Apple M1 Microarchitecture Research by Dougall Johnson

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

USHL (vector, 8B)

Test 1: uops

Code:

  ushl v0.8b, v0.8b, v1.8b
  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)033a3f4e51inst 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
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371516116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420731506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715010316872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  ushl v0.8b, v0.8b, v1.8b
  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)03080b18191e1f3f4e51inst 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)9fa0a1a6a7a8a9acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037155000000821968725101001041000010010000500284768002001820037200371842231874510100200102932002000020037200371110201100991001001000010000001000000710116311979100100001002003820038200382003820038
1020420037156000000611968725101001001000010010000500284768002001820037200371842231874510100204100002002000020037200371110201100991001001000010000000000000710116111979100100001002003820038200382003820038
1020420037156000000891968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000030000710116111979100100001002003820038200382003820038
1020420037156000030611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000020830000710116111979100100001002003820038200382003820038
10204200371550000001031968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000000000710116111979100100001002003820038200382003820038
1020420037155000000611968725101001001000010010000500284768002001820037200371842231874510100200101662022000020037200371110201100991001001000010040000000000710116111979100100001002003820038200382003820038
1020420132155000000611968725101161001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000000000710116111979100100001002003820038200382003820038
1020420037155000000611968725101001001000010010000536284768002001820037200371842231874510100200100002002000020037200372110201100991001001000010000000000000710116111979100100001002003820038200382003820038
102042003715500000021431966582101171271001213210152652285001402001820037200371842531874510100200100002002000020037200371110201100991001001000010020000000000710116111979100100001002003820038200382003820038
10204200371550000270611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010004000000000710116111979100100001002003820038200382003820038

1000 unrolls and 10 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)9fa1a6a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715500000003461968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010001006404162219785010000102003820038200382003820038
10024200371550000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371560000000611968725100101010000101000050284768012001820131200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371550000000611968725100101010000101000050284768002001820037200371844471876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371550000000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371550000000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371560000000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371550000000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
100242003715500000001031968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010001006402162219785010000102003820038200382003820038
10024200371550000000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010001006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  ushl v0.8b, v1.8b, v0.8b
  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)9fa0a1a6a8a9aaacc2branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037155000027006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715600000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715600000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
1020420037155000018006119687251010010010000100100005002847680020018200372003718422318745101002001116523422646203732027631102011009910010010000100000000000071011611197910100001002003820038200382003820038
1020420037155000024006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715500000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000100000071011611197910100001002003820038200382003820038
102042003715600000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
1020420037155000015006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000020603100071011611197910100001002003820038200382003820038
10204200371550000333006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715500000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000080311611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03070a181e1f3f4e51inst 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)9fa8a9acc2branch mispredict (cb)cfd2icache miss (d3)d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037155100120611968725100101010000101000060284768002001802003720037184443187671001020100002020000200372003711100211091010100001000000640002162219785010000102003820038200382003820038
100242003715510000611968725100121210000121000050284768002001802003720037184443187671001020100002020000200372003711100211091010100001000000640003166619787210000102003820038200382003820038
10024200371560005280611968725100101010000101000050284768002001802003720037184443187671001220100002020000200372003711100211091010100001000000640004163219785010000102003820038200382003820038
100242003715600000611968725100101010000101000050284768002001802003720037184443187671001220100002020000200372003711100211091010100001000000640004165319787210000102003820038200382003820038
100242003715500000611968725100101010000101000050284768002005402003720037184443187671001020100002020000200372003711100211091010100001000000640002162219785010000102003820038200382003820038
100242003715600000611968725100121210000121000060284768002001802003720037184443187671001220100002020000200372003711100211091010100001000300640002162219785010000102003820038200382003820038
1002420037155100120611968725100101010000101000050284768012001802003720037184443187671001020100002020000200372003711100211091010100001000000642004163319785010000102003820038200382003820038
100242003715510000611968725100101010000101000060284768012001802003720037184447187671016620101622020998200372003751100211091010100001000000642002162219785010000102003820038200382003820038
100242003715600000611968725100101010000101000050284768002001802003720037184443187671001020100002020000200372003711100211091010100001000000640002163319785010000102003820038200382003820038
1002420037155000006311968725100121010000121000060284768002001802003720037184443187671001020100002020000200372003711100211091010100001000000640002162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  ushl v0.8b, v8.8b, v9.8b
  ushl v1.8b, v8.8b, v9.8b
  ushl v2.8b, v8.8b, v9.8b
  ushl v3.8b, v8.8b, v9.8b
  ushl v4.8b, v8.8b, v9.8b
  ushl v5.8b, v8.8b, v9.8b
  ushl v6.8b, v8.8b, v9.8b
  ushl v7.8b, v8.8b, v9.8b
  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)030818191e1f3a3f51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420057156000540040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000030511021611200350800001002003920039200392003920039
802042003815500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511011611200350800001002003920039200392003920039
802042003815500000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000200000511011611200350800001002003920039200392003920039
8020420038156000600230258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511011611200350800001002003920039200392003920039
802042003815500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511011611200350800001002003920039200392003920039
802042003815600000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000001000511011611200350800001002003920039200392003920039
802042003815500000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511011611200350800001002003920039200392003920039
802042003815500400040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511011611200350800001002003920039200392003920039
802042003815500000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511011611200350800001002003920039200392003920039
802042003815500000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000001000511011611200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030818191e1f3f51inst 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
800242004715500000392580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000502013161782003580000102003920039200392003920039
8002420038155000240392580010108000010800005064075612001920038200389996310018800102080000201600002003820038118002110910108000010003502017161762003580000102003920039200392003920039
80024200381560000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050208161782003580000102003920039200392003920039
8002420038155000240392580010108000010800005064000002001920038200389996310018800102080098201600002003820038118002110910108000010010502015288172003580000102010420090200912003920104
80024200381550005040704258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001002050207161782003580000102003920039200392003920039
80024200381560002370392580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000502017161572003580000102003920039200392003920039
8002420038155000240812580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010010502017161762003580000102003920039200392003920039
80024200381550004260252258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001001050208161782003580000102003920039200392003920039
800242003816000028203925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100005020171617172003580000102009120039200392003920039
800242003815612013288392580010108000010800985064154412005920090201019996310018800102080000201600002003820038118002110910108000010000502017301762003580000102003920039200392003920039