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, 4H)

Test 1: uops

Code:

  facge 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)03080b18191e1f3a3f4e51inst 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)a0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150000000611687251000100010002646801201820372037157231895100010002000203720371110011000000000073216221787100020382038203820382038
100420371500000006116872510001000100026468002018203720371572318951000100020002037203711100110000000090073216221787100020382038203820382038
10042037150000000611687251000100010002646801201820372037157231895100010002000203720371110011000000003073216221787100020382038203820382038
10042037150000000611687251000100010002646801201820372037157231895100010002000203720371110011000000000073216221787100020382038203820382038
10042037150000900611687251000100010002646800201820372037157231895100010002000203720371110011000000001818073216221787100020382038203820382038
100420371500000001561687251000100010002646800201820372037157231895100010002000203720371110011000000009073216221787100020382038203820382038
10042037150000000611687251000100010002646801201820372037157231895100010002000203720371110011000000000073216221787100020382038203820382038
10042037150000120061168725100010001000264680120182037203715723189510001000200020372037111001100000000135073216221787100020382038203820382038
10042037150000000821687251000100010002646801201820372037157231895100010002000203720371110011000000000073216221787100020382038203820382038
100420371500000001031687251000100011522646801201820372084157271895100010002000203720841110011000002040073216221787100020852038203820382085

Test 2: Latency 1->2

Code:

  facge 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)030b3f4e5051inst 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)9fa6a8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150061196872002125101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000171011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100010071011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020054200372003718422318745101002001000020020000200372003711102011009910010010000100010071011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611198590100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150044119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715006119687025101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038201332003820038

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)60696b6d6edispatch 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)9fa0a1a6a7a8a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000000000084196872510010101000010100005028476801200180200372003718444031876710010201000020200002003720037111002110910101000010000030006402162219785010000102003820038200382003820038
1002420037150000000000061196872510010101000010100005028476800200180200372003718444031876710010201000020200002013220084311002110910101000010000010006402162219785010000102003820038200382003820038
100242003715000000000007261968725100101010000101000050284768012001802003720037184440318767100102010000202000020037200371110021109101010000100000004506402162219785010000102003820038200382003820038
1002420037150000000000061196872510010101000010100005028476800200180200372003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
10024200371500000000000103196872510010101000010100005028476800200180201792003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000000061196872510010101000010100005028476801200180200372003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820132
1002420037150000000000061196872510010101000010100005028476800200180200372003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000000061196872510010101000010100005028476800200180200372003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000000061196872510010101000010100005028476800200180200372003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000000061196872510010101000010100005028476800200180200372003718444031876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  facge 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)030818191e1f3a3f4e51inst 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)9fa0a1a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
1020420037150089792006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000024071011611197910100001002003820038200382003820038
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
102042003715000000072619687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204202341500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0308091e3f4e51inst 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)9fa8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010100640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010003640216221978510000102008420085200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010000640216331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000821968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150010611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010006640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768020018020037200371844431876710010201000020200002003720037111002110910101000010003640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  facge v0.4h, v8.4h, v9.4h
  facge v1.4h, v8.4h, v9.4h
  facge v2.4h, v8.4h, v9.4h
  facge v3.4h, v8.4h, v9.4h
  facge v4.4h, v8.4h, v9.4h
  facge v5.4h, v8.4h, v9.4h
  facge v6.4h, v8.4h, v9.4h
  facge 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)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)9fa1a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420058151040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051102161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150063258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120120800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
80204200381500610258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038149040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150040258010010080080100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
80204200381500515258010010080000100800005006400001200192003820038998139996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030b18191e1f3a3f51inst 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)9fa0a1a6a7a8a9accficache miss (d3)d5d6dbddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242004815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031603620035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031603320035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020021603420035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031603420035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031603320035080000102003920039200392003920039
800242003815000000039258001010800001080096506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020021603220035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031605520035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031602320035080000102003920039200392003920039
8002420038150000000355258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020041603320035080000102003920039200392003920039
800242003815000000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001000000005020031606720035080000102003920039200392003920039