Apple M1 Microarchitecture Research by Dougall Johnson

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

FMINNM (vector, 4H)

Test 1: uops

Code:

  fminnm v0.4h, v0.4h, v1.4h
  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)0308090b18191e1f3a3f4e51inst 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)a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150000012006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203716000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203716000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110000000073116111787100020382038203820382038
1004203715000000006116872510001000100026468012018203720371572318951000100020002037203711100110002000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  fminnm v0.4h, v0.4h, v1.4h
  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)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)9fa1a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000047101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010003007101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680120018200372003718422318745101002001000020020000200372003711102011009910010010000100020607101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010007007101161119791100001002003820038200382003820038
10204200371500000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000307101161119791100001002003820038200382003820038
102042003715000001741968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010005007101161119791100001002003820038200382003820038

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)9fa1a6a8acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002008420037211002110910101000010004306402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844531876710010201016920200002003720037111002110910101000010001306402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010002016402162219785010000102003820038200382003820038
10024200371500611967625100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010006306402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  fminnm v0.4h, v1.4h, v0.4h
  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)03080b18191e3f4e51inst 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)9fa1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037149000001031968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011622197910100001002003820038200382003820038
102042003715100000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000100071011611197910100001002003820038200382003820038
102042008415000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000076011611197910100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000071011611197910100001002003820038200382003820038
102042003715000000611968743101351001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000100071011611197910100001002003820038200382003820085
102042003715000000611968725101001001000010610152500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000100071011621197910100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000080511611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03080b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006404162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000010006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382008620038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000030006402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000010306402162219785010000102003820038200382003820038
1002420037150000000061196872510010101000010100005028476802001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  fminnm v0.4h, v8.4h, v9.4h
  fminnm v1.4h, v8.4h, v9.4h
  fminnm v2.4h, v8.4h, v9.4h
  fminnm v3.4h, v8.4h, v9.4h
  fminnm v4.4h, v8.4h, v9.4h
  fminnm v5.4h, v8.4h, v9.4h
  fminnm v6.4h, v8.4h, v9.4h
  fminnm v7.4h, v8.4h, v9.4h
  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)031e1f3f51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042003815000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100000511031623200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100003511021623200350800001002003920039200392003920039
8020420038150004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001000543511031633200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100000511031633200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100010511031623200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100000511021623200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100000511021623200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100000511031633200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100000511031633200350800001002003920039200392003920039
802042003815000402580100100800001008000050064000002001920038200389973399968010020080000200160000200382003811802011009910010080000100020511031633200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03181e1f3f51inst 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)5f60696b6d6edispatch 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)9fa8accfd2icache miss (d3)d5d6daddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242004815000039258001010800001080000506400000020019020038200389996310018800102080000201600002003820038118002110910108000010005020001216064200350080000102003920039200392003920039
800242003815000039258001010800001080000506400000020019020038200389996310018800102080000201600002003820038118002110910108000010005020006160572003515580000102003920039200392003920039
800242003815001503925800101080000108000050640000002001932003820038999631001880010208000020160000200462003811800211091010800001000502000416177200350080000102003920039200392003920039
80024200381500003925800101080000108000050640000002001902003820038999631001880010208000020160000200382003811800211091010800001000502001616056200350080000102003920039200392003920039
80024200381500003925800101080000128000050640000002001902003820038999631001880010208000020160000200382003811800211091010800001000502000416064200350080000102003920039200392003920039
80024200381500003925800101080000108000050640000012001902003820038999631001880010208000020160000200382003811800211091010800001000502000716056200350080000102003920039200392003920039
80024200381500003925800101080000108000050640000012001902003820038999631001880010208000020160000200382003811800211091010800001000502000416075200350080000102003920039200392003920039
80024200381500003925800101080000108000050640000112001902003820038999631001880010208000020160000200382003811800211091010800001000502000716047200350080000102003920039200392003920039
80024200381500003925800101080000108000050640000012001902003820038999631001880010208000020160000200382003811800211091010800001000502000616064200350080000102003920039200392003920039
80024200381500003925800101080000108000050640000012001902003820038999631001880010208000020160000200382003811800211091010800001003502000616056200350080000102003920039200392003920039