Apple M1 Microarchitecture Research by Dougall Johnson

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

SMLAL (by element, 4S)

Test 1: uops

Code:

  smlal v0.4s, v1.4h, 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)030b18191e1f3f4e51inst 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)a0a1a6a7a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037230000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073216112630100030383038303830383038
100430372200015061254825100010001000398313130183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
10043037230000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
10043037230000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
100430372300030612548251000100010003983130305830853037241532895100010003000303730371110011000000004273116112630100030383038303830383038
10043037230000061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030853086308530863038
10043084230003061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116222630100030383038303830383038
10043037230003061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
10043037230003061254825100010001000398313030183037303724153289510001000300030373037111001100000000073116112630100030383038303830383038
10043037230003061254825100010001000398313030183085303724153289510001000300030373037111001100000000073216112630100030383038303830383038

Test 2: Latency 1->1

Code:

  smlal v0.4s, v1.4h, 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): 3.0037

retire (01)cycle (02)03080b181e1f3a3f4e5051inst 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)9fa8a9accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000006129548025101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010010371002162329634100001003003830038300383003830038
10204300372250100006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000371002162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010010071012162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010010071012162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010010071012162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010010071012162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000671012162229634100001003003830038300383003830038
10204300372240000006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010020071012162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010010071012162229634100001003003830038300383003830038
10204300372250000006129548025101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010010071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0307080a0b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243008423200000000061295482510010101000010100005042773130300180300373003728287328767100102010000203000030037300371110021109101010000100000003006403163329630010000103003830038300383003830038
10024300372250000000001011295482510010101000010100005042773130300650300373003728287328767100102010000203000030037300371110021109101010000100000000006403163329630010000103003830038300383003830038
10024300372250000003300612954825100101010000101000050427731303001803003730037282873287671001020100002030000300373003711100211091010100001000008024006693163329630010000103003830038300383003830038
1002430037239000000000612954825100101010000101000050427731303001803003730037282873287671001020100002030000300373003711100211091010100001000002072006403163329630010000103003830038300383003830038
1002430037225000000000942954825100101010000101000050427731303001803003730037282873287671001020100002030000300373003711100211091010100001000000090006403163329630010000103003830038300383003830038
1002430037225000000000822954825100101010000101000050427731303001803003730037282873287671001020100002030000300373003711100211091010100001000000081006403163329630010000103003830038300383003830038
10024300372250100001328807422954825100101010000101000050427731303001803003730037282873287671001020100002030000300373003711100211091010100001000000030006403163329630010000103003830038300383003830038
1002430037225000000000822954825100101010000101000050427731303006503003730037282873287671016020100002030000300373003711100211091010100001000000021006403163329630010000103003830038300383003830038
1002430037232000000000612954825100101010000101000050427731303001803003730037282873287671001020100002030000300373003711100211091010100001000007018006403163329630010000103003830038300383003830038
10024300372330000005760061295482510017101000010100005042777970300180300373003728287328767100102010000203000030037300371110021109101010000100000009006403163329630010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  smlal v0.4s, v0.4h, 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): 3.0037

retire (01)cycle (02)03081e3f4e51inst 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037224006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011621296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011711296340100001003003830038300383003830038
10204300372250021229548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611297420100001003003830038300383003830038
10204300372250010629548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037224006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100071011602296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b191e1f3f4e51inst 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)9fa1a8acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001002606402162229630010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000006402162229630010000103003830038300383003830038
10024300372240000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001001006402162229630010000103003830038300383003830038
100243003722500000536295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006402162229630010000103003830038300383003830038
100243003722400000726295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000006402162229630010000103003830038300383003830038
10024300372250000072629548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100561206402162229630010000103003830038300383003830038
10024300372250009061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000306402162229630010000103003830038300383003830038
10024300372261000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000006402162229630010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000006402162229630010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  smlal v0.4s, v1.4h, 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): 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)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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037224000000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000100071011611296340100001003003830038300383003830038
102043003722500000840006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296342100001003003830038300383003830038
1020430037225000005220025329548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300183003730037282653287451010020410000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
102043003722500000237006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
1020430037224000000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
102043003722500000699006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0307080b18191e1f3f4e51inst 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)9fa0a1a6a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000043806129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103008530038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373008428287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001012100081010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722400000006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000006402162229630010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  smlal v0.4s, v8.4h, v9.h[1]
  movi v1.16b, 0
  smlal v1.4s, v8.4h, v9.h[1]
  movi v2.16b, 0
  smlal v2.4s, v8.4h, v9.h[1]
  movi v3.16b, 0
  smlal v3.4s, v8.4h, v9.h[1]
  movi v4.16b, 0
  smlal v4.4s, v8.4h, v9.h[1]
  movi v5.16b, 0
  smlal v5.4s, v8.4h, v9.h[1]
  movi v6.16b, 0
  smlal v6.4s, v8.4h, v9.h[1]
  movi v7.16b, 0
  smlal v7.4s, v8.4h, 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)03081e3a3f51inst 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)60696d6escheduler 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? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200911510003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641510003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010021011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
1602042006415100070425801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030b1e3a3f51inst 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)9fa0a1a8acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)dfe0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600242006415000045258001212800001280000626400001120031200502005032280012208000020240000200462004611160021109101016000010000001003531116202117100200432150160000102004720047200472004720047
160024200461500004525800121280000128000062640000112002720046200463228001220800002024000020046200461116002110910101600001000000100353118202111090200432150160000102004720047200472004720047
16002420046150072045258001212800001280000626400001120027200462004632280012208000020240000200462004611160021109101016000010000001003131110202118100200432150160000102004720047200472004720047
1600242004615006304525800121280000128000062640000112002720046200463228001220800002024000020046200461116002110910101600001000000100333111220211690200432150160000102004720047200472004720047
16002420046150063045258001212800001280000626400001120027200462004632280012208000020240000200462004611160021109101016000010000001003431192021110120200432150160000102004720047200472004720047
160024200461500111145258001212800001280000626400001120027200462004632280012208000020240000200462004611160021109101016000010000001003331112202118110200432150160000102004720047200472004720047
16002420046150000928258001212800001280000626400001120027200462004632280012208000020240000200462004611160021109101016000010000001003331111202111480200432150160000102004720117200472004720047
160024200461500004525800121280000128000062640000112002720046200463228001220800002024000020046200461116002110910101600001000000100353117202117120200432150160000102004720047200472004720047
160024200461500004525800121280000128000062640000112002720046200463228001220800002024000020046200461116002110910101600001000000100303117202111070200432150160000102004720047200472004720047
1600242004615000045258001212800001280209626416641120027200462004632280012208000020240315201272004611160021109101016000010420021003331172021112130200432150160000102004720047200472046420047

