Apple M1 Microarchitecture Research by Dougall Johnson

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

SHA512H

Test 1: uops

Code:

  sha512h q0, q1, v2.2d
  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)03080b18191e1f3f4e51inst 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)a0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
1004203816000000619372510001000100068984201920382038178831896100010003000203820381110011000000100073116111970100020392039203920392039
1004203816000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000300073116111970100020392039203920392039
1004203815000000619372510001000100068984201920382038178831896100010003000203820381110011000000000073116111970100020392039203920392039
10042038150000006193725100010001000689842019203820381788318961000100030002038203811100110000001700073116111970100020392039203920392039

Test 2: Latency 1->1

Code:

  sha512h q0, q1, v2.2d
  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): 2.0038

retire (01)cycle (02)030818191e1f3a3f4e51inst 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)9fa0a1a8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003815000090061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071212162219969100001002003920039200392003920039
102042003815000000061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
1020420038150000330061993725101001001000010010000500707984020019200382003818638318777101002001000020030000200382003811102011009910010010000100001371012162219969100001002003920039200392003920039
1020420038150000360061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
102042003815000000061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
102042003815000000061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
102042003815000000061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
10204200381500002488061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
102042003815000030061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039
102042003815000000061993725101001001000010010000500707984020019200382003818638318746101002001000020030000200382003811102011009910010010000100000071012162219969100001002003920039200392003920039

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0038

retire (01)cycle (02)0318191e1f3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200381500048061993725100101010000101000050707984120019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039
1002420038150003061993725100101010000101000050707984120019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039
1002420038150000061993725100101010000101000050707984020019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039
1002420038150000015899372510010101000010100005070798412001920038200381866031876810010201000020300002003820038111002110910101000010000209350640316331996910000102003920039200392003920039
1002420038150000061993725100101010000101000050707984020019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039
10024200381500000493993725100101010000101000050707984120019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039
1002420038150006061993725100101010000101000050707984120019200382003818660318768100102010000203000020038200381110021109101010000100001000640316341996910000102003920039200392003920039
1002420038150000061993725100101010000101000050707984120019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039
100242003815000935261993725100101010000101000050707984120019200382003818660318768100102010059223000020038200381110021109101010000102001002671316332000710000102003920039200392003920039
100242003815000390390989925100101010000101000050707984120019200382003818660318768100102010000203000020038200381110021109101010000100000000640316331996910000102003920039200392003920039

Test 3: Latency 1->2

Code:

