Apple M1 Microarchitecture Research by Dougall Johnson

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

SMLSL (by element, 2D)

Test 1: uops

Code:

  smlsl v0.2d, v1.2s, v2.s[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)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
10043037220842548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
1004303722216612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110001073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037226612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100030003037303711100110000073216222630100030383038303830383038

Test 2: Latency 1->1

Code:

  smlsl v0.2d, v1.2s, v2.s[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)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)9fa8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722530061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013162229634100001003003830038300383008430038
1020430037225282061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012162229634100001003003830038300383003830038
1020430037224258061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012163229634100001003003830038300383003830038
1020430037225372061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071212162229634100001003003830038300383003830038
1020430037225450061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012162229634100001003003830038300383003830038
10204300372240061295384510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012162229634100001003003830038300383003830038
1020430037225444061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012162229634100001003003830038300383003830038
1020430037225426061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012162229634100001003003830038300383003830038
102043003722539601177295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071012162229634100001003003830038300383003830038
10204300372253030612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010048671012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b1e1f3f4e5051inst 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)9fa6a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000061295480251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722500150612954830021251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722500150612954821119251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250024061295480251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
1002430037225000061295480251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216232963010000103003830038300383003830038
10024300372250024061295480251001010100001010000504277313030018300373003728287328767100102010000203000030037300372110021109101010000100000640216222963010000103003830038300383003830038
10024300372250027061295480251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250015061295480251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250015061295480251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216232963010000103003830038300383003830038
1002430037225000061295480251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  smlsl v0.2d, v0.2s, v1.s[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)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)9fa0a8acc5branch mispredict (cb)cdcfd6e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372258461295482510100100100001001000050042773130300183003730037282727287411010020010008200300243003730037111020110099100100100001000001117171629647100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282726287401010020010008200300243003730037111020110099100100100001000001117171629647100001003003830038300383003830038
10204300372252161295482510100100100001001000050042773131300183003730037282727287411010020010008200300243003730037111020110099100100100001000101117181629646100001003003830038300383003830038
1020430037225661295482510100100100001001000050042773130300183003730037282726287411010020010008200300243003730037111020110099100100100001000001117171629647100001003003830038300853003830038
1020430037225961295482510100100100001001000050042773130300183003730037282897287571010020010008202300243003730037211020110099100100100001002001117181629646100001003003830038300383003830038
10204300372250103295482510100100100001001000050042773131300183003730037282727287401010020010008200300243003730037111020110099100100100001000001117181629647100001003003830038300383003830038
1020430037225682295482510100100100001001000050042773130300183003730037282726287411010020010008200300243003730037111020110099100100100001000001117171629646100001003003830038300383003830038
10204300372251261295482510100100100001001000050042773130300183003730037282726287411010020010008200300243003730037111020110099100100100001000031117171629646100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282726287401010020010008200320163003730037111020110099100100100001000061117172429646100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773130300183003730037282726287401010020010008200300243003730037111020110099100100100001000001117181629646100001003003830038300383003830038

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)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
10024300372254261295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722527726295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722521726295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372259536295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250251295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010106402162229630010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372251861295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250536295482510010101000010100005042773133001830037300372828732878610010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010106402162229630010000103003830038300383003830038
1002430037224061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  smlsl v0.2d, v1.2s, v0.s[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)03191e3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100107101161129634100001003003830038300383003830038
10204300372250216129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037224006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037224006129548251010012110000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225096129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0309181e1f3f4e51inst 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)9fa0a1a6a8acc3cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
100243003722500006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100005030013640216222963010000103003830038300383003830038
10024300372250060612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
1002430037225000132612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372240000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372260000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372250000892954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  smlsl v0.2d, v8.2s, v9.s[1]
  movi v1.16b, 0
  smlsl v1.2d, v8.2s, v9.s[1]
  movi v2.16b, 0
  smlsl v2.2d, v8.2s, v9.s[1]
  movi v3.16b, 0
  smlsl v3.2d, v8.2s, v9.s[1]
  movi v4.16b, 0
  smlsl v4.2d, v8.2s, v9.s[1]
  movi v5.16b, 0
  smlsl v5.2d, v8.2s, v9.s[1]
  movi v6.16b, 0
  smlsl v6.2d, v8.2s, v9.s[1]
  movi v7.16b, 0
  smlsl v7.2d, v8.2s, v9.s[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)03080918191e1f3f51inst 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)9fa0a6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200891500000300622580100100800001008000050064000012004520064200643228042720080000200240000200642006411160201100991001001600001000017091011111611200611600001002006520065200652006520065
16020420064150000000622580100107803131098000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000018061011111611200611600001002006520065200652006520065
16020420064150010021083258010010080000106800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100002061011111611200611600001002006520065200652006520065
16020420064150000021085258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100000061011111611200611600001002006520065200652006520065
160204200641510000912108498258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100002061011111611200611600001002006520065200652006520065
16020420064150000021085258010010080000116800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100002061011111611200611600001002006520065200652006520065
16020420064150000021039258010010080000125800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100005001011114811200611600001002006520065200652006520065
16020420064150000015083258010010080000100800006446400001200452006420064322801002008000020024000020064200641116020110099100100160000100002061011111611200611600001002006520065200652006520065
160204200641510000162085258010010080000100800005006400000200452006420064322802072008000020024000020064200641116020110099100100160000100002031011115011200611600001002006520065200652006520065
16020420064151000021085258010010080000100800005006400001200452006420064322801002008000020024000020064200641116020110099100100160000100000031011111611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030a1e3a3f51inst 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)9fa6accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420078150000452780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100010029311825211610200482201160000102005220052200522005220052
160024200511500004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000100293111025211610200482201160000102005220052200522005220052
160024200511500004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000100336111025211106200482201160000102005220052200522005220061
160024200511500005127800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000100293111025211106200482201160000102005220052200522005220052
1600242005115100045278001212800001280000626400000120032200512005132280012208000020240000200512005111160021109101016000010001003331110252111010200482201160000102005220061200522005220052
16002420051150000452780012128000012800006264083211200322005120051322800122080000202400002006020051111600211091010160000100010033311625211610200482201160000102005220052200522005220052
16002420051150200512780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100010033311625211610200482201160000102005220052200522005220052
16002420051150000452780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100010029311625211610200482201160000102005220052200522005220052
160024200511500004527800121280000128000062640000112003220051200513228001220800002024000020051200511116002110910101600001000100293111025211610200482201160000102005220052200522005220052
16002420051150060452780012128000012800006264000011200412005120051322800122080000202400002005120051111600211091010160000100010033311625211610200482201160000102005220061200522006120052

