Apple M1 Microarchitecture Research by Dougall Johnson

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

UABA (vector, 4H)

Test 1: uops

Code:

  uaba v0.4h, 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)031e3f4e51inst 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)a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372206125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
100430372306125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303722061254825100010001000398313301830373037241532895100010003000303730371110011000100273116112630100030383038303830383038
100430372308225482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
100430372306125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303722057125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
100430372206125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303722061254825100010001000398313301830373037241532895100010003000303730371110011000033073116112630100030383038303830383038
100430372206125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
100430372306125482510001000100039831330183037303724153289510001000300030373037111001100003073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  uaba v0.4h, 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): 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)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)9fa0a1a6a7a8a9acc2cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000000822954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003013230038300383003830038
102043003722500000000612954825101001001000010010000500427867003001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121623296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000002000710121622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000300612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121623296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000710121622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b181e3f4e51inst 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)9fa7a8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100090640316222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018300373003728287328767100102010000203051630037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640316222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830085
10024300372250006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640217222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  uaba v0.4h, 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): 3.0037

retire (01)cycle (02)03080b1e3f4e51inst 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
10204302292261108229548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300863003830038
1020430037225000215929548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
102043003722500014729548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830227300383003830086
102043003722400186129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0307080a3a3f4e51inst 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)9fa9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037224101273329548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100064481611102963010000103003830038300383003830038
1002430037225101268295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001020644101615102963010000103003830038300383003830038
1002430037225101268295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000644111610102963010000103003830038300383003830038
1002430037225101289295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000644101611112963010000103003830038300383003830038
100243003722510126829548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100064411165102963010000103003830038300383003830038
100243003722510126829548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100064410161052963010000103003830038300383003830038
100243003722510126829548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100064410161092963010000103003830038300383003830038
1002430037225101213129548251001010100001010000504277313130018300373003728287328767100102010000203000030037300371110021109101010000100064410165102963010000103003830038300383003830038
1002430037225101268295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000644111610102963010000103003830038300383003830038
1002430037225101268295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000644101610102963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  uaba v0.4h, 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): 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)696b6d6edispatch 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)9fa0a1a6a8a9acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071021611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100006234277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038
1020430037225000000006129548251010010010000100100005004277313300180300373003728265328745101002001000020030000300373003711102011009910010010000100000000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b1e3f4e51inst 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)9fa0a1a7a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372240061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640416222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10025300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10024300372250082295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10024300372240061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640316222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  uaba v0.4h, v8.4h, v9.4h
  movi v1.16b, 0
  uaba v1.4h, v8.4h, v9.4h
  movi v2.16b, 0
  uaba v2.4h, v8.4h, v9.4h
  movi v3.16b, 0
  uaba v3.4h, v8.4h, v9.4h
  movi v4.16b, 0
  uaba v4.4h, v8.4h, v9.4h
  movi v5.16b, 0
  uaba v5.4h, v8.4h, v9.4h
  movi v6.16b, 0
  uaba v6.4h, v8.4h, v9.4h
  movi v7.16b, 0
  uaba v7.4h, 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)03081e1f3f51inst 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
160204200651500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011121611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
16020420064150047403925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641510003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
160204200641500003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010001011131611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03040718191e1f3a3f51inst 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)5f6061696d6escheduler 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)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420087150100000375278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010053842128252122427200482201160000102005220052200522005220052
160024200511501000003455278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010049712125252112323200482201160000102005220052200522005220052
16002420051150000000375278001212800001280000626409161152003220051200513228001220800002024000020051200511116002110910101600001000010046682126252111922200482201160000102005220052200522005220052
16002420051150100000375278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010052712123252112521200482201160000102005220052200522005220052
16002420051150100000375278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010046772121252112323200482201160000102005220052200522005220293
160024200511511000003740278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000310050742123252112023200482201160000102005220052200522005220052
16002420051150100000375278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010051712124252112323200482201160000102005220052200522005220052
16002420051151101000375278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010046742124252111426200482201160000102005220052200522005220052
16002420051150100000375278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010049742124252112922200482201160000102005220052200522005220052
16002420051150100015903170278001212800001280000626400001152003220051202133228001220800002024000020051200511116002110910101600001000010051722123252112621200482201160000102005220052200522005220052

