Apple M1 Microarchitecture Research by Dougall Johnson

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

URHADD (vector, 4H)

Test 1: uops

Code:

  urhadd 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)033f4e51inst 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
100420371610316872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371515616872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  urhadd 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)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)9fa0a1a6a8accdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500000000611968725101001001000010010000500284768012005420037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
10204200371500000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
102042003715000000006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100000000806256321993126100001002023320276202312013320231
10204202661521144396440120781963243101961351007213810760679285502312019820321203131844031188531105221810827216219922027720324411020110099100100100001000021100380824165222000727100001002032220281203272027820228
10204203261521166936606130201962110210194137100721321091264228553781202342032420278184158188541104621810997220204442032620322711020110099100100100001007000120250842166311986127100001002032620276203222003820324
102042032715210666665280611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
10204200371500000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
10204200371500000000611968725101001001001210710000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
10204200371500000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
10204200371490000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030b18191e1f3f4e51inst 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)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000120611968725100101010000101000050284768002001802003720037184443187671001020100002020000200372003711100211091010100001000010196206404162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001802003720037184443187671001020100002021326200372003711100211091010100001000000306402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768012001802003720037184443187671001020100002020000200372003711100211091010100001000000006402163219785010000102003820038200382003820086
10024200371500000352611968725100101010000101000050284768012001802003720037184443187671001020100002020000202292003711100211091010100001000000006402162219785010000102003820038200382003820038
100242003715000400611968725100101010000101000050284768012001802003720037184443187671001020100002020000200372022611100211091010100001000000006402162219785010000102003820038200382003820038
10024200371500001680611968725100101010000101000050284768002001802003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768012001802003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001832003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768012001802003720037184443187671001020100002020000200372003711100211091010100001000050606402162219785010000102003820038200382003820038
1002420037150000120611968725100101010000101000050284768012001802003720037184443187671001020100002020000200372003711100211091010100001000000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  urhadd 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)030b18191e1f3f4e51inst 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)9fa1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150000237061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
10204200371560000061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
102042003715000063061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
1020420037150000480061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000100071011611197910100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
10204200371500002580726196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)031e1f3f4e51inst 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150492065519687251001010100001010152552847680020018200852003718444318767100102010000202000020037200371110021109101010000100026640316521978510000102003820038200382008520038
10024200371500063119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037149006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
10024200371501206119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100003640216221978510000102003820038200382003820038
1002420037150006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000119019687811001010100241010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100100640216221978510000102003820038200382003820038
1002420037150006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
1002420037150006119687251001012100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  urhadd v0.4h, v8.4h, v9.4h
  urhadd v1.4h, v8.4h, v9.4h
  urhadd v2.4h, v8.4h, v9.4h
  urhadd v3.4h, v8.4h, v9.4h
  urhadd v4.4h, v8.4h, v9.4h
  urhadd v5.4h, v8.4h, v9.4h
  urhadd v6.4h, v8.4h, v9.4h
  urhadd 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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420049150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000511031611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400000200192003820038997379996802122008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010006511011611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010003511012911200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038998139996801002008000020016019420097200381180201100991001008000010000511011611200350800001002003920039200392003920039
80204200381500822580100101800001008000060264000012001920038200939989310022801002008009620216000020038200381180201100991001008000010003511011611200350800001002003920039200392003920039
8020420038150040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000511011611200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030a1e3f51inst 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)9faccfd5d6dbddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481501213925800101080000108000050640000312001920038200389996310018800102080000201600002003820038118002110910108000010050203160242003580000102003920039200392003920039
80024200381500013425800101080000108000050640000512001920038200389996310018800102080000201600002003820038118002110910108000010050203160322003580000102003920039200392003920039
8002420038150003925800101080000108000050640000612001920038200389996310018800102080000201600002003820038118002110910108000010050204160322003580000102003920039200392003920039
8002420038150003925800101080000108000050640000812001920038200389996310018800102080000201600002003820038118002110910108000010050202160322003580000102003920039200392003920039
8002420038150003925800101080000108000050640000712001920038200389996310018800102080000201600002003820038118002110910108000010050203160432003580000102003920039200392003920039
8002420038150003925800101080000108000050640000712001920038200389996310018800102080000201600002003820038118002110910108000010050202160232003580000102003920039200392003920039
8002420038150003925800101080000108000050640000712001920038200389996310018800102080000201600002003820038118002110910108000010050203160232003580000102003920039200392003920039
8002420038150003925800101080000108000050640000712001920038200389996310018800102080000201600002003820038118002110910108000010050203160332003580000102003920039200392003920039
8002420038150008125800101080000108000050640000712001920038200389996310018800102080000201600002003820038118002110910108000010050203160332003580000102003920039200392003920039
8002420038150063925800101080000108000050640000712001920038200389996310018800102080000201600002003820038118002110910108000010050203160342003580000102003920039200392003920039