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, 4S)

Test 1: uops

Code:

  fcmlt v0.4s, v0.4s, #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)accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715061168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203715061168625100010001000264521201820372037157131895100010001000203720371110011000373116111786100020382038203820382038
1004203715061168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203715061168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203715061168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203716061168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203715061168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203715961168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203716082168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038
1004203715961168625100010001000264521201820372037157131895100010001000203720371110011000073116111786100020382038203820382038

Test 2: Latency 1->2

Code:

  fcmlt v0.4s, v0.4s, #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)0318191e3a3f4e51inst 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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150001320664196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150110261196862510134100100001041015250028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
1020420037150000084196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500012061196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001001007101161119791100001002003820038200382003820038
10204200371500000620196862510143131100001001000050028475211200902003720037184213187451010020010000200100002003720037111020110099100100100001002007101161119791100001002003820038200382003820038
1020420037150003061196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500012061196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
102042003715000180975196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500021082196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002008520038200382003820038
1020420037150000061196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03071e3f4e51inst 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)9fa8acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715004261196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000006404162219786210000102003820038200382003820038
1002420037150045611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010112006402162219786010000102003820038200382003820038
10024200371500061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
10024200371500061196862510010101000010100005028475211200182003720037184433187671001220100002010000200372003711100211091010100001000006422162219786010000102003820038200382003820038
10024200371500061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
10024200371500061196862510010101000010100005028475211200182003720037184433187861001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
10024200371500061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382003820038
10024200371500061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000206402162219786010000102003820038200382003820038
100242003715002461196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000006402162219786010000102003820038200382008520038
10024200371500061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000006402163219786210000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  fcmlt v0.4s, v8.4s, #0
  fcmlt v1.4s, v8.4s, #0
  fcmlt v2.4s, v8.4s, #0
  fcmlt v3.4s, v8.4s, #0
  fcmlt v4.4s, v8.4s, #0
  fcmlt v5.4s, v8.4s, #0
  fcmlt v6.4s, v8.4s, #0
  fcmlt v7.4s, v8.4s, #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)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)9fa8acc2c5branch mispredict (cb)cdcfd5d6e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420038150029258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511821620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
80204200381500163258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
80204200381491829258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080130200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039
8020420038150029258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100000111511801620035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030418191e3f51inst 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)9fa8accfd2itlb miss (d4)d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420051150000123925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000502101191614172003580000102003920039200392003920039
800242003815010003925800101080000108000050640000020019200382003899963100188001020800002080000200382003811800211091010800001000502101141617142003580000102003920039200392003920039
800242003815010003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000502101161617172003580000102003920039200392003920039
800242003815010068125800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001010502101171617172003580000102003920039200392003920039
800242003815010003925800101080000108000050640764120019200382003899963100188001020800002080000200382003811800211091010800001000502101171614172003580000102003920039200392003920039
8002420038150100393925800101080000108000050640000120019200382003899963100438001020801962080000200382003811800211091010800001000502101171614172003580000102003920039200392003920039
80024200381501000392580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000100050210191617162003580000102003920039200392003920039
8002420038150100123925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000502101171610172003580000102003920039200392003920039
800242003815010003925800101080000108000050640000020019200382003899963100188001020800002080097200902003811800211091010800001000502101141614172007680000102003920039200392003920039
800242003815010008125800101080000108000050640000020019200382003899963100188001020800002080000200382003811800211091010800001000502101131617172003580000102003920039200392003920039