Apple M1 Microarchitecture Research by Dougall Johnson

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

CMGT (register, 2D)

Test 1: uops

Code:

  cmgt v0.2d, v0.2d, v1.2d
  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)031e1f3f4e51inst 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
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715606116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037159886116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037160015616872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  cmgt v0.2d, v0.2d, v1.2d
  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)030b181e1f3f4e51inst 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
1020420037150009061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000051128476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
10204200371500012061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037211020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030a1e1f3f4e51inst 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)696b6d6edispatch 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? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150078082196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715000061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
1002420037150012061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715000061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001002640216221978510000102003820038200382003820038
100242003715000066196872510010101000010100005028476802001802003720037184443187671001020101782020000200372003711100211091010100001000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010150640216221978510000102003820038200382003820038
100242003715000061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
1002420037150057061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715010061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
100242003715000061196872510010101000010100005028476802001802003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  cmgt v0.2d, v1.2d, v0.2d
  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)03091e3f4e51inst 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)9fa0a6a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000611968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979101100001002003820038200382003820038
1020420037150018611968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003714900611968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
10204200371500240611968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038
102042003715000821968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000000710116111979100100001002003820038200382003820038

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)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)9fa0a1a6a7a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037149000054006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000000006402162519785010000102003820038200382003820038
100242003715000000006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
1002420037156000000012419687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000000016402162219785010000102003820038200382003820038
100242003715000000006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200374110021109101010000100000800006402162219785010000102003820038200382003820038
100242003715000009006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000103006402162219785010000102003820038200382003820038
1002420037150000012006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000103006402162219785010000102003820038200382003820038
100242003715000000006119687251001010100001010000502848786120018202262003718444318767100102010000202000020037200371110021109101010000100001200006402162219785010000102003820038200382003820038
1002420037150000000067819687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000103006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  cmgt v0.2d, v8.2d, v9.2d
  cmgt v1.2d, v8.2d, v9.2d
  cmgt v2.2d, v8.2d, v9.2d
  cmgt v3.2d, v8.2d, v9.2d
  cmgt v4.2d, v8.2d, v9.2d
  cmgt v5.2d, v8.2d, v9.2d
  cmgt v6.2d, v8.2d, v9.2d
  cmgt v7.2d, v8.2d, v9.2d
  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)03181e1f3f51inst 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)9fa6a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420058150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100010511021611200350800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010001477511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100013511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100013511011611200350800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010002042511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100013511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100010511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100013511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080193200160194200982003811802011009910010080000100013511011611200350800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100000511011611200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03081e1f3f51inst 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)696b6d6edispatch 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)e0ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420048150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010001050202163320035080000102003920039200392003920039
8002420038150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050202163320035080000102003920039200392003920039
80024200381500003925800101080000108000050640000200190200382003899963100188001020800002016000020038200381180021109101080000100001250202163320035080000102003920039200392003920039
8002420038150100392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050203162320035080000102003920039200392003920039
8002420038150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050203162320035080000102003920039200392003920039
8002420038150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050203163320035080000102003920039200392003920039
8002420038150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050202162320035080000102003920039200392003920039
8002420038150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050203163320035080000102003920039200392003920039
8002420038150090392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010000050203163220035080000102003920039200392003920039
800242003815000032425800101080000108000050640000200190200382003899963100188001020800002016000020038200381180021109101080000100018050203163220035080000102009020039200392003920039