Apple M1 Microarchitecture Research by Dougall Johnson

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

ADDHN2 (vector, 8H)

Test 1: uops

Code:

  addhn2 v0.16b, v1.8h, v2.8h
  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)696d6edispatch 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
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230612548251000100011493996703018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110001073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037236612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
100430372312612548251000100010003983133018303730372415328951000100030003037303711100110001373216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038

Test 2: Latency 1->1

Code:

  addhn2 v0.16b, v1.8h, v2.8h
  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)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)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)9fa0a1acbranch mispredict (cb)cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722566061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121620296340100001003003830038300383003830038
102043003722512061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121622296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000712121622296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121622296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282793287451010020010000200300003003730086111020110099100100100001000000710123222296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121622296340100001003003830038300383003830038
10204300372250161295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121622296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121632296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710121622296340100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282773287451010020010000200300003003730037111020110099100100100001000000710121622296340100001003003830038300383003830038

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)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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722502512954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000666316332963010000103003830038300383003830038
100243003722507262954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010200640316332963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
100243003722507262954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
100243003722507262954825100101010000101000050427731330090030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038
10024300372240612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640316332963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  addhn2 v0.16b, v0.8h, v1.8h
  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)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
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129672100001003003830038300383003830038
1020430037225071295482510100100100001001000050042786701300183003730037282653287451026220210000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e1f3a3f4e51inst 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)60696b6d6edispatch 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000000061295482510010101000010100005042773131300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
100243003722500000000350295482510010101000010100005042773131300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042773130300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042773130300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042786700300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042773130300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830086300383003830038
10024300372240000000061295482510010101000010100005042773130300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042773130300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042773130300180300373003728287032876710010201000020300003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000000061295482510010101000010100005042773130300370300373003728287032876710010201000020300003003730037111002110910101000010000110006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  addhn2 v0.16b, v1.8h, v0.8h
  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)033f4e51inst 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)9fa0a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372254142954825101001001000010010000500427731330018300373003728265328745101002001016720030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722520829548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000207102162229634100001003003830038300383003830038
10204300372254592954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722511032954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
102043003722510332954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
10204300372251452954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100407102162229634100001003003830038300383003830038
10204300372259382954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
10204300372259182954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229691100001003003830038300383003830038
10204300372257262954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038
10204300372259652954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100007102162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030809181e3a3f4e51inst 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)9fa0a1a6a7a8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000298329548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000644111610112963010000103003830038300383003830038
10024300372250000283029548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100031000644101610112963010000103003830038300383003830038
10024300372250000287729548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000644101610102963010000103003830038300383003830038
10024300372250000212529548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000644101610102963010000103003830038300383003830038
10024300372240000297629548251001010100001010000504277313130018300373003728287328767100102010000203054030037300371110021109101010000100000001644101610102963010000103003830038300383003830038
1002430037224000026229548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000644101610102963010000103003830038300383003830038
1002430037225000027542954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000000064410168102963010000103003830038300383003830038
10024300372250000293029548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000644101610102963010000103003830038300383003830038
10024300372250000296029548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000000644101610102963010000103003830038300383003830038
10024300372250000287129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000000644101610102963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  addhn2 v0.16b, v8.8h, v9.8h
  movi v1.16b, 0
  addhn2 v1.16b, v8.8h, v9.8h
  movi v2.16b, 0
  addhn2 v2.16b, v8.8h, v9.8h
  movi v3.16b, 0
  addhn2 v3.16b, v8.8h, v9.8h
  movi v4.16b, 0
  addhn2 v4.16b, v8.8h, v9.8h
  movi v5.16b, 0
  addhn2 v5.16b, v8.8h, v9.8h
  movi v6.16b, 0
  addhn2 v6.16b, v8.8h, v9.8h
  movi v7.16b, 0
  addhn2 v7.16b, v8.8h, v9.8h
  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)03080b18191e3f51inst 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)9fa8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602042009115000000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001001001011111611200611600001002006520065200652006520065
16020420064151000007432580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065
1602042006415010110392580100100800001008000050064000002004520145200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065
1602042006415000000392580100100802081008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065
1602042006415000000392580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065
16020420064150000001672580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065
160204200641510000039258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100024931011115611200611600001002006520065200652014520155
1602042006415011000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011118011201351600001002014620145200652017120065
1602042006415000000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065
1602042006415000000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011111611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03080b18191e1f3a3f51inst 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)9fa1a8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420088150000000019327800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000010029311102521197200482201160000102005220061200522005220052
1600242005115000000008422780012128000012800006264000011200322005120051322800122080000202400002005120060111600211091010160000100001003062162521196200572602160000102006120061200522005220052
160024200511510000000452780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001003231110252111010200482201160000102005220052200522005220052
160024200511500000003100927800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001001010029311102521169200482401160000102005220052200522005220052
1600242005115000000002362780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001003032192521179200572202160000102005220061200612005220061
160024200601500000000572780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100201003231162521169200482402160000102005220052200522005220052
160024200511500000000512780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001003231162521179200482201160000102005220052200522006120061
1600242005115000000002362780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001003231172521179200482201160000102005220052200522005220052
16002420051150000001080452780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001002931192521197200482201160000102005220052200522005220052
1600242005115000006002572780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100131003031192521196200482201160000102005220052200522005220052

