Apple Microarchitecture Research by Dougall Johnson

M1/A14 P-core (Firestorm): Overview | Base Instructions | SIMD and FP Instructions
M1/A14 E-core (Icestorm):  Overview | Base Instructions | SIMD and FP Instructions

FMULX (vector, 2D)

Test 1: uops

Code:

  fmulx v0.2d, v0.2d, v1.2d
  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 uop (01)cycle (02)031e3f4e51schedule uop (52)schedule simd uop (54)dispatch simd uop (57)ldst uops in schedulers (5b)60696d6emap stall dispatch (70)map rewind (75)map stall (76)dispatch uop (78)map simd uop (7e)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst simd alu (9a)l1d cache writeback (a8)accfd5map dispatch bubble (d6)ddfetch restart (de)e0? simd retires (ee)f5f6f7f8fd
1004403730010334072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373006134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
1004403730015634072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373106134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373106134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
1004403730015634072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
1004403730014734072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373106134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373006134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373006134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038

Test 2: Latency 1->2

Code:

  fmulx v0.2d, v0.2d, v1.2d
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire uop (01)cycle (02)031e3f4e51schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)ldst uops in schedulers (5b)60696b6d6emap stall dispatch (70)map rewind (75)map stall (76)dispatch uop (78)map int uop (7c)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst branch cond (94)inst int alu (97)inst simd alu (9a)9fl1d cache writeback (a8)accfd5map dispatch bubble (d6)ddfetch restart (de)e0? simd retires (ee)? int retires (ef)f5f6f7f8fd
10204400373000613940725101001001000010010000500570690804001804003740037381081238745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037300061394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037300061394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037299061394072510100100100001001000050057069080400180400374003738143338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037300061394072510100100100001001000050057069081400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037300061394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037299061394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161039479100001004003840038400384003840038
1020440037299061394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037300061394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400374003711102011009910010010000100007101161139479100001004003840038400384003840038
1020440037300084394072510100100100001001000050057069080400180400374003738108338745101002001000020020000400814003711102011009910010010000100007101161139479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire uop (01)cycle (02)03mmu table walk instruction (07)l2 tlb miss instruction (0a)1e1f3a3f4e51schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)ldst uops in schedulers (5b)60696d6emap stall dispatch (70)map rewind (75)map stall (76)dispatch uop (78)map int uop (7c)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst branch cond (94)inst int alu (97)inst simd alu (9a)9fl1d cache writeback (a8)accfd5map dispatch bubble (d6)ddfetch restart (de)e0? simd retires (ee)? int retires (ef)f5f6f7f8fd
10024400373001100289393982510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000644101610103947310000104003840038400384003840038
10024400373001100268394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000644101610113947310000104003840038400384003840038
1002440037300110026839407251001010100001010000505706908140018400374008438130338767100102010000202000040037400371110021109101010000100064410161053947310000104003840038400384003840038
100244003730011002733394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001020644101610103947310000104003840038400384003840038
10024400372991100268394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000644101610103947310000104003840038400384003840038
1002440037299110026839407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100064410161083947310000104003840038400384003840038
10024400373001100268394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000644101610103947310000104003840038400384003840038
10024400373001100268394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000644101610103947310000104003840038400384003840038
10024400373001100268394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000644101610103947310000104003840038400384003840038
100244003729911002683940725100101010000101000050570690814001840037400373813033876710010201000020200004003740037111002110910101000010006448168103947310000104003840038400384003840038

