Apple M1 Microarchitecture Research by Dougall Johnson

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

CMLE (zero, 8B)

Test 1: uops

Code:

  cmle v0.8b, v0.8b, #0
  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)a1a6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150841686251000100010002645211201820372037157131895100010001000203720371110011000000073116111786100020382038203820382038
10042037150611686251000100010002645210201820372037157131895100010001000203720371110011000000073216111786100020382038203820382038
10042037150611686251000100010002645210201820372037157131895100010001000203720371110011000000073116111786100020382038203820382038
10042037150611686251000100010002645210201820372037157131895100010001000203720371110011000020373116111786100020382038203820382038
10042037160611686251000100010002645210201820372037157131895100010001000203720371110011000000073116111786100020382038203820382038
10042037150611686251000100010002645210201820372037157131895100010001000203720371110011000000073116111786100020382038203820382038
10042037150821686251000100010002645210201820372037157131895100010001000203720371110011000000073116111786100020382038203820382038
100420371527611686251000100010002645210201820372037157131895100010001000203720841110011000000073116111786100020382038203820382038
10042037160611686251000100010002645210201820372037157131895100010001000203720371110011000000075216111786100020382038203820382038
10042037150611686251000100010002645211201820372037157131895100010001000203720371110011000000073116111786100020382038203820382038

Test 2: Latency 1->2

Code:

  cmle v0.8b, v0.8b, #0
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03081e3f4e51inst 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)9fa0a1a8a9acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150006119686251010010010000100100005002847521200182003720037184287187411010020010008200100082003720037111020110099100100100001000000011171801600198010100001002003820038200382003820038
1020420037150006119686251010010010000100100005002847521200182003720037184287187411010020010008200100082003720037111020110099100100100001000000011171701600198000100001002003820038200382003820038
10204200371500061196862510100100100001001000050028475212001820037200371842871874110100200100082001000820085200371110201100991001001000010000004511171801600198010100001002003820038200382003820038
1020420037150006119686251010010010000100100005002847521200182003720037184286187401010020010008200100082003720037111020110099100100100001000000011171701600198010100001002003820038200382003820038
1020420037150006119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000010000071011611197910100001002003820038200382003820038
1020420037150006119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000000071011611197910100001002003820038200382003820038
10204200371500061196862510100100100001001000050028475212001820037200371842131874510100200100002001000020037200371110201100991001001000010000006300071011611197910100001002003820038200382003820038
1020420037150066119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000200000071011611197910100001002003820038200382003820038
10204200371500025119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000000071011611197910100001002003820038200382003820038
1020420037150006119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0309191e3f4e51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371560090611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640416221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
10024200371500008219686251001010100001010000502847521020018200372003718443318767100102010000201000020037200371110021109101010000100063640216221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037155000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037150000611968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  cmle v0.8b, v8.8b, #0
  cmle v1.8b, v8.8b, #0
  cmle v2.8b, v8.8b, #0
  cmle v3.8b, v8.8b, #0
  cmle v4.8b, v8.8b, #0
  cmle v5.8b, v8.8b, #0
  cmle v6.8b, v8.8b, #0
  cmle v7.8b, v8.8b, #0
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)030708090a1e3f51inst 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)9fa0a7a8acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005715010010292580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151182161120035800001002003920039200392003920039
802042003815010010292580108100800081008002050064013202001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
8020420038150100102192580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
8020420038150100115292580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392009220039
802042003815010010712580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
8020420089150100101132580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
8020420038150100102192580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
802042003815010010292580108100800081008002050064013202001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
802042003815010010292580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039
802042003815010010292580108100800081008002050064013202001920038200389977699898012020080032200800322003820038118020110099100100800001000000011151181161120035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03191e3f51inst 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)9fa8a9acc5cfd5d6ddinst fetch restart (de)e0ec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242005015000419258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010000050206167520035080000102003920039200392003920039
80024200871500039258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010303050206167520035080000102003920039200392003920039
80024200381500039258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010000050207165420035080000102003920039200392003920039
80024200381500039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010000050205166820035080000102003920039200392003920039
8002420038150003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000159050207167520035080000102003920039200392003920039
800242003815010514258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010000050208166520035080000102003920039200392003920039
8002420038150003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000120050207165520035080000102003920039200392003920039
80024200381500039258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010000050206165620035080000102003920039200392003920039
80024200381500039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010309050207166620035080000102003920039200392003920039
800242003815000392580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000102209050209166720035080000102003920039200392003920039