Apple M1 Microarchitecture Research by Dougall Johnson

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

FMLSL2 (vector, 4S)

Test 1: uops

Code:

  fmlsl2 v0.4s, v1.4h, v2.4h
  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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004403730001033407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
1004403730048613398251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110001073116113473100040384038403840384038
1004403730001473407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
1004403730002703407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038

Test 2: Latency 1->1

Code:

  fmlsl2 v0.4s, v1.4h, v2.4h
  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)03080b18191e1f3a3f4e51inst 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)9fa0a1a6a8acc2c5branch mispredict (cb)cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730000000001263940725101001001000010010000500570690840018400374003738115738740101002001000820030024400374003711102011009910010010000100000000111720001600394900100001004003840038400384003840038
10204400373000000001613940725101001001000010010000500570690840018400374003738115638740101002001000820030024400374003711102011009910010010000100000000111718001600394900100001004003840038400384003840038
10204400373000000000613940725101001001000010010000500570690840018400834003738115738741101002001034620430024400374003711102011009910010010000100000000111717001600394890100001004003840038400384003840038
10204400373000000000613940725101001001000010010000500570690840018400374003738115638740101002001000820030024400374003711102011009910010010000100000000111717001600394890100001004003840038400384003840038
10204400373000000000613940725101001001000010010000500570690840018400374003738115638741101002001000020030000400374003711102011009910010010000100000000000710121622394790100001004003840038400384003840038
10204400833000000000613940725101001001000010010000500570690840018400374003738108338745101002001000020030000400374003711102011009910010010000100000000000710121622394790100001004003840038400384003840038
10204400373000000000613940725101001001000010010000500570690840018400374003738108338745101002001000020030000400374003711102011009910010010000100000000000710121622394790100001004003840038400384003840038
10204400373000000000613940744101001001000010410000500570690840018400374003738108338745101002001000020030000400374003711102011009910010010000100000000000710131622394790100001004003840038400384003840038
10204400373000000000613940725101001001000010010000500570690840018400374003738108338745101002001000020030000400374003711102021009910010010000100000000000710121622394790100001004003840038400384003840038
10204400373100000000613940725101001001000010010000500570690840018400374003738108338745101002001000020030000400374003711102011009910010010000100000000000710121623394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03181e3f4e5051inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002440037300006139407400212510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640616343947310000104003840038400384003840038
1002440037300006139407400212510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640316343947310000104003840038400384003840038
100244003730000613940702510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000640316433947310000104003840038400384003840038
100244003730010613940702510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000640416433947310000104003840038400384003840038
100244003730000613940702510010101000010100005057069080400184003740037381303387811001020100002030000400374003711100211091010100001000640316343947310000104003840038400384003840038
100244003729900613940702510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001003640316333947310000104003840038400384003840038
100244003730000613940702510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640416433947310000104003840038400384003840038
100244003729900613940702510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640316433947310000104003840038400384003840038
100244003730000613940702510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640316343947310000104003840038400384003840038
100244003730000613940702510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640316343947310000104003840038400384003840038

Test 3: Latency 1->2

Code:

  fmlsl2 v0.4s, v0.4h, v1.4h
  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)03090b18191e1f3f4e51inst 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)9fa0a1a6a7a8a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730000000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
102044003730000000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
102044003730000000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
102044003730000000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004022940038400384003840038
102044003730000000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
102044003730000000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
102044003729900000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
102044003729900000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611395000100001004003840038400384003840038
102044003730000000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840181400384003840038
102044003730000000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03181e3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002440037300001158394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000961394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400372990061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001013640216223947310000104003840038400384008440038
10024400373002156251393802510010101000010100006057083040400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384022840038
100244003729901261394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001010640216223947310000104003840038400384003840038
100244003730000613940725100101010000101000050570690804001840230400373815116387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000361394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000640216223947310000104003840038400384003840038

