Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRDMLSH (by element, S)

Test 1: uops

Code:

  sqrdmlsh s0, s1, v2.s[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)0308191e3f4e51inst 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)a1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037220006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037220006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230066125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230006125482510001000100039831330183037303724153289510001000300030373037111001100008073216222630100030383038303830383038
10043037230096125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037220006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038
10043037230006125482510001000100039831330183037303724153289510001000300030373037111001100000073216222630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqrdmlsh s0, s1, v2.s[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)031e3a3f4e51inst 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)9fbranch mispredict (cb)cdcfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710131622296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038
1020430037225001242954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296342100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372827332874510100200100002003000030037300371110201100991001001000010000710021622296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038
1020430037225001052954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000710121622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b0e181e1f3a3f4e51inst 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)9fa0a1a6a8acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225002000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006404163329630010000103003830038300383003830038
1002430037224000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006403162229630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006403162229630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402164329630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383022930038
10024300372250000300126295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006403162329630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402162229630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006402163329630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000006403163329630010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqrdmlsh s0, s0, v1.s[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)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)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)9fa0a1a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830038300383003830038
10204300372250000000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830038300383003830038
10204300372250000000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830038300383003830038
10204300372250000000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830234300383003830038
10204300372250000000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830038300383003830038
10204300372250000000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830038300383003830038
102043003722500000012006129548251010010010000100100005004277313030018300373003728265328745101002041000020030000300373003711102011009910010010000100420002853071021622296340100001003003830038300383003830038
1020430037225000000300077729530251011811910000100100006214277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000071021622296340100001003003830038300383003830038
102043003722500000012006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000003071021622296340100001003008630086300383003830038
102043003722500000000161295482510100100100001001000050042773130300183003730037282651728745101002001000020030000300373003711102011009910010010000100000200071021622296340100001003003830038301343013630038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103022830038300383003830038
10024300372240061295482510010101000010100005042786700300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001003640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183008430037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250084295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqrdmlsh s0, s1, v0.s[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)03071e3a3f4e51inst 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)9fa0a1a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225001612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011710296342100001003003830038300383003830038
1020430037225100612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011615296340100001003003830038300383003830038
10204300372250660612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000627427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225001612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071051711296340100001003003830038300383003830038
1020430037225100612954825101001001000012510000626427731330018300373003728265328744101252001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225001612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038

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)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
1002430037225000000000170295482510010101000010100005042773133001830037300372828732876710161201000020300003003730037111002110910101000010000000006404165629630010000103003830038300383003830038
1002430037225000000000126295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006404165529630010000103003830038300383003830038
1002430037225000000000147295482510010101000010100005042773133001830037300372828732876710010201049220300003003730037111002110910101000010240000026406164529630010000103003830038300383003830038
1002430037224000000000145295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006404165529630010000103003830038300383003830038
1002430037225000000000126295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006405166529630010000103003830038300383003830038
10024300372250000000006522295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006406166529630010000103003830038300383003830038
100243003722500000000061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006404164429630010000103003830038300383003830038
1002430037225000000000174295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006405165529630010000103003830038300383003830038
1002430037224200000000530295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006404165629630010000103003830038300383003830038
1002430037225000000000170295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000000006404165629630010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqrdmlsh s0, s8, v9.s[1]
  movi v1.16b, 0
  sqrdmlsh s1, s8, v9.s[1]
  movi v2.16b, 0
  sqrdmlsh s2, s8, v9.s[1]
  movi v3.16b, 0
  sqrdmlsh s3, s8, v9.s[1]
  movi v4.16b, 0
  sqrdmlsh s4, s8, v9.s[1]
  movi v5.16b, 0
  sqrdmlsh s5, s8, v9.s[1]
  movi v6.16b, 0
  sqrdmlsh s6, s8, v9.s[1]
  movi v7.16b, 0
  sqrdmlsh s7, s8, v9.s[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)030b1e3f51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200781500417392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
160204200641500018132580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
16020420064150005102580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
160204200641510675392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
16020420064150036392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
16020420064150042392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222032801600001002006520065200652006520065
1602042006415000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065
1602042006415000392580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000010112216222006101600001002006520065200652006520065

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2507

retire (01)cycle (02)03070a0b18191e1f3a3f51inst 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)9fa0a1a6a8a9acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600242007015022000003812580010108000010800005064000011200342005320053322800102080000202400002005320053111600211091010160000100000000100473112320211251820050015160000102005420054200542005420054
1600242005315033000003522580010108000010800005064000011200342005320053322800102080000202400002005320053111600211091010160000100000000100463111720221161720050015160000102005420054200542005420054
1600242005315033000003872580010108000010800005064000011200342005720053322800102080000202400002005720057111600211091010160000100000000100466211924221172020050030160000102005420058200542005820058
1600242005315032000003702580010108000010800005064000011200382005720057322800102080000202400002005720057111600211091010160000100000000100476212020422201920050030160000102005820058200582005820054
1600242005715133000003702580010108000010800005064000001200382005320057322800102080000202400002005720053111600211091010160000100002000100406121824421221720054030160000102005820058200582005420058
1600242005715033000003702580010108000010800005064000011200382005320057322800102080000202400002005720057111600211091010160000100000000100496211824412231920054030160000102005820058200582005820058
16002420057150330000031792580010108000010800005064000001200382005720057322800102080000202400002005720057111600211091010160000100000000100496211724422122020054030160000102005420054202122005420054
1600242005315133000003872580010108000010800005064000011200342005320053322800102080000202403122005320053111600211091010160000100000000100463111920211222020050015160000102005420054200542005420054
1600242005315133000003642580010108000010800005064000011200342005320053322800102080000202400002005320053111600211091010160000100000000100453111920211262020050015160000102005420054200542005420054
1600242005315033000303642580010108000010800005064000011200342005320053322800102080000202400002005320053111600211091010160000100000000100443111920211232120050015160000102005420054200542005420054