  sha512h q0, q0, v1.2d
  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)030718191e3f4e51inst 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)9fa8cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500006119901251010010010000100100005001067500030018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
1020430037225000012219901251010010010000100100005001067500030018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
102043003722500006119901251010010010000100100005001067500130018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
102043003722500006119901251010010010000100100005001067500130018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
1020430037224000246119901251010010010000100100005001067500030054300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
1020430037225000216119901251010010010000100100005001067500130018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
102043003722500006119901251010010010000100100005001067500130018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
102043003722500006119901251010010010000100100005001067500030022300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
102043003722500006119901251010010010000100100006201067500030018300373003728589328745101002001000020030000300373003711102011009910010010000100071021622299200100001003003830038300383003830038
1020430037225000644119901251010010010000100100005001067500030018300373003728589328745101002021000020030000300373003711102011009910010010000100074021622299200100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03070a18191e1f3a3f4e51inst 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
100243003722500003180061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006403162229920010000103003830038300383003830038
10024300372250000150061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037225000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037224000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037225000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037225000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037225000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037224000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430084224000000061199012510015101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038
1002430037225000000061199012510010101000010100005010675001300183003730037286113287671001020100002030000300373003711100211091010100001000000006402162229920010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sha512h q0, q1, v0.2d
  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)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)9fa0a1a6a8a9acc2branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430084225000003004451990125101001001000010010000500106750003001830226300372858932874510100200100002003000030037300371110201100991001001000010000020000071011611299200100001003003830038300383022530038
1020430037225010000004691990125101001001000010010000500106750013001830037300372858932874510145200100002003000030037300371110201100991001001000010000000300071011611299200100001003003830038300383003830038
1020430037225000002100611990125101001001000010010000500106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000020600071011611299200100001003003830038300383003830038
1020430037225000003000611990125101001001000010010000500106750003001830037300372860632874510100200100002003000030037300851110201100991001001000010000010900071011611299200100001003003830038300383003830038
1020430037224000009001451990125101001001000010010000500106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000030000071011611299200100001003003830038300383003830038
10204300372250000024001241990125101001001000810010000500106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000000000071011611299200100001003003830038300383003830038
102043003722500000300611990125101001001000010010000550106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000000000071011611299200100001003003830038300383003830038
102043003722500000000611990125101001001000010010000500106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000000000071011611299200100001003003830038300383003830038
102043003722500000000611990125101001001000010010000500106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000000000071011611299200100001003003830038300383003830038
102043003722500000000841990125101001001000010010000500106750003001830037300372858932874510100200100002003000030037300371110201100991001001000010000000300071011611299200100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e3f4e51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640316432992010000103003830038300383003830038
1002430037224000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640416432992010000103003830229300383003830038
1002430037225000000611990125100101010000101000050106750013001803003730037286113287671001020100002030000300373003711100211091010100001000640316342992010000103003830038300383003830038
1002430037225000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640316442992010000103003830038300383003830038
1002430037225000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640316342992010000103003830038300383003830038
1002430037225000000611990125100101010000101000050106750013001803003730037286113287671001020100002030000300373003711100211091010100001000640416342992010000103003830038300383003830038
1002430037225000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640316342992010000103003830227300853003830038
1002430037225000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640416442992010000103003830038300383003830038
1002430037225000000611990125100101010000101000050106750003001803003730037286113287671001020100002030000300373003711100211091010100001000640416342992010000103003830038300383003830038
1002430037225000000611990125100101010000101000050106750013001803003730037286113287671001020100002030000300373003711100211091010100001000640416442992010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sha512h q0, q8, v9.2d
  movi v1.16b, 0
  sha512h q1, q8, v9.2d
  movi v2.16b, 0
  sha512h q2, q8, v9.2d
  movi v3.16b, 0
  sha512h q3, q8, v9.2d
  movi v4.16b, 0
  sha512h q4, q8, v9.2d
  movi v5.16b, 0
  sha512h q5, q8, v9.2d
  movi v6.16b, 0
  sha512h q6, q8, v9.2d
  movi v7.16b, 0
  sha512h q7, q8, v9.2d
  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): 2.0005

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)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)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020416003811990000978352617993749801001008000010080000500567798401600190160038160038139907313999680100200800002002400001600381600381116020110099100100160000100000000000010110116111599810160000100160039160039160039160039160039
16020416003811990000119106179937258010010080000100800005005677984016001901600381600381398883313999680100200800002002400001600381600381116020110099100100160000100000000000010135116111599810160000100160039160039160039160039160039
1602041600381199000046506179937258010010080000100800005005677984016001901600381600381398883139996801002008000020024000016003816003811160201100991001001600001000000110000010110116111599810160000100160039160039160039160039160039
16020416003811990000114909437993725801001008000010080000500567798401600190160038160038139888313999680100200800002002400001600381600381116020110099100100160000100000000000010110116111599810160000100160039160039160039160039160039
16020416003811990000107407267993725801001008000010080000500567798401600190160038160038139888313999680100200800002002400001600381600381116020110099100100160000100000000000010110116111599810160000100160039160039160039160039160039
1602041600381198000012060617993725801001008000010080000500567798401600190160038160038139888313999680100200800002002400001600381600381116020210099100100160000100000000000010110116111599810160000100160039160039160039160039160039
160204160038119900004170617993725801001008000010080000500567798401600190160038160038139888313999680100200800002002400001600381600381116020110099100100160000100000000000010110116111599810160000100160039160039160099160039160039
1602041600381199000047106179937258011210080000100800005005677984016001901600381600381398883139996801002008000020024000016003816003811160201100991001001600001000000000001101101161115998111160000100160039160039160039160039160039
1602041600381198000013290617993725801001008000010080000500567798401600190160038160038139888313999680100200800002002400001600381600381116020110099100100160000100000000000010110116111599810160000100160039160039160039160039160039
1602041600381198000011010617993725801001008000010080000500567798401600190160038160038139888313999680100200800002002400001600381600381116020110099100100160000100000010000010110116111599810160000100160279160039160039160039160135

