Apple M1 Microarchitecture Research by Dougall Johnson

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

FMLS (vector, 8H)

Test 1: uops

Code:

  fmls v0.8h, v1.8h, v2.8h
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

retire (01)cycle (02)030b191e1f3f4e51inst issue (52)~issue fp/simd (54)~dispatch fp/simd (57)huge thing fp/simd (5b)60696d6edispatch 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)a8acc2cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037300000863407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000900730116113473100040384038403840384038
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037310000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037300000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038
10044037310000613407251000100010005319081401840374037325833895100010003000403740371110011000000730116113473100040384038403840384038

Test 2: Latency 1->1

Code:

  fmls v0.8h, v1.8h, v2.8h
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)033a3f4e51inst 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)9fa6a8a9acbranch mispredict (cb)cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020440037300020783940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071003162239479100001004003840038400384003840038
102044003730001873940725101001001000010010000500570690814001804003740037381163387451010020010000200300004003740037111020110099100100100001000000071012162239479100001004003840038400384003840038
1020440037300010923940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071012162239479100001004003840038400384003840038
102044003730003763940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071012162239479100001004003840038400384003840038
102044003730002503940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071012162339479100001004003840038400384003840038
1020440037300113103940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001001000071012162239479100001004003840038400384003840038
1020440037300012163940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071012162239479100001004003840038400384003840038
1020440037299011973940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071010162339479100001004003840038400384003840038
1020440037300013143940725101001001000010010000500570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071012162239479100001004003840038400384003840038
1020440037300012323940725101001001000010010000522570690814001804003740037381083387451010020010000200300004003740037111020110099100100100001000000071012162239479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)0309181e1f3a3f4e51inst 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)9fa0a1a8a9accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000000072639407251001010100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010001000640316333947310000104003840038400384003840038
100244003730000000121939407251001010100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010000030640316333947310000104003840038400384003840038
100244003730000000124939407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640416333947310000104003840038400384003840038
1002440037300000008039407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038
1002440037300004200120539407251001010100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038
100244003729900000136339407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038
100244003730000000106639407251001010100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038
100244003730000000122639407251001710100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038
10024400373000000027539407251001010100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038
1002440037300000006139407251001010100001010000505707439140018040037400373813033876710010201000020300004003740037111002110910101000010000000640316333947310000104003840038400384003840038

Test 3: Latency 1->2

Code:

  fmls v0.8h, v0.8h, v1.8h
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

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)9fa6a8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204400372990613940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139479100001004003840038400384003840038
102044003730001313940725101001001000010010000500570690804001840037400373810833874510100200100002003000040078400371110201100991001001000010000007101161139479100001004003840038400384003840038
102044003730007263940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139479100001004003840038400384003840038
10204400373000613940725101001001000010010000500570690804001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139535100001004003840038400384003840038
10204400373000613940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000307101161139553100001004003840038400384003840038
10204400373000613940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139479100001004003840038400384003840038
10204400373000613940725101001001000010010000500570690804001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139479100001004003840038400384003840038
1020440037299010633940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139479100001004003840038400384003840038
10204400372990613940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000007101161139479100001004003840038400384003840038
10204400373000613940725101001001000010010000500570690814006540037400373810833874510100200100002003000040037400371110201100991001001000010000007101163139479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

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)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)9fa0a1a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
1002440037300000090106839407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400372990000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400372990000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400373000000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400373000000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400373000000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400373000000009439407251001010100001010000505706908140018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400373000000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
10024400372990000006139407251001010100001010000505706908040018040037400373813033876710010201000020300004003740037111002110910101000010000000640216223954510000104003840038400384003840038

Test 4: Latency 1->3

Code:

  fmls v0.8h, v1.8h, v0.8h
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

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)9fa6a8a9acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730006139407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000000007101161139479100001004003840038400384003840038
102044003730006139407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000000007101161139479100001004003840038400384003840038
1020440037299015639407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000003007101161139479100001004003840038400384003840038
102044003730006139407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000200007101161139479100001004003840038400384003840038
102044003730006139407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000000007101161139479100001004003840038400384003840038
102044003730006139407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000000007101161139479100001004003840038400384003840038
1020440037300072639407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000000007101411139479100001004003840038400384003840038
1020440037300061394072510100100100001001000050057069084001840037400373810833874510100200100002003000040037400371110201100991001001000010005409007101161139479100001004003840038400384003840038
102044003729906139407251010010010000100100005005706908400184003740037381083387451010020010000200300004003740037111020110099100100100001000000007101161139479100001004003840038400384003840038
102044003730001677394072510100100100001001029658357083044001840037400373810813387821042020610167200300004003740037111020110099100100100001007020207101321139479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03080b1e1f3a3f4e51inst 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)9fa0a1a6a8a9accdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000061394072510010101000610100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
100244003730000000908394072510010101000010100005057069080400184003740037381303387671015820100002030000400374003711100211091010100001000000006402162239473010000104008240038400384003840038
10024400373000000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000000006402512239473010000104003840038400384003840038
10024400373000000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  fmls v0.8h, v8.8h, v9.8h
  movi v1.16b, 0
  fmls v1.8h, v8.8h, v9.8h
  movi v2.16b, 0
  fmls v2.8h, v8.8h, v9.8h
  movi v3.16b, 0
  fmls v3.8h, v8.8h, v9.8h
  movi v4.16b, 0
  fmls v4.8h, v8.8h, v9.8h
  movi v5.16b, 0
  fmls v5.8h, v8.8h, v9.8h
  movi v6.16b, 0
  fmls v6.8h, v8.8h, v9.8h
  movi v7.16b, 0
  fmls v7.8h, v8.8h, v9.8h
  movi v8.16b, 9
  movi v9.16b, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2508

retire (01)cycle (02)030b181e1f3f51inst 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)696d6escheduler 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
16020420077150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001001010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238012420080000200240000200652006511160201100991001001600001001010111116112006201600001002006620066200662006620066
16020420065150000082258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066
16020420065150000040258010010080000100800005006400002004620065200653238010020080000200240000200652006511160201100991001001600001000010111116112006201600001002006620066200662006620066