Test 6: throughput

Count: 12

Code:

  sqrdmlsh s0, s12, v13.s[1]
  sqrdmlsh s1, s12, v13.s[1]
  sqrdmlsh s2, s12, v13.s[1]
  sqrdmlsh s3, s12, v13.s[1]
  sqrdmlsh s4, s12, v13.s[1]
  sqrdmlsh s5, s12, v13.s[1]
  sqrdmlsh s6, s12, v13.s[1]
  sqrdmlsh s7, s12, v13.s[1]
  sqrdmlsh s8, s12, v13.s[1]
  sqrdmlsh s9, s12, v13.s[1]
  sqrdmlsh s10, s12, v13.s[1]
  sqrdmlsh s11, s12, v13.s[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)0307080a0b18191e1f373a3f4e51inst 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)9fa0a6a7a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
12020430052230000000630004102512015310012000010012000050099000013002030042300391497331499712010020012000020036000030042300391112020110099100100120000100000007610116113094001200001003004030043300403004030040
12020430039225000000005304102512010110012000010012000050099000013002030039300391497331500012010020012000020036000030039309431112020110099100100120000100000007610116113094001200001003004330040300403004330040
12020430943225000000001041025120117100120000100120000500428953913002030039300391497331499712010020012000020036000030039300391112020110099100100120000100000007610116113003601200001003004030040300413004030040
12020430039225000000001704102512015310012000010012000050099000013002030039300391497331499712010020012000020036000030039300391112020110099100100120000100000007610116113003601200001003004030043300403004030040
12020430039225000000000041025120101100120000100120000500428340013002031749300391497331499712010020012000020036000030039300391112020110099100100120000100000007610116113003701200001003004030040300403004330040
1202043094322500000000004434052512010010012000010012000050099000013002030039300391497331499712010020012000020036000030039300391112020110099100100120000100000007610116113003601200001003004030944300403004030040
120204300392250000006010410251201001001200001001200005009600001300203003930042149731014997120100202120741200363126312443133781120201100991001001200001002514474077803567630659211200001003123331428311573078831133
120204302052366101991206704234166046911971210691251209241241214825963145488130943315273121415649581579812178120212122820236501930610304571011202011009910010012000010020005068778071094731289211200001003118230769318573166631087
12020430144235001010101206792104102512011710012000010012000050096000013090330039300421497331499712010020012000020036000030042300391112020110099100100120000100000007610116113003601200001003004030040309443004030944
1202043003922400000000004402512010010012000110012000050096000013002130039300391497331499712010020012000020036000030042300391112020110099100100120000100000007610116113003601200001003004330040300403004130040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307080a0b18191e1f373a3f4e51inst 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)9fa8acc5cfd0d2d5d6d9ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1200243005022510100090024702512002710120000101200005096000000300203003930039149953150191200102012000020360000300393003911120021109101012000010000752402171601817300360120000103004030040300403004030040
120024300392251010004590026802512001010120000101200005096000000300203003930039149953150191200102012000020360000300393003911120021109101012000010001752400141601614300360120000103004030040300403004030040
12002430039225101000150024702512001010120000101200005096000000300203003930039149953150191200102012000020360000300393003911120021109101012000010000752400171601613300360120000103004030040300403004030040
1200243003922510100000024702512001010120000101200005096000000300203003930039166703150191200102012000020360000300393003911120021109101012000010000752402141601614300360120000103004030040300403004030040
120024300392251010004110024702512001010120000101200005096000005300203003930039149953150191200102012000020360000300393092211120021109101012000010000752450151601416300360120000103004030040300403004030040
12002430039225101000180024713275251200101012000010120000509600000030020300393003914995315019120010201200002036000030039300391112002110910101200001000075240014160169300360120000103004030040300403004030040
12002430039225101000240024702512001010120000101200005096000000300203003930039149943150191200102012000020360000300393003911120021109101012000010000752400171601717300360120000103004030040300403004030040
1200243003922510100018002470251200101012000010120000509600000030020300393003914995315019120010201200002036000030039300391112002110910101200001000075245018160915300360120000103004030040300403004030040
12002430039225101000180024702512001010120000101200005096000000300203003930039166703150191200102012000020360000300393003911120021109101012000010000752400161601716300360120000103004030040300403004030040
120024300392251010002430024702512001010120000101200005096000000300203003930039149953150191200102012000020360000300393003911120021109101012000010000752450151601417300360120000103004030040300403004030040