Test 3: Latency 1->3

Code:

  fmulx v0.2d, v1.2d, v0.2d
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 4.0037

retire uop (01)cycle (02)03mmu table walk data (08)09l2 tlb miss data (0b)18191e3a3f4e51schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)ldst uops in schedulers (5b)696b6d6emap stall dispatch (70)map rewind (75)map stall (76)dispatch uop (78)map int uop (7c)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst branch cond (94)inst int alu (97)inst simd alu (9a)9fl1d cache writeback (a8)accfd5map dispatch bubble (d6)ddfetch restart (de)e0? int output thing (e9)? simd retires (ee)? int retires (ef)f5f6f7f8fd
10204400373000000000147394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840038400384003840038
1020440037300001000061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840038400384003840038
1020440037300000000061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840038400384003840038
1020440085300010000061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000371011611394790100001004003840038400384003840038
1020440037300000000061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840038400384003840038
10204400373000000000726394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840229400384003840038
1020440037300000000061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840038400384003840038
10204400373000000012061394072510100100100001001000050057069084001804003740037381083387451010020210000204200004003740037311020110099100100100001000071011611394790100001004003840038400384003840038
10204400373000000012061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001000071011611394790100001004003840038400384003840038
1020440037300000000061394072510100100100001001000050057069084001804003740037381083387451010020010000200200004003740037111020110099100100100001001071011611394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire uop (01)cycle (02)03mmu table walk data (08)1e1f3f4e51schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)ldst uops in schedulers (5b)6061696d6emap stall dispatch (70)map rewind (75)map stall (76)dispatch uop (78)map int uop (7c)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst branch cond (94)inst int alu (97)inst simd alu (9a)9fl1d cache writeback (a8)acbranch cond mispred nonspec (c5)cfd0d2d5map dispatch bubble (d6)ddfetch restart (de)e0? simd retires (ee)? int retires (ef)f5f6f7f8fd
100244003730000013483940725100101010000101000050570690815400184003740037381303387671001020100002020000400374003711100211091010100001000064053216223947310000104003840038400384003840038
1002540037300000613940725100101010000101000050570690800400184003740037381303387671001020100002020000400374003711100211091010100001000064000216223947310000104003840038400384003840038
1002440037300000613940725100101010000101000050570690800400184003740037381303387671015820100002020000400374003711100211091010100001000064000216223947310000104003840038400384003840038
10024400373000005553940725100101010000101000050570690800400184003740037381303387671001020100002020000400374003711100211091010100001000064000216223947310000104003840038400384003840038
100244003730000014253940725100101010000101000050570690800400184003740037381303387671001020100002020000400374003711100211091010100001000064000216223947310000104003840038400384003840038
100244003729900013793940725100101010000101000050570690805400184003740037381303387671001020100002020000400374003711100211091010100001000064000216223947310000104003840038400384003840038
10024400373000008943940725100101010000101000050570690800400184003740037381303387671001020100002020000400374003711100211091010100001000064053216223947310000104003840038400384003840038
1002440037300000613940725100101010000101000050570690815400184003740037381303387671001020100002020000400374003711100211091010100001000064053216223947310000104003840038400384003840038
100244003730000018663940725100101010000101000050570690815400184003740037381303387671001020100002020000400374003711100211091010100001000064053216223947310000104003840038400384003840038
1002440037300000613940725100101010000101000050570690815400184003740037381303387671001020100002020000400374003711100211091010100001000064053216223947310000104003840038400384003840038

Test 4: throughput

Count: 8

Code:

