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, 4H)

Test 1: uops

Code:

  saba 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)03091e3f4e51inst 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)a0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004303723006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303723006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303723006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303723006125482510001000100039831330223037303724153289510001000300030373037111001100000673116112630100030383038303830383038
1004303723006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303722006125482510001000100039831330183037303724153289510001000300030373037111001100000973116112630100030383038303830383038
1004303723006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
10043037230061254825100010001000398313301830373037241532895100010003000303730371110011000020073116112630100030383038303830383038
1004303722006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038
1004303723006125482510001000100039831330183037303724153289510001000300030373037111001100000073116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  saba 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)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)9fa0a1a8acc5cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500124295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071013162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071012163229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071012162329634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282423287451010020010000200300003003730037111020110099100100100001000000071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001002400071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071012162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071013162229634100001003003830038300383003830038
10204300372250061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000071012162229634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030a0b1e1f3f4e51inst 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
10024300372250018061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640416332963010000103003830038300383003830038
100243003722500297061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640316342963010000103003830038300383003830038
100243003722400264061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640316332963010000103003830038300383003830038
100243003722500426061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640316332963010000103003830038300383003830038
10024300372250024061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640316432963010000103003830038300383003830038
100243003722500285061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640416432971510000103003830038300383003830038
100243003722500282061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640316332963010000103003830038300383003830038
100243003722400429061295482510010101000010100005042773131300183003730037282873287671001020100002030000300373003711100211091010100001000640316332963010000103003830038300383003830038
10024300372250018061295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640416332963010000103003830038300383003830038
100243003722500278861295482510010101000010100005042773130300183003730037282873287671001020100002030000300373003711100211091010100001000640316332963010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  saba 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)030b1e1f3f4e51inst 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)9fa6accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722501506129548251010010010000100100005004277313130018300373003728265328745101002001000020030501300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250606129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372250303076295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000247101161129634100001003003830038300383003830038
1020430037225034206129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
102043003722401506129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225063526129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225020706129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225026106129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
1020430037225033006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100007101161129634100001003003830038300383003830038
10204300372240336010429548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100107101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030a0b191e1f3f4e51inst 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)91inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa6a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372240007207262954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
1002430037225000001452954864100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
1002430037224000360612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
100243003722511000612954825100101010000101000050427867003001830037300372828772876710010201000020304893003730037111002110901010100001030006402162229630010000103003830038300383003830038
100243003722400045013642953925100101010016101000050427731313001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
1002430037225000150612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
1002430037225000330612954825100101010000101000050427731303001830037300372828732876710010201016320300003003730037111002110901010100001000306402162229630010000103003830038300383003830038
10024300372250001807722954825100101010008101000050427731313001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
10024300372250003240612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110901010100001000006402162229630010000103003830038300383003830038
100243003722500000612954825100101010000101000050427731303001830037300372828732876710010201000020300003003730037111002110911010100001000006402162229630010000103003830038300383003830038

Test 4: Latency 1->3

Code:

  saba 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)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)9fa0a1a6a7a8a9acc2c5cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000018006129548251010010010000100100005004277313130018300373003728265328744101002001000020030000300373003711102011009910010010000100000000000710061722296340100001003003830038300383003830038
1020430037224000000061295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000007140616222963425100001003003830038300383003830038
10204300372250000300061295482510100100100001001000050042773131300183003730037282652228744101252001000020030492300843003711102011009910010010000100000000000710021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101252001000020030000300373003711102011009910010010000100000000000710021622296340100001003003830038300383003830038
10204300372250000180064295482510100100100001001000050042773131300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000007140617222963425100001003003830038300383003830038
1020430037225000012006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000001710021622296340100001003003830038300383003830038
1020430037224000045006129548251010010010000100100005004277313030018300373003728265328745101002001000020030000300373003711102011009910010010000100000000000710021622296340100001003003830038300383003830038
102043003722500000006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000000710021622296340100001003003830038300383003830038
1020430037225000015006129548251010010010000100100005004277313130018300373003728265328745101002001000020030000300373003711102011009910010010000100000000000710021622296340100001003003830038300383003830038
102043003722400003180061295482510100100100001001000050042773130300183003730037282653287451010020010000200300003003730037111020110099100100100001000000000007100216222963425100001003003830038300383003830038

