Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMLAL (by element, S)

Test 1: uops

Code:

  sqdmlal s0, h1, 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)03181e3f4e51inst 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)a0a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037220061254825100010001000398313301830373037241532895100010003000303730371110011000010073216112630100030383038303830383038
10043037230061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
10043037220061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
10043037220061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
10043037230061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
1004303723024695254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
10043037230082254825100010001000398313301830373037241532895100010003000303730371110011000200073116112630100030383038303830383038
10043037220061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
10043037230061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038
10043037230061254825100010001000398313301830373037241532895100010003000303730371110011000000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqdmlal s0, h1, 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)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)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)9fa0a1a6a7a8a9acc2branch mispredict (cb)cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000000612949412410143117100401191089454442813120301263013330169282751628798104042101049720831485301783013141102011009910010010000100220000000712121622296340100001003008730038300383003830038
1020430037225000000006129548251010010010008100100005004277313030018300373003728265328745101002001000020230000300373003711102011009910010010000100000000000712131622296340100001003003830038300383003830038
1020430037226000001320079429548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000010300712121632296340100001003003830038300383003830038
1020430037225000010008229548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000010000710131622296340100001003003830038300383003830038
10204300372240000008806129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000010000710121622296340100001003003830038300383003830038
1020430037225000010008229548251011010010000100100005004280027030054300373008428265328745101002001000020030000300373003711102011009910010010000100000010000712121622296340100001003003830038300383003830133
1020430131226100000006129548251010010710000107100005114277313030018301313003728271328782101002001033320030483300373008411102011009910010010000100000000000712121622296343100001003003830038301373003830133
1020430037225000000006129548251010010010000100100005004277313130018300373013328265328745101002001000020030000300373003711102011009910010010000100003020000712133222296340100001003003830038300893003830038
1020430037225100000006129548251011010010000100100005004280027030018300373003728265928745104052001016620030000300373003711102011009910010010000100000002279320710121632298184100001003008630038300853003830038
102043008422500000000543929548251010010010000100100005114277313130018300373003728269328745101002001000020030000300373003711102011009910010010000100000100000763131632296340100001003003830038300383003830038

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)9fc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100640316222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722406129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100640216232963010000103003830038300383003830038
100243008222506129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038
10024300372250674629548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722406129548251001010100001010000604277313130018300373003728287328767100102010000203000030037300371110021109101010000100640316222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqdmlal s0, h0, 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)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)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)9fa0a1a6a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000495061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000015061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773133001830037300372826532874510100200100002003049830226300371110201100991001001000010000013071011611296340100001003003830038300383003830038
10204300372250000339061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500007320536295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037224000045304412954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000450071011611296340100001003003830038300383003830038
102043003722500004320251295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000000726295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372250000507061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071042511296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773133001830037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030918191e1f3f4e51inst 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)9fa8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300832261231832641032954825100101010032101000050427731330018030037300842828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010100640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216322963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222970010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400000612954825100101010000101000050427731330018030037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqdmlal s0, h1, 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)0307081e3f4e51inst 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)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)9fa6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250008229548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037224000612954825101341001000010010000500427731313001803003730037282653287451010020010000200300003003730037111020110099100100100001000010123071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001803003730037282653287451010020010000200300003003730037111020110099100100100001002030012071011611296340100001003003830038300383003830038
10204300372240006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372240006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372240006129548251010010010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038

