Apple M1 Microarchitecture Research by Dougall Johnson

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

TRN2 (vector, 2S)

Test 1: uops

Code:

  trn2 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)031e3f4e51inst 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)a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100060073116111787100020382038203820382038
10042037150611687251000100010002646800201820372037157231895100010002000203720371110011000160073116111787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100050073116111787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100090073116111787100020382038203820382038
10042037150611687251000100010002646800201820372037157231895100010002000203720371110011000170073116111787100020382038203820382038
100420371502711687251000100010002646800201820372037157231895100010002000203720371110011000190073116111787100020382038203820382038
10042037150611687251000100010002646800201820372037157231895100010002000203720371110011000150073116111787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  trn2 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)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)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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000307102162219791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184413187451010020010000200200002003720037111020110099100100100001000407102162219791100001002003820038200382003820038
1020420037150961196872510116100100001081000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000207102162219791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000267102162219791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010001207102162219791100001002003820038200382003820038
102042003716001181196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102162219791100001002003820038200382003820038
1020420037150061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001000197102162219791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0308090b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500000020400611968725100101010000101015250284768020018200372003718444318767100102010000202000020037200371110021109101010000100000006906403162319785010000102003820038200382003820038
100242003715000000000611968725100101110000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100200001506402162219785010000102003820038200382003820038
1002420037150000000005571968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000008406402162219785010000102003820038200382003820038
1002420037150000000006119687811001010100001010000502847680200182003720037184443187671001020100002020000200842008411100211091010100001000000010806402162219785010000102003820038200382003820038
1002420037150100000006119687251001010100001010000502847680200182003720037184441218767100102010000202000020037200371110021109101010000100000009906402162219785010000102003820038200382003820038
100242003715000000000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000001206402162219785010000102003820038200382003820038
100242003715000000000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000008106402162219785010000102003820038200382003820038
1002420037150000000001031968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000008406402162219785010000102003820038200382003820038
10024200371500000021006119687251001010100001010000502847680200182003720037184443187671001020100002020338200372003711100211091010100001000000011106402162219785010000102003820038200382003820038
100242003715000000000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000100006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  trn2 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)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)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)9fa0a1a6a8a9acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500000000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715000000001241968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715000000001051968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000300000071011611197910100001002003820038200382003820038
10204200371500000000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
10204200371500000090611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038
102042003715000000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001002220299530000802158121993323100001002027620276202772027520278
102042023015210055540440250319643991019414110060136107496952848883200902027520273184372218835102772181081721621658202772026351102011009910010010000100000200000071011611197910100001002003820038200382003820038
10204200371500003212010319687251010010010000100100005002847680200182003720085184223187451010021210166200200002003720037211020110099100100100001000200057011171801600198010100001002003820038200382003820038
10204200371550105579235235451968782101001001003611210760587284768020090200372027718441331878110100226109972162265220132204697110201100991001001000010000000147000171011611197910100001002003820038200382003820038
10204200371500000000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100000000000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03191e3f4e51inst 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)9fa0a7a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000621196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
100242003715000267196872510010101000010101525028476800200182003720037184443187671001020100002020000200372003711100211091010100001000200640216221978510000102013220038200382003820038
100242003715000611968725100101010000101000050284768002009020037200371844431876710010201000020200002003720037111002110910101000010002903640216221978510000102003820038200382003820038
100242003715000124196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
1002420037150001051968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010002400640216221978510000102003820038200382003820038
100242003715000126196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500061196872510010101000010100005028476800200182003720037184443187861001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010002300640216221978510000102003820038200382003820038
100242003715000170196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  trn2 v0.2s, v8.2s, v9.2s
  trn2 v1.2s, v8.2s, v9.2s
  trn2 v2.2s, v8.2s, v9.2s
  trn2 v3.2s, v8.2s, v9.2s
  trn2 v4.2s, v8.2s, v9.2s
  trn2 v5.2s, v8.2s, v9.2s
  trn2 v6.2s, v8.2s, v9.2s
  trn2 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)030408191e1f3f51inst 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)9fa0a1a8a9acc2c5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420038150100004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511031611200350800001002003920039200392003920039
80204200381500000080725801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011621200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200992003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000001511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011611200350800001002003920039200392003920039
8020420038150000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511011621200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03191e3f51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420048150001327258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020216642003580000102003920039200392003920039
8002420038150001010258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020716432003580000102003920039200392003920039
800242003815002139258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020416432003580000102008820039200392003920039
80024200381500039258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020416442003580000102003920039200392003920039
80024200381500039488001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020216762003580000102003920039200392003920039
800242003815000916258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020416672003580000102003920039200392003920039
80024200381500039258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020416342003580000102003920039200392003920039
800242003815003639258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020316432003580000102003920039200392003920039
80024200381510039258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000100005020816672003580000102003920039200392003920039
80024200381501039258001010800001080000506400000200192003820038999603100188001020800002016000020038200381180021109101080000102005020316442003580000102003920039200392003920039