Apple M1 Microarchitecture Research by Dougall Johnson

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

FCMLT (vector, zero, 4H)

Test 1: uops

Code:

  fcmlt v0.4h, v0.4h, #0
  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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203716060016862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
100420371506116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
100420371606116862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
100420371606116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
1004208416033716754310121000115226452102054208520731571318951000100010002037203711100110000073116111786100020382038203820382038
100420371606116862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
100420371506116862510001000100026452102018203720371571318951000100010002037203711100110001073116111786100020382038203820382038
100420371506116862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
100420371506116862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
100420371506116862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038

Test 2: Latency 1->2

Code:

  fcmlt v0.4h, v0.4h, #0
  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)03080918191e1f3f4e5051inst 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
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001001371021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000071021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000071021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200902003720037184213187451010020010000200100002003720037111020110099100100100001000371021622197910100001002003820085200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000071021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001001371021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000071021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001002371021622197910100001002003820038200382003820038
10204200371500000001031968602510151100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000071021622197910100001002003820038200382003820038
1020420037150000000611968602510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000071021622197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)0308090b1e1f3f4e51inst 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)9fa6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003717300000611968625100101010048101000050284752102001820037200371844331876710620201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
100242003717300000611968625100101010000101000050284752102001820085200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037174000006119686251001010100001010000502847521020018200372003718443318767100102010000201000020037200371110021109101010000100253994640216221978610000102003820038200382003820038
1002420037173000006119686251001010100001010000502847521020018200372003718443318767100102010000201000020037200371110021109101010000100253640216221978610000102003820038200382003820038
1002420037174000006119686251001010100001010000502847521120018200372003718443318767100102010000201000020037200371110021109101010000100336640216221978610000102003820038200382003820038
10024200371610001806119686251001010100001010000502847521020018200372003718443318767100102010000201000020037200371110021109101010000100370640216221978610000102003820038200382003820038
100242003716000100611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
1002420037161000002511968625100101010000101000050284752102001820037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
10024200371600000010319653251001010100001010000502847521020018200372003718443318767100102010000201000020037200371110021109101010000100393640216221978610000102003820038200382003820038
10024200371610000053619686251001010100001010000502847521020018201772003718443318767100102010000201000020037200371110021109101010000100603640216221978610000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  fcmlt v0.4h, v8.4h, #0
  fcmlt v1.4h, v8.4h, #0
  fcmlt v2.4h, v8.4h, #0
  fcmlt v3.4h, v8.4h, #0
  fcmlt v4.4h, v8.4h, #0
  fcmlt v5.4h, v8.4h, #0
  fcmlt v6.4h, v8.4h, #0
  fcmlt v7.4h, v8.4h, #0
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)03070a181e3f51inst 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)9fa1a8acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420059150110029258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100000011151181161120086800001002003920039200392003920039
802042003815011002945801081008000810080020500640132120019200382003899776998980120200800322008003220038200381180201100991001008000010004312011151181161120035800001002003920039200392003920039
80204200381501100292580108100800081008002050064013212001920038200389977699898012020080032200800322003820038118020110099100100800001000350011151181161120035800001002003920039200392003920039
8020420038150110029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000011151181161120035800001002003920039200392003920039
8020420038150110029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100013011151181161120035800001002003920039200392003920039
8020420038150110029258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100020011151182161120035800001002003920039200392003920039
8020420038150110029648010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100020011151181161120035800001002003920039200392003920039
8020420038150110029258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100000011151181161120035800001002003920039200392003920039
802042003815011002925801081008000810080020500640132020019200382003899772799898012020080032200800322003820038118020110099100100800001000293011151181161120035800001002003920039200392003920039
8020420038150110629258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100000011151181161120035800001002003920039200392003920039

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)9fa0a7a8a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200511500392580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000100000005020616442003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100020005020416422003580000102003920039200392003920039
800242003815002292580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000005020216422003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000005020416542003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000005020416242003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000005020416442003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100030005020216242003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100010005020416242003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200871180021109101080000100000015020416442003580000102003920039200392003920039
80024200381500392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000005020216242003580000102003920039200392003920039