Test 6: throughput

Count: 16

Code:

  uaba v0.4h, v16.4h, v17.4h
  uaba v1.4h, v16.4h, v17.4h
  uaba v2.4h, v16.4h, v17.4h
  uaba v3.4h, v16.4h, v17.4h
  uaba v4.4h, v16.4h, v17.4h
  uaba v5.4h, v16.4h, v17.4h
  uaba v6.4h, v16.4h, v17.4h
  uaba v7.4h, v16.4h, v17.4h
  uaba v8.4h, v16.4h, v17.4h
  uaba v9.4h, v16.4h, v17.4h
  uaba v10.4h, v16.4h, v17.4h
  uaba v11.4h, v16.4h, v17.4h
  uaba v12.4h, v16.4h, v17.4h
  uaba v13.4h, v16.4h, v17.4h
  uaba v14.4h, v16.4h, v17.4h
  uaba v15.4h, v16.4h, v17.4h
  movi v16.16b, 17
  movi v17.16b, 18

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)030407080a0b18191e1f373a3f51inst 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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602044004830001010000017249251601171001600171001600005001280000400294003940048199730320006160100200160000200480000400484003911160201100991001001600001000001011414161414400451600001004004040049400404004940040
1602044003930001010000002166251601001001600001001600005001280000400204003940048199730320006160100200160000200480000400394004811160201100991001001600001000001011412161412400361600001004004040049400404004940041
1602044004830001010001200269251601171001600011001600005002398999400204003940048199730320007160100200160000200480000400394004811160201100991001001600001000001011413161314400361600001004004940040400414004040049
1602044004930001010000017257251601171001600171001600005002398999400294004840048199730320006160100200160000200480000400484003911160201100991001001600001000001011414161314400361600001004004040041400404005040040
1602044003930001010000012582516011710016006510016000050023989994003040039400391997303200061601002001600002004800004004840048111602011009910010016000010000210114616816400451600001004004040049400404004040049
1602044004830001010000002523251601171001600171001600005001319998400294003940039199730319997160100200160000200480000400484003911160201100991001001600001000001011413161214400451600001004004040049400404004040040
1602044003930001010002100248251601011001600001001600005001280000400204003940039199730319997160100200160000200480000400394003911160201100991001001600001000001011414161414400361600001004004040040400494004040040
1602044003930001010000017248251601171001600171001600005002399055400204004840039199730319997160100200160000200480000400484003911160201100991001001600001000001011412161414400361600001004005040040400404004040050
1602044003930001010000017248251601171001600171001600005001280000400204003940048199730320006160100200160000200480000400394004811160201100991001001600001000001011414161415400361600001004004040102400494004040040
16020440039300010100000172922251601171001600001001600005001280000400204004840039199730319997160100200160000200480000400394004811160201100991001001600001000001011412161412400361600001004004940040400494004040049

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0308091e1f373f4e51inst 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)9fa8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400493000000053025160010101600001016000050128000011400204003940039199963200191600102016000020480000400394003911160021109101016000010001002231161621188400450206160000104004040049400404004040110
160024400392990000052025160027101600011016000050128000011400204004040039199963200191600102016000020480000400394004911160021109101016000010001002231141641266400360206160000104004040040400404007240040
160024400393000090046025160027101600001016000050239908211400204003940048199963200191600102016000020480000400394005111160021109101016000010001002231151621147400360206160000104004040040400404007240040
1600244003930000000460251600711016000010160000501280000114002040039400391999632001916001020160000204800004003940039111600211091010160000100310022311716211454003602018160000104004940040400404004040040
1600244003930000001755025160010101600001016000050239899911400204003940071199963200281600102016000020480000400394003911160021109101016000010001002231271621153400360206160000104004040040400404004040040
160024400393000000046025160010101600001016000050128000011400294004840039199963200191600102016000020480000400394003911160021109101016000010001002231141622174400360206160000104007240040400724004040040
1600244007129900000460251600101016000110160000501280000014005240039400391999632002016001020160000204800004003940039111600211091010160000100010024612516422774003604012160000104004040040400724004040040
1600244003930000000460251600101016000010160000501280000014002040039400391999632002016001020160000204800004003940071111600211091010160000100010022622516411534004502014160000104004140050400404004040040
160024400483000000046025160010101600001016000050128000011400204003940039199963200511600102016000020480000400394003911160021109101016000010001002232251642255400460209160000104004040041400404007240040
160024400393000000046025160010101600001016000050128000011400204003940039199963200191600102016000020480000400394003911160021109101016000010001002461151622184400680206160000104004040040400404004040040