Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMLAL2 (vector, 4S)

Test 1: uops

Code:

  sqdmlal2 v0.4s, v1.8h, v2.8h
  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)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)accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372301562548251000100010003983131301830373037241532895100010003000303730371110011000073216112630100030383038303830383038
10043037220612548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
10043037230612548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
10043037230612548251000100010003983130301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
10043037230612548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
100430372201052548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
10043037230822548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
10043037230822548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
100430372312612548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038
10043037230822548251000100010003983131301830373037241532895100010003000303730371110011000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  sqdmlal2 v0.4s, v1.8h, v2.8h
  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)030b18191e1f3a3f4e51inst 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)9fa6a8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204304062270763996160395929492180101781221004811211192648428813113009030417304692829543289121146922011327226339783046630417101102011009910010010000100202223371002162229634100001003003830038300383003830038
1020430037224000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162329634100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162229634100001003003830038300383003830038
1020430037225100000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012163229634100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162229634100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162229634100001003003830038300383003830038
10204300372250000007472954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162329634100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162229634100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010001071012162229634100001003003830038300383003830038
10204300372240000007262954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0318191e3f4e51inst 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)9fa1a6a8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216232963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313006503003730037282873287671001020100002030000300373003711100211091010100001000000640216222968810000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
1002430037224000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
1002430037224000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038
10024300372250002512954825100101010000101000050427731313001803003730037282873287671001020100002030000300373003711100211091010100001000000640216222963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  sqdmlal2 v0.4s, v0.8h, v1.8h
  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)030b18191e3f4e51inst 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372330000612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500002322954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722400004412954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000073211611296340100001003003830038300383003830038
1020430085226000558612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100201071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100010071011611296340100001003003830038300383003830038
10204301322250000612954825101001001000010010000500427731330018300373003728265328745101002001000020030498300373003711102011009910010010000100000071012511296340100001003003830038301333003830038
102043003722400001032954825101001001000010010000500427731330018300373003728265328745102642001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313300183003730037282653287451010020010000200300003003730037111020110099100100100001000000710125212963420100001003003830038300863003830038
1020430037225100018512953925101201041000011010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731330018300373003728265328745101002001000020030000300373003711102011009910010010000100000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03181e1f3f4e51inst 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)9fa0a1a7a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000206402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402172229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722500061295394410019101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722502413261295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
1002430037225018061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
1002430037225000536295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722500061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722510061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  sqdmlal2 v0.4s, v1.8h, v0.8h
  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)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)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)9fa0a1a6a7a8a9acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000000822954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000000071021611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000003000071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731313001830037300372826532874510100200100002003000030037300371110201100991001001000010000001000071011711296340100001003003830038300383003830038
10204300372240000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000001000071011611296340100001003003830038300383003830038
1020430037225000000056712954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000001000071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000001000071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000001000071011611296340100001003003830038300383003830038
102043003722501000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000023000071011611296340100001003003830038300383003830038
10204300372250000000822954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000000030071011611296340100001003003830038300383003830038
10204300372250000000612954825101001001000010010000500427731303001830037300372826532874510100200100002003000030037300371110201100991001001000010000001000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b18191e3a3f4e51inst 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)9fa6a7a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000726295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000590640316332963010000103003830038300383003830038
10024300372240000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000100640316332963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000640316332963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000640316332963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000100640316332963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001001130640316332963010000103003830038300383003830038
10024300372240000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000100640316332963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000640316332963010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010006830640316332963010000103003830038300383003830038
100243003722500000536295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000000640316332963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  sqdmlal2 v0.4s, v8.8h, v9.8h
  movi v1.16b, 0
  sqdmlal2 v1.4s, v8.8h, v9.8h
  movi v2.16b, 0
  sqdmlal2 v2.4s, v8.8h, v9.8h
  movi v3.16b, 0
  sqdmlal2 v3.4s, v8.8h, v9.8h
  movi v4.16b, 0
  sqdmlal2 v4.4s, v8.8h, v9.8h
  movi v5.16b, 0
  sqdmlal2 v5.4s, v8.8h, v9.8h
  movi v6.16b, 0
  sqdmlal2 v6.4s, v8.8h, v9.8h
  movi v7.16b, 0
  sqdmlal2 v7.4s, v8.8h, v9.8h
  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)03070a191e3a3f51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602042009015011003462580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001012412161311200611600001002006520065200652006520065
1602042006415000004582580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001011912161113200611600001002006520065200652006520065
16020420064150000031412580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000001012614161412200611600001002006520065200652006520065
1602042006415022004582580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001012714161315200611600001002006520065200652006520065
1602042006415001004582580100100800001008000050064000002004520064200643228010020080000200240000200642006411160201100991001001600001000101012712171512200611600001002006520065200652006520065
1602042006415010004582580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001012515161214200611600001002006520065200652006520065
1602042006415010005462580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001012714161414200611600001002006520065200652006520133
16020420064150110045225801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010002431012210161513200611600001002006520065200652006520065
1602042006415000004582580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000031012415161314200611600001002006520065200652006520065
1602042006415011004582580100100800001008000050064000012004520064200643228010020080000200240000200642006411160201100991001001600001000001012713161215200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)031e3f51inst 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)9fa8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420085150045278001212800001280000626400001102003220051200513228001220800002024000020051200511116002110910101600001000100348311225211116200482201160000102005220052200522006120052
16002420051150045298001212800001280000626400000152004120051200603228001220800002024000020060200601116002110910101600001000100311152534222611200572202160000102005220052200522005220052
160024200511500512980012128000012800006264000001520041200602006032280012208000020240000200602005111160021109101016000010101002884110252111111200482201160000102006120052200612006120061
16002420060150051278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001010100348411025211129200482201160000102005220052200522006120061
1600242005115005129800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000101010033114211342111010200482402160000102005220052200522005220052
16002420051151045298001212800001280000626400000152004120060200513228001220800002024000020051200511116002110910101600001000100328421034211116200572402160000102005220052200522005220052
160024200511500452980012128000012800006264000001520032200512005132280012208000020240000200602006011160021109101016000010001002811511034221107200482402160000102005220052200522005220052
16002420051150045298001212800001280000626400001152004120060200513228001220800002024000020051200511116002110910101600001000100281151534411126200482201160000102005220061200522006120052
16002420060150045278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001015710033851625211611200482201160000102005220052200612005220052
16002420060150051278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000100338511025211105200482201160000102005220052200522006120052

