Apple M1 Microarchitecture Research by Dougall Johnson

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

UQSUB (scalar, B)

Test 1: uops

Code:

  uqsub b0, b0, b1
  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)03080b18191e1f3a3f4e51inst 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)91inst neon or fp (9a)a0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037220000000612548251000100010003983133018303730372415328951000116820003037303711100130211000000000073116112630100030383038303830383038
100430372200007500612548251000100010003983133018303730372415328951000100020003037303711100130211000000000073116112630100030383038303830383038
100430372300009300612548251000100010003983133018303730372415328951000100020003037303711100130211000000000073116112630100030383038303830383038
100430372300007200612548251000100010003983133018303730372415328951000100020003037303711100101000000000073116112630100030383038303830383038
10043037230000000612548251000100010003983133018303730372415328951000100020003037303711100101000000000073116112630100030383038303830383038
10043037230000000612548251000100010003983133018303730372415328951000100020003037303711100101000000009073116112630100030383038303830383038
10043037220000000612548251000100010003983133018303730372415328951000100020003037303711100101000000000073116112630100030383038303830383038
10043037220000000612548251000100010003983133018303730372415328951000100020003037303711100101000000000073116112630100030383038303830383038
100430372200000002102548251000100010003983133018303730372415328951000100020003037303711100101000000000073116112630100030383038303830383038
100430372300003000612548251000100010003983133018303730372415328951000100020003037303711100101000000000073116112630100030383038303830383038

Test 2: Latency 1->2

Code:

  uqsub b0, b0, b1
  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)031e1f3f4e51inst 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)9fa6a8accfd5d6ddinst fetch restart (de)e0ec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612952125101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
1020430037225008282954844101001001000010010000500427731313001830086300372826932874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002036030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010003071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010001071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b18191e1f3a3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa1a6a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250001800612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250005100612954825100101010000101000050427731303001830037300372828732884010010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
100243003722500000061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000017406402163329630010000103003830038300383003830038
1002430037225000000612954825100101010000101014950427731303001830037300372828732876710010201000020200003003730037111002110910101000010000906402162229630010000103003830038300383003830038
1002430037224000000612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000906402162229630010000103003830038300383003830038
10024300372250000006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100008406402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000306402162229630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  uqsub b0, b1, b0
  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)03191e3f4e51inst 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)9fa0a1a8a9acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500662954825101001001000010010000500427731313001830037300372826532874510100208100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500428274113001830037300372826532874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200104942002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612951125101001001000010010000500427731313001830037300372826532874510100200104962002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000000071014111296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010022000071011611296340100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313012630037300372826532874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 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)9fa0a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000030006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006407166629630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006406166629630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006406166729630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100010006406166629630010000103003830038300383003830038
100243003722500000008929548251001010100001010000504277313130018300373003728287328767100102010000202000030037300372110021109101010000100000006406166629630010000103003830038300383003830038
100243003722500000006129548251001010100001110000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100030006406166729630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006407166729630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110022109101010000100000006405166729630010000103003830038300383003830038
100243003722500000006129539251001010100001010000504277313030018300373003728287328767100102010000202000030037300372110021109101010000100000006406166629630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000006407166729630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  uqsub b0, b8, b9
  uqsub b1, b8, b9
  uqsub b2, b8, b9
  uqsub b3, b8, b9
  uqsub b4, b8, b9
  uqsub b5, b8, b9
  uqsub b6, b8, b9
  uqsub b7, b8, b9
  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)03070a0b191e1f3a3f51inst 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)9fa0a6a8a9acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005815011000021242580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511411161011200360800001002004020040200402004020040
80204200391501100002482580100100800001008000050064000012002020039200399973399978010020080000200160000200392003911802011009910010080000100000030511481688200360800001002004020040200402004020040
80204201031501100002482580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100003000511410161011200360800001002004020040200402004020040
8020420039150110000248258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000051141116810200360800001002004020040200402004020040
802042003915011000024825801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000005114516116200360800001002004020040200402004020040
802042003915011000024825801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000005114616116200360800001002004020040200402004020040
802042003915011000024825801001008000010080000500640000120020200392003999733999780100200800002001600002003920039118020110099100100800001000000005114816105200360800001002004020040200402004020040
80204200391501100002482580100100800001008000050064000012002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511411161111200360800001002004020040200402004020040
80204200391501100002482580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511410161110200360800001002004020040200402004020040
8020420039150110000248258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000051141016510200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)033f51inst 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)9fa8acc2cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242003915040258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915040258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
8002420039150515258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915040258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915040258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915040978001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915040258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915040258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
8002420039150325258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040
800242003915061258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000050200116112003680000102004020040200402004020040