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03191e3f51inst 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)9fa0cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600242006615000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100373111420211281220044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100483112620211282820044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100513112820211282820044215160000102004820048200482004820048
1600242004715000672580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100503112720211142420044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002005120047111600211091010160000100100343111320211271120044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100503112420211152520044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100503112720211142420044215160000102004820048200482004820048
16002420047150027462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100503112720211272720044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100503112720211272720044215160000102004820048200482004820048
1600242004715000462580012128000012800006264000011200282004720047323800122080000202400002004720047111600211091010160000100100503112720211272720044215160000102004820048200482004820048

Test 6: throughput

Count: 16

Code:

  fmls v0.8h, v16.8h, v17.8h
  fmls v1.8h, v16.8h, v17.8h
  fmls v2.8h, v16.8h, v17.8h
  fmls v3.8h, v16.8h, v17.8h
  fmls v4.8h, v16.8h, v17.8h
  fmls v5.8h, v16.8h, v17.8h
  fmls v6.8h, v16.8h, v17.8h
  fmls v7.8h, v16.8h, v17.8h
  fmls v8.8h, v16.8h, v17.8h
  fmls v9.8h, v16.8h, v17.8h
  fmls v10.8h, v16.8h, v17.8h
  fmls v11.8h, v16.8h, v17.8h
  fmls v12.8h, v16.8h, v17.8h
  fmls v13.8h, v16.8h, v17.8h
  fmls v14.8h, v16.8h, v17.8h
  fmls v15.8h, v16.8h, v17.8h
  movi v16.16b, 17
  movi v17.16b, 18

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)030818191e373f4e51inst 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)9fa8c2cfd2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440051300000004202516010010016000010016000050058560570040021400404231819973322267160100200160000200480000422944004011160201100991001001600001000010110011611400371600001004229540041400414004140041
16020442294300000004202516013510016000010016000050012800000140021400404231822222319998160100200160000200480000400404004011160201100991001001600001000010110111611400371600001004004140041400414229540041
16020440040300001004288442516010010016000010016000050057210730042299423184004019973322239160100200160000200480000423184004011160201100991001001600001000010110011611400371600001004004142295412464004142295
160204400402990000072802516010010016000010016000050012800000040021400404004019973319998160100200160000200480000412454004011160201100991001001600001000010110011611400371600001004119141220400414004140041
160204400403000001208402516013510016000010016000050012800000141183400404004019973319998160100200160000200480000400404004011160201100991001001600001000010110011611400371600001004004140041400414004142295
160204412453000000354202516010010016000010016000050012800000140021422944004021151322252160100200160000200480000423184004011160201100991001001600001000010110011611422911600001004004140041400414228640041
16020440040317000004202516010010016003610016000050012800000040021412454124519973319998160100200160000200480000400404004011160201100991001001600001000010110011611400371600001004124640041400414004140041
160204400402990000356146952516013510016003510016000050058655260140021422944124522198319998160100200160000200480000422944124511160201100991001001600001000010110011611412421600001004229541246400414229540041
16020440040300000004202516010010016000010016000050058655260141226400404004019973322254160100200160000200480000400404004011160201100991001001600001000010110011611422781600001004229540041422954124640041
16020440040300000006102516010010016003510016000050012800000140021400404004019973319998160100200160000200480000422944004011160201100991001001600001000010110011611400371600001004229541246400414229540041

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2645

retire (01)cycle (02)031e373f4e51inst 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)9fa1a8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002440040300080478955251600901016008012160000505868895110423080423274004019996032120916001020160000204800004232740040111600211091010160000100001002231191641110742324208160000104232840041423284004142328
16002440040317080478955251600101016000010160000505868895115422990423274232722237032119416001020160000204800004004042327111600211091010160000100001002284151621111740037209160000104004142328400414232840041
1600244004031700670251600901016000010160000505868895115400210400404232722237333223071600102016000020480000423274004011160021109101016000010000100228417272119342324208160000104232842328423284004142328
160024423273000010902516001010160080101600005058688951154002104232742327222373332002016001020160000204800004004042327111600211091010160000100001002284110162119342324208160000104004142409407614232840041
1600244232729900678955251600901016000010160000501280000115423080423274004019996333200201600102016000020480000423274004011160021109101016000010003100228415162119840037209160000104232840041423284004142328
1600244004031700670251600101016000010160000501280000115423080400404232722237333223071600102016000020480000423274004011160021109101016000010003100228414162119540037209160000104232840041423154004142328
16002442327300120470251600101016008010160000505868895115404090400924232722237032230716001020160000204800004004042314111600211091010160000102231002211525162115542324208160000104004142328423284004142328
16002442327300080478955251600101016000010160000501280000115423080400404232722237333200201600102016000020480000400404232711160021109101016000010000100228415162113640037208160000104232840041423284232842328
160024423273000804789552516009010160080101600005058688951154230804232740040222373332002016001020160000204800004232740040111600211091010160000100001002284131621131242324208160000104232840041423284004142328
160024400403170806789552516009010160080101600005012800001154230804232740040199963332002016001020160000204800004232742327111600211091010160000100001002284151621151140037209160000104232840041423284232840041