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)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)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000000612954825100101010000101000050427731303001830037300372828732876710010201000020304413008430037111002110910101000010000010606402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000020306402162229630010000103003830038300383003830038
1002430037224000000006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000101506402162229630010000103003830038300383003830038
1002430037225000000006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000470006402162229630010000103003830038300383003830038
100243003722500000000612954844100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000030606402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000010306402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000010306402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000020306402162229630010000103003830038300383003830038
10024300372250000000010962954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000010306402162229630010000103003830038300383003830038
100243003722500000000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000000306402162229630010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlal s0, h8, v9.h[1]
  movi v1.16b, 0
  sqdmlal s1, h8, v9.h[1]
  movi v2.16b, 0
  sqdmlal s2, h8, v9.h[1]
  movi v3.16b, 0
  sqdmlal s3, h8, v9.h[1]
  movi v4.16b, 0
  sqdmlal s4, h8, v9.h[1]
  movi v5.16b, 0
  sqdmlal s5, h8, v9.h[1]
  movi v6.16b, 0
  sqdmlal s6, h8, v9.h[1]
  movi v7.16b, 0
  sqdmlal s7, h8, 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)03090b18191e1f3a3f51inst 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)60696b6d6escheduler 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)9fa0a1a6a7a8a9acc2c5branch mispredict (cb)cdcficache miss (d3)d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200651500000000257258010010080000100800005006400001200453200642006432280100200800002002400002006420064111602011009910010016000010000000000000101130416332006101600001002006520065200652006520065
16020420064150000000039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001000000101130316332006101600001002006520065200652006520065
1602042006415000000001148258010010080000100800005006400001200450200642006432280100200800002002400002006420064111602011009910010016000010000000030000101130316332006101600001002006520065200652006520065
1602042006415100000001139258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001000000101130316332006101600001002006520065200652006520065
16020420064150000000039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000000000000101130316232006101600001002006520065200652006520065
160204200641500000000230258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000000000000101140316332006101600001002006520065200652006520065
160204200641500000000293258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001000000101130316332006101600001002006520065200652006520065
16020420064150000000039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000000000000101130316332006101600001002006520065200652006520065
160204200641500000000274258010010080000100800005006400001200450200642006432280100200800002002400002006420064111602011009910010016000010000000000000101130316332006101600001002013220065200652006520065
1602042006415000000005142580100100800001008000050064000012004502006420064322801002008000020024000020064200641116020110099100100160000100202004147520001016104494520301231600001002022620307203332031020308

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03080b191e1f3f51inst 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)5f60696d6e74scheduler 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)9fa0a1a8a9acbranch mispredict (cb)cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420061150000008552780012128000012800006264000011200322005120051032280012208000020240000200512005111160021109101016000010000000100343111125411910200482201160000102005220052200522005220052
16002420051150000008242780012128000012800006264000011200322005120051032280012208000020240000200512005111160021109101016000010000000100353111225211129200482211160000102005220052200522005220052
16002420051150000007132780012128000012800006264000011200322005120051032280012208000020240000200512005111160021109101016000010000000100313111125211710200482201160000102005220052200522005220061
1600242005115000000737278001212800001280000626400001120032200512005103228001220800002024000020051200511116002110910101600001000000010034311925211810200482201160000102005220052200522005220052
1600242005115000000691278001212800001280000626400001120032200512005103228001220800002024000020051200511116002110910101600001000000010033311725211118200482201160000102005220052200522005220052
16002420051150000006322780012128000012800006264000011200322005120051032280012208000020240000200512005111160021109101016000010000000100343111025211108200482201160000102005220052200522005220162
160024200511510000072227800121280000128000062640000112003220051200510322800122080000202400002005120051111600211091010160000100010001003531110252111111200502221160000102005220052200522005220052
1600242005116300000490278001212800001280000626400001120032200512005103228001220800002024000020051200511116002110910101600001020200010034311825211912200482201160000102005220052200522005220052
1600242005115010014117665278001212800001280000626400001120032200512005103228001220800002024000020051200511116002110910101600001000000010030311725211710200482201160000102005220052200522005220052
16002420051150000004527800121280000128000062640000112003220051200510322800122080000202400002005120051111600211091010160000100000001003531110252111212200482201160000102005220052200522005220052

