Apple M1 Microarchitecture Research by Dougall Johnson

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

TRN1 (vector, 4H)

Test 1: uops

Code:

  trn1 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)031e3f4e51inst issue (52)~issue fp/simd (54)~dispatch fp/simd (57)huge thing fp/simd (5b)6061696d6edispatch 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)a8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
100420371569611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
100420371501031687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038
10042037150611687251000100010002646801020182037203715723189510001000200020372037111001100000730116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  trn1 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)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)9fa8accdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150001911968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715000821968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715590821968725101001001000010010150500284768012001820037200371842231874510100200100002002000020037200372110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010006071021622197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010013071021622197910100001002003820038200382003820038
102042003715000821968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038
102042003715000821968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000071021622197910100001002003820038200382003820038

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 ld/st (80)~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)9fa6a8a9accdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371560241119687251001010100001010000502847680020018200372003718444318767100102010000200200002003720037111002110910101000010000006441016101319785010000102003820038200862003820038
100242003716102661968725100101010000101000050284768012001820037200371844431876710010201000020413200002003720037111002110910101000010000006441016111019785010000102003820038200382003820038
100242003715002661968725100101010000101000050284768012001820037200371844431876710010201000020020000200372003711100211091010100001000100644516111119785010000102003820038200382003820038
100242003715002661968725100101010000101000050284768012001820037200371844431876710010201000020020000200372003711100211091010100001000000644101610619785010000102003820038200382003820038
1002420037150028719687251001010100001010000502847680020018200372003718444318767100102010000200200002003720037111002110910101000010000006441016111019785010000102003820038200382003820038
1002420037150026619687251001010100001010000502847680120018200372003718444318767100102010000200200002003720037111002110910101000010000630644111661119785010000102003820038200382003820038
100242003715002661968725100101010000101000050284768002001820037200371844431876710010201000020020000200372003711100211091010100001000000644111651019785010000102003820038200382003820038
10024200371500210819687251001010100001010000502847680020018200372003718444318767100102010000200200002003720037111002110910101000010000006441116111119785010000102003820038200382003820038
10024200371500273119687251001010100001010000502847680020018200372003718444318767100102010000200200002003720037111002110910101000010001006441016111019785010000102003820038200382003820038
1002420037150026619687251001010100001010000502847680120018200372003718444318767100102010000200200002003720037111002110910101000010000006441116121119785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  trn1 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)031e3f4e51inst 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)9facc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500296196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
10204200371500212196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150082196872510100100100001001000050028476801200180200372003718426318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
10204200371500631196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715001241968725101001001000010010000500284768012001802003720037184223187451010020010000200200002003720037111020110099100100100001001500071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200180200372003718422318745101002001000020020000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03080a18191e1f3f4e51inst 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)9fa1a6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500000001031968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
1002420085150000000821968725100101010000101000050284768012001820037200371844431876710164201000020200002003720084111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000001241968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200852003820038
1002420037150000000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000004591968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010001006402162219785010000102003820038200382003820038
10024200371500000001031968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
1002420037150000000821968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
1002420037150000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500000001031968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010004066402162219785010000102003820038200382003820038
10024200371500000005531968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  trn1 v0.4h, v8.4h, v9.4h
  trn1 v1.4h, v8.4h, v9.4h
  trn1 v2.4h, v8.4h, v9.4h
  trn1 v3.4h, v8.4h, v9.4h
  trn1 v4.4h, v8.4h, v9.4h
  trn1 v5.4h, v8.4h, v9.4h
  trn1 v6.4h, v8.4h, v9.4h
  trn1 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)0307080b18191e1f3f51inst 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)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042006015000000001712580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511021611200350800001002003920039200392003920039
802042003815000000003322580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
802042003815010000002612580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011621200350800001002003920039200392003920039
802042003815000000001282580125100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200353800001002003920039200392003920039
80204200381500000000636580100100800951008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000002103394480100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000001000511011611200350800001002003920039200392003920039
80204200381500000000402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011624200350800001002003920039200392003920039
8020420038155000001801242580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511021611200350800001002003920039200392003920039
80204200381640000000402580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002014020039200392003920039
802042003815000000001452580100100800001008000050064000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039

1000 unrolls and 10 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)9fa8accfd2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242004015003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100050200616632003580000102003920039200392003920039
800242003815008125800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100050200216222003580000102003920039200392003920039
8002420038150012525800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100050200216222003580000102003920039200392003920039
800242003815003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100050200616622003580000102003920039200392003920039
800242003815003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100050200616322003580000102003920039200392003920039
800242003815003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100050200616362003580000102003920039200392003920039
800242003815003925800101080000108000050640000020019200382003899963100188001020800002016019820038200381180021109101080000100050200216222003580000102003920039200392003920039
800242003815003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100050200616622003580000102003920039200392003920039
8002420038150051425800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100050200616632003580000102003920039200392003920039
800242003815003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100050200216222003580000102003920039200392003920039