Apple M1 Microarchitecture Research by Dougall Johnson

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

UABD (vector, 8H)

Test 1: uops

Code:

  uabd v0.8h, v0.8h, v1.8h
  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)03081e1f3f4e51inst 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
10043037230006125482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372300015625482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372301206125482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037230006125482510081000100039831313018303730372415328951000100020003084303711100110002073216222630100030383038303830383038
10043037230906125482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100020003037308511100110000073216222630100030383038303830383038
10043037220006125482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372200010325482510001000100039831313018303730372415328951000100020003037303711100110000073216222630100030383038303830383038

Test 2: Latency 1->2

Code:

  uabd 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)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)9fa7a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722507262954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038

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)5e60696d6edispatch 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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722506129548251001010100001010000504277313013001830037300372828732876710010201000020200003003730037111002110910101000010001640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313013001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313003001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722406129548251001010100001010000504277313013001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313013001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722406129548251001010100001010000504277313013001830037300372828772876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
10024300372240612954825100101010000101000050427731301300183003730037282873287671001020100002020000300373003711100211091010100001001320640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313013001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313013001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313003001830037300372828732876710010201000020200003003730037111002110910101000010000703216222963010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  uabd 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)03081e3f4e5051inst 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
1020430037225006129548025101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250061295483002125101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037225006129548025101001001000010010000500427731313001830037300842826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037225006129548025101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383008530038
1020430037225006129548025101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037225006129548025101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954802510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000014771011611296340100001003003830038300383003830038
1020430037225006129548025101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010010071011611296340100001003003830038300383003830038
1020430037225006129548025101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250156129548025101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038

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)9fa6a8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372240612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018300373008528287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018300373003728287328786100102010000202000030037300371110021109101010000100030640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
1002430037225061295482510010101000010100005042773133001830037300372828732876710010201000020200003003730037111002110910101000010002310640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100100640216222963010000103003830038300383003830038
100243003722421612954825100101010000101000050427731330018300373008428287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
10024300372250612954825100101010000101000050427867230018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  uabd v0.8h, v8.8h, v9.8h
  uabd v1.8h, v8.8h, v9.8h
  uabd v2.8h, v8.8h, v9.8h
  uabd v3.8h, v8.8h, v9.8h
  uabd v4.8h, v8.8h, v9.8h
  uabd v5.8h, v8.8h, v9.8h
  uabd v6.8h, v8.8h, v9.8h
  uabd 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): 0.2505

retire (01)cycle (02)030b18191e1f3f51inst 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)9fa8acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005815000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100000511031633200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100000511021623200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100030511031632200360800001002004020040200402004020040
8020420039150000304125801001008000010080000500640000200202003920039997303999780100200800002001600002003920039118020110099100100800001000305127216232011621800001002004020040200402004020040
8020420039150000330412580100109800961148021050064000020020200392003999730399978010020080000200160000200392003941802011009910010080000100000511021623200360800001002004020094200922004020040
8020420039150001006425801001008000010080000500640000200202003920039997303999780100200800002001600002003920039118020110099100100800001000150511021623200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100000511021623200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100030511021623200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100030511031633200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000020020200392003999730399978010020080000200160000200392003911802011009910010080000100000511021633200360800001002004020040200402004020040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2505

retire (01)cycle (02)030b18191e1f3f51inst 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)9fa1a8acc5cficache miss (d3)d5d6ddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420048150000009002580090108000010800005064000000200202003920039999631001980010208000020160000200392003911800211091010800001000905020015167122003658080000102004020040200402004020040
8002420039150000004025800101080000108000050640000012002020039200399996310019800102080000201600002003920039118002110910108000010010050200111610102003620080000102004020040200402004020040
800242003915000000402580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000005020011161011200360080000102004020040200402004020040
800242003915000000402580010108000010800005064000000200202003920039999631001980010208010620160198201172003911800211091010800001020005020013161313200360080000102004020051200402004020040
8002420039150111010440258001010800001080000506400000020020200392003999963100198001020800002016000020039200391180021109101080000100000502001216910200360080000102004020040200402004020040
8002420039150000004025800101080000108000050640000012002020039200399996310019800102080000201600002003920039118002110910108000010000050200916116200360680000102004020040200402004020040
80024200391500000082258001010800001080000506400000120020200392003999963100198001020800002016000020039200391180021109101080000100100502001216712200360080000102004020040200402004020040
800242003915000090402580010108000010800005064000000200202003920039999631001980010208000020160000200392003911800211091010800001000005020011161211200360080000102004020040200402004020040
8002420039150000210402580010108000010800005064000000200202003920039999631001980010208000020160000200392003911800211091010800001000005020012161210200360080000102004020040200402004020040
800242003915000000402580010108000010800005064000001200202003920039999631001980010208000020160000200892003911800211091010800001000005020012161112200360080000102004020040200402004020040