Apple M1 Microarchitecture Research by Dougall Johnson

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

FCMEQ (scalar, S)

Test 1: uops

Code:

  fcmeq s0, s0, s1
  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)a0a6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
1004208415611687251000100010002646801201820372037157231895100010002000203720371110011000006073116111787100020382038203820382038
1004203716611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000673116111787100020382038203820382038
1004203715611687251000100010002646801201820372037157231895100010002000203720371110011000000073116111787100020382038203820382038
10042037151051687251000100010002646801201820372037157231895100010002000203720371110011000002073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  fcmeq s0, s0, s1
  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)030b181e1f3f4e51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000008219687251010010010000100100005002847680020018020037200371842291874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
1020420037150000053619687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197911100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000071011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)033f4e51inst 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)9fa8acc2c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715014519687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000640316231978510000102003820038200382003820038
10024200371506119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000101000640216231978510000102003820038200382003820038
100242003715010319687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000640216231978510000102003820038200382003820038
100242003715082196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001050600640216221978510000102003820038200382003820038
100242003715036519687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000640216231978510000102003820038200382003820038
10024200371506119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000640216231978510000102003820038200382003820038
10024200371506119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000640316221978510000102003820038200382003820038
100242003715026519687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216231978510000102003820038200382003820038
10024200371506119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
10024200371506119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100000640216231978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  fcmeq s0, s1, s0
  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)03041e3f4e51inst 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)9fa6a7accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371501010319687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007102161119791100001002003820038200382003820038
1020420037150008219687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150008419687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500072619687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150008219687251011510010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500012419687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0308091e1f3f4e51inst 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)9fa8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003714900906119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715000006119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150001506119687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010030640216221978510000102003820038200382003820038
1002420037150001807519687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715000006119687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150001206119687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150003006119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150001806119687251001010100001010000502847680120018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371501027053619687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150001506119687251001010100001010000502847680020018020037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  fcmeq s0, s8, s9
  fcmeq s1, s8, s9
  fcmeq s2, s8, s9
  fcmeq s3, s8, s9
  fcmeq s4, s8, s9
  fcmeq s5, s8, s9
  fcmeq s6, s8, s9
  fcmeq s7, s8, s9
  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)9fa8acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511031631200350800001002003920039200392003920039
802042003815004025801001168000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
802042003815004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200392003920039
8020420038150124025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000000511011611200350800001002003920039200892003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030b18191e3f51inst 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)6061696d6edispatch 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)9fa0a1a6a8a9acc2cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420039150000039258001010800001080000506400001220019200382003899963100188001020800002016000020038200381180021109101080000100001000502001161120035080000102003920039200392003920039
8002420038150000039258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502011161120035080000102003920039200392003920039
8002420038150000039258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502011161120035080000102003920039200392003920039
800242003815000022239258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502011161120035080000102003920039200392003920039
8002420038150000039258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502011161120035080000102003920039200392003920039
800242003815000022239258001010800001080000506400000120019200382003899963100188001020800002016000020038200381180021109101080000100000000502001161120035080000102003920039200392003920039
8002420038150000039258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502011161120035080000102003920039200392003920039
8002420038150000039258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502001161120035080000102003920039200392003920039
8002420038150000039258001010800001080000506400001120019200382003899963100188001020800002016000020038200381180021109101080000100000000502011161120035080000102003920039200392003920039
800242003815000018939258001010800001080000506400001020019200382003899963100188001020800002016000020038200381180021109101080000100000000502001161120035080000102003920039200392003920039