Apple M1 Microarchitecture Research by Dougall Johnson

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

FMINNMP (vector, 4H)

Test 1: uops

Code:

  fminnmp v0.4h, v0.4h, v1.4h
  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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001012100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371566116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468002018203720371572318951000100020002037203711100110001673116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  fminnmp v0.4h, v0.4h, v1.4h
  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)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)9facc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150061196872510100100100001001000050028476801200182003720037184223187451057620010000204200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
102042003715029161196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150306408196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451025420010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030b18191e3f4e51inst 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)9fa1a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006403162219785010000102003820038200382003820038
1002420037150000375611968725100101010000101000050284768012001820037200371844481880510010201000020200002003720037111002110910101000010000006402162219785110000102003820038200382003820038
10024200371500000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
1002420037150000234611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371490000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  fminnmp v0.4h, v1.4h, v0.4h
  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)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)9fa0a1branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150661196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037149061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500425196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150061196872510100104100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037149061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)60696b6d6edispatch 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
100242003715006119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
10024200371503816119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010006402162219837010000102003820038200382003820038
100242003715006119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010106402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680020018020037200371844431882210010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
1002420084150025119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010006402162219785110000102003820038200382003820038
100242003715006119687251001010100001010000502847680020018020037200371844431876710161201000020200002003720037111002110910101000010006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  fminnmp v0.4h, v8.4h, v9.4h
  fminnmp v1.4h, v8.4h, v9.4h
  fminnmp v2.4h, v8.4h, v9.4h
  fminnmp v3.4h, v8.4h, v9.4h
  fminnmp v4.4h, v8.4h, v9.4h
  fminnmp v5.4h, v8.4h, v9.4h
  fminnmp v6.4h, v8.4h, v9.4h
  fminnmp v7.4h, v8.4h, v9.4h
  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)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)9fa6a7a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005815004025801001008000010080000500640000200192003820038998839996801002008000020016000020038200381180201100991001008000010000036511021611200350800001002003920039200392003920039
80204200381500402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000006511011611200350800001002003920039200392003920039
80204200381500402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
80204200381500822580202122800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000016511011611200350800001002003920039200392003920039
8020420189150040258010010080000100800005006407762001920038200389973399968010020080000200160000200382003811802011009910010080000100001155110116112003519800001002003920039200392003920039
802042003815004722580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
80204200381500402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
80204200381500402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000003511011611200350800001002003920039200392003920039
80204200381500402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000006511011611200350800001002003920039200392003920039
802042003815024402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039

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)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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200471500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001000050202163320035080000102008720039200392003920039
80024200381500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001000050202163220035080000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001009050203163320035080000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001000050203162320035080000102003920039200392003920039
8002420038150039258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100138050202163220035080000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001000250203163320035080000102003920039200392003920039
800242003815003925800101080000108000050640000020019200382003899960310018800102080000201600002003820038118002110910108000010012050203163320035080000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996731001880010208000020160000200382003811800211091010800001000050203162320035080000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001000050203163320035080000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996031001880010208000020160000200382003811800211091010800001009050203163320035080000102003920136200922009620039