Apple M1 Microarchitecture Research by Dougall Johnson

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

UADDW2 (vector, 2D)

Test 1: uops

Code:

  uaddw2 v0.2d, v0.2d, 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)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)a1a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037160611687251000100010002646800201820372037157231895100010002000203720371110011000000073216111787100020382038203820382038
100420371604321687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150611687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150611687251000100010002646800201820372037157271895100010002000203720371110011000000073116111787100020382038203820382038
10042037150611687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
100420371502511687251000100010002646800201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037160611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037150611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  uaddw2 v0.2d, v0.2d, 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): 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)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)9fa8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100320710116111979126100001002003820038200382003820038
102042003715000611968725101001001001210010000500284768002001820037200371842231874510100200101672042000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071011611198590100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
10204200371500215801968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020086200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382007520038
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000611964325101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010030071011611197910100001002003820038200382003820038

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)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)9fa0a1acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715003705196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
10024200371500168196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715001131196872510010101000010100005028476800200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
10024200371500251196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150061196872510010101000010100005028481821200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150061196872510010101000010100005028476800200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
10024200371500156196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
10024200371500126196872510010101000010100005028476800200182003720037184440318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  uaddw2 v0.2d, v1.2d, 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): 2.0037

retire (01)cycle (02)030918191f3f4e51inst 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)9facc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071021611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
10204200371500000149196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000084196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000084196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038

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)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)9fa1a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006404162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
10024200371501261196872510010101000010100005028476801200182003720037184443187671016220100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150361196872510022101000010100006028476800200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
10024200371500536196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
100242003715001053196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
10024200371551261196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001020006402242219823110000102003820038200382008620038
1002420037150061196872510010101000012100005028489630200182008520037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  uaddw2 v0.2d, v8.2d, v9.4s
  uaddw2 v1.2d, v8.2d, v9.4s
  uaddw2 v2.2d, v8.2d, v9.4s
  uaddw2 v3.2d, v8.2d, v9.4s
  uaddw2 v4.2d, v8.2d, v9.4s
  uaddw2 v5.2d, v8.2d, v9.4s
  uaddw2 v6.2d, v8.2d, v9.4s
  uaddw2 v7.2d, v8.2d, 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)0318191e3f51inst 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)9faccfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051102161120035800001002003920039200392003920039
8020420038150000404980100100800001008000050064000012001920038200389988399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038150009402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039
8020420038151000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100051101161120035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0318191e3a3f51inst 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)9fa0a8a9acc2cdcfd5d6d9ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200471510045244258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000002050241116014162003580000102003920039200392003920039
80024200381500018244258001010800001080000506400001200192011220098999631001880010208000020160000200382003811800211091010800001000000050241416015122003580000102003920039200392003920039
80024200381500002519258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000000050241416017122003580000102003920039200392003920039
8002420038150000244258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000000050241716016102003580000102003920039200392003920039
8002420038150000244258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000000050241616017162003580000102003920039200392003920039
8002420038150000244258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000000050241616017172003580000102003920039200392003920039
80024200381500015244258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000000050241616015162003580000102003920039200392003920039
80024200381500021244258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000000050241716015162003580000102003920039200392003920039
8002420038150000244258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000000050241616016172003580000102003920039200392003920039
800242003815000024425800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100000005024151601592003580000102003920039200392003920039