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 2.0005

retire (01)cycle (02)03090b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600241600381198000060073799372580010108000010800005056779840116001916003816003813991031400188005620800002024018616003816003811160021109101016000010000000001002262251621164159983020171816000010160039160039160039160039160039
1600241600381199000000067799372580010108000010800005056779841016001916003816003813991031400188001020800002024000016003816003811160021109101016000010000000001002261121621142159983020341816000010160039160039160039160039160039
1600251600381199000000025779937258001010800001080000505677984111600191600381600381399103140018800102080000202400001600381600381116002110910101600001000000001100783114242112415998322017916000010160039160039160039160039160039
160024160086120010000008879937258001010800001080000505677984111600191600381600381399103140018800102080000202400001600381600381116002110910101600001000000000100223116162113615998302017916000010160039160039160039160039160039
1600241600381199000021006779937258001010800001080000505677984111600191600381600381399103140018800102080000202400001600381600381116002110910101600001000000000100223114162114215998302017916000010160039160039160039160039160039
1600241600381199100018006779937258001010800001080000505677984111600191600381600381399103140018800102080000202400001600381600381116002110910101600001000000000100223112162214215998302017916000010160039160039160039160039160039
160024160038119900000006779937258001010800001080000505677984111600191600381600381399103140018800102080000202400001600381600381116002110910101600001000002002100223116242112416001802017916000010160039160039160039160039160039
1600241600381199000012007657993725800101080000108000050567798411160019160038160038139910314001880010208000020240000160038160038111600211091010160000100021109410100223114162114615998302017916000010160039160039160039160039160039
160024160038119900000006779918258001010800001080000505677984111600191600381600381399103140018800102080000202400001600381600381116002110910101600001000000000100223114162114216004902017916000010160039160039160039160039160039
160024160038119900000006779937258001010800001080000505677984111600191600381600381399103140018800102080062202400001600381600381116002110910101600001000000000100223112162112415998302017916000010160039160039160039160039160039

Test 6: throughput

Count: 16

Code:

  sha512h q0, q16, v17.2d
  sha512h q1, q16, v17.2d
  sha512h q2, q16, v17.2d
  sha512h q3, q16, v17.2d
  sha512h q4, q16, v17.2d
  sha512h q5, q16, v17.2d
  sha512h q6, q16, v17.2d
  sha512h q7, q16, v17.2d
  sha512h q8, q16, v17.2d
  sha512h q9, q16, v17.2d
  sha512h q10, q16, v17.2d
  sha512h q11, q16, v17.2d
  sha512h q12, q16, v17.2d
  sha512h q13, q16, v17.2d
  sha512h q14, q16, v17.2d
  sha512h q15, q16, v17.2d
  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): 2.0002

