Apple M1 Microarchitecture Research by Dougall Johnson

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

UABAL (vector, 2D)

Test 1: uops

Code:

  uabal v0.2d, v1.2s, v2.2s
  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)030b181e3a3f4e51inst 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)a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004303723000210725482510001000100039831330183037303724153289510001000300030373037111001100000075516662630100030383038303830383038
100430372200018425482510001000100039831330183037303724153289510001000300030373037111001100000075616662630100030383038303830383038
100430372300018225482510001000100039831330183037303724153289510001000300030373037111001100002730075516662630100030383038303830383038
1004303723000110225482510001000100039831330183037303724153289510001000300030373037111001100006075616552630100030383038303830383038
1004303722000116425482510001000100039831330183037303724153289510001000300030373037111001100000075616662630100030383038303830383038
100430372200026125482510001000100039831330183037303724153289510001000300030373037111001100010075516662630100030383038303830383038
100430372200018425482510001000100039831330183037303724153289510001000300030373037111001100010075616552630100030383038303830383038
100430372200018425482510001000100039831330183037303724153289510001000300030373037111001100000075616772630100030383038303830383038
100430372300018225482510001000100039831330183037303724153289510001000300030373037111001100000075616662630100030383038303830383038
1004303723000110725482510001000100039831330183037303724153289510001000300030373037111001100010075616662630100030383038303830383038

Test 2: Latency 1->1

Code:

  uabal v0.2d, v1.2s, v2.2s
  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)030708090a0b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2c5branch mispredict (cb)cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722510010000006129548251010010010000100100005004277313030018030037300372827262874010100200100082003054030037300371110201100991001001000010000000000111766021611296490100001003003830038300383003830038
102043003722510010000006129548251010010010000100100005004277313130018030037300372827262874110100200100082003002430037300371110201100991001001000010000000030111720011611296490100001003003830038300383003830038
102043003722510010000008229548251010010010000100100005004277313030018030037300372827272874010100200100082003002430037300371110201100991001001000010000000000111717011611296480100001003003830038300383003830038
102043003722510010000006129548251010010010000100100005114277313030018030037300372827262874110100200100082003002430037300371110201100991001001000010000000000111717011611296480100001003003830038300383003830038
10204300372251001000000142629548251010010010000100100005004277313130018030037300372827262874110100200100082003002430037300371110201100991001001000010000001000000710131633296340100001003003830038300383003830038
102043003722500000000006129548251010010010000100100005004277313030018030037300372826532874510100200100002003000030037300371110201100991001001000010000000000000712131633296340100001003003830038300383003830038
102043003722500000000006129548251010010010000100100005004277313030018030037300372826532874510100200100002003000030037300371110201100991001001000010000000000000712131633296340100001003003830038300383003830038
102043003722500000000008629548251010010410000100100005004277313030018030037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710131633296340100001003003830038300383003830038
102043003722500000000006129548251011210010000100100005004277313130018030037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710131634296340100001003003830038300383003830038
102043003722500000000006129548251010010010000100100005004277313030018030037300372826532874510100200100002003000030037300371110201100991001001000010000000000000710131633296340100001003003830038300383003830038

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)9fa8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722502512954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372330612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162329630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722505362954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000006402163229630010000103003830038300383003830038
100243003722504412954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  uabal v0.2d, v0.2s, v1.2s
  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)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)9fa8a9acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250019992954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
1020430037224009782954825101001001000010010000500427731313001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
1020430037225009722954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500822954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
1020430037225002082954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372240016762954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372410012672954825101001001000010010000500427731313001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372240013372954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250010612954825101001001000010010000500427731303001803003730037282653287451010020010000200300003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038

