Apple M1 Microarchitecture Research by Dougall Johnson

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

FCMGE (vector, zero, 4H)

Test 1: uops

Code:

  fcmge 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)0308191e3f4e51inst 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
100420371500010516862510001000100026452102018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
100420371600061168625100010001000264521120182037203715713189510001000100020372084111001100002473216221786100020382038203820382038
100420371501186116862510001000100026452112018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000073216221852100020382038203820382038
100420371500156116862510001000100026452112018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000373216221786100020382038203820382038
10042037160006116862510001000100026452112018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
100420371500156116862510001000100026452112018203720371571318951000100010002037203711100110000073216221786100020382038203820382038
10042037150096116862510001000100026452102018203720371571318951000100010002037203711100110000073216221786100020382038203820382038

Test 2: Latency 1->2

Code:

  fcmge 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)0318191e1f3f4e51inst 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)9fa0a1a6a8a9accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000300611968625101001001000010010000500284752112001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007102161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752102001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752112001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752112001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500090611968625101001001000010010000500284752102001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752112001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752102001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752112001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752102001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
10204200371500000611968625101001001000010010000500284752102001820037200371842131874510100200100002001000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038

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)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
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640616661978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640616651978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640616461978610000102003820038200382003820038
1002420037150061196862510010121001210100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640516551978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640616561978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640516651978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640516641978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640616561978610000102003820038200382003820038
10024200371500611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010015640516541978610000102003820038200382003820038
100242003715038761196862510010101000010100005028475211200182003720037184433187671001020100002010000200372003711100211091010100001000640516561978610000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  fcmge v0.4h, v8.4h, #0
  fcmge v1.4h, v8.4h, #0
  fcmge v2.4h, v8.4h, #0
  fcmge v3.4h, v8.4h, #0
  fcmge v4.4h, v8.4h, #0
  fcmge v5.4h, v8.4h, #0
  fcmge v6.4h, v8.4h, #0
  fcmge 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)0308191e3a3f5051inst 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)9fa8acc5branch mispredict (cb)cdcfd6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042006115000002920022258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160120035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401321200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381491000290258010810080008100800205006409201200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039
80204200381500000290258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100001115118160020035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0308091e3f51inst 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)9fa8accficache miss (d3)d5d6ddinst fetch restart (de)e0eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420039150020392580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000103657502003162320035080000102003920039200392003920039
80024200381500003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000502004162320035080000102003920039200392003920039
80024200381500003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000502002162220035080000102003920039200392003920039
800242003815000039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010351502002162220035080000102003920039200392003920039
80024200381500003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001010502002163420035080000102003920039200392003920039
8002420038150001239258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010153502002162220035080000102003920039200392003920039
800242003815000039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010353502002162320035080000102003920039200392003920039
800242003815000039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010160502002164220035080000102003920039200392003920039
800242003815000039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010260502002164420035080000102003920039200392003920039
800242003815000039258001010800001080000506400001200192003820038999631001880010208000020800002003820038118002110910108000010263502003162220035080000102003920088200392003920039