Test 4: Latency 1->3

Code:

  fmlsl2 v0.4s, v1.4h, v0.4h
  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)03080a0b1e1f3f4e51inst 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)9fa0a1a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020440037300000007893940725101001001000010010000500570690814001840037400373810833874510100200100002123000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038
102044003730000000843937125101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038
1020440037300001001263940725101001001000010010000500570690814001840037400373810833874510252200100002003000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038
1020440037300000002333940725101001211000010010000623570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038
1020440037300000002583940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071023322394790100001004003840038400384003840038
10204400373000000019139407251010010010000100100005005706908140018400374003738108338745101002001000020030000400374003711102011009910010010000100000004710216223947925100001004003840038400384003840038
102044003730010000613940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038
102044003730000000613940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038
1020440037300000002793940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071011622394790100001004003840038400384003840038
1020440037300000001913940725101001001000010010000500570690814001840037400373810833874510100200100002003000040037400371110201100991001001000010000000071021622394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)031e1f3f4e51inst 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)9fa0a8acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100244003729900124394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001001006402162239473210000104003840038400384003840038
100244003730000211394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000061394072510010101000010100005057069081400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
100244003730000160394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000082394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002030000400374003711100211091010100001000006402162239473010000104003840038400384003840038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  fmlsl2 v0.4s, v8.4h, v9.4h
  movi v1.16b, 0
  fmlsl2 v1.4s, v8.4h, v9.4h
  movi v2.16b, 0
  fmlsl2 v2.4s, v8.4h, v9.4h
  movi v3.16b, 0
  fmlsl2 v3.4s, v8.4h, v9.4h
  movi v4.16b, 0
  fmlsl2 v4.4s, v8.4h, v9.4h
  movi v5.16b, 0
  fmlsl2 v5.4s, v8.4h, v9.4h
  movi v6.16b, 0
  fmlsl2 v6.4s, v8.4h, v9.4h
  movi v7.16b, 0
  fmlsl2 v7.4s, v8.4h, v9.4h
  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)033f51inst 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? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200911504025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020110099100100160000100001011121611200621600001002006620066200662006620066
160204200651514025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515012425801001008000010080000500640000200462006520065323801002008000020024000020065200891116020110099100100160000100001011111611200621600001002006620066200662006620066
160205200761514025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515140258010010080000100800005006400002004620065200651223801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
160204200651504025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
160204200651504025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
160204200651504025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020210099100100160000100001011111611200621600001002006620066200662006620066
160204200651504025801001008000010080000500640000200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
160204200651504025801001008000010080000500640000200462015720065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0304070a191e3a3f51inst 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)9fa0a8accfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600242007315001000088258001212800001280000626400001120030200492004903238001220800002024000020049200491116002110910101600001000010060311030221311342120046216160000102005020050200502005020050
1600242004915010000146258001212800001280000626400001120030200492004903238001220800002024000020049200491116002110910101600001000010061311121221281273420046216160000102005020050201192005020050
1600242004915000000158258001212800001280000626400001120030200472004903238001220800002024000020049200491116002110910101600001000010057311136221281353720046216160000102005020050200502005020050
1600242004915110000146258001212800001280000626400001120149200492004903238001220800002024000020049200491116002110910101600001000010043311020221321372620046216160000102005020050200502005020050
16002420049150001001167258001212800001280000626400001120030200492004903238001220800002024000020049200491116002110910101600001000010044311037221251363720046216160000102022320050200502005020050
1600242004915000100058258001212800001280000626400001020030200492004903238001220800002024032420049201351116002110910101600001000010057311036221261362620046216160000102005020050200502005020048
1600242004915001000158258001212800001280000626400001120030200492004903238001220800002024000020049200491116002110910101600001000010063311036223221343720046216160000102005020050200502005220050
1600242004915001100158258001212800001280000626400001120034200492005303238001220800002024000020049200491116002110910101600001000010060312036261211383820046216160000102005020050200502005020054
1600242004915000100158258001212800001280000626400001020030200492004903238001220800002024000020049200491116002110910101600001000010058621037223242383720046216160000102005020050200502005020054
1600242004915001000158258001212800001280000626400001020030200492004903238001220800002024000020049200491116002110910101600001000010060311035221221363620046216160000102005020050200502005020133