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
10024300372250300295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722501200295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722501118295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250683295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722561026295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722501129295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250612954864100221010008141014950427731330018301333008528296122878510010201000020300003003730037111002110910101000010006402162229702310000103003830038300383003830085
100243003722501339295482510010111000810100005042773133005430085300832829282878610161201000020300003003730085111002110910101000010006402162229630010000103003830038300383003830038
1002430037225661295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722401175295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  uabal v0.2d, v1.2s, v0.2s
  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)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)9fa0a1a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000023129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000180071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
10204300372250000008929548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
10204300372250000006129548251011010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100001000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
10204300372250000006129548251010010010000100100005004277313130018300373003728293328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
102043003722500009025129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e1f3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000007242162229630010000103003830038300383003830038
1002430037225000000612954825100101010000131000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000001206402162229630010000103003830038300383003830038
1002430037225000012061295482510010101000010100005042773131300183003730228282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100006142773131300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
100243003722400000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
1002430037224000000726295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001002000006402162229630010000103003830038300383003830038
1002430037225010000536295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000906402162229630010000103003830038300383003830038
10024300372250000907262954825100101010000101000050427731313001830037300372829132876710010201000020300003003730037211002110910101000010000102790508343973329992410000103051330558305493054830510
1002430558229009111329880594429458239101022410088231141893429088313041430402305432832551289731150220118062435388305593046213110021109101010000104224227835085931042430055210000103050830558304643046130512
10024305572281089118852853462947620010090151006410113418142881690303423046730415283193628917112062211316223393330462304594110021109101010000100021225145085051214330061510000103046130556305563051030543

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  uabal v0.2d, v8.2s, v9.2s
  movi v1.16b, 0
  uabal v1.2d, v8.2s, v9.2s
  movi v2.16b, 0
  uabal v2.2d, v8.2s, v9.2s
  movi v3.16b, 0
  uabal v3.2d, v8.2s, v9.2s
  movi v4.16b, 0
  uabal v4.2d, v8.2s, v9.2s
  movi v5.16b, 0
  uabal v5.2d, v8.2s, v9.2s
  movi v6.16b, 0
  uabal v6.2d, v8.2s, v9.2s
  movi v7.16b, 0
  uabal v7.2d, v8.2s, v9.2s
  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)031e3f51inst 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)9fa0a8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020420088150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064156039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150054258010010080000100800005006400001200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065
16020420064150039258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111613200611600001002006520065200652006520065
16020420064151081258010010080000100800005006400000200450200642006432280100200800002002400002006420064111602011009910010016000010000001011111611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03070a1e3a3f51inst 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)9fa8a9accfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024201081512000572780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001004131280242521121212004822001160000102005220052200522005220052
160024200511502200572780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001004731240192521222192004824001160000102005220061200522005220052
160024200511501100512780012128000012800006264000011200322005120051322800122080000202400002005120060111600211091010160000100001004431250192521118212004822001160000102005220052200522005220052
160024200511501100572780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001004531240202521122182004822001160000102005220052200522005220052
160024200511501100572780012128000012800006264000011200322005120051322800122080000202400002005120051111600211091010160000100001004331230203422222212004824002160000102005220061200522006120052
1600242005115011002902780012128000012800006264000011200322005120060322800122080000202400002005120060111600211091010160000100001004462230143441222232004824002160000102006120061200612005220061
1600242006015011002922780012128000012800006264000001200322006020060322800122080000202400002006020051111600211091010160000100001004462200183422220192005724002160000102006120052200612005220061
16002420060150110011082780012128000012800006264000011200412006020051322800122080000202400002006020060111600211091010160000100001004932240203441221222004822002160000102006120061200612006120052
160024200511501100512780012128000012800006264000011200322005120051322800122080000202400002006020060111600211091010160000103001003631200202521118202004822001160000102005220052200522005220052
160024200601502200512980012128000012800006264000011200412005120060322800122080000202400002006020060111600211091010160000100001004562200192541221202005722001160000102006120052200612006120052