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)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)9fa8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722501562954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372240612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372240612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722504412954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722407262954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010300640216222963010000103003830038300383003830038
100243003722527612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731313001830037300372828732876710010201000020300003003730037111002110910101000010000640216222963010000103003830038300383003830038

Test 5: throughput

Count: 8

Code:

  movi v0.16b, 0
  saba v0.4h, v8.4h, v9.4h
  movi v1.16b, 0
  saba v1.4h, v8.4h, v9.4h
  movi v2.16b, 0
  saba v2.4h, v8.4h, v9.4h
  movi v3.16b, 0
  saba v3.4h, v8.4h, v9.4h
  movi v4.16b, 0
  saba v4.4h, v8.4h, v9.4h
  movi v5.16b, 0
  saba v5.4h, v8.4h, v9.4h
  movi v6.16b, 0
  saba v6.4h, v8.4h, v9.4h
  movi v7.16b, 0
  saba 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)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)696b6d6escheduler 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)9fc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602042009115003925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
16020420064150393925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
16020420064150273925801001008000010080000500640000200453200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
1602042006415003925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002013420065200652006520065
1602042006415003925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
1602042006415103925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065
16020420064150273925801001008000010080000500640000200450200642006432280100200800002002400002006420064111602011009910010016000010001011111611200611600001002006520065200652006520065

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03090b181e1f3f51inst 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)9fa8acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0? int output thing (e9)eaec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024200781500002104527800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100001003183172521158200482201160000102005220052200522005220052
16002420051150000004527800121280000128000062640000115200322005120051322800122080000202400002005120051111600211091010160000100001003283162521164200482201160000102005220052200522005220052
160024200511500000045278001212800001280000626400001152003220051200513228001220800002024000020051200511116002110910101600001000010031135152521164200482201160000102005220052200522005220052
16002420051150000120452780012128000012800006264000011102003220051200513228001220800002024000020051200511116002110910101600001000010027135152521185200482201160000102005220052200522005220052
16002420051151000006627800121280000128000062640000111020032200512005132280012208000020240000200512005111160021109101016000010000100271351412321146200482201160000102005220052200522005220052
16002420051150000180452780012128000012800006264000011102003220051200513228001220800002024000020051200511116002110910101600001000010027135182521146200482201160000102005220052200522005220052
1600242005115000000452780012128000012800006264000011102003220051200513228001220800002024000020051200511116002110910101600001000010031136152521164200482201160000102005220052200522005220052
1600242005115100000452780012128000012800006264000011102003220051200513228001220800002024000020051200511116002110910101600001000010027135142521146200482201160000102005220052200522005220052
1600242005115000000452980012128000012800006264000011102004120051200603228001220800002024000020051200601116002110910101600001000010031164183422168200482201160000102005220052200522005220052
16002420051150000480452780012128000012800006264000011102003220051200603228001220800002024000020051200511116002110910101600001000010031134162521188200482201160000102005220052200522005220052

