Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRDMLSH (vector, 4S)

Test 1: uops

Code:

  sqrdmlsh v0.4s, v1.4s, v2.4s
  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
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110000073316222630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100030003037303711100110003073216222630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110001373216222630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
1004303723077325482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100030003037303711100110002073216222630100030383038303830383038
1004303723925125482510001000100039831303018303730372415328951000100030003037303711100110000073316222630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqrdmlsh v0.4s, v1.4s, v2.4s
  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)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)9fa0a1a6a8a9acc2cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000090019329548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710021622296340100001003003830038300383003830038
102043003722500000006129548461010010210000100100005004277313030018300373003728265032874510100200100002023000030037300371110201100991001001000010000000000710021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710021632296340100001003003830038300383003830038
1020430037225000021016129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
1020430037224000000072629548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710121722296340100001003003830038300383003830038
1020430037225000000061929548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265032874510100200100002003000030037300371110201100991001001000010000000000710121632296340100001003003830038300383003830038

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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722505412954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010006403162229630010000103003830038300383003830038
100243003722401262954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722501562954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722505362954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630110000103003830038300383003830038
100243003722501472954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722501262954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqrdmlsh v0.4s, v0.4s, v1.4s
  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)0318191e1f3f4e51inst 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)9faccfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225006606129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225003906129548251010010010000100100005004277313030018300373003728265328750101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
10204300372250042606129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
10204300372250019806129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225111206129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037224002406129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
10204300372250012306129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
10204300372250026106129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225003906129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038

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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010036402162229630010000103003830085300383003830038
100243003722400061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229668010000103003830038300383003830038
1002430037225000441295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
1002430037225000726295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqrdmlsh v0.4s, v1.4s, v0.4s
  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)03080b1e3f4e51inst 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)9fa8c2cfd0d2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225004561295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
1020430037225004261295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
102043003722400061295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
102043003722500661295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
1020430037225001561295482510100100100001001000050042773130030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
1020430037224003661295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
1020430037225001861295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
102043003722400661295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
1020430037225006361295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038
102043003722500361295482510100100100001001000050042773131030018300373003728265328745101002001000020030000300373003711102011009910010010000100007100011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0318191e1f3f4e5051inst 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)9fa0a8acc3cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225002610612954802510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000006404164429630010000103003830038300383003830038
1002430037225005406129548300212510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006403164329630010000103003830038300383003830038
1002430037224001506129548300212510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006404164429630010000103003830038300383003830038
100243003722500570612954802510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006403164429630010000103003830038300383003830038
100243003722500510612954802510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006404163429630010000103003830038300383003830038
100243003722500150612954802510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000006403164429630010000103003830038300383003830038
100243003722500210612954802510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006404163429630010000103003830038300383003830038
10024300372250030011022954802510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006404164329630010000103003830038300383003830038
100243003722500360612954802510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006403164329630010000103003830038300383003830038
100243003722500210612954802510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006404164429630010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqrdmlsh v0.4s, v8.4s, v9.4s
  movi v1.16b, 0
  sqrdmlsh v1.4s, v8.4s, v9.4s
  movi v2.16b, 0
  sqrdmlsh v2.4s, v8.4s, v9.4s
  movi v3.16b, 0
  sqrdmlsh v3.4s, v8.4s, v9.4s
  movi v4.16b, 0
  sqrdmlsh v4.4s, v8.4s, v9.4s
  movi v5.16b, 0
  sqrdmlsh v5.4s, v8.4s, v9.4s
  movi v6.16b, 0
  sqrdmlsh v6.4s, v8.4s, v9.4s
  movi v7.16b, 0
  sqrdmlsh v7.4s, v8.4s, v9.4s
  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)030818191e1f3f51inst 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)9fa0a6a8a9accfd0d2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020420089151000903925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000001011150116112006101600001002006520065200652006520065
160204200641500002403925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000001011151116112006101600001002006520065200652006520065
16020420064151000003925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000001011151116112006101600001002006520065200652006520065
160204200641510009003925801001008000010080000500640000015201342006420064322801002008000020024000020064200641116020110099100100160000100000001011150116112006101600001002006520065200652006520065
16020420064150000003925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000001011151116112006101600001002006520065200652006520065
16020420064150000003925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000031011150116212006101600001002006520065200652006520065
160204200641500002403925801001008000010080000500640000115200452006420064322801002008000020024000020064200641116020110099100100160000100000001011150116112006101600001002006520065200652006520065
16020420064150000903925801001008000010080000500640000115200452006420064322801002008000020024000020064200641116020110099100100160000100000001011150116112006101600001002006520065200652006520065
160204200641500008703925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000001011150116112006101600001002006520065200652006520065
16020420064150000903925801001008000010080000500640000015200452006420064322801002008000020024000020064200641116020110099100100160000100000001011150116112006101600001002006520065200652006520065

