Apple M1 Microarchitecture Research by Dougall Johnson

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

FMLA (by element, H)

Test 1: uops

Code:

  fmla h0, h1, v2.h[1]
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

retire (01)cycle (02)031e1f3f4e51inst 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)a6a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004403730420613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113547100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000173116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
100440373000613407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038
10044037321201893407251000100010005319084018403740373258338951000100030004037403711100110000073116113473100040384038403840384038

Test 2: Latency 1->1

Code:

  fmla h0, h1, v2.h[1]
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03081e3a3f4e51inst 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)9fa0a1a8a9acbranch mispredict (cb)cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000610710121622394790100001004003840038400384003840038
1020440037300000595394072510100100100001001000050057069081400184003740037381083387451010020410000200300004003740037111020110099100100100001000000000710021622394790100001004003840038400384003840038
102044003730000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000710021622394790100001004003840038400384003840038
102044003730000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000710021622394790100001004003840038400384003840038
102044003730000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000007000710131632394790100001004003840038400384003840038
102044003730000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000710131622394790100001004003840038400384003840038
1020440037299000726394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000710131622394790100001004003840038400384003840038
102044003730000061394072510100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000710121622394790100001004003840038400384003840038
102044003730000061394072510100100100001001000050057069081400184003740037381083387451025120010000200300004003740037111020110099100100100001000000000710121622394790100001004003840038400384003840038
102044003730000061394074410100100100001001000050057069081400184003740037381083387451010020010000200300004003740037111020110099100100100001000000000710121620394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

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)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)9fa8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000013383940725100101010000101000050570690804001840037400373813033876710010201000020300004003740037111002110910101000010000640516343947310000104003840038400384003840038
10024400373000012133940725100101010000101000050570690814001840037400373813033876710010201000020300004008440037111002110910101000010000640316333947310000104003840038400384003840038
10024400373000012103940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000640316333947310000104003840038400384003840038
10024400373000011713940725100101010000101000050570690804001840037400373813033876710010201000020300004003740037111002110910101000010000640316333947310000104003840038400384003840038
10024400373000012523940725100101010000101000050570690804001840037400373813033876710010201000020300004003740037111002110910101000010000640316343947310000104003840038400384003840038
10024400373000013533940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000640316333947310000104003840038400384003840038
10024400373000012233940725100101010000101000050570690804001840037400373813033876710010201000020300004003740037111002110910101000010000640316333947310000104003840038400854003840038
10024400372990013303940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000640316333947310000104003840038400384003840038
10024400372990912463940725100101010000101000050570690804006540037400373813033876710010201000020300004003740037111002110910101000010000640316333947310000104003840038400384003840038
10024400373000012883940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000640316433947310000104003840038400384003840038

Test 3: Latency 1->2

Code:

  fmla h0, h0, v1.h[1]
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03080b1e3a3f4e51inst 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)6061696d6edispatch 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)9fa8cfd0d2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020440037299011208239407251010010010000100100005005706908104001840037400373810833874510100200100002003000040037400371110201100991001001000010007100011611394790100001004003840038400384003840038
102044003730000006139407251010010010000100100005005706908104001840037400373810833874510100200100002003000040037400371110201100991001001000010007105111611394790100001004003840038400384003840038
10204400373000000328439407251010010010000100100005005706908154001840037400373810833874510100200100002003000040037400371110201100991001001000010007105111611394790100001004003840038400384003840038
1020440037299000021239407251010010010000100100005005706908154001840037400373810833874510100200100002003000040037400371110201100991001001000010007105111611395620100001004003840038400384003840038
1020440037299000016839407251010010010000100100005005706908154001840037400373810833874510100200100002003000040037400371110201100991001001000010007105111611394790100001004003840038400384003840038
1020440037300000080439407251010010010000100100005005706908054001840037400373810833874510100200100002003000040037400841110201100991001001000010007105111611394790100001004003840038400384003840038
1020440037299000117039407251010010010000100100005005706908154001840037400373810833874510100200100002003000040037400372110201100991001001000010007105111611394790100001004003840038400384003840038
1020440037300000032139407251010010010000100100005005706908154001840037400373810833874510100200100002003000040037400371110201100991001001000010017100011611394790100001004003840038400384003840038
102044003730000006139407251010010010000100100005005706908004001840037400373810833874510100200100002003000040037400371110201100991001001000010007100011611394790100001004003840038400384003840038
1020440037299009014939407251010010010000100100005005706908004001840037400373810833874510100200100002003000040037400371110201100991001001000010007100011611394790100001004003840038400384003840038

