Apple M1 Microarchitecture Research by Dougall Johnson

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

FRSQRTS (vector, 4S)

Test 1: uops

Code:

  frsqrts v0.4s, v0.4s, v1.4s
  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)0308091e3f4e51inst 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
10044037300006134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300006134072510001000100053190804018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300066134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300006134072510001000100053190804018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300006134072510001000100053190804018403740373258338951000100020004037403711100110000073116113510100040384038403840384038
10044037300006134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
100440373100156134072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300008234072510001000100053190814018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300006134072510001000100053190804018403740373258338951000100020004037403711100110000073116113473100040384038403840384038
10044037300006134072510001000100053190804018403740373258338951000100020004037403711100110000073116113473100040384038403840384038

Test 2: Latency 1->2

Code:

  frsqrts v0.4s, v0.4s, v1.4s
  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 (01)cycle (02)03070a181e1f3f4e51inst 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)91inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa1a8acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020440037300110006139407251010010010000100100005005706908140018400374003738115638741101002001000820020016400374003711102011009911001001000010000601117181161139493100001004003840038400384003840038
10204400373001100061394072510100100100001001000050057069081400184003740037381156387411010020010008200200164003740037111020110099010010010000100030301117181161139494100001004003840038400384003840038
1020440037299110006139407251010010010000100100005005706908040018400374003738108338745101002001000020020000400374003711102011009901001001000010002000007102162239479100001004003840038400384003840038
10204400373000000061394072510100136100001001000050057069081400184003740037381083387451010020010000200200004003740037111020110099010010010000100059600007102162239479100001004003840038400384003840038
1020440037300000006139407251010010010000100100005005706908140018400374003738108338745101002001000020020000400374003711102011009901001001000010003000007102162239479100001004003840038400384003840038
10204400373000000044139407251010010010000100100005005706908140018400374003738108338745101002001000020020000400374003711102011009901001001000010009000007102162239479100001004003840038400384003840038
1020440037300000006139407251010010010000100100005005706908140018400374003738108338745101002001018020020000400374003711102011009901001001000010004000007102162239479100001004003840038400384003840038
1020440037300000006139407251010010010000100100005005706908040018400374003738108338745101002001000020020000400374003711102011009901001001000010001000007102162239479100001004003840038400384003840038
1020440037300000006139407251014410010000100100005005706908140018400374003738108338745101002001000020020000400374003711102011009901001001000010001000007102162239479100001004003840038400384003840038
1020440037300000006139407251010010010000100100005005706908140018400374003738108338745101002001000020020000400374003711102011009901001001000010008600007102162239479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03181e3a3f4e51inst 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
100244003731200061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001020640216223947310000104003840038400384003840038
100244003729900061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003730000061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003730000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003730000061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001020640216223947310000104003840038400384003840038
100244003730006061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003730000061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001040640216223947310000104003840038400384003840038
100244003729900061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003729900061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003721100211091010100001000640216223947310000104003840038400384003840038
100244003730000061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038

Test 3: Latency 1->3

Code:

  frsqrts v0.4s, v1.4s, v0.4s
  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 (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)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)9fa1a6a8acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730000613940710510100100100001001000050057069080400184003740037381156387411010020010008200200164003740037111020110099100100100001002000011171701600394900100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381157387411010020010008200200164003740037111020110099100100100001000000011171701600394890100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
10204400372990061394072510100100100001001000050057069081400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
10204400373000061393792510100100100001001000050057069080400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
10204400372990061394072510100100100001001000050057069081400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069081400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038
10204400372990061394072510100100100001001000050057069081400184003740037381083387451010020010000200200004003740037111020110099100100100001000010000071011611394790100001004003840038400384003840038
10204400373000061394072510100100100001001000050057069080400184003740037381083387451010020010000200200004003740037111020110099100100100001000000000071011611394790100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)030b1e3f4e51inst 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)9fa6a8acc5branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000300640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038
100244003730000726394072510010101000010100005057069080400184003740037381303387671001020100002020000400374008421100211091010100001000000640216223947310000104003840038400384003840038
100244003730001561394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038
10024400372990061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038
100244003730000726394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003721100211091010100001000000640216223947310000104003840038400384003840038
10024400373000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000000640216223947310000104003840038400384003840038

Test 4: throughput

Count: 8

Code:

  frsqrts v0.4s, v8.4s, v9.4s
  frsqrts v1.4s, v8.4s, v9.4s
  frsqrts v2.4s, v8.4s, v9.4s
  frsqrts v3.4s, v8.4s, v9.4s
  frsqrts v4.4s, v8.4s, v9.4s
  frsqrts v5.4s, v8.4s, v9.4s
  frsqrts v6.4s, v8.4s, v9.4s
  frsqrts v7.4s, v8.4s, v9.4s
  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)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)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)9faccdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420049150042258010010080000100800005006400001200210200402004099733999880100200800002001600002004020040118020110099100100800001000051102161120037800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
8020420040150042258010010080000100800005006400001200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
8020420040150042258010010080000100800005006400001200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200210200932004099733999880100200800002001600002004020040118020110099100100800001000051301161120037800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
8020420040150042258010010080000100800005006400000200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
80204200401500529258010010080000100800005006400000200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
80204200401500772258010010080000100800005006400001200210200402004099733999880100200800002001600002004020040118020110099100100800001000051101161120037800001002004120041200412004120041
8020420040150084258010010080000100800005006400001200210200402004099733999880100200800002001600002004020040118020110099100100800001000051102161120037800001002004120041200412004120041

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03191e3f51inst 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)9fa0a1a6a8accfitlb miss (d4)d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200491500025125800101080000108000050640000020021200402004099963100208001020800002016000020040200401180021109101080000100000050201216322003780000102004120041200412004120041
8002420040150004125800101080000108000050640000020021200402004099963100208001020800002016000020040200401180021109101080000100000050200316332003780000102004120041200412004120041
8002420040150004125800101080000108000050640000120021200402004099963100208001020800002016000020040200401180021109101080000100000050200316332003780000102004120041200412004120041
80024200401500041258001010800001080000506400000200212004020040999626100208001020800002016000020040200401180021109101080000100000050200316332003780000102004120041200412004120041
8002420040150004125800101080000108000050640000020021200402004099963100208001020800002016000020040200401180021109101080000100000050200316332003780000102004120041200412004120041
80024200401500067225800101080000108000050640000020021200402004099963100208001020800002016000020040200401180021109101080000100000050200216332003780000102004120041200412004120041
8002420040150004125800101080000108000050640000120021200402004099963100208001020800002016000020040200401180021109101080000100000050200216322003780000102004120041200412004120041
8002420040150004125800101080000108000050640000120021200402004099963100208001020800002016000020040200401180021109101080000100000050200316232003780000102004120041200412004120041
8002420091150004125800101080000108000050640000120021200402004099963100208001020800002016000020040200401180021109101080000100000050200316332003780000102004120041200412004120041
8002420040150004125800101080000108000050640000120021200402004099963100208001020800002016000020040200401180021109101080000100000050200216322003780000102004120041200412004120041