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2507

retire (01)cycle (02)03041e3a3f51inst 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)9fa8cdcfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024200761511615127800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010050861146271111726200502211160000102005420054200542005420054
1600242005315013614527800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010048871125271112527202482211160000102005420054200542005420054
1600242005315017205127800121280000128000062640000115200342005320053322800122080000202400002005320062111600211091010160000100010048871126271111926200502211160000102005420054200542005420054
1600242005315001215127800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010048871126271112625200502211160000102005420054200542005420054
1600242005315113014527800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010038871115271112521200502211160000102005420054200542005420054
1600242005315011515127800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010040871126271111625200502211160000102005420054200542005420054
1600242005315011805127800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010041871123271112520200502211160000102005420054200542005420054
16002420053150122815127800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010040871121271112521200502211160000102005420054200542005420054
1600242005315011215127800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010048871125271111928200502211160000102005420054200542005420054
160024200531501017227800121280000128000062640000115200342005320053322800122080000202400002005320053111600211091010160000100010048871125271112628200502211160000102005420054200542005420054

Test 6: throughput

Count: 16

Code:

  sqrdmlsh v0.4s, v16.4s, v17.4s
  sqrdmlsh v1.4s, v16.4s, v17.4s
  sqrdmlsh v2.4s, v16.4s, v17.4s
  sqrdmlsh v3.4s, v16.4s, v17.4s
  sqrdmlsh v4.4s, v16.4s, v17.4s
  sqrdmlsh v5.4s, v16.4s, v17.4s
  sqrdmlsh v6.4s, v16.4s, v17.4s
  sqrdmlsh v7.4s, v16.4s, v17.4s
  sqrdmlsh v8.4s, v16.4s, v17.4s
  sqrdmlsh v9.4s, v16.4s, v17.4s
  sqrdmlsh v10.4s, v16.4s, v17.4s
  sqrdmlsh v11.4s, v16.4s, v17.4s
  sqrdmlsh v12.4s, v16.4s, v17.4s
  sqrdmlsh v13.4s, v16.4s, v17.4s
  sqrdmlsh v14.4s, v16.4s, v17.4s
  sqrdmlsh v15.4s, v16.4s, v17.4s
  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)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)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)9fa8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602044005930000175125160100100160000100160000500128000040020400394003919973320010160100200160000200480000400394003911160201100991001001600001000001011011611400361600001004004040040400404004040040
160204400393000004125160100100160000100160000500128000040030400494004919973320007160100200160000200480000400494004911160201100991001001600001000001011011611400361600001004004040040400404004040040
160204400393000004125160100100160000100160000500128000040020400394003919973319997160100200160000200480000400394003911160201100991001001600001000001011011611400361600001004004040049400404004040040
1602044003930000185125160100100160001100160000500128000040020400494004919973319997160100200160000200480000400394003911160201100991001001600001000001011011611400361600001004004040040400404005040040
16020440039300001851625160100100160000100160000500128000040020400394003919973319997160100200160000200480000400394003911160201100991001001600001001001011011611400461600001004004040050400504005040050
1602044004930000184125160101100160000100160000500132000040029400394003919973319997160100200160000200480000400494004911160201100991001001600001000001013511611400361600001004004040040400494004040040
160204400393000004125160100100160000100160000500128000040020400494004919973320007160100200160000200480000400394004811160201100991001001600001000011011011611400461600001004004040040400404004040040
1602044003930000185125160118100160018100160000500243886540020400394003919973319997160100200160000200480000400394003911160201100991001001600001000001011011611400461600001004004040040400404004040040
160204400393000004125160101100160000100160000500128000040030400394003919973319997160100200160000200480000400394003911160201100991001001600001000001011011611400461600001004005040050400504005040040
1602044003930000032625160100100160000100160000500128000040030400394003919989320007160100200160000200480000400494004911160201100991001001600001000001011011611400361600001004004040040400404004040040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307080a1e373a3f51inst 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)5f60696d6edispatch 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)9fa8a9acc2branch mispredict (cb)cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400493000005110462516001010160017101600005013199991140029400404004819996320019160010201600002048000040048400391116002110910101600001000000100223112416221252640037207160000104004140040400414004040041
16002440040299101310462516001010160000101600005012800001140021400394003919996320020160010201600002048000040040400391116002110910101600001000000100223111916211272240036206160000104004140040400504005040041
16002440040300000001622516002710160017101600005012800001140020400394004819996320020160010201600002048000040040400391116002110910101600001000000100223112716211222740037207160000104004040041400404004140040
160024400403000000170552516001010160000101600005023989991140021400494004919996320058160010201600002048000040039400481116002110910101600001000000100223112716211272740046206160000104004940049400404004940041
16002440117300000010462516001110160001101600005024388651140020400404004919996320029160010201600002048000040039400491116002110910101600001000000100223112716211282840037206160000104005040041400404004140040
16002440039300000000462516002710160017101600005012800001140020400394004819996320034160010201600002048000040040400391116002110910101600001030000100223112716211162640045209160000104004940040400494004040049
16002440040299000010472516001010160000101600005023990271140021400394004919996320029160010201600002048000040049400401116002110910101600001000000100223111616211272340045206160000104004040049400404004940040
160024400392990000170552516001010160018101600005012800001140020400494003919996320019160010201600002048000040039400491116002110910101600001000000100223112216211272240036207160000104004940049400494004940049
160024400483000000170562516002710160017101600005012800001140020400404003919996320020160010201600002048000040048400481116002110910101600001000000100223111416211272240045206160000104004040049400404004940040
160024400393000000170462516001010160000101600005013199971140021400484003919996320020160010201600002048000040049400401116002110910101600001000000100823112116211272240036209160000104004040052400504004040041