Test 6: throughput

Count: 12

Code:

  smlal v0.4s, v12.4h, v13.h[1]
  smlal v1.4s, v12.4h, v13.h[1]
  smlal v2.4s, v12.4h, v13.h[1]
  smlal v3.4s, v12.4h, v13.h[1]
  smlal v4.4s, v12.4h, v13.h[1]
  smlal v5.4s, v12.4h, v13.h[1]
  smlal v6.4s, v12.4h, v13.h[1]
  smlal v7.4s, v12.4h, v13.h[1]
  smlal v8.4s, v12.4h, v13.h[1]
  smlal v9.4s, v12.4h, v13.h[1]
  smlal v10.4s, v12.4h, v13.h[1]
  smlal v11.4s, v12.4h, 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): 0.2503

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)9fa8a9acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1202043084322500410251201001001200001001200005009600001309323003930039149733149971201002001200002003600003003930039111202011009910010012000010000007610316433003601200001003004030357300403004030040
1202043003923200410251201001001200001001200005009600001300203003930039149733167061201002001200002003600003003930039111202011009910010012000010000017610416443003601200001003004030040300403004030040
1202043003922500410251201001001200001001200005009600001300203003930039149733159091201002001200002003600003003930039111202011009910010012000010000007610416453094801200001003095230040300403004030040
12020430039224004102512010010012000010012000050042861841300203003930039149733149971201002001204242003600003003930039111202011009910010012000010020007610516553003601200001003004030952300403095230040
1202043003922400610251201001001200001001200005009600001300203003930039149733149971201002001200002003600003003930039111202011009910010012000010000007610416533003601200001003095230040309523004030040
1202043003922501410251201001001200521001200005009600001300203003930039149733149981201002001200002003600003003930039111202011009910010012000010012307610516453003601200001003004030040300403004030040
12020430039231004102512010010012000010012000050042861841300643003930039149738158081201002001200002003600003003930039111202011009910010012000010000007610516543003601200001003004030040300403004030952
1202043003922500410251201001001200001001200005009600001300203004130039149733149971201002001200002003600003003930039111202011009910010012000010010007610416443003601200001003004030040300403004030952
1202043003922500416710251201001001200521001200005009600001300203003930039149733149971201002001205102003600003003930039111202011009910010012000010000007610416443008801200001003004030040300403004030040
12020430039224004102512010010012000010012000050042861841300203003930039158573149971201002001200002003600003003930039111202011009910010012000010000007610516553003601200001003095230040300403004030040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2503

retire (01)cycle (02)031e1f373a3f4e51inst 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)9fa8acc2cfd0d2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
120024303382250002450251200101012000010120000509600001030020300403003914996315019120012201200002036000030039300391112002110910101200001000075240213161618300360120000103004030040300403004030040
1200243003923200172660251200271012005210120000509600000030020300393003914996315019120012201200002036000030039300391112002110910101200001000075245014161614300370120000103004030040309433004130040
1200243004023800352450251200101012000010120000509600001030020300393003914996315019120012201207332036000030040304491112002110910101200001000075240317161416300360120000103004030040300403175130040
1200243003922500172660251200101012003510120000509600000530023300393003915850315902120012201200002036000030039300391112002110910101200001000075245315161816300360120000103004030040300413004030040
120024309512250002450251200101012000010120000509600000530021300393003914996315931120012201200002036000030039300391112002110910101200001000075240316161513300360120000103095230040300403004030952
120024300392250012256025120010101200001012000050960000103002030039309511587931501912001220120000203600003003930039111200211091010120000100007524031616916300360120000103175130041300403004030040
120024300392250002450251200101012000010120000509600000530020300393003914996315019120012201200002036000030039309971112002110910101200001000075240316161616300360120000103004030040309233004030040
1200243003922500352450251200101012000010120000509600000530020300393092214996315019120012201201062036000030039300391112002110910101200001000075245016161516309480120000103004030040300403004030040
120024317502250002450251200621012000010120000509600001030020300393003914996315019120012201200002036000030039300391112002110910101200001000075240315161614300370120000103004030040300403004030040
1200243003923800172450251200101012000010120000509600001030020309223003914996315019120012201200002036000030039300391112002110910101200001000075245016161717300360120000103004030040300403004030041