Test 6: throughput

Count: 16

Code:

  sqdmlal2 v0.4s, v16.8h, v17.8h
  sqdmlal2 v1.4s, v16.8h, v17.8h
  sqdmlal2 v2.4s, v16.8h, v17.8h
  sqdmlal2 v3.4s, v16.8h, v17.8h
  sqdmlal2 v4.4s, v16.8h, v17.8h
  sqdmlal2 v5.4s, v16.8h, v17.8h
  sqdmlal2 v6.4s, v16.8h, v17.8h
  sqdmlal2 v7.4s, v16.8h, v17.8h
  sqdmlal2 v8.4s, v16.8h, v17.8h
  sqdmlal2 v9.4s, v16.8h, v17.8h
  sqdmlal2 v10.4s, v16.8h, v17.8h
  sqdmlal2 v11.4s, v16.8h, v17.8h
  sqdmlal2 v12.4s, v16.8h, v17.8h
  sqdmlal2 v13.4s, v16.8h, v17.8h
  sqdmlal2 v14.4s, v16.8h, v17.8h
  sqdmlal2 v15.4s, v16.8h, v17.8h
  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)031e373f51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602044008230001710202516010010016001710016000050023989991400204004840039199733199971601002001600002004800004004840039111602011009910010016000010021010110116114004501600001004004040049400404004940040
16020440048300007062516010010016000010016000050023989991400294003940048199733199971601002001600002004800004003940048111602011009910010016000010000010110116114004501600001004004040049400404004940049
16020440048300017412516011710016000010016000050012800001400204003940039199733200061601002001600002004800004003940048111602021009910010016000010000010110116114003601600001004004040049400404004940049
16020440048300017502516010010016000010016000050023989991400294003940048199733199971601002001600002004800004003940039111602011009910010016000010000010110116114003601600001004004940040400494004040049
16020440048300017502516011710016001710016000050012800001400294004840039199733199971601002001600002004800004003940048111602011009910010016000010000010110116114003601600001004004040049400404004940040
1602044003930000412516010010016001710016000050012800001400204003940048199733200061601002001600002004800004004840039111602011009910010016000010002010110116114003601600001004004040049400404005040049
1602044004830000502516011710016001710016000050012800001400204003940048199733200061601002001600002004800004004840048111602011009910010016000010000010110116114004501600001004004940040400494004040049
16020440048300017502516010010016000010016000050023989991400204003940039199733200061601002001600002004800004004840039111602011009910010016000010000010110116114004501600001004004940040400494004040049
1602044003930000512516010010016000010016000050012800001400204004840039199733199971601002001600002004800004004840039111602011009910010016000010000010110116114004501600001004004940040400494004040049
16020440048300017412516010010016000010016000050023989991400204004840039199733200061601002001600002004800004003940048111602011009910010016000010001010110116114003601600001004004040040400494004040049

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03080a0b18191e1f373f51inst 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)9fa1a6a8acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024400393000000000174625160027101600001016000050239899911400294003940052199963200191600102016000020480000400494004811160021109101016000010001001002231161621144400360155160000104004940040400494005040049
16002440048300000000014725160027101600011016000050239908211400204004040052199963200191600102016000020480000400394004811160021109101016000010001001002231131621144400450155160000104004140040400404004040050
16002440052300000000005925160011101600001016000050239899911400214003940039199963200281600102016000020480000400494004811160021109101016000010000001002231141621144400490155160000104004140040400404005040049
16002440040300000000015625160011101600001016000050131999711400204004940048199963200191600102016000020480000400524004911160021109101016000010000001002231131621144400490155160000104005340053400404004940050
16002440040300000000014625160011101600001016000050251912901400294003940048199963200281600102016000020480000400484003911160021109101016000010002001002231141621134400360155160000104004040040400494004040040
16002440039300000000005525160010101600001016000050255896211400214003940039199963200291600102016000020480000400394003911160021109101016000010000001002431141621143400360155160000104004140041400414004040040
16002540039300000000014625160010101600171016000050128000011400204005240039199963200281600102016000020480000400394004811160021109101016000010000001002262131642144400450155160000104005340049400414004040040
160024400492990000000184625160027101600171016000050128000001400304004940048199963200281600102016000020480000400484004811160021109101016000010000001002432241621143400360155160000104004040040400494004940040
160024400403000000000175925160027101600001016000050239908211400204004940039199963200321600102016000020480000400394004011160021109101016000010000001002431131621234400460305160000104004040049400504004940041
16002440040300000000005525160027101600171016000050128000011400204003940052199963200191600102016000020480000400404003911160021109101016000010000001002231141621143400450155160000104004140053400414004040040