000020024000020064200642116020110099100100160000100000010111116112006101600001002006520065200652006520065
16020420064150000039258010010080000100800005006400000200452006420064322801002008000020024000020064200641116020110099100100160000100000010111116112006101600001002006520065200652006520065
160204200641500021039258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100000010112116112006101600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0304081e1f3a3f51inst 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)9fa0a8accfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420077150200006827800121280000128000062640000112004320051200513228001220800002024000020062200621116002110910101600001000010063652139363221936200592412160000102006320063200632006320063
16002420062150200018029800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000010059652120363223636200592412160000102006320063200632006320063
160024200621502000079298001212800001280000626400000120043200622006232280012208000020240000200622006211160021109101016000010000100591052133363223531200592412160000102006320063200632006320063
16002420062151209008029800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000010063652136363223728200592412160000102006320063200632006320063
16002420062151200018029800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000610060652133363222035200592412160000102006320063200632006320063
16002420062150200017429800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000010060652135363223536200592412160000102006320063200632006320063
16002420062151200007429800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000010063652137363223629200592412160000102006320063200632006320063
16002420062150200017429800121280000128000062640000012004320062200623228001220800002024000020051200511116002110910101600001000010063652136363223132200592412160000102006320063200632006320063
16002420062151100008029800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000010048652124363223336200592412160000102006320063200632006320063
16002420062150200007429800121280000128000062640000012004320062200623228001220800002024000020062200621116002110910101600001000010062652135363223528200592412160000102006320063200632006320063

Test 6: throughput

Count: 12

Code:

  sqdmlsl v0.4s, v12.4h, v13.h[1]
  sqdmlsl v1.4s, v12.4h, v13.h[1]
  sqdmlsl v2.4s, v12.4h, v13.h[1]
  sqdmlsl v3.4s, v12.4h, v13.h[1]
  sqdmlsl v4.4s, v12.4h, v13.h[1]
  sqdmlsl v5.4s, v12.4h, v13.h[1]
  sqdmlsl v6.4s, v12.4h, v13.h[1]
  sqdmlsl v7.4s, v12.4h, v13.h[1]
  sqdmlsl v8.4s, v12.4h, v13.h[1]
  sqdmlsl v9.4s, v12.4h, v13.h[1]
  sqdmlsl v10.4s, v12.4h, v13.h[1]
  sqdmlsl v11.4s, v12.4h, v13.h[1]
  movi v12.16b, 13
  movi v13.16b, 14

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2503

retire (01)cycle (02)031e1f373f4e51inst 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)9fa0a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1202043005922700041025120152100120000100120000500960000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000007610616333003601200001003004030040300403004030040
1202043003922490041025120100100120000100120000500960000300703003930039149733149971201002001200002003600003003930039111202011009910010012000010000007610516543003601200001003004030923300403004030040
12020430039225210041025120100100120018100120000500960000300203003930039149733149971201002001200002003600003003930427211202011009910010012000010001007610516353003601200001003004030040300403004030952
1202043095122500061025120100100120000100120000500960000300203003930039149733159091201002001200002003600003003930039111202011009910010012000010000007610416443091901200001003095230040300403004030040
12020430039225210064352325120100100120000100120000500960000300203003930951149733149981201002001200002003600003003930039111202011009910010012000010000007610416543003601200001003095230040309523004030040
1202043003922500041352325120100100120052100120000500960000300203003930041158573159091201002001200002003600003003930039111202011009910010012000010000007610516453003601200001003004030040300403004030040
12020430039225150041025120100100120000100120000500960000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000007610316443003601200001003004030952300403004030923
12020430039225120082025120100100120000100120000500960000300203174830039149733149971201002001200002003600003003930039111202011009910010012000010000007610416443094801200001003004030040300403004030040
1202043003923260041025120100100120000100120000500960000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000007610516553003601200001003004030040300403004030040
12020430039225510061025120185100120000100120000500960000300203003930039149733149971201002001200002003600003003930951111202011009910010012000010000007610416553003601200001003004030040300403004030040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2503

retire (01)cycle (02)03041e1f373f4e51inst 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)cfd0d5d6dbddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12002430039225124006102512002710120001101200005096000013002030039300391499631501912001020120000203600003003930039111200211091010120000100007520011602330036120000103004030040300403004030040
120024300392250180040025120010101200001012000050990000130021300403004014996315902120010201200002036000030040300391112002110910101200001000075201241602230036120000103004030040300403004030040
12002430039225021004002512001010120000101200005096000013002030039300391499631501912001020120000203600003003930039111200211091010120000100007520031633230036120000103004030040300403004030040
120024300392250261014002512001010120000101200005096000013002130040300401499631501912001020120000203600003003930039111200211091010120000100007520021602230036120000103175130040309233004030040
120024300392250300004502512001010120000101200005096000013002030039300391499631501912001020120000203600003003930039111200211091010120000100007520041604430036120000103004030040300403004031751
120024300392250369004002512004510120001101200005099000013090330040309971499631501912001020120000203600003003930039111200211091010120000100007520031603230037120000103175130041300413004130041
12002431750238098804002512001010120001101200005096000013002030039300391499631501912001020120000203600003003930177111200211091010120000100007520021602230036120000103004030040300403004030040
12002430039225021004002512001010120000101200005096000013002030039300391499631501912001020120000203600003003930039111200211091010120000100007520031603230919120000103004030040300403004030040
1200243003922500004002512001010120000101200005096000013002030039300391499631502012001020120000203600003175030040111200211091010120000100007520021603230036120000103004030040300403004030040
120024300392250300004002512001010120000101200005096000013002030039300391499631501912001020120000203600003003930039111200211091010120000100007520041604330948120000103004030040300403004030923