Apple M1 Microarchitecture Research by Dougall Johnson

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

CMHS (register, 2D)

Test 1: uops

Code:

  cmhs 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)030b1e3f4e51inst 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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100420371500821687251000100010002646802018203720371572318951000100020002037203711100110002073216221787100020382038203820382038
100420371500611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371500611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371500611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371500611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371600611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371500611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371600611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371500821687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371500611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038

Test 2: Latency 1->2

Code:

  cmhs 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)03191e1f3f4e51inst 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)9fa0a6a8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150021606119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100035007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371500606119687251010010010000100100005002847680020018200372003718422318745101002041000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371501006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
102042003715000061196872510100100100001001000050028476800200182003720037184222618745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
102042003715009906119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371500306119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000307101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03191e1f3f4e51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500630611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150027264611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150460611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
10024200371500002511968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640316331978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  cmhs 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)03181e3f4e51inst 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)9fa0a6a7a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150001040196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020210167200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
1020420037150020461196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
102042003715002761196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0318191e3f4e51inst 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)9fcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150110611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
10024200371500018611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150009611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  cmhs v0.2d, v8.2d, v9.2d
  cmhs v1.2d, v8.2d, v9.2d
  cmhs v2.2d, v8.2d, v9.2d
  cmhs v3.2d, v8.2d, v9.2d
  cmhs v4.2d, v8.2d, v9.2d
  cmhs v5.2d, v8.2d, v9.2d
  cmhs v6.2d, v8.2d, v9.2d
  cmhs 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)03181e3f51inst 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
80204200581500040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000511021611200350800001002003920039200392003920039
80204200381500040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
80204200381500040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200357800001002003920039200392003920039
80204200381500040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
802042003815041240258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
80204200381500040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
8020420038150033640258010010080000100800005006400000200692003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
80204200381500040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
80204200381500040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
802042003815002440258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03080a0b18191e1f3a3f51inst 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)9fa0a8a9acc2cfd5d6dbddinst fetch restart (de)e0? int output thing (e9)eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420039150000000003925800101080000108000050640000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020316033200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000005020316033200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000012001920038200389996710018800102080000201600002003820038118002110910108000010000005020316033200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020216033200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020216023200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000005020216033200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000112001920038200389996310018800102080000201600002003820038118002110910108000010000005020316033200350080000102003920039200392003920039
8002420038150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000005020316033200350080000102003920039200392003920039
8002420086150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000005020316033200350080000102003920089200392003920039
8002420038150000000003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010000005020216023200350080000102003920039200392003920039