Apple M1 Microarchitecture Research by Dougall Johnson

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

FMAXNM (scalar, H)

Test 1: uops

Code:

  fmaxnm h0, h0, h1
  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)030818191e3f4e51inst 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)a6a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150000611687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150000611687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037160000611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037160000611687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150000821687251000100010002646801201820372037157231895100010002000203720371110011000200073116111787100020382038203820382038
10042037150000611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
100420371500001331687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150000611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037160000611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
100420371500006116872510001000100026468012018203720371572318951000100020002037203711100110000240073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  fmaxnm h0, h0, h1
  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)0318191e3a3f4e51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500000541196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007103162219791100001002003820038200382003820038
10204200371500000237196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150009061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500018061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037149000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)033f4e51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371505521968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010136402162219785010000102003820038200382003820038
10024200371501451968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
1002420037150611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
10024200371491491968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010106402162219785010000102003820038200382003820038
1002420037150611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
1002420037150611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
1002420037150611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010006402242219785010000102003820038200382003820038
10024200371501451968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
10024200371501701968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
10024200371501451968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010206402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  fmaxnm h0, h1, h0
  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)030a191e1f3a3f4e51inst 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)9fa6a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150000001261968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
1020420037150000001241968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
102042003715000000821968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
102042003715000000841968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
1020420037150000001241968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
1020420037150000001241968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010002007102162219791100001002003820038200382003820038
1020420037150000001931968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010002007102162219791100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
1020420037150000001661968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010001007102162219791100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030b191e3f4e51inst 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)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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
1002420037150000166196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196672510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
1002420037150000229196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196874210022101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  fmaxnm h0, h8, h9
  fmaxnm h1, h8, h9
  fmaxnm h2, h8, h9
  fmaxnm h3, h8, h9
  fmaxnm h4, h8, h9
  fmaxnm h5, h8, h9
  fmaxnm h6, h8, h9
  fmaxnm h7, h8, h9
  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)03090b181e1f3a3f51inst 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)9fa6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005915000014430040258010010080000100800005006400002001920038200389973310023801002008000020016000020038200381180201100991001008000010031000511021611200350800001002003920039200392003920039
802042003815000015004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000200511011611200350800001002003920039200392003920039
80204200381500000004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000030511011611200350800001002003920039200392003920039
80204200381490000004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002024420039200392003920039
80204200381500000004063801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
80204200381500100006125801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
80204200381500000004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200383180201100991001008000010000000511011611200350800001002003920039200392003920039
80204200381500000004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002019220039200392003920039
802042003815000000025825801001008000010080000500640000201002003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
80204200881500000004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010003000511011611200350800001002003920039201472003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0308181e3a3f51inst 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)6061696d6edispatch 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)9fa6a8accfd0d2icache miss (d3)d5d6ddinst fetch restart (de)dfe0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200471500017703925800101080000108000050640000102001920038200389996310123800102080000201600002003820038118002110910108000010000502640051653320035080000102003920039200392003920039
8002420038150000333525800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502694051653320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502694051653320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201601982009020038118002110910108000010000502694051653320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502694051635320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502694051653320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502691051635320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502691051653320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502691041648320035080000102003920039200392003920039
800242003815000035025800101080000108000050640000152001920038200389996310018800102080000201600002003820038118002110910108000010000502693031655320035080000102003920039200392003920039