Apple M1 Microarchitecture Research by Dougall Johnson

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

FRECPS (scalar, D)

Test 1: uops

Code:

  frecps d0, d0, d1
  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)03181e3f4e51inst 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
1004403730006134072510001000100053190804018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
1004403731006134072510001000100053190804018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
10044037300012634072510001000100053190804018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
1004403730006134072510001000100053190814018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
100440373004686134072510001000100053190804022403740373258338951000100020004037403711100110000073216223473100040384038403840384038
1004403731034861340725100010001000531908040184037403732583389510001000200040374037111001100001273216223473100040384038403840384038
1004403730006134072510001000100053190804018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
1004403730006134072510001000100053190804018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
10044037300186134072510001000100053190814018403740373258338951000100020004037403711100110000073216223473100040384038403840384038
100440373002046134072510001000100053190814018403740373258338951000100020004037403711100110000073216223473100040384038403840384038

Test 2: Latency 1->2

Code:

  frecps d0, d0, d1
  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)03081e3a3f4e51inst 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)9fa8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204400372990001263940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400373000001243940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
102044003730005708313940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400373000002293940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400373000001243940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
1020440037300001613940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400373000001473940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400372990001493940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400373000002083940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038
10204400373000001453940725101001001000010010000500570690804001840037400373810833874510100200100002002000040037400371110201100991001001000010007102162239479100001004003840038400384003840038

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)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
10024400373000027439407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000000640316333947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000000640316333947310000104003840038400384003840038
10024400373000053639407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000000640316333947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100001030678316333947310000104003840038400384003840038
100244003730015018939389251003310100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000001640316333947310000104003840038400384003840038
1002440037300006139407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000030640316333947310000104003840038400384003840038
10024400373000072639407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000000640316333947310000104003840038400384003840038
1002440037300008439407251001010100001010000505706908140018400844003738130338767100102010000202000040037400371110021109101010000100000000640316333947310000104003840038400384003840038
1002440037299006139407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000000640433333950910000104003840038400384003840038
1002440037300008739407251001010100001010000505706908140018400374003738130338767100102010000202000040037400371110021109101010000100000030640316333947310000104003840038400384003840038

Test 3: Latency 1->3

Code:

  frecps d0, d1, d0
  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)030b191e1f3f4e51inst 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)9fa6a8acc2c5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102044003730000006139407251010010010000100100005005706908400184003740037381082638745101002001000020020000400374003711102011009910010010000100000007102161139479100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690840018400374003738108338745101002001018020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100003007101161139479100001004003840038400384003840038
10204400373000000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400372990000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400373001000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038
10204400372990000613940725101001001000010010000500570690840018400374003738108338745101002001000020020000400374003711102011009910010010000100000007101161139479100001004003840038400384003840038

1000 unrolls and 10 iterations

Result (median cycles for code): 4.0037

retire (01)cycle (02)03080b181e1f3f4e51inst 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)9fa1a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640616223947310000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400372990000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003730000000156394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000000061394072510010101000010100005057069081400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
10024400373000000061394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038
100244003730001000631394072510010101000010100005057069080400184003740037381303387671001020100002020000400374003711100211091010100001000640216223947310000104003840038400384003840038

Test 4: throughput

Count: 8

Code:

  frecps d0, d8, d9
  frecps d1, d8, d9
  frecps d2, d8, d9
  frecps d3, d8, d9
  frecps d4, d8, d9
  frecps d5, d8, d9
  frecps d6, d8, d9
  frecps d7, d8, d9
  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)0309191e1f3f51inst 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)9fa0a6a7a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80204200591500036042258010010080000100800005006400001200622004020040997339998801002008000020016000020040200401180201100991001008000010000000051102161120037800001002004120041200412004120041
80204200401500024042258010010080000100800005006400001200212004020040997339998801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041
8020420040150100084258010010080000100800005006400000200212004020040997339998801002008000020016000020040200401180201100991001008000010000000351101161120037800001002004120041200412004120041
8020420040150000084258010010080000100800005006400000200212004020040997339998801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041
8020420040150000042258010010080000100800005006400000200212004020040997339998801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041
8020420106150006042258010010080000100800005006400001200212004020040997339998801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041
8020420040150000042258010010080000100800005006400001200232004020040997339998801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041
8020420040150006042258010010080000100800005006400000200212004020040997339998801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412009820041
8020420040150000042258010010080000100800005006400000200212004020040997339998802242008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041
80204200401500000422580100100800001008000050064000002002120040200409973310082801002008000020016000020040200401180201100991001008000010000000051101161120037800001002004120041200412004120041

1000 unrolls and 10 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200491500412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050207162420037080000102004120041200412004120041
80024200401500412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050204162420037080000102004120041200412004120041
80024200401500412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000350204162420037080000102004120041200412004120041
80024200401500412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050206164220037080000102004120041200412004120041
80024200401500412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050204164220037080000102004120041200412004120041
800242004015042412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050204164220037080000102004120041200412004120041
800242004015012412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050202162420037080000102004120041200412004120041
80024200401500412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000350204162420037080000102004120041200412004120041
800242004015004125800101080000108000050640000120021200402004099963100208001020800002016000020040200401180021109101080000101100050202164220037080000102004120041200412004120096
800242004015033412580010108000010800005064000012002120040200409996310020800102080000201600002004020040118002110910108000010000050204164320037080000102004120041200412004120041