retire (01)cycle (02)0307090a1e3f4e51inst 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)9fa0a8a9acc5cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204320038239801118242815993725160100100160000100160000500113579840320019032003832003829988803299996160100200160000200480000320038320038111602011009910010016000010001200001011021624319969160000100320039320039320039320039320039
160204320038239800012611599372516010010016000010016000050011357984032001903200383200382998880329999616010020016000020048000032003832003811160201100991001001600001000000011011021622319969160000100320039320039320039320039320039
160204320038239700097261599372516010010016000010016000050011357984032001903200383200382998880329999616010020016000020048000032003832003811160201100991001001600001000003001011031622320043160000100320039320039320039320039320039
16020432003823970000611599372516010010016000011716000050011357984032001903200383204202998880329999616010020016028620248000032003832003821160201100991001001600001000020001011021622319969160000100320039320039320039320039320039
1602043200382398000072615993725160100100160000100160000500113579840320019332003832003829988803299996160100200160000200480000320038320038111602011009910010016000010000012001011021622319969160000100320039320039320039320039320039
16020432003823970000611599372516010010016000010016000050011357984032001903200383200382998880329999616010020016000020048000032003832003811160201100991001001600001000000001013221622319969160000100320039320039320039320039320039
160204320038239800007261599372516010010016000010016000050011357984132001903200383200382998880329999616010020016000020048000032003832003811160201100991001001600001000006001011021622319969160000100320087320039320039320039320039
16020432003823970000941599372516010010016000010016000050011357984032001903200383200382998880329999616010020016000020048000032003832003811160202100991001001600001000000001011021622320004160000100320039320039320039320039320039
160204320038239800007261599375316010010016000010016000050011357984132001903200383200382998880329999616010020016000020048000032003832003811160201100991001001600001000000001011021622319969160000100320039320039320039320039320039
16020532003823970000611599372516010010016000010016000050011357984032001933200383200382998880330003216010020016000020048000032003832003811160201100991001001600001000000101011021622319969160000100320039320039320039320039320039

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 2.0002

retire (01)cycle (02)03040708090b18191e1f3a3f4e51inst 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)5f6061696b6d6edispatch 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)9fa0a6a8a9acc2c5cdcfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024320038239800000000006715993725160010101600001016000050113579841103200190320038320038299910033000541600102016000020480000320038320038111600211091010160000100000000010023851144162223130319969021382016000010320039320039320039320039320039
160024320038239710010001200738159937251600101016000010160000501135798401032001903200383200382999100330001816001020160000204802043200383200381116002110910101600001000000000100241622031162221930319969042382016000010320039320039320039320039320039
1600243200382397000000000011615993744160010101600001016000050113579840110320067332003832003829991003300018160010201600682048000032003832003821160021109101016000010000018000100251672118162212917319969021382016000010320039320039320039320039320039
16002432003823971001000000116159937251600101016000010160000501135798401103200190320038320038299910033000181600102016000020480000320038320038111600211091010160000100000000010025682120162223121319969042382016000010320039320039320039320039320039
160024320038239710010000007381599372516001010160000101600005011357984111032001903200383200382999100330001816001020160000204800003200383200381116002110910101600001000003000100251622116162222917319969042601016000010320039320039320039320039320039
160024320038239710010000001161599372516001010160012101600005011357984111032001903200383200382999100330001816001020160000204800003200383200381116002110910101600001000100000100231351125161112718319969021191016000010320039320039320039320039320039
160024320038239810010000001161599372516001010160000101600005511357984111032001903200383200382999100330001816001020160000204800003200383200381116002110910101600001000000000100231311140161112126320017021191016000010320086320039320039320039320039
160024320038239710010000001161599372516001010160000101600005011357984111032001903200383200382999100330001816005720160000204800003200383200381116002110910101600001000000000100231311127161112828319969021191016000010320039320039320039320039320039
16002432008723981001000000116159937251600101016000010160000501135798401032001903200383200382999100330001816001020160000204800003200383200381116002110910101600001000000010100231311127161111827320017021191016000010320039320039320039320039320039
160024320038239710010000007811599372516001010160000101600005011357984111032001903200863200382999100330001816001020160000204800003200383200381116002110910101600001000100000100231311119161112819319969021191016000010320039320039320039320039320039