Test 6: throughput

Count: 12

Code:

  sqdmlal s0, h12, v13.h[1]
  sqdmlal s1, h12, v13.h[1]
  sqdmlal s2, h12, v13.h[1]
  sqdmlal s3, h12, v13.h[1]
  sqdmlal s4, h12, v13.h[1]
  sqdmlal s5, h12, v13.h[1]
  sqdmlal s6, h12, v13.h[1]
  sqdmlal s7, h12, v13.h[1]
  sqdmlal s8, h12, v13.h[1]
  sqdmlal s9, h12, v13.h[1]
  sqdmlal s10, h12, v13.h[1]
  sqdmlal s11, h12, 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)0308090b18191e1f373f4e51inst 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)9fa0a1a6a8a9acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1202043003923200000132004930251202351001200001171200005009600003002030039317481497331525412010020212000020036062730089300931112020110099100100120000100110200945076102161130036201200001003019730758300403174930040
12020430039233000000011710251201181001200181171200005004399524317293003930039149733149971201002001200002003603093003930039111202011009910010012000010000000007610116113091901200001003092330040317493004031606
1202043174823300000000416713251201011001200001001200005001566598317293003930039149733149971201002001200002003600003003930040111202011009910010012000010020000475076101161130038221200001003004030040300913004030040
1202043174823300010000646713251201181001201151001200005009600003172931748300391498231502312010020012000020036071430039317481112020110099100100120000100000103076101161130037101200001003004031749300403174930040
1202043003923300000900416713251201181001200181001200005004399524313243004030039149733149971201002001200002003600003003931748111202011009910010012000010000000607610116113003701200001003004030041300413004031749
1202043003924600000001610251201001001200001001200005004399524317293174830039149733149971201002001200002003600003003931748111202011009910010012000010000000007610116113003601200001003174930040317493004031749
120204317482330000000183760251201001001200001001200005004399524300203003930040149733149971201002021200002003600003003930039111202011009910010012000010000000007610116113003601200001003174930040317493004030041
1202043004023300000000251671325120118100120018100120000500960000300203003931748166533167061201002001200002003600003174830039111202011009910010012000010000000007645116113003601200001003004031749302083094231749
1202043003924100000000416713251202381001200181001200985004399524300203003930042149733149971201002001200002003600003003930097111202011009910010012000010000000007610116113003601200001003174930040317493009030040
1202043174824100001132016202512011810012000110012000050096000030062300393174814973315593120100200120000200360000301993003911120201100991001001200001002021238000777211332130705201200001003073131783312333105230147

1000 unrolls and 10 iterations

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

retire (01)cycle (02)031e373a3f4e51inst 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)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)9fa0a6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12002430922225000400251200101012000010120000509600000300200300393003914996315019120010201200002036000030039300391112002110910101200001000007520151611630036120000103004030040300403004030040
1200243003922500069025120010101200001012000050960000130020030039300391499631501912001020120000203600003003930039111200211091010120000100000752061612730036120000103004030040300403004030040
120024300392250004002512001010120000101200005096000013002003003930039149963150191200102012000020360000300393003911120021109101012000010000075206165630036120000103004030040300403004030040
1200243003922500010302512001010120000101200005096000013002003003930039149963150191200102012000020360000300393003911120021109101012000010001075205165630036120000103004030040300403004030040
120024300392250004002512001010120000101202065096000003002003003930039149963150191200102012000020360000300393003911120021109101012000010000075206165630036120000103004030040300403004030040
120024300392250004002512001010120000101200005096000003002003003930039149963150191200102012000020360000300393003911120021109101012000010020224075206167630036120000103004030040300403004030040
120024300392250004002512001010120000101200005096000003002003003930039149963150191200102012000020360000300393003911120021109101012000010000675206164530036120000103004030040300403004030040
120024300392250004002512001010120000101200005096000013090303003930039149963150191200102012000020360000300393003911120021109101012000010000075206166630039120000103004030040300403004030040
12002430922225010710251200101012000010120000502555521030020030039300391499631501912001020120000203600003003930039111200211091010120000100000752011165630036120000103004030040300403004030040
120024300392250004002512001010120000101200005096000003002003003930039149963150191200102012000020360000300393003911120021109101012000010000075204165530036120000103004030040300403004030040