Test 6: throughput

Count: 16

Code:

  addhn2 v0.16b, v16.8h, v17.8h
  addhn2 v1.16b, v16.8h, v17.8h
  addhn2 v2.16b, v16.8h, v17.8h
  addhn2 v3.16b, v16.8h, v17.8h
  addhn2 v4.16b, v16.8h, v17.8h
  addhn2 v5.16b, v16.8h, v17.8h
  addhn2 v6.16b, v16.8h, v17.8h
  addhn2 v7.16b, v16.8h, v17.8h
  addhn2 v8.16b, v16.8h, v17.8h
  addhn2 v9.16b, v16.8h, v17.8h
  addhn2 v10.16b, v16.8h, v17.8h
  addhn2 v11.16b, v16.8h, v17.8h
  addhn2 v12.16b, v16.8h, v17.8h
  addhn2 v13.16b, v16.8h, v17.8h
  addhn2 v14.16b, v16.8h, v17.8h
  addhn2 v15.16b, v16.8h, v17.8h
  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)0308181e1f373a3f51inst 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)9fa6a7a8acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440048299000000240251601001001600171001600005001280000040021400494004919973319997160100200160000200480000400394004911160201100991001001600001000000010110216224003601600001004004140040400404004140040
1602044004930000000042251601011001600171001600005001319998040029400394003919973320006160100200160000200480000400404003911160201100991001001600001000000010110216224003701600001004004040041400404004140040
1602044004029900001041251601001001600001001600005002399027140030400394004019973319998160100200160000200480000400394004011160201100991001001600001000009010110216224003601600001004005040040400404004940040
1602044004829900000041251601011001600001001600005002398999040029400484003919973319997160100200160000200480000400394004011160201100991001001600001000000010110216224004501600001004004040040400494004040040
16020440039300000017050251601171001600171001600005001319998140020400484003919973320006160100200160000200480000400394004811160201100991001001600001000000010110216224003601600001004005040049400414004040041
16020440049300000017050251601171001600001001600005001280000040021400484003919973319997160100200160000200480000400484003911160201100991001001600001000000010110216224004601600001004004040041400404004140049
1602044004930000000041251601001001600001001600005001319998140020400394003919973320006160100200160000200480000400484003911160201100991001001600001000000010145216224009601600001004004040041400404004140040
1602044004029900000041251601001001600001001600005001280000040020400404003919973319997160100200160000200480000400404004811160201100991001001600001000000010110216224004501600001004004940040400494004040040
16020440039300000017041251601001001600171001600005001280000140020400484003919973319997160100200160000200480000400484003911160201100991001001600001000000010110416224003601600001004004940040400494004040040
1602044004830000001050251601011001600001001600005001280000140029400394003919973319997160100200160000200480000400484003911160201100991001001600001000000010110216224003601600001004004040040400414004040049

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0304080b1e1f373f51inst 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)9fa8a9acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600244004930000000046251600101016001710160000501280000114002040049400491999632001916001020160000204800004003940049111600211091010160000101000100223122816211231640036305160000104004040040400404004040050
1600244003930000000046251600101016000010160000501280000114002040039400391999632001916001020160000204800004003940048111600211091010160000100000100223112416211222240046156160000104004040040400404004040040
1600244003930000000046251600101016000010160000501280000114002040039400391999632001916001020160000204800004004040039111600211091010160000100000100223112416211232440046155160000104004040040400404004040040
1600244003929900000056251600101016000010160000502398999114002040039400391999632001916001020160132204800004003940039111600211091010160000100000100223112216211242440036155160000104004040040400404004040040
16002440039300000000462516002710160000101600005012800001140020400394003919996272001916001020160000204800004003940048111600211091010160000100000100223112216211222340046155160000104005040040400404004040040
1600244003930000000046251600101016000010160000501280000114002040039400391999632001916001020160000204800004003940039111600211091010160000101000100223112316211232340036155160000104004040040400404004040040
16002440039300000001846251600101016000010160000501280000114003040048400391999632001916001020160000204800004003940039111600211091010160000100000100223112316211232340036155160000104004040040400404004040040
16002440039300000000462516001010160000101600005012800001140020400394003919996320019160010201600002048000040049400391116002110910101600001000001002462222164222322400363010160000104004940040400504005040040
16002440039300000000462516001010160000101600005012800000140020400394003919996320019160010201600002048000040039400391116002110910101600001000001002232222164222222400363010160000104004040041400404004040040
16002440039299000000462516002810160000101600005012800000140030400394003919996320019160010201600002048000040039400391116002110910101600001000001002232221164222221400363010160000104004040040400404004040040