Test 6: throughput

Count: 12

Code:

  fmlsl2 v0.4s, v12.4h, v13.4h
  fmlsl2 v1.4s, v12.4h, v13.4h
  fmlsl2 v2.4s, v12.4h, v13.4h
  fmlsl2 v3.4s, v12.4h, v13.4h
  fmlsl2 v4.4s, v12.4h, v13.4h
  fmlsl2 v5.4s, v12.4h, v13.4h
  fmlsl2 v6.4s, v12.4h, v13.4h
  fmlsl2 v7.4s, v12.4h, v13.4h
  fmlsl2 v8.4s, v12.4h, v13.4h
  fmlsl2 v9.4s, v12.4h, v13.4h
  fmlsl2 v10.4s, v12.4h, v13.4h
  fmlsl2 v11.4s, v12.4h, v13.4h
  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): 0.3337

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)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)9faccfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1202044003930000619961251201001001200001001200005005630640041672416864003924932324997120100200120000200360000416874003911120201100991001001200001000761041635400301200001004169240040416924004040040
12020441686300002519961251201011001200011001200005005630640040020400394003926582324997120100200120000200360000400394169111120201100991001001200001000761021623400301200001004004041687400404169240040
1202044003931200899961251201001001200001001200005005851869040020400394003926582324997120100200120000200360000416864003911120201100991001001200001000761021623416831200001004004041692400404169240040
1202044003931300619961251201001001200001001200005005630640040020416914003924932324997120100200120000200360000400394169111120201100991001001200001000761031632400301200001004004040040416924004041692
1202044168630000619961251201001001200001001200005005851869040020400394003924932324997120100200120000200360000400394169121120201100991001001200001000761031632400301200001004004041692400404169240040
12020440039312016137966251201011001200011001200005005630640041672416914003924932324997120100200120000200360000424704003911120201100991001001200001000761031632406031200001004004041687400404169240040
1202044003930000619961251201001001200001001200005005851869040020400394168626582326644120100200120000200360000400394009111120201100991001001200001000761031623400301200001004004041687400404169240040
1202044003930000619961251201001001200001001200005005851869040020400394003924932324997120100200120000200360000400394003911120201100991001001200001000761031633400301200001004247140040416924004040040
12020441686300181619961251201001001200031001200005005630640041667400394003924932324997120100200120000200360000400394003911120201100991001001200001000761021632400301200001004169240040416924004041687
120204416913000158637966251201011001200011001200005005630640040020400394003924932324997120100200120000200360000400394169111120201100991001001200001000761031623400301200001004169240040400404004040040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.3337

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)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)9fa6a8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12002440039316006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004168640039111200211091010120000100007522311516211754003020105120000104004040040400404004040040
12002440039300006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004003941691111200211091010120000100007522311716211774003020105120000104004040040400404004040040
12002440039300006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004170240039111200211091010120000100007522311516211774003020105120000104004040040400404004040040
12002440039312006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004168640039111200211091010120000100007522311716211574003020105120000104004041692400404169240040
12002440039299006799612512001010120000101200005056306401104245140039400392660432501912001020120000203600004003941701111200211091010120000100007522311716421574003020105120000104004040040400404004040040
12002440039300006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004003940039111200211091010120000100007524611716421754003020105120000104004040040400404004040040
12002440039300006799612512001010120000101200005056306401104002040039400902495532501912001020120000203600004003940039111200211091010120000100007524622716412574003020205120000104004040040400404004040040
12002440039300006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004003940039111200211091010120000100007522311516222574003020105120000104004040040400404004040040
12002440039299006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004003940039111200211091010120000100007522621716212754003020105120000104004040040400404004040040
12002440039300006799612512001010120000101200005056306401104002040039400392495532501912001020120000203600004003940039111200211091010120000100007522312716412774003020109120000104004040040400404004040040