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)696d6edispatch 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)9fa0a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100244003730010002402143940725100101010000101000050570690840123400374008438169026388431075220106422031929402734022761100211091010100001000000006402162239473010000104003840038400384003840038
1002440037300000000105394072510010101000010100005057069084001840037400373813003387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
1002440084300000000147394072510010101000010100005057069084001840037400373813003387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400372990000003639394072510010101000010100005057069084001840037400373813003387671001020100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
10024400373000000001703940725100101010000101000050570690840018400374003738130033876710010201000020300004003740037111002110910101000010000002488506403494439619010000104041340275404634032240227
10024402263041179119144073253934314510055161004815100005057069084001840037400853813003387671001020100002030000400374003711100211091010100001042112364348764962240034010000104003840038400384003840038
1002440037300000014618806139407251001710100001010000505706908400184003740037381300453896111645201178220347794051440559111100211091010100001000000364806402162239512010000104003840038400384003840038
1002440037300008119279689615393531881008215100362010888103571807640403403684035738166055389411134620100002030000400374003711100211091010100001000000006402162239473010000104003840038400384003840038
100244003729900691065968275394072510010101000010100005057069084001840037400373813003387671001020100002030000400374003711100211091010100001000000006402162239473110000104003840038400384003840038
100244003729900000061394072510010101000010100005057069084001840037400373813003387671001020100002030000401314003711100211091010100001000000007032162239473010000104003840038400384003840038

Test 4: Latency 1->3

Code:

  fmla h0, h1, v0.h[1]
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 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)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)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020440037300000000072639407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010006100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004008040038400384003840038
102044003730000000008239407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400184003740037381087338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400884003740037381080338763101002021000021030000400374017911102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908400184003740037381080338745101002001000020030000400374003711102011009910010010000100000000071011611394790100001004003840038400384003840038

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)9fa8cfd2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400372993420613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
10024400373003390613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002193239473010000104003840038400384003840038
1002440037300300613940725100101010000101000050570690814001840037400843813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
1002440037300420613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
1002440037300510613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
1002440037300300613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
1002440037299300613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
100244003730000613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
1002440037300540613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104003840038400384003840038
1002440037299300613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010064002162239473010000104007140038400384003840038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  fmla h0, h8, v9.h[1]
  movi v1.16b, 0
  fmla h1, h8, v9.h[1]
  movi v2.16b, 0
  fmla h2, h8, v9.h[1]
  movi v3.16b, 0
  fmla h3, h8, v9.h[1]
  movi v4.16b, 0
  fmla h4, h8, v9.h[1]
  movi v5.16b, 0
  fmla h5, h8, v9.h[1]
  movi v6.16b, 0
  fmla h6, h8, v9.h[1]
  movi v7.16b, 0
  fmla h7, h8, v9.h[1]
  movi v8.16b, 9
  movi v9.16b, 10

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602042008915000402580100100800001008000050064000011200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515000402580100100800001008000050064000001200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515000402580100100800001008000050064000001200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
160204200651500225402580100100800001008000050064000001200462006520065323801002008000020024000020149200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515000402580100100800001008000050064000011200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
160204200651510040258010010080000100800005006400001120046200652006532380100200800002002400002006520065111602011009910010016000010002311011111611200621600001002006620066200662006620066
16020420065150001832580100100800001008000050064000011200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515000402580100100800001008000050064000001200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
16020420065151054402580100100800001008000050064000001200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066
1602042006515000402580100100800001008000050064000011200462006520065323801002008000020024000020065200651116020110099100100160000100001011111611200621600001002006620066200662006620066

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03090a0b18191e1f3f51inst 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)5f6061696d6e74scheduler 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)9fa0a1a6a8accdcfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420098150000000046298001212800001280000626400001152003320052200520323800122080000202400002005220052111600211091010160000100005001003083162521169200492201160000102005320053200532005320053
16002420052150000000046298001212800001280000626400001152003320052200520323800122080000202400002005220052111600211091010160000100002001003283192521169200492201160000102005320053200532005320053
160024200521500000000462980012128000012800006264000011520033200522005203238001220800002024000020052200521116002110910101600001000000010029831102521196200492201160000102005320053200532005320053
160024200521500000000462980012128000012800006264000011520033200522005203238001220800002024000020052200521116002110910101600001000010010032831102521196200492201160000102005320053200532005320053
16002420052150000000046298001212800001280000626400001152003320052200520323800122080000202400002005220052111600211091010160000100006001003283172521179200492401160000102005320053200532005320053
16002420052150000000046298001212800001280000626400001152003320052200520323800122080000202400002005220052111600211091010160000100006001002984192522297200492402160000102005320062200532006220053
160024200611500000000462980012128000012800006264000001520033200522006103238001220800002024000020061200611116002110910101600001000050135010032113162521197200492201160000102005320053200532005320053
16002420052150000000046298001212800001280000626400001152004220052200520323800122080000202400002005220052111600211091010160000100000001003283162521179200492201160000102005320053200532005320053
16002420052150000000046298001212800001280000626400001152003320052200520323800122080000202400002005220052111600211091010160000100002016801003083192521169200492201160000102005320053200532005320053
160024200521500000018046298001212800001280000626400001152003320052200520323800122080000202400002005220052111600211091010160000100007001002984192521179200492201160000102005320053200532005320053