Test 6: throughput

Count: 16

Code:

  uabal v0.2d, v16.2s, v17.2s
  uabal v1.2d, v16.2s, v17.2s
  uabal v2.2d, v16.2s, v17.2s
  uabal v3.2d, v16.2s, v17.2s
  uabal v4.2d, v16.2s, v17.2s
  uabal v5.2d, v16.2s, v17.2s
  uabal v6.2d, v16.2s, v17.2s
  uabal v7.2d, v16.2s, v17.2s
  uabal v8.2d, v16.2s, v17.2s
  uabal v9.2d, v16.2s, v17.2s
  uabal v10.2d, v16.2s, v17.2s
  uabal v11.2d, v16.2s, v17.2s
  uabal v12.2d, v16.2s, v17.2s
  uabal v13.2d, v16.2s, v17.2s
  uabal v14.2d, v16.2s, v17.2s
  uabal v15.2d, v16.2s, v17.2s
  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)03081e373f4e51inst 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)9fa8accfd0d2icache miss (d3)d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440076300001741025160101100160000100160000500128000004002940071400491997331999716010020016000020048000040039400481116020110099100100160000100001011000026411400451600001004004940040400494004040040
16020440039300101741025160101100160000100160000500131999804002940039400391997331999816010020016000020048000040039400481116020110099100100160000100001011000011611400371600001004004940050400404004040040
1602044004830009050025160100100160000100160000500239899904002040039400401997331999716010020016000020048000040048400391116020110099100100160000100001011000021611400461600001004004140049400414005040040
160204400483000017502525160100100160000100160000500128000004002040039400481997331999716010020016000020048000040048400481116020110099100100160000100001011000011611400361600001004004040040400494004040049
1602044003930000041025160100100160000100160000500239899904002040039400391997331999716010020016000020048000040049400391116020110099100100160000100001011000011611400361600001004004040049400404004940074
16020440039300001750025160117100160000100160000500128000004002140039400481997331999716010020016000020048000040048400401116020110099100100160000100001011000011611400361600001004004140040400494004140049
1602044003930000150025160101100160017100160000500128000004002140039400401997332000616010020016000020048000040048400391116020110099100100160000100001011000011611400361600001004004940040400494004040040
1602044004830000071025160100100160000100160000500128000004002940039400391997331999816010020016000020048000040039400491116020110099100100160000100001011000011611400361600001004004140049400414004040040
16020440049300001741025160100100160017100160000500131999804002140040400981997331999816010020016000020048000040048400391116020110099100100160000100001011000111611400461600001004004040049400404004940040
1602044003930000050025160100100160017100160000500128000004002040040400481997332000616010020016000020048000040039400391116020110099100100160000100001011000011611400371600001004004040049400404004940040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307080a0b18191e1f373f51inst 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)5f6061696d6edispatch 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)9fa0a1a6a8a9acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600244005730000000012004625160010101600001016000050128000011040029400394004819996320028160010201600002048000040039400481116002110910101600001000204001002285134162111310400360206160000104004040049400404004040040
1600244004829900000000175525160027101600001016000050128000011540020400394004819996320028160010201600002048000040039400481116002110910101600001000000001002285113162111914400450209160000104004940040400494004040049
160024400483000000000004625160027101600001016000050239899911540020400394004819996320019160010201600002048000040048400481116002110910101600001000000001002285113164111213400450206160000104004040049400404005040040
160024400483000000000004625160027101600171016000050128000011540020400394004819996320028160010201600002048000040048400391116002110910101600001000000001002485112162111914400360409160000104004940040400494004040049
160024400482990000000005422516001010160000101600005012800000154002040039400481999632001916001020160000204800004004840039111600211091010160000100001060100228511216211149400881206160000104004040049400404004940049
1600244004830000000000055251600101016000010160000501280000115400204003940039199963200281600102016000020480000400484003911160021109101016000010000100010022851916211812400450206160000104004940040400494004040040
160024400393000000000004625160010101600001016000050239899911540029400394004819996320019160010201600002048000040039400481116002110910101600001000000021002285110162111012400360209160000104004040040400494004040040
1600244003930000000036051552516002710160017101600005012800001154002040090400391999632001916001020160000204800004003940048111600211091010160000100000000100228511416211129400450206160000104004940040400494004040040
16002440039300000000000552516002710160000101600005023989991154002040039400391999632002816001020160000204800004004840039111600211091010160000100000000100228511316211917400360206160000104004940040400404004940049
160024400483000000000017552516002710160017101600005012800001154002040039400391999632001916001020160000204800004003940049111600211091010160000100000000100228511216211128400360209160000104004040040400404004940040