Apple M1 Microarchitecture Research by Dougall Johnson

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

CMEQ (register, 2S)

Test 1: uops

Code:

  cmeq v0.2s, v0.2s, v1.2s
  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)030b1f3f4e51inst 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)a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100420371600611687251000100010002646801201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500611687251000100010002646800201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500611687251000100010002646800201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715088611687251000100010002646800201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500611687251000100010002646800201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500611687251000100010002646801201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203716001051687251000100010002646801201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500611687251000100010002646801201820372037157231895100010002000203720371110011000073124111787100020382038203820382038
100420371500611687251000100010002646801201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
100420371500611687251000100010002646801201820372037157231895100010002000203720371110011000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  cmeq v0.2s, v0.2s, v1.2s
  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)03080b181e3a3f4e51inst 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)9fa8a9accdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500001206119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119676251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715000000438719687251010010010000100100005002847680020018200372003718422318745101002041000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100101525002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)031e3a3f4e51inst 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)9fa0a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102008620038200382003820038
10024200371500061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000640216221978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  cmeq v0.2s, v1.2s, v0.2s
  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)0308191e3f4e51inst 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)6061696b6d6edispatch 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)9fa0a8cfd0d2d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150000611968725101001001000010010000500284768000200180200372003718422318745101002001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680002001802003720037184223187451010020010000200200002003720037111020110099100100100001000071000116111979125100001002003820038200382003820038
1020420037150000611968725101001001000010010000500284768010200180200372003718422318745101002001000020020000200372003711102011009910010010000100007100011612197910100001002003820038200862003820038
1020520037150000611968725101001001000010010000500284768000200180200372003718422318744101002001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
1020420037150000611968725101001001000010010000500284768010200180200372003718422318745101002001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
1020420037150002071031968725101001001000010010000500284768000200180200372003718422318745101002001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
1020420037150000611968725101001001000010010000500284768000200180200372003718422318745101252001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
1020420037150000611968725101001001000010010000500284768000200180200372003718422318745101002001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
1020420037150000611968725101001001000010010000500284768000200180200372003718422318745101002001000020020000200372003711102011009910010010000100007100011611197910100001002003820038200382003820038
10204200371500039611968725101001001000010010000500284768000200180200372003718422318745101002001000020020000200372003711102011009910010010000100017100011611197910100001002003820038200382003820038

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
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000100061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
10024200371500000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000002706402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000095196672510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196672510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402163219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  cmeq v0.2s, v8.2s, v9.2s
  cmeq v1.2s, v8.2s, v9.2s
  cmeq v2.2s, v8.2s, v9.2s
  cmeq v3.2s, v8.2s, v9.2s
  cmeq v4.2s, v8.2s, v9.2s
  cmeq v5.2s, v8.2s, v9.2s
  cmeq v6.2s, v8.2s, v9.2s
  cmeq v7.2s, v8.2s, v9.2s
  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)9fa1a6accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042003815000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051104162220035800001002003920039200392003920039
802042003815000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051102162220035800001002003920039200392003920039
802042003815000082258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000651102162220035800001002003920039200392003920039
802042003815000940258010010080000100800005006400000200192007520038997339996801002008000020016000020038200381180201100991001008000010000051102162220035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100001551102162220035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100001551102162220035800001002003920039200392003920039
80204200381500015402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100006351102162220035800001002003920039200392003920039
802042003815000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051102162220035800001002003920039200392003920039
802042003815000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000051102162220035800001002003920039200392003920039
802042003815000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000951102162220035800001002003920039200392003920039

1000 unrolls and 10 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)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)9faccfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242004815000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001005020716342003580000102003920039200392003920039
800242003815000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001095020416432003580000102003920039200392003920039
800242003815000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001005020416432003580000102003920039200392003920039
800242003815013039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001005020416432003580000102003920039200392003920039
80024200381490003925800101080000108000050640000200190200382003899963100188001020800002016000020038200381180021109101080000101505020416542003580000102003920039200392003920039
800242003815000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001005020416432003580000102003920039200392003920039
800242003815000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001035020316342003580000102003920039200392003920039
8002420038150002739258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001005020416472003580000102003920039200392003920039
800242003815000039258001010800001080000506400002001902003820038999631001880010208000020160000200382003811800211091010800001005020516462003580000102003920039200392003920039
8002420038150000392580010108000010800005064000020019020038200389996310018800102080000201600002003820038118002110910108000010965020416432003580000102003920039200392003920039