Apple M1 Microarchitecture Research by Dougall Johnson

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

FACGE (vector, 2D)

Test 1: uops

Code:

  facge 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)030b181e3f4e51inst 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
10042037150006116872510001000100026468002018203720371572318951000100020002037203711100110000373116111787100020382038203820382038
10042037150006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037150006116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037150006116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037150006116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037150006116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037150006116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020842038203820382038
100420371500061168725100010001000264680020182037203715723189510001000200020372037111001100001873116111787100020382038203820382038
10042037150006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037160006116872510001000100026468002018203720371572318951000100020002037203711100110000673116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  facge 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)0308191e3a3f4e51inst 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)9fa8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150001201721968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
102042003715000001261968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500100611968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150000061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100015307102162219791100001002003820038200382003820038
102042003715000007681968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768002001820037200371842203187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038

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)9fa6a8acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001041006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150082196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
10024200371500145196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038
1002420037150084196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001003006402162219785010000102003820038200382003820038
1002420037150061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  facge 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)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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150027519687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161219882100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150014919687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100107101161119791100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150027719687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0308090b18191e1f3f4e51inst 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)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000000082196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000020006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
10024200371500000000536196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  facge v0.2d, v8.2d, v9.2d
  facge v1.2d, v8.2d, v9.2d
  facge v2.2d, v8.2d, v9.2d
  facge v3.2d, v8.2d, v9.2d
  facge v4.2d, v8.2d, v9.2d
  facge v5.2d, v8.2d, v9.2d
  facge v6.2d, v8.2d, v9.2d
  facge 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)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)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)9fa8cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420038150013125801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051104163320035800001002003920039200392003920039
802042003815006125801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103162320035800001002003920039200392003920039
8020420038150012825801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051102163220035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103163220035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103163320035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103163220035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103163220035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001001051103163220035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051103163320035800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000051102163320035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0308181e3a3f51inst 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)9fa0a8a9accfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200391500000392580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000050200111166132003580000102003920039200392003920039
80024200381500000392580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000050200015161782003580000102003920039200392003920039
80024200381500000812580010108000010800005064000002001920038200389996310018800102080000201600002003820038518002110910108000010000050200016161762003580000102003920039200392003920039
800242003815000001442580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000050200016165162003580000102003920039200392003920039
800242003815000003925800101080000108000050640000120023200382003899963100188001020800002016000020038200381180021109101080000100000502000171617172003580000102003920039200392003920039
800242003815000003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100000502000131617172003580000102003920039200392003920039
80024200381500000392580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000050200017166172003580000102003920039200392003920039
80024200381500000392580010108000010800005064000002001920038200389996310018800102080000201607762003820038118002110910108000010000050200017161782003580000102003920039200392003920039
800242003815000005142580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000050200017161782003580000102003920039200392003920039
80024200381500000762580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000050200017166172003580000102003920039200392003920039