Apple M1 Microarchitecture Research by Dougall Johnson

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

CMGT (zero, 2D)

Test 1: uops

Code:

  cmgt v0.2d, v0.2d, #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)696d6edispatch 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)a0a6cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037160611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
100420371601031686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150611686251000100010002645212018203720371571318951000100010002037203711100110000073216221786100020382038203820382038

Test 2: Latency 1->2

Code:

  cmgt v0.2d, v0.2d, #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)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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150072619686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500611968625101001001000010010000500284752112001820037200371842131874510100200100002001000020037200371110201100991001001000010001777101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150061196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000847101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100007101161119791100001002003820038200382003820038

1000 unrolls and 10 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)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)9fa0a1a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150010319686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001024006403162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
100242003715007261968625100101010000101000050284752120018200372003718443318767100102010000201000020037200371110021109101010000100001596402162219786010000102003820038200382003820038
10024200371500611968625100101010000101000050284752120018200372003718443318767100102010000201000020037200371110021109101010000100001926402162219786010000102003820038200382003820038
100242003715006119686251001010100001010000502847521200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  cmgt v0.2d, v8.2d, #0
  cmgt v1.2d, v8.2d, #0
  cmgt v2.2d, v8.2d, #0
  cmgt v3.2d, v8.2d, #0
  cmgt v4.2d, v8.2d, #0
  cmgt v5.2d, v8.2d, #0
  cmgt v6.2d, v8.2d, #0
  cmgt v7.2d, v8.2d, #0
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2506

retire (01)cycle (02)03073f51inst 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)9fa0a6a8acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80204200481500292580108100800081008002050064013202001920038200389977699898012020080032200800322003820038118020110099100100800001000000111511801600200350800001002003920039200392003920039
80204200381500292580108100800081008016650064019612002820049200499976999868012820080038200800382004920048118020110099100100800001000000222512812311200450800001002004920049200492004920049
802042004915006426801161008001610080028500640196120028200482004999761099868012820080038200800382004920048118020110099100100800001000000222512812311200460800001002004920049200492004920049
8020420048151064268011610080016100800285006401960200282004920048997699986801282008003820080038200492004811802011009910010080000100000105222512812311200450800001002004920049200492005020050
802042004815006426801161008001610080028500640196120028200482004899761099868012820080038200800382004920048118020110099100100800001000000222512912311200450800001002004920049200492004920050
80204200491500642780116100800161008002850064019602002820049200489976999868012820080038200800382004820048118020110099100100800001000000222512812311200450800001002005020050200502004920049
8020420048150055726801161008001610080028500640952120028200482004899769998680128200800382008003820049200481180201100991001008000010000090222513412311200450800001002004920049200492004920050
80204200491500642680116100800161008002850064019612002820048200489976999868012820080038200800382004820048118020110099100100800001000011105222512912311200460800001002004920050200502005020049
80204200481500642780116100800161008002850064019602002820048200489976999868012820080038200800382004820048118020110099100100800001000000222512812311200460800001002004920049200492005020050
802042004815006426801161008001610080028500640196020028200492004899761099868012820080038200800382004820048118020110099100100800001000000222512812311200460800001002004920049200502005020050

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030a1e3f51inst 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? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242003915000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000106435020416222003580000102003920039200392003920039
80024200381500039258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010035020216222003580000102003920039200392003920039
80024200381490039258001010800001080000506400001200192024520038999631001880010208000020800002003820038118002110910108000010005020216222003580000102003920039200392003920039
80024200381500039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010065020216222003580000102003920039200392003920039
800242003815000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000101235020216222003580000102003920039200392003920039
800242003815010392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000104235020216222003580000102003920039200392003920039
80024200381500039258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010035020216222003580000102003920039200392003920039
80024200381500039258001010800001080000506400000200192003820038999631001880010208000020800002003820038118002110910108000010005020216222003580000102003920039200392003920039
80024200381500039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010005020216222003580000102003920039200392003920039
800242003815000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100155020216222003580000102003920039200392003920039