Apple M1 Microarchitecture Research by Dougall Johnson

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

SMAXP (vector, 2S)

Test 1: uops

Code:

  smaxp v0.2s, v0.2s, v1.2s
  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)030b1e3f4e51inst 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
10042037150038116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203716006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203716006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110001073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203716006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  smaxp v0.2s, v0.2s, v1.2s
  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)0308090b18191e1f3a3f4e51inst 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)9fa0a6a7a8a9acc2cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500000000082196872510100100100001001000050028476801200182003720037184223187601025620010000200200002003720037111020110099100100100001000301000779024123198610100001002003820182200872018320085
102042013415011121408176023741966579101661061001213510608522284768012001820037200371842227187631042120410166200200002003720037111020110099100100100001000200002710021622198610100001002003820038200382003820085
1020420179150011005798807121966525101001001000010010000500285024612012620085200371843331874510100200100002002000020037200371110201100991001001000010022000007100224221979124100001002008520038200382003820038
10204200371510001192640789196766210151124100361001000050028476800200182003720037184223187821057920010000200200002003720037111020110099100100100001000200019930710021622199020100001002003820038200382003820038
10204200371490000000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000000733021622197910100001002003820038200382003820038
10204200371500000000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000000710021622197910100001002003820038200382003820038
102042003715000004417264161119654101101631231000010010000500284768002001820037200371842212187451010020610000200200002003720037611020110099100100100001000000059450710021622197910100001002003820038200382003820038
10204200371500000060082196874910116122100361001045650028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000002000710021622197910100001002003820038200382003820038
10204200371500000000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000001000710021622197910100001002003820038200382003820038
102042003715000000000212196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000000710021622197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000000006619687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001002406402162219785010000102003820038200382003820038
10024200371500000000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000101006402162219785010000102003820038200382003820038
10024200371500000000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100006402242219785010000102003820038200382003820038
10024200371500000000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000000003321968725100101010000101000050284768020018200372003718447318767100102010000202000020037200371110021109101010000100006402162219785010000102008620038200382003820038
100242003715000000001261968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000000001241968725100101010000101000050284768020018200852003718444318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
1002420037150000008801241968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100006402162219785010000102003820038200382003820038
100242003715000000001051968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000101006402162219785010000102003820038200382003820038
10024200371500000000821968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000101006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  smaxp v0.2s, v1.2s, v0.2s
  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)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)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)9fa8c2c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001060850028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038

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)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? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371503876119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000727224221978510000102003820038200382003820038
100242003715041725119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
1002420037150516119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715096119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  smaxp v0.2s, v8.2s, v9.2s
  smaxp v1.2s, v8.2s, v9.2s
  smaxp v2.2s, v8.2s, v9.2s
  smaxp v3.2s, v8.2s, v9.2s
  smaxp v4.2s, v8.2s, v9.2s
  smaxp v5.2s, v8.2s, v9.2s
  smaxp v6.2s, v8.2s, v9.2s
  smaxp v7.2s, v8.2s, v9.2s
  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)03181e3f51inst 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? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80204200581500664025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051104164320035800001002003920039200392003920039
8020420038149094025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051105164520035800001002003920039200392003920039
8020420038150004062801001188000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051105165420035800001002003920039200392003920039
8020420038150094025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051105165520035800001002003920039200392003920039
8020420038150004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051104164420035800001002003920039200392003920039
8020420038150004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051105165320035800001002003920039200392003920039
8020420038150004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051104165420035800001002003920039200392003920039
802042003815001551525801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051104164520035800001002003920039200392003920039
8020420038150094025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051104164420035800001002003920039200392003920039
8020420038150004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103164420035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03181e3f51inst 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)9fa8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481500546025800101080000108000050640000120021200382003899963100188001020800002016000020038200381180021109101080000100015020616552003580000102003920039200392003920039
8002420038150003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100005020516352003580000102003920039200392003920039
8002420038150004192580010108000010800005064000012001920038200389996201001880010208000020160000200382003811800211091010800001000050203281532003580000102003920039200392003920039
800242003815002283943800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100005020516552003580000102003920039200392009020346
8002420038150063925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100005020316352003580000102003920039200392003920039
8002420038150063925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100005020516552003580000102003920039200392003920039
8002420038150003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100005020516532003580000102003920039200392003920039
800242003815000116825800101080000108000050640000120019200382008999963100188001020800002016000020038200381180021109101080000100005037516352003580000102003920039200392003920039
8002420038150003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100005020316352003580000102003920039200392003920039
8002420038150003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100005020516552003580000102003920039200392003920039