Apple M1 Microarchitecture Research by Dougall Johnson

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

URSRA (vector, D)

Test 1: uops

Code:

  ursra d0, d1, #3
  movi v0.16b, 1
  movi v1.16b, 2

(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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831303018303730372415328951000100020003084307311100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100020003037303711100110003073116112630100030383038303830383038
100430372306125484410001000100039831313018303730372415328951000100020003037303711100110001373116112630100030383038303830383038
100430372206125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831313018303730372415328951000100020003037303711100110000373116112630100030383038303830383038

Test 2: Latency 1->1

Code:

  ursra d0, d1, #3
  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)031e1f3a3f4e51inst 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)9fa8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500012222954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037225000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250002352954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372240007942954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830083
10204300372250001872954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037224000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372240009432954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372240002752954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000971011611296340100001003003830038300383003830038
102043003722500013122954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250001072954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0318191e1f3f4e51inst 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)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)9fa0a1a6a7a8acc2cdcficache miss (d3)d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000612954825100101010000101000050427731313001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
100243003722500005572954825100101010000101000050427731303001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
100243003722500001682954825100101010000101000050427731303001803003730080282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
100243003722500001242954825100101010000101000050427731313001803003730037282870328767100102010000202000030037300371110021109101010000100000100064002162229630010000103003830038300383008530038
10024300372250000842954825100101010000101000050427731303001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
10024300372250000842954825100101010000101000050427731303001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
10024300372250000822954825100101010000101000050427731303001803003730037282870328767100102010000202000030037300371110021109101010000100000000164002162229630010000103003830038300383003830038
10024300372250000612954825100101010000101000050427731303001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
10024300372250000822954825100101010000101000050427731313001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038
10024300372250000822954825100101010000101000050427731313001803003730037282870328767100102010000202000030037300371110021109101010000100000000064002162229630010000103003830038300383003830038

Test 3: Latency 1->2

Code:

  ursra d0, d0, #3
  movi v0.16b, 1

(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)9fa0a1a6a7a8a9acb5c2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000000612954725101001001000010010000500427716003001830037300372827172874010100200100082002001630037300371110201100991001001000010000000000011171801600296450100001003003830038300383003830038
1020430037225000000006129547251010010010000100100005004277160030018300373003728271122875910100200100082002001630037300371110201100991001001000010000000000011171801600296460100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716003001830037300372827172874010100200100082002001630037300371110201100991001001000010000000000011171801600296450100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716013001830037300372827172874110100200100082002001630037300371110201100991001001000010000000000011171701600296450100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716003001830037300372827172874010100200100082002001630037300371110201100991001001000010000000000011171701600296460100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716013001830037300372827172874010100200100082002001630037300371110201100991001001000010000000000011171801600296450100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716003001830037300372827172874010100200100082002001630037300371110201100991001001000010000000000011171701602296460100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716003001830037300372827162874010100200100082002001630037300371110201100991001001000010000000000011171801600296450100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716003001830037300372827172874010100200100082002001630037300371110201100991001001000010000001060011171801600296460100001003003830038300383003830038
102043003722500000000612954725101001001000010010000500427716013001830037300372827172874110100204100082002001630037300371110201100991001001000010000000000011171801600296460100001003003830038300383003830038

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)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)9fa6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250612954725100181010000101000050427716030018300373003728291032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372250612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372250612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372250612954725100101010000101014850427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372240612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372250612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000642216222962910000103003830038300383003830038
10024300372250612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372250612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216222962910000103003830038300383003830038
10024300372250612954725100191010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216422962910000103003830038300383003830038
10024300372240612954725100101010000101000050427716030018300373003728286032876710010201000020200003003730037111002110910101000010000640216342962910000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  movi v0.16b, 0
  ursra d0, d8, #3
  movi v1.16b, 0
  ursra d1, d8, #3
  movi v2.16b, 0
  ursra d2, d8, #3
  movi v3.16b, 0
  ursra d3, d8, #3
  movi v4.16b, 0
  ursra d4, d8, #3
  movi v5.16b, 0
  ursra d5, d8, #3
  movi v6.16b, 0
  ursra d6, d8, #3
  movi v7.16b, 0
  ursra d7, d8, #3
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2508

retire (01)cycle (02)0309191e1f3f51inst 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)9fa0a8a9acc5branch mispredict (cb)cdcfd6e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020420065151003605025801161008001610080028500640196120045200652006561280128200800282001600562006520065111602011009910010016000010009401111011916200621600001002006620066200662006620066
1602042006515110005225801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
1602042006515000002925801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000031111011916200621600001002006620066200662006620066
160204200651500030602925801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
1602042006515000002925801161008001610080028500640196120045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
16020420065150004205025801161008001610080028500640196120045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
1602042006515000002925801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
1602042006515000002925801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
1602042006515000002925801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066
1602042006515000002925801161008001610080028500640196020045200652006561280128200800282001600562006520065111602011009910010016000010000001111011916200621600001002006620066200662006620066