Test 6: throughput

Count: 12

Code:

  fmla h0, h12, v13.h[1]
  fmla h1, h12, v13.h[1]
  fmla h2, h12, v13.h[1]
  fmla h3, h12, v13.h[1]
  fmla h4, h12, v13.h[1]
  fmla h5, h12, v13.h[1]
  fmla h6, h12, v13.h[1]
  fmla h7, h12, v13.h[1]
  fmla h8, h12, v13.h[1]
  fmla h9, h12, v13.h[1]
  fmla h10, h12, v13.h[1]
  fmla h11, h12, v13.h[1]
  movi v12.16b, 13
  movi v13.16b, 14

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)031e1f373f4e51inst 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
1202044003930000361996125120100100120000100120000500585199314002040039416862657725266491201002001200002003600004168640039111202011009910010012000010007610116114167701200001004004040040400404169240040
1202044003931200061402742512010010012000010012000050058519930416674168640039249323249971201002001200002003600004003941686111202011009910010012000010007610116114003001200001004004041692416924004041687
120204400393120036199612512010010012000010012000050058519930416724168640039249323249971201002001200002003600004168640039111202011009910010012000010007610116114167701200001004169241692424714004041687
1202044168630000061356892512010010012000010012000050058519930416724168640039249323249971201002001200002003600004003941686111202011009910010012000010007610116114167701200001004004040040416874004041692
1202044168629900361379662512010010012000010012000050058518690416674003941686265773266441201002001200002003600004168640039111202011009910010012000010007610116114003001200001004004040040416924004040040
1202044003931000015999382512020011612000010012000050056306401416674168640039249323266441201002001200002003600004168640039111202011009910010012000010007610116114167701200001004004040040400404168740040
120204416863000036199612512010010012000010012000050056306401400204003941691265773249971201002001200002003600004169140039111202011009910010012000010067610116114003001200001004004040040400404168740040
120204416863000036199612512010310012000310012000050056306400400204003941686265823266491201002001200002003600004168640039111202011009910010012000010007610116114003001200001004004040040416924004041692
120204416913000006199612512010010012000010012000050058519930400204003941688249323249971201002001200002003600004003941686111202011009910010012000010027610116114003001200001004004041692416874004040040
1202044003931200061356892512010310012000110012019650056306400400204003941686265773266441201002001200002003600004168640039111202011009910010012000010007610116114167701200001004004040040400404171241692

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03080b18191e1f373f4e51inst 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)9fa0a1a8acc2c5branch mispredict (cb)cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
120024424703180000000903796625120010101200001012000050563064011416724169140039249553250191200102012000020360000400394003911120021109101012000010001300075223211416211394167801594120000104004040040400404171040040
12002440039300000012016799612512001110120000101200005056306401041672400394003924955326667120010201200002036000040039400391112002110910101200001000100007522311316211534003003090120000104004040040400404004040040
1200244003930001003600673568925120010101200001012000050585199311416674003940039249553250191200102012000020360000416914003911120021109101012000010000000075223113162125340030015174120000104004041688400404169240040
1200244003931200000006799612512001310120000101200005056306401140020400394168624955325019120010201200002036000040039424701112002110910101200001000000007522311516211534003001594120000104004040040400404004041687
12002440039300000000067379662512001310120001101200005056306401140020400394168624955325019120010201200002036000040039400391112002210910101200001000000007522311516211354003001594120000104168740040400404004040040
1200244003930000000006799612512001010120000101200005056306401140020400394003924955325019120010201200002036000040039400391112002110910101200001000000107522311516211534003001594120000104004040040400404163940040
1200244003930000000006799612512001010120000101200005056306401140020400394003924955325019120010201200002036000041686400391112002110910101200001000000007522311416211554003001594120000104004040040416924004040040
12002440039299000000044799612512001010120000101200005056306401140020400394003924955325019120010201200002036000040039400391112002110910101200001000000007522311516211394003001594120000104004040040400404004040040
12002440039300000000067996125120010101200001012000050563064010400204003940039249553250191200102012000020360000400394003911120021109101012000010000000075223111116211454003001590120000104004040040400404004040040
1200244003930000000006799612512001010120000101200005056306401040020400394168724955325019120010201200002036000040039400391112002110910101200001000000017522311416211944003001594120000104168940040400404004040040