Apple M1 Microarchitecture Research by Dougall Johnson

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

FRINTN (vector, 4H)

Test 1: uops

Code:

  frintn v0.4h, v0.4h
  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)030b191e3f4e51inst issue (52)~issue fp/simd (54)~dispatch fp/simd (57)huge thing fp/simd (5b)696d6edispatch 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)a1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037230008225472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037230006125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037240008225472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037230006125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037240006125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037230008425472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
100430372300126125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037240006125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037240006125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038
10043037230006125472510001000100039816030183037303724143289510001000100030373037111001100000073116112629100030383038303830383038

Test 2: Latency 1->2

Code:

  frintn v0.4h, 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): 3.0037

retire (01)cycle (02)030918191e1f3a3f4e51inst 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)9fa0a1a6a7a8acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003723200000061295472510100100100001001000050042771600300183003730037282643287451010020010000204100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
1020430037232000840061295472510100100100001001000050042771600300183003730084282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
1020430037233000000131295472510100100100001001000050042771600300183003730037282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
102043003723200000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020210099100100100001000000000071011611296330100001003003830038300383003830038
102043003723300000061295472510100100100001001000050042771601300183003730037282763287451010020010000200101663003730037111020110099100100100001000000057710171011611296330100001003003830038300383003830038
102043003723200000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
1020430037233000000475295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
102043003723200000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
102043003723300000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038
102043003723300000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000000071011611296330100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03071e3f4e51inst 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225021961295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006405162229631010000103003830038300383003830038
100243003722503361295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830083
10024300372250661295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629210000103003830038300383003830038
10024300372240061295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830038
10024300372250061295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830038
10024300372240061295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830038
10024300372250061295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830038
100243003722502761295472510012101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830038
1002430037225022861295472510010101000710100005042771601300183003730037282863288421001020100002010000300373003711100211091010100001006402162229629010000103003830038300383003830038
10024300372250661295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006402164229629010000103003830038300383003830038

Test 3: throughput

Count: 8

Code:

  frintn v0.4h, v8.4h
  frintn v1.4h, v8.4h
  frintn v2.4h, v8.4h
  frintn v3.4h, v8.4h
  frintn v4.4h, v8.4h
  frintn v5.4h, v8.4h
  frintn v6.4h, v8.4h
  frintn v7.4h, v8.4h
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)03080b18191e1f3f51inst 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)9fa0a1a6a7a8a9acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80204200681550000903025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391550000003025801081008000810080020500640132020020200392019699776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391550000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391550000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391550000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391550000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391610000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391560000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200392180201100991001008000010000000030111511801600200360800001002004020040200402004020040
80204200391560000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040
80204200391550000003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000000111511801600200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030a1e3f51inst 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)9fa8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200511550040258001010800001080000506400001200202003920039999631001980010208000020800002003920039118002110910108000010000502091612122003680000102004020040200402004020040
800242003915600402580010108000010800005064000002002020039200399996310019800102080000208000020039200391180021109101080000100005020171612122003680000102004020040200402004020040
8002420039156015402580010108000010800005064000002002020039200399996310019800102080000208000020039200391180021109101080000100005020151612122003680000102004020040200402004020040
80024200391550151102580010108000010800005064000002002020039200399996310019800102080000208000020039200391180021109101080000100005020121612122003680000102004020040200402004020040
8002420039156004202580010108000010800005064000012002020039200399996310019800102080000208000020039200391180021109101080000100005020171617132003680000102004020040200402004020040
8002420039155042402580010108000010800005064000012002020193200399996310019800102080000208000020039200391180021109101080000100005020141614162003680000102004020040200402004020040
800242003915506402580010108000010800005064000002002020039200399996310019800102080000208000020039200391180021109101080000100005020111615112003680000102004020040200402004020040
8002420039155127402580010108000010800005064000002002020039200399996310019800102080000208000020039200391180021109101080000100005020121611112003680000102004020040200402004020040
800242003916000402580010108000010800005064000002002020039200399996310019800102080000208000020039200391180021109101080000100005020111611122003680000102004020040200402004020040
8002420039155039402580010108000010800005064000012002020039200399996121001980010208000020800002003920039118002110910108000010003502013169142003680000102004020040200402004020040