Test 6: throughput

Count: 12

Code:

  smlsl v0.2d, v12.2s, v13.s[1]
  smlsl v1.2d, v12.2s, v13.s[1]
  smlsl v2.2d, v12.2s, v13.s[1]
  smlsl v3.2d, v12.2s, v13.s[1]
  smlsl v4.2d, v12.2s, v13.s[1]
  smlsl v5.2d, v12.2s, v13.s[1]
  smlsl v6.2d, v12.2s, v13.s[1]
  smlsl v7.2d, v12.2s, v13.s[1]
  smlsl v8.2d, v12.2s, v13.s[1]
  smlsl v9.2d, v12.2s, v13.s[1]
  smlsl v10.2d, v12.2s, v13.s[1]
  smlsl v11.2d, v12.2s, v13.s[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)03373f4e51inst 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)9fa8c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1202043034622501060251201001001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611300361200001003004030040300403004130040
120204300402251410251201001001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611300371200001003004030040300403004030040
120204300392250410251201621001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611309191200001003004030040300403004031775
120204300392250620251201001001200001001200005009600000300203003930090149733149971201002001200002003600003003930040111202011009910010012000010000761011611317331200001003009331023300403004030040
1202043003922501250251201001001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611300361200001003004030040300403004030041
1202043003922501290251201001001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611300391200001003004030040300403004030040
120204300392250410251201001001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611317501200001003004030040300403004130042
120204300392250640251201001001200811001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611309191200001003004030041300403004030040
1202043003922511060251201001001200001001200005009600000300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000761011611317501200001003004030041300403004030040
120204300392250410251201001001200001001200005009600000300203003930039149733149971201002001200002003600003004130039111202011009910010012000010000761011611300391200001003004030040300403004030040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307080a0b18191e1f373a3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12002430050225000000000245025120010101200001012000050960000130020300393003914996315074120010201200002036000030039300391112002110910101200001000000000752481615133003600120000103004030040300403004030040
120024300392250000030002450251200101012000010120000509600000300203003930039149963150191200102012000020360000300393003911120021109101012000010000000007524171618143003600120000103004030040300403004030040
120024300392250000000002450251200101012000010120000509600000300203003930039149963150191200102012000020360000300393003911120021109101012000010000000007524151616183003600120000103004030040300403004030040
120024300392250000000002450251200101012000010120000509600001300203003930039149963150191200102012000020360000300393003911120021109101012000010000000007524141616153003600120000103004030040300403004030040
120024300392250000000002450251200281012000010120000509600000300203019631010149963150191200102012000020360000300393003911120021109101012000010000000007524151614163003600120000103004030040300403004030040
120024300392250000000002450251200101012000010120000509600001300203003930039149963150191200102012000020360000300393003911120021109101012000010000000007524121615143007800120000103004030040300403004030040
12002430039225000000000245025120010101200001012000050960000030020300393003914996315019120534201200002036000030039300391112002110910101200001000000000752411168153003600120000103004030040300403064130040
120024300392250000000002450251200101012000010120000509600000300203003930039149963150191200102012000020360000300393003911120021109101012000010000000007524151616173003600120000103004030040300403004030040
1200243003922500000000024502512001010120000101200005096000003002030039300391499631501912001020120000203600003003930039111200211091010120000100000001560752415169153003600120000103004030040300403004030040
120024300412250000000002450251200101012000010120000509600000300203003930039149963150191200102012000020360000300393003911120021109101012000010000000007524121620183003600120000103004030040300403004030040