Apple M1 Microarchitecture Research by Dougall Johnson

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

FNMSUB (scalar, D)

Test 1: uops

Code:

  fnmsub d0, d0, d1, d2
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(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)033f4e51inst 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)a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004403730251340725100010001000531908040184037403732583389510001000300040374037111001100000073224223473100040384038403840384038
100440373082340725100010001000531908040184037403732583389510001000300040374037111001100002073216223473100040384038403840384038
1004403730105340725100010001000531908140184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403731367340725100010001000531908040184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403730103340725100010001000531908040184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403730105340725100010001000531908140184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403730127340725100010001000531908040184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403730207340725100010001000531908140184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403730103340725100010001000531908140184037403732583389510001000300040374037111001100000073216223473100040384038403840384038
1004403730105340725100010001000531908040184037403732583389510001000300040374037111001100000073216223473100040384038403840384038

Test 2: Latency 1->2

Code:

  fnmsub d0, d0, d1, d2
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)030b191e3a3f4e51inst 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)60696b6d6edispatch 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)9fa6a8acbranch mispredict (cb)cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204400372990000613940725101001001000010010000500570690804001804003740037381083387451010020010000200300004003740037111020110099100100100001000000710031622394790100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690804001804003740037381083387451010020010000200300004003740037111020110099100100100001000000710121632394790100001004003840038400384003840038
10204400373000001613940725101001001000010010000500570690804001804003740037381083387451010020010000200300004003740037111020110099100100100001000000710121622394790100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690804001804003740037381083387451010020010000200300004003740037111020110099100100100001000000710121622394790100001004003840038400384003840038
102044003730000007263940725101001001000010010000500570690804001804003740037381083387451010020010000200300004018040037111020110099100100100001000000710121622394790100001004003840038400384003840038
10204400372990000613940725101001001000010010000500570690804001804003740037381083387641010020010000200300004003740037111020110099100100100001000000710121622394790100001004003840038400384003840038
10204400373000000823940725101001001000010010000500570690804005404003740037381083387451010020010000200300004003740037111020110099100100100001000000710121622394790100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690804001804003740037381083387451010020410000200300004003740037111020110099100100100001000000710121622394790100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690804001804003740037381083387451010020010000200300004003740037111020110099100100100001000000712121622394790100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690804001804003740037381083387451010020010000200300004008540037111020110099100100100001000000710121622394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)030818191e1f3f4e51inst 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)9fa0a6a8acc2cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100244003729900000613940725100101010000101000050570690814001840037400373813033876710010201000020300004022740037111002110910101000010000000640316223947310000104003840038400384003840038
100244003729900000613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000000640316223947310000104003840038400384003840038
100244003730000000613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
100244003730000030613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000000640216323947310000104003840038400384003840038
100244003730000000613940725100101010000101000050570690804001840037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
100244003729908810597927203393352021006012100241411194615719472140298404164046138161403893911198241128822339004046040179121100211091010100001003132493008283101353961910000104041840509404624046240489
1002440462303149106879267943932620410066141005410113327257194721403684049840466381663838960114902011457243436540523404639110021109101010000100003257020890388323947310000104003840038400384003840038
100244003730000000613940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010000000640316223947310000104003840038400384003840038
100244003730000000613940725100101010000101000050570690804001840037400373813033876710010201000020300004003740037111002110910101000010000000640216223947310000104003840038400384003840038
1002440037300000007263940725100101010000101000050570690814001840037400373813033876710010201000020300004003740037111002110910101000010400000640216223947310000104003840038400384003840038

Test 3: Latency 1->3

Code:

  fnmsub d0, d1, d0, d2
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)031e3a3f4e51inst 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)9fa8accfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001001071002162239479100001004003840038400384003840038
10204400372999061393712510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071012162239479100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071012162239479100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071012162339479100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071012162239479100001004003840038400384003840038
10204400373000161394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071012162239479100001004003840038400384003840038
10204400373000061393702510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071012162239479100001004003840038400384003840038
1020440037300006526394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071212162239479100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001000071013162239479100001004003840038400384003840038
10204400373000161394072510100100100001001000050057069080400184003740037381083387451010020010000200300004003740037111020110099100100100001002071012162239479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03081e3f4e51inst 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)72scheduler 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)9fa8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002440037300006139407251001010100001010000505706908140018400374003738130733876710010201000020300004003740037111002110910101000010000640316233947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004007540037111002110910101000010000640216223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640216223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640216223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640316223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640216223947310000104003840038400384003840038
10024400373000072639407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010001640216223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640216223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640216223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130033876710010201000020300004003740037111002110910101000010000640216223947310000104003840038400384003840038

