Apple M1 Microarchitecture Research by Dougall Johnson

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

ADDP (vector, 4H)

Test 1: uops

Code:

  addp 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)030b181e1f3f4e51inst issue (52)~issue fp/simd (54)~dispatch fp/simd (57)huge thing fp/simd (5b)696d6edispatch 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)a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500132061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203716000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371600010861168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715000061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  addp 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)030a181e3f4e51inst 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)9fa8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150000797196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002008620038200862008520038
102042003715000061196872510100100100001001000060128476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150000149196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150008161196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150000883196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030708090a0b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acafc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000000000094319687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038
1002420037149000000012006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038
100242003715000000000006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038
100242007015000000000006119687251001010100001010152602851529200182003720037184443187671001020100002020324201322003711100211091010100001000000000016402162219785010000102003820038200382003820038
100242003715000000000006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006682162219785010000102003820038200382003820038
100242003715000000000006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038
100242003715000000000006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038
10024200371500000000000371719687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038
100242003715000000000006119687251001010100001010000502847680200182003720037184443187671001020100002020000202242003711100221091010100001000000000006402162219785010000102003820038200382003820038
100242003715000000000006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000000000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  addp 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)03090b181e1f3f4e51inst 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)9fa0a7a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150000006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372022811102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371500000035719687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371500000026619687251010010010000100100005002847680020018200372008418423718762102782001000020020000200852003711102011009910010010000100000032507342161119791100001002008620038200852003820038

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)9fa0a1a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150186119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420085150961196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000186402162219785010000102003820038200382003820038
1002420037150652319687251001010100001010000502847680200182003720085184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000156403162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150426119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150276119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150396119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  addp v0.4h, v8.4h, v9.4h
  addp v1.4h, v8.4h, v9.4h
  addp v2.4h, v8.4h, v9.4h
  addp v3.4h, v8.4h, v9.4h
  addp v4.4h, v8.4h, v9.4h
  addp v5.4h, v8.4h, v9.4h
  addp v6.4h, v8.4h, v9.4h
  addp 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)6061696d6edispatch 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)9fa8branch mispredict (cb)cfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420049150040258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110202161120035800001002003920039200392003920039
8020420038150640258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001161320035800001002003920039200392003920039
802042003815014440258010010080000100800005006400001020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110221161120035800001002003920039200392003920039
80204200381500515258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001162120035800001002003920039200392003920039
8020420038150645258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005126001161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400000020019200382003899730399968010020080000200160000200382003811802011009910010080000100005110001161120035800001002003920039200972003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307080b18191e1f3a3f51inst 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)5f60696d6edispatch 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)9fa0a1a7a8a9acc2branch mispredict (cb)cficache miss (d3)d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481500000000039258001010800001080000506400000120019200382003899963100188001020800002016000020038200381180021109101080000100000000050200171616620035080000102003920039200392003920039
80024200381500000000039258001010800001080000506400000120019200382003899963100188001020800002016000020038200381180021109101080000100000000050200161616620035080000102003920039200392003920039
80024200381500000030039258001010800001080000506400000020019200382003899963100188001020800002016000020038200381180021109101080000100000000050200161616620035080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000000005020061616620035080000102003920039200392003920039
80024200381500000000039258001010800001080000506400000020019200382003899963100188001020800002016000020038200381180021109101080000100000000050200161661620035080000102003920039200392003920039
800242003815000000150039258001010800001080000506400000020019200382003899963100188001020800002016000020038200381180021109101080000100006609005020061616520035080000102003920039200392003920039
80024200381500000000039258001010800001080000506400000120019200382003899963100188001020800002016000020038200381180021109101080000100000000050200161661620035080000102003920039200392003920039
80024200381500000000039258001010800001080000506400000120019200382003899963100188001020800002016000020038200381180021109101080000100000000050200161616620035080000102003920039200392003920039
80024200381500000000039258001010800001080000506400000120019200382003899963100188001020800002016000020038200381180021109101080000100000000050200161661620035080000102003920039200882003920039
8002420038150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000000005020061661520035080000102003920039200392003920039