Apple M1 Microarchitecture Research by Dougall Johnson

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

FSUB (vector, 4S)

Test 1: uops

Code:

  fsub v0.4s, v0.4s, v1.4s
  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)03181e3f4e51inst 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)a0a1a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372300612548251000100010003983133018303730372415328951000100020003037303711100110000000373116112630100030383038303830383038
100430372200612548251000100010003983133018303730372415328951000100020003037303711100110000030073116112630100030383038303830383038
100430372200612548251000100010003983133018303730372415328951000100020003037303711100110000000973116112630100030383038303830383038
100430372300612548251000100010003983133018303730372415328951000100020003037303711100110000000973116112630100030383038303830383038
100430372300612548251000100010003983133018303730372415328951000100020003037303711100110000000973116112630100030383038303830383038
100430372300612548251000100010003983133018303730372415328951000100020003037303711100110000000073116112630100030383038303830383038
100430372300612548251000100010003983133018303730372415328951000100020003037303711100110000000973116112630100030383038303830383038
100430372300612548251000100010003983133018303730372415328951000100020003037303711100110000000373116112630100030383038303830383038
100430372200612548251000100010003983133018303730372415328951000100020003037303711100110000000259373116112630100030383038303830383038
100430372200612548251000100010003983133018303730372415328951000100020003037303711100110000000073116112630100030383038303830383038

Test 2: Latency 1->2

Code:

  fsub v0.4s, v0.4s, v1.4s
  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)03080b1e1f3f4e51inst 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)9fa0a6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000105295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000371011611296340100001003003830038300383003830038
1020430037226000061295482510100100100001001000058342773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003008130228300383003830038
1020430037225000061295392510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000671011611296340100001003003830038300383003830038
1020430037225000010329548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100021012371011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300375110201100991001001000010000008171011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000101571011611296340100001003003830038300383003830038
1020430037225000352578295482510100100100001001000067642773131300183003730037282773287451010020010000200200003003730037111020110099100100100001000000371011611296340100001003003830038300383003830038
10204300372250000103295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500006129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000011171011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000008771011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080a18191e1f3f4e51inst 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)9fa6a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
10024300372250000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100200006402162229630010000103003830038300383003830038
10024300372250000006129548251001010100001010149504277313030018300373003728287328767100102010000202000030037300371110021109101010000100100006402242229630010000103003830038300383003830038
10024300372240000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100100006402162229630010000103003830038300383003830038
10024300372250000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100103006402162229630010000103003830038300383003830038
10024300372240000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100300006402162229630010000103003830038300383003830038
10024300372250000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100400006402162229630010000103003830038300383003830038
100243003722500000061295482510043101000010100005042773130300183003730037282873287671001020100002020000300373003711100211091010100001002900006402162229630010000103003830038300383003830038
10024300372250000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100100006402162229630010000103003830038300383003830038
100243003722500000056429548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100200006402162229630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  fsub v0.4s, v1.4s, v0.4s
  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)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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037224006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037224006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001009071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038
1020430037225008229512251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001001071011611296340100001003003830038300383003830038
1020430037224006129548251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080918191e3f4e51inst 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)9fa8acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001000006405162229630010000103003830038300383003830038
1002430037224000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001000006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001000006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001000006402162229630010000103003830038300383003830038
1002430037224000008229530251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001006006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001000006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001030006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001010006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313300183003730037282873287671001020100002020000300373003711100211091010100001000006402162229630010000103003830038300383003830038
1002430037224000006129548251001010100001010000504277313300623003730037282873287671001020100002020000300373003711100211091010100001000006402162229630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  fsub v0.4s, v8.4s, v9.4s
  fsub v1.4s, v8.4s, v9.4s
  fsub v2.4s, v8.4s, v9.4s
  fsub v3.4s, v8.4s, v9.4s
  fsub v4.4s, v8.4s, v9.4s
  fsub v5.4s, v8.4s, v9.4s
  fsub v6.4s, v8.4s, v9.4s
  fsub v7.4s, v8.4s, v9.4s
  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)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)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)9fa8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420060150041258010010080096100800005006400001200202003920039997339997801002008000020016000020039200391180201100991001008000010010051103162120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400001200202003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010023111051101161120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400001200202003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
8020420039150041258010012580000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010023051101421120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400001200202003920039997339997801002008000020016000020039200391180201100991001008000010010051101161120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400001200202003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
80204200391500412580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100313051101161120036800001002004020040200402004020040
8020420039150041258010010080000100800005006400001200202003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040

1000 unrolls and 10 iterations

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

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)5f60696d6edispatch 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)9fa8accfd5d6dbddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242004815004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001000502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001010502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001000502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000012002020039200399996031001980010208000020160000200392003911800211091010800001010502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001000502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001010502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001010502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000012002020039200399996031001980010208000020160000200392003911800211091010800001000502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001000502011601120036080000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996031001980010208000020160000200392003911800211091010800001000502011601120036080000102004020040200402004020040