Test 4: Latency 1->4

Code:

  fnmsub d0, d1, d2, d0
  movi v0.16b, 1
  movi v1.16b, 2
  movi v2.16b, 3

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

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)60696b6d6edispatch 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)9fa0a1a6a7a8a9acc2cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730000000008239407251011010010000104101485005709700140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000000735021623394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000000710121622394750100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908140018040037400373810833874510100200101642003050140085400841110201100991001001000010000000000710121622394790100001004003840038400384003840038
102044003729900000006139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000000710121623394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040083400371110201100991001001000010000001030710121622394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000001230732121622394790100001004003840038400384013540038
102044003730001011710010339407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000030710131622394790100001004003840038400384003840038
1020440037300000012006139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000000710121623394790100001004003840038400384003840038
1020440037300000000025139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000000710121622394790100001004003840038400384003840038
102044003730000000006139407251010010010000100100005005706908140018040037400373810833874510100200100002003000040037400371110201100991001001000010000000000710121622394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)031e1f3f4e51inst 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)696b6d6edispatch 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)9fa8acc5cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000010339407251001010100001010000505712492400180400374003738130338767100102010000203000040037400371110021109101010000100000640316223947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640216223947310000104003840038400384003840038
10024400373000045239407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640316333947310000104003840038400384003840038
10024400373000027539407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640316333947310000104003840038400384003840038
10024400373000010339407251001010100001010000505706908400180400374003738146338767100102010000203000040037400371110021109101010000100000640316333947310000104003840038400384003840038
1002440037300008439407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640216223947310000104003840038400384003840038
1002440037299008439407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640216233947310000104003840038400384003840038
10024400372990016639407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640316333947310000104003840038400384003840038
10024400373000042939407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640216223947310000104003840038400384003840038
10024400373000010739407251001010100001010000505706908400180400374003738130338767100102010000203000040037400371110021109101010000100000640316223947310000104003840038400854003840038

Test 5: throughput

Count: 8

Code:

  fnmsub d0, d8, d9, d10
  fnmsub d1, d8, d9, d10
  fnmsub d2, d8, d9, d10
  fnmsub d3, d8, d9, d10
  fnmsub d4, d8, d9, d10
  fnmsub d5, d8, d9, d10
  fnmsub d6, d8, d9, d10
  fnmsub d7, d8, d9, d10
  movi v8.16b, 9
  movi v9.16b, 10
  movi v10.16b, 11

(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)9fa8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80204200491500218258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000512711611200370800001002004120041200412004120041
80204200401500105258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041
80204200401500126258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010003511011611200370800001002004120041200412004120041
80204200401500707258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041
8020420040150042258010010080000100801045006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041
8020420040150642258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010003511011611200370800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200212004020040997339998801002008000020024000020040200401180201100991001008000010000511011611200370800001002004120041200412004120041

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0318191e3a3f51inst 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)9fa9accfd5d6d9daddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420050150000041258001010800001080000506400000120021020040200409996310020800102080000202400002004020040118002110910108000010005020121610016182003780000102004120041200412004120041
800242004015000008325800101080000108000050640000002002102004020040999631002080010208000020240000200402004011800211091010800001000502016168015152003780000102004120041200412004120041
800242004015000604125800101080000108000050640000002002102004020040999631002080010208000020240000200402004011800211091010800001000502016168011152003780000102004120041200412004120041
800242004015000004125800101080000108000050640000012002102004020040999631002080010208000020240000200402004011800211091010800001000502016166015152003780000102004120041200412004120041
80024200401500000412580010108000010800005064000001200210200402004099963100208001020800002024000020040200401180021109101080000100050209166013132003780000102004120041200412004120041
800242004015000004125800101080000108000050640000002002102004020040999631002080010208000020240000200402004011800211091010800001000502012166013132003780000102004120041200412004120041
800242004015000004125800101080000108000050640000012002102004020040999631002080010208000020240000200402004011800211091010800001000502014166017122003780000102004120041200412004120041
800242004015000004125800101080000108000050640000012002102004020040999631002080010208000020240000200402004011800211091010800001000502018168014142003780000102004120041200412004120090
800242004015000004125800101080000108000050640000012002102004020040999631002080010208000020240000200402004011800211091010800001000502013168016142003780000102004120041200412004120041
800242004015000004125800101080000108000050640000012002102004020040999631002080010208000020240000200402004011800211091010800001000502014168012152003780000102004120041200412004120041