Apple M1 Microarchitecture Research by Dougall Johnson

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

SABA (vector, 8H)

Test 1: uops

Code:

  saba v0.8h, 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)030b1e3f4e51inst 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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
1004303722006125482510001000100039831313018303730372415328951000100030003037303711100110000073116112630100030383038303830383038
10043037230015625482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000300030373037111001100005773116112630100030383038303830383038
100430372301116125482510001000100039831313018303730372415328951000100030003037303711100110000073216222713100030383038303830383038
1004303723066125482510001000100039831313018303730372415328951000100030003037303711100110000073216212630100030383038303830383038
10043037220186125482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
100430372302676125482510001000100039831313018303730372415328951000100030003037303711100110000073216122630100030383038303830383038
100430372303426125482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038
1004303723006125482510001000100039831313018303730372415328951000100030003037303711100110000073216222630100030383038303830383038

Test 2: Latency 1->1

Code:

  saba v0.8h, 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)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)9fa8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013243329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000074113163329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001001071013163329634100001003003830038300383003830038
10204300372250161295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013163429634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071014163329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013163329634100001003003830038300383003830038
10204300372240061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013163329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013163329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020410000200300003003730037111020110099100100100001000071013163329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000071013163329634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030408090b18191e3a3f4e51inst 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)9fa6a7a8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000000017829548251001010100001010000504277313030018300373003728287328767100102010000203000030037300371110021109101010000100003064211169112963010000103003830038300383003830038
10024300372251000000178295482510010101000010100005042786701300183008430084282873287671001020100002030000300373003711100211091010100001000100642616892963010000103003830038300383003830038
10024300372251000000178295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000642111610102963010000103003830038300383003830038
100243003722510000092782954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110910101000010000006421016992963010000103003830038300383003830038
100243003722510000012178295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000200642716682963010000103003830038300383003830038
100243003722510000001206295482510010101000810100005042800270300183003730180282873287671016120100002230000300843003711100211091010100001000000642924892963010000103003830038300383003830038
10024300372241000000178295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000642916792963010000103003830038300383003830038
10024300372251000003178295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000642716672963010000103003830038300383003830038
100243003722510000001782954825100101010000101000050427731303001830037301312828732876710515201000020300003003730037111002110910101000010002306429161082963010000103003830038300383003830038
10024300372251000000178295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000000642816882963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  saba v0.8h, 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)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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372240000000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372240000000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000000061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000000061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372240000090061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000100071011611296340100001003003830038300383003830038
10204300372250000000061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
102043003722500000000726295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000071011611296340100001003003830038300383003830038
10204300372250000000061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000100071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e3f4e51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010009640216222963010000103003830038300383003830038
1002430037225061295482510019101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640116222963010000103003830038300383003830038
1002430037225961295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  saba v0.8h, 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)0307080918191e3a3f4e51inst 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)9fa0a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000071021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000071021622296340100001003003830038300383003830038
102043003722500010006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100016071021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000071021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071021622296340100001003003830038300383003830038
1020430037225000000010329548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100010173221624296340100001003003830038300383003830038
102043003722500000316129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100010071021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100046071021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030498300373003721102011009910010010000100000071021622296340100001003003830038300383003830038
102043003722500000552025129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000071021622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e3f4e51inst 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
1002430037225000006192954825100101010000101000050427731303001830037300372828726287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300842240000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830083
100243003722500000124295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038
10024300372250000061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  saba v0.8h, v8.8h, v9.8h
  movi v1.16b, 0
  saba v1.8h, v8.8h, v9.8h
  movi v2.16b, 0
  saba v2.8h, v8.8h, v9.8h
  movi v3.16b, 0
  saba v3.8h, v8.8h, v9.8h
  movi v4.16b, 0
  saba v4.8h, v8.8h, v9.8h
  movi v5.16b, 0
  saba v5.8h, v8.8h, v9.8h
  movi v6.16b, 0
  saba v6.8h, v8.8h, v9.8h
  movi v7.16b, 0
  saba v7.8h, 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)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)9fa1a8accfd5d6ddinst fetch restart (de)e0ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204200651510003925801001008000010080000500640000120045200642006432280100200800002002400002006420064111602011009910010016000010000010111116112006101600001002006520065200652006520065
16020420064150000176525801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000010111116112006101600001002006520065200652006520065
1602042006415000094825801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000010111116112006101600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001610111116112006101600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010021010111116112006101600001002006520065200652006520065
160204200641510003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000010111116112006101600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000010111116112006101600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000010111116112006101600001002006520065200652006520065
160204200641510003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010000310111116112006101600001002006520065200652006520065
160204200641500003925801001008000010080000500640000020045200642006432280100200800002002400002006420064111602011009910010016000010001010111116112006101600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0304070a1e3a3f51inst 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)9fa0a8acc5cfd0d2icache miss (d3)itlb miss (d4)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002420086150100013752580012128000012800006264000011520029200482004832280012208000020240000200482004811160021109101016000010000010055811141263223737200602330160000102005320053200532005320053
16002420052150110014042580012128000012800006264000010520033200482005232280012208000020240000200522005211160021109101016000010000010065621138263222937200492330160000102005320053200532005320133
160024202171521100030725800121280000128000062640000100200332005220052322800122080000202400002004820052111600211091010160000100000100621122137263222536200492480160000102005320053200532005320053
160024200481501110029225800121280000128000062640000010200332005220052322800122080000202400002005220052111600211091010160000100000100621122035263223636200492480160000102005320053200492005320053
16002420052150111013682580012128000012800006264000010520029200482004832280012208000020240000200482004811160021109101016000010000010060321137221112537200452480160000102004920049200492004920049
16002420048151111012572580012128000012800006264000010520029200482004832280012208000020240000200482004811160021109101016000010000010060321138221113939200452330160000102004920049200492004920049
16002420048150111013342580012128000012800006264000010520029200482004832280012208000020240000200482004811160021109101016000010006010046311137221133135200452330160000102004920049200492004920049
16002420048150110005122580012128000012800006264000010520029200482004832280012208000020240000200482004811160021109101016000010000010047321126221113838200452480160000102004920049200492004920049
16002420048150100013362580012128000012800006264000010520029200482004832280012208000020240000200482004811160021109101016000010000010061321130221113436200452330160000102004920049200492004920049
160024200481501100114002580012128000012800006264000010020029200482004832280012208000020240000200482004811160021109101016000010003010061811137221112237200452330160000102004920049200492004920049