Test 6: throughput

Count: 16

Code:

  saba v0.4h, v16.4h, v17.4h
  saba v1.4h, v16.4h, v17.4h
  saba v2.4h, v16.4h, v17.4h
  saba v3.4h, v16.4h, v17.4h
  saba v4.4h, v16.4h, v17.4h
  saba v5.4h, v16.4h, v17.4h
  saba v6.4h, v16.4h, v17.4h
  saba v7.4h, v16.4h, v17.4h
  saba v8.4h, v16.4h, v17.4h
  saba v9.4h, v16.4h, v17.4h
  saba v10.4h, v16.4h, v17.4h
  saba v11.4h, v16.4h, v17.4h
  saba v12.4h, v16.4h, v17.4h
  saba v13.4h, v16.4h, v17.4h
  saba v14.4h, v16.4h, v17.4h
  saba 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)0308090b1e1f373f4e51inst 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)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)9fa8acc2c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020440049300000000716025160100100160000100160000500239902704002040039400491997303199971601002001600002004800004003940039111602011009910010016000010000201011011611400361600001004004140040400404004140041
16020440039300000000410251601001001600001001600005001280000040020400484003919973025199971601002001600002004800004003940048111602011009910010016000010000001011011611400681600001004007240049400724004940072
1602044004930000000050025160100100160061100160000500128000004002040071400481997303200061601002001600002004800004003940048111602011009910010016000010000001011011611400361600001004004940072400404007240049
1602044003929900000041025160101100160017100160000500239908204002040039400491997303199981601002001600002004800004003940040111602011009910010016000010000001011011611400361600001004004040072400404004040040
1602044003930000000051025160100100160000100160000500128000004002040040400391997303200071601002001601432004800004003940049111602011009910010016000010000001011011611400361600001004005040040400404005040050
16020440071300000001851025160101100160001100160000500128000014002140049400391997303199981601002001600002004800004004840039111602011009910010016000010000001011011611400361600001004004940072400494004040040
1602044004930000000041025160117100160017100160000500239899914002940039400491998003200071601002001600002004800004003940039111602011009910010016000010000001011011611400361600001004005040040400404005040040
16020440049300000000502525160100100160000100160000500128000004002940048400391997303199971601002001600002004800004004040049111602011009910010016000010000001011011611400371600001004005040040400404005040050
1602044003930000000141025160100100160001100160000500128000004002040049400391997303199971601002001600002004800004007140048111602011009910010016000010000001011011611400681600001004005040040400494004040040
160204400392990000017430025160100100160017100160000500132000004002940049400491997303200071601002001600002004800004004940049111602011009910010016000010000001011011611400361600001004004040040400494004040050

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03080a1e373f51inst 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)5f6061696d6edispatch 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)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002440040300000174625160010101600001016000050239899910040029400394003919996320019160010201600002048000040039400391116002110910101600001000100223112162114740045206160000104004040040400404004040040
16002440039300000174625160010101600001016000050239899911540020400484003919996320019160010201600002048000040039400391116002110910101600001000100223416382114440045206160000104004040040400404004040040
1600244003930000004625160010101600171016000050128000011540020400394003919996320019160010201600002048000040089400481116002110910101600001000100228313162114740036206160000104004940040400494004040040
1600244003930000005525160010101600171016000050239899911540029400484003919996320019160010201600002048000040039400391116002110910101600001000100228326162116640036206160000104004040040400404004040040
16002440039300000174625160010101600171016000050128000011540020400394003919996320019160010201600002048000040039400391116002110910101600001020100228313162113540036206160000104004040040400404004040040
16002440039300000174625160010101600001016000050128000011540029400394003919996320019160010201600002048000040039400391116002110910101600001000100228313162114740045206160000104004940040400494004040040
16002440048300000174625160010101600001016000050128000011540020400394003919996320019160010201600002048000040039400391116002110910101600001000100228313162114440036206160000104004040040400404004040049
1600244003930000004625160010101600001016000050128000011540029400394004819996320019160010201600002048000040048400391116002110910101600001000100223314162116640045206160000104004040040400404004040040
1600244003929900004625160010101600001016000050239899910540029400394003919996320019160010201600002048000040039400391116002110910101600001000100228317162114440045206160000104004040040400404004040040
1600244003930000004625160027101600001016000050128000011540020400394003919996320028160010201600002048000040039400391116002110910101600001000100228314162115740036206160000104004040040400404004040049