  fmulx v0.2d, v8.2d, v9.2d
  fmulx v1.2d, v8.2d, v9.2d
  fmulx v2.2d, v8.2d, v9.2d
  fmulx v3.2d, v8.2d, v9.2d
  fmulx v4.2d, v8.2d, v9.2d
  fmulx v5.2d, v8.2d, v9.2d
  fmulx v6.2d, v8.2d, v9.2d
  fmulx v7.2d, v8.2d, v9.2d
  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 uop (01)cycle (02)033a3f51schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)ldst uops in schedulers (5b)696b6d6emap stall dispatch (70)map rewind (75)map stall (76)dispatch uop (78)map int uop (7c)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst branch cond (94)inst int alu (97)inst simd alu (9a)9fl1d cache writeback (a8)accfd5map dispatch bubble (d6)ddfetch restart (de)e0? int output thing (e9)? simd retires (ee)? int retires (ef)f5f6f7f8fd
802042005815021332580100100800001008000050064000020021020040200409973399988010020080000200160000200402004011802011009910010080000100005114101669200370800001002004120041200412004120041
8020420040150247258010010080000100800005006400002002102004020040997339998801002008000020016000020040200401180201100991001008000010000511451659200370800001002004120041200412004120041
8020420040150270258010010080000100800005006400002002102004020040997339998801002008000020016000020040200401180201100991001008000010000511471695200370800001002004120041200412004120041
80204200401502702580100100800001008000050064000020021020040200409973399988010020080000200160000200402004011802011009910010080000100005114111649200370800001002004120041200412004120041
8020420040150247258010010080000100800005006400002002102004020040997339998801002008000020016000020040200401180201100991001008000010000511471674200370800001002004120041200412004120041
80204200401502472580100100800001008000050064000020021020040200409973399988010020080000200160000200402004011802011009910010080000100005114916105200370800001002004120041200412004120041
802042004015024725801001008000010080000500640000200210200402004099733999880100200800002001600002004020040118020110099100100800001000051141116114200370800001002004120041200412004120041
8020420040150247258010010080000100800005006400002002102004020040997331002680100200800002001600002004020040118020110099100100800001000051141016114200370800001002004120041200412004120041
802042004015024725801001008000010080000500640000200210200402004099733999880100200800002001600002004020040118020110099100100800001000051141016104200370800001002004120041200412004120041
8020420040150247258010010080000100800005006400002002102004020040997339998801002008000020016000020040200401180201100991001008000010000511410161092003722800001002004120041200412004120041

1000 unrolls and 10 iterations

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

retire uop (01)cycle (02)031e1f3f51schedule uop (52)schedule int uop (53)schedule simd uop (54)dispatch int uop (56)dispatch simd uop (57)int uops in schedulers (59)ldst uops in schedulers (5b)60696d6emap stall dispatch (70)simd prf full (72)map rewind (75)map stall (76)dispatch uop (78)map int uop (7c)map simd uop (7e)map int uop inputs (7f)map simd uop inputs (81)8283flush restart other nonspec (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst branch cond (94)inst int alu (97)inst simd alu (9a)9fl1d tlb access (a0)l1d tlb miss (a1)ld unit uop (a6)l1d cache writeback (a8)accfd2l1i cache miss demand (d3)d5map dispatch bubble (d6)d9dadbddfetch restart (de)e0? int output thing (e9)eaeb? simd retires (ee)? int retires (ef)f5f6f7f8fd
800242004915000622580010108000010800005064000012002120040200409996031002080010208000020160000200402004011800211091010800001000000502000151600016142003700080000102004120041200412004120041
80024200401500046925800101080000108000050640000120021200402004099960310020800102080000201600002004020040218002110910108000010000005020001616000642003700080000102004120041200412004120041
80024200401500041258001010800001080000506400000200212004020040999603100208001020800002016000020040200401180021109101080000100000050200016160006142003700080000102004120041200412004120041
800242004015000412580010108000010800005064000002002120040200409996031002080010208000020160000200402004011800211091010800001000000502000161600016420037020080000102004120041200412004120041
800242004015000412580010108000010800005064000012002120040200409996031002080010208000020160000200402004011800211091010800001000010502000616000632003700080000102004120041200412004120041
8002420040150001502580010108000010800005064000002002120040200409996031002080010208000020160000200402004011800211091010800001000000502000161600016320037020880000102004120041200412004120041
8002420040150004125800101080000108000050640000020021200402004099960310020800102080000201600002004020040118002110910108000010000005020006160006132003700080000102004120041200412004120041
8002420040150006225800101080000108000050640000120021200402004099960310020800102080000201600002004020040118002110910108000010000005020006160001632003700080000102004120041200412004120041
800242004015000104258001010800001080000506400000200212004020040999603100208001020800002016000020040200401180021109101080000100000050200014160001332003700080000102004120041200412004120041
80024200401500041258001010800001080000506400000200212004020040999603100208001020800002016000020040200401180021109101080000100000050200016160001632003700080000102004120041200412004120041