Test 6: throughput

Count: 16

Code:

  saba v0.8h, v16.8h, v17.8h
  saba v1.8h, v16.8h, v17.8h
  saba v2.8h, v16.8h, v17.8h
  saba v3.8h, v16.8h, v17.8h
  saba v4.8h, v16.8h, v17.8h
  saba v5.8h, v16.8h, v17.8h
  saba v6.8h, v16.8h, v17.8h
  saba v7.8h, v16.8h, v17.8h
  saba v8.8h, v16.8h, v17.8h
  saba v9.8h, v16.8h, v17.8h
  saba v10.8h, v16.8h, v17.8h
  saba v11.8h, v16.8h, v17.8h
  saba v12.8h, v16.8h, v17.8h
  saba v13.8h, v16.8h, v17.8h
  saba v14.8h, v16.8h, v17.8h
  saba v15.8h, 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)0308090b181e373f51inst 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)9fa7a8a9acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160204400593000000004125160100100160000100160000500128000040020400484003919973320009160100200160000200480000400484003911160201100991001001600001000000010110216224004501600001004004940049400404004040040
160204400493000000004125160117100160000100160000500239899940020400394004919973319997160100200160000200480000400394004811160201100991001001600001000100010110216224003601600001004004040041400494004040040
1602044003929900000174225160117100160000100160000500128000040020400394003919973320006160100200160000200480000400484003911160201100991001001600001000000010110216224003601600001004004940040400404004940040
160204400483000000004125160100100160017100160000500128000040020400394004819973320006160100200160000200480000400394003911160201100991001001600001000000010110216224003701600001004004040041400494004940050
16020440039300000001750251601171001600001001600005001280000400304004040040199733199971601002001600002004800004004840039111602011009910010016000010000021010110124224003601600001004004040040400414004940040
1602044004829900000175125160117100160000100160000500128000040029400394003919973319997160100200160000200480000400394004811160201100991001001600001000003010110217224003601600001004005040040400504004040049
160204400392990000004225160100100160017100160000500128000040020400394003919973319997160100200160000200480000400394003911160201100991001001600001000000010110216224004501600001004004040049400494004040049
16020440039300000015161225160117100160017100160000500239899940020400394004819973319997160100200160000200480000400394004811160201100991001001600001000000010110216224004501600001004004040049400494004040040
1602044004830000000176225160117100160000100160000500128000040029400484004819973320006160100200160000200480000400394003911160201100991001001600001000000010110216224003601600001004004140040400414004040049
160204400393000000004125160100100160000100160000500128000040029400484004819973320007160100200160000200480000400394004811160201100991001001600001000000010110216224004601600001004004040040400404004940040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030b18191e1f373f51inst 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)5f6061696b6d6edispatch 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)9fa6a8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600244004930000000175525160010101600001016000050128000011540029040048400391999632001916001020160000204800004004840048111600211091010160000100001002213415164114440045209160000104004940049400494004940049
16002440048299000001755251600101016000010160000502398999111040020040048400521999632002816001020160000204800004003940052111600211091010160000100001002213614168213440036209160000104004940049400404004940040
160024400483000000017552516001010160017101600005023989991110400290400394004819996320028160010201600002048000040048400481116002110910101600001000010022136141641144400462018160000104004940040400494004940040
1600244003930000000055251600101016000010160000501280000111040029040039400481999632002916001020160000204800004004840048111600211091010160000100001002213613168114440045209160000104004040049400404004940049
16002440048300000001746251600101016000010160000502398999311040029040039400391999632001916001020160000204800004004840043111600211091010160000100001002213614168114440045209160000104004940040400494004040040
16002440048300000001746251600101016000010160000502398999111040020040039400481999632002816001020160000204800004003940043111600211091010160000100001002213614164114340036209160000104004040049400494004040040
16002440048300000001755251600101016000010160000501280000111040029040039400481999632001916001020160000204800004004840048111600211091010160000100001002213614164114440036209160000104004940040400494004940040
1600244003930000000055251600271016000010160000501280000111040029040039400481999632002816001020160000204800004004840043111600211091010160000100001002213614164113440045209160000104004940040400494004040040
16002440039300000001776251600271016001710160000501280000111040030040039400521999632002816001020160000204800004004840043111600211091010160000100001002213613168114440045209160000104004940049400494004940049
1600244003930000000046251600271016001710160000502398999111040020040048400481999632001916001020160000204800004003940045111600211091010160000100001002213614168114340045209160000104004940049400404004940040