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03041e3a3f51inst 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)9faccfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
160024200511501003892580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010040841192842118202004315160000102004720047200472004720047
16002420050150000452580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010039841192022118192004315160000102004720051200472004720047
16002420046150000452580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010042841202022116172004315160000102004720047200472004720047
16002420046150000512580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010043841182422213172004330160000102004720047200472004720047
160024200461500004525800101080000108000050640000115200272004620046322800102080000201600002004620046111600211091010160000100100391141172021117172004315160000102004720047200472004720047
16002420046150000452580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010042841192421118212004315160000102004720047200472004720047
16002420046150001512580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010042841172041119142004315160000102004720047200472005120051
16002420046150000512580010108000010800005064000011520027200462004632280010208000020160000200462004611160021109101016000010010043841122021114192004315160000102004720047200472004720047
16002420046150000452580010108000010800005064000011520027200502004632280010208000020160000200462004611160021109101016000010010044842192442120192004330160000102005120047200512004720047
160024200501500014525800101080000108000050640000015200312005020046322800102080000201600002004620050111600211091010160000100100401151192422119172004330160000102004720051200472005120051

Test 5: throughput

Count: 16

Code:

  ursra d0, d16, #3
  ursra d1, d16, #3
  ursra d2, d16, #3
  ursra d3, d16, #3
  ursra d4, d16, #3
  ursra d5, d16, #3
  ursra d6, d16, #3
  ursra d7, d16, #3
  ursra d8, d16, #3
  ursra d9, d16, #3
  ursra d10, d16, #3
  ursra d11, d16, #3
  ursra d12, d16, #3
  ursra d13, d16, #3
  ursra d14, d16, #3
  ursra d15, d16, #3
  movi v16.16b, 17

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)0318191e3f51inst 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)9fa0a6a8a9acc2c5branch mispredict (cb)cdcfd2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1602044004930000213025160108100160112102160020500128013210400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000106011110118001600400361600001004004040040400404004040040
1602044003930000030251601081001600081001600205001280132004002340039400391997711199901601202001600322003200644003940039111602011009910010016000010000100011110118001600400361600001004004040040400404004040040
160204400393000003025160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001600400361600001004004040040400404004040040
160204400393000003025160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001700400361600001004004040040400404004040040
16020440039300000127425160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001600400361600001004004040040400404004040040
1602044003930004243025160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001600400361600001004004040040400404004040040
160204400393000003025160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001600400361600001004004040040400404004040040
160204400393000003025160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010002000011110118001600400361600001004004040040400404004040040
160204400393000003025160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001600400361600001004004040040400404004040040
1602044003930000073725160108100160008100160020500128013200400204003940039199776199901601202001600322003200644003940039111602011009910010016000010000000011110118001600400361600001004004040040400404004040040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030818191e1f3f51inst 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)9fa0a1a7a8a9acc2cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600244003929900000462516001010160000101600005012800001154002040039400391999632001916001020160000203200004003940039111600211091010160000100000000100758415162119740036206160000104004040040400404004040040
1600244003930000000462516001010160000101600005012800001154002040039400391999632001916001020160000203200004003940039111600211091010160000100001018760100228419162117740036206160000104025140040400404004040040
1600244003930000039071125160010101600001016000050128000001540020400394003919996320019160010201600002032000040039400391116002110910101600001000000001002284291642275400362012160000104004040040400404004040040
16002440039300000121764625160010101600001016000050128000001540143400394003919996320019160010201600002032000040039400391116002110910101600001000000001002411417164118740036406160000104004040040400404004040040
1600244003930000012046251600101016000010160000501280000015400824003940039199963200191600102016000020320000400394003911160021109101016000010000000010024115291642296400364012160000104004040040400404004040145
160024400393000000046251600101016000010160000501280000115400204003940039199963200191600102016000020320000400394003911160021109101016000010000429010022841101621299400364012160000104004040040400404004040040
160024400392990000046251600101016000010160000501280000115400204003940039199963200191600102016000020320000400394003911160021109101016000010000000010024115291642296400364012160000104004040040400404004040040
160024400393000000052251600101016000010160000501280000015400204003940039199963200191600102016000020320000400394003911160021109101016000010000000010024115271642279400364012160000104004040040400404004040040
16002440039300000159046251603031016000010160000501280000115400204003940039199963200191600102016000020320000400394003911160021109101016000010000103010024115271642275400364012160000104004040040400404004040040
160024400393000004650711251600101016000010160000501280000115400204003940039199963200191600102016000020320000400394003911160021109101016000010000000010024115291641275400364012160000104004040040400404004040040