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

FSQRT (vector, 2D)

Test 1: uops

Code:

  fsqrt v0.2d, v0.2d
  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)696d6emap 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
10041303798061115402510001000100045795313018130371303712248312895100010001000130371303711100110005373116111243110001303813038130381303813038
10041303798061115402510001000100045795313018130371303712248312895100010001000130371303711100110006073116111243110001303813038130381303813038
10041303798061115402510001000100045795313018130371303712248312895100010001000130371303711100110000073116211243110001303813038130381303813038
100413037970611154025100010001000457953130181303713037122483128951000100010001303713071111001100004873116111243110001303813038130381303813038
1004130379815611154025100010001000457953130181303713037122483128951000100010001303713037111001100004273116111243110001303813038130381303813038
10041303797061115402510001000100045795313018130371303712248312895100010001000130371303711100110000073116111243110001303813038130381303813038
10041303798061115402510001000100045795313018130371303712248312895100010001000130371303711100110000073116111243110001303813038130381303813038
10041303797061115402510001000100045795313018130371303712248312895100010001000130371303711100110000073116111243110001303813038130381303813038
10041303798061115402510001000100045795313018130371303712248312895100010001000130371303711100110000073116111243110001303813038130381303813038
10041303797961115402510001000100045795313018130371303712248312895100010001000130371303711100110000073116111243110001303813038130381303813038

Test 2: Latency 1->2

Code:

  fsqrt v0.2d, 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): 13.0037

retire uop (01)cycle (02)0309l2 tlb miss data (0b)1e3f4e51schedule 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)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)9fld unit uop (a6)l1d cache writeback (a8)acbranch cond mispred nonspec (c5)cdcfd5map dispatch bubble (d6)ddfetch restart (de)e0ea? simd retires (ee)? int retires (ef)f5f6f7f8fd
102041300379740007011195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037211020110099100100100001002000071011611129431010000100130038130038130038130038130038
102041300379740001871195402510100100100001001000050046609530130018013007013003712809803128745101002001000020010000130037130037111020110099100100100001000000071011611129431010000100130038130038130038130038130038
102041300379740007261195402510100100100001001003750046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000030071011611129431010000100130038130038130038130038130038
102041300379740005691195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000000071011611129431010000100130038130038130038130038130038
102041300379740002561195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000000071011611129431010000100130038130038130038130038130038
10204130037974000611195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000000071011611129431010000100130038130038130038130038130038
10204130037974000611195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000000071011611129431010000100130038130038130038130038130038
10204130037974000611195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000000171011611129431010000100130038130038130038130038130038
10204130037974000611195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000030071011611129431010000100130038130038130038130038130038
102041300379740002161195402510100100100001001000050046609530130018013003713003712809803128745101002001000020010000130037130037111020110099100100100001000000071011611129431010000100130038130038130038130038130038

1000 unrolls and 10 iterations

Result (median cycles for code): 13.0037

retire uop (01)cycle (02)03mmu table walk data (08)181e1f3a3f4e51schedule 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 tlb access (a0)l1d tlb miss (a1)ld unit uop (a6)st unit uop (a7)l1d cache writeback (a8)acc2branch cond mispred nonspec (c5)cdcfd5map dispatch bubble (d6)ddfetch restart (de)e0? int output thing (e9)? simd retires (ee)? int retires (ef)f5f6f7f8fd
10024130037974000008211954025100101010000101000050466095311300183130037130037128120312879910010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101210000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000102031000006402162212943101000010130038130038130038130038130038
1002413003797400138006111954025100101010000101000050466095311300180130226130225128191312876710010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312887610010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000100000000006402162212943101000010130038130038130038130038130038
10024130037974000006111954025100101010000101000050466095311300180130037130037128120312876710010201000020100001300371300371110021109101010000100000000006402162112943101000010130038130038130038130038130038

Test 3: throughput

Count: 8

Code:

  fsqrt v0.2d, v8.2d
  fsqrt v1.2d, v8.2d
  fsqrt v2.2d, v8.2d
  fsqrt v3.2d, v8.2d
  fsqrt v4.2d, v8.2d
  fsqrt v5.2d, v8.2d
  fsqrt v6.2d, v8.2d
  fsqrt v7.2d, v8.2d
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 2.0006

retire uop (01)cycle (02)03mmu table walk instruction (07)l2 tlb miss instruction (0a)191e3f4e51schedule 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)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)acbranch cond mispred nonspec (c5)branch mispred nonspec (cb)cdcfd5map dispatch bubble (d6)ddfetch restart (de)e0? int output thing (e9)? simd retires (ee)? int retires (ef)f5f6f7f8fd
80204160048119801030617993725801001008000010080000500567798411600290160048160048149895061500028010020080008200800081600481600481180201100991001008000010000003111511711611159984080000100160049160049160049160049160049
8020416004811991100617993725801001008000010080000500567798401602640160048160048149895061500028010020080008200800081600481600482180201100991001008000010002003360111511711611159984080000100160049160049160049160049160049
80204160048119911015617993725801001008000010080000500567798411600290160048160048149895061500028010020080008200800081600481600481180201100991001008000010000000111511711611159984080000100160049160049160049160049160049
8020416004811991100617993725801001008000010080000500567798411600290160048160048149895061500028010020080008200800081600481600482180201100991001008000010000000111511711611159984080000100160049160049160049160049160049
8020416004811991100617993725801001008000010080000500567798411600290160048160341149895061500028010020080008200800081600481600483180201100991001008000010000000111511711611159984080000100160049160049160049160049160049
8020416004811991100617993725801001008000010080000500567798411600290160048160048149895061500028010020080072200800081600481600481180201100991001008000010000003111511711611159984080000100160049160049160049160049160049
802041600481199110822617993725801001008000010080000500567798411600290160048160048149895061500028010020080008200800801600481600481180201100991001008000010000000111511711611159984080000100160049160049160049160049160049
8020416004811981100617993725801001008000010080000500567798411600290160048160048149895061500028010020080008200800801600481600481180201100991001008000010000000111511711611159984080000100160049160049160049160049160105
8020416004811991100617993725801001008000010080000500567798411600290160048160048149895061500028010020080008200800081600481600481180201100991001008000010000000111511711611159984080000100160049160049160049160049160049
80204160048119911018617992125801001008000010080000500567798411600290160048160048149895061500028010020080008200800721600481600481180201100991001008000010000000111511711611159984080000100160049160049160049160049160049

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 2.0006

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)acc2branch mispred nonspec (cb)cdcfd5map dispatch bubble (d6)ddfetch restart (de)e0? int output thing (e9)? simd retires (ee)? int retires (ef)f5f6f7f8fd
8002416004811990617993725800101080000108000050567798411600290160048160048149910315002880010208000020800001600481600481180021109101080000100000050201416121215997108000010160049160049160049160049160049
8002416004811980617993725800101080000108000050567798411600290160048160048149910315002880010208000020800001600481600481180021109101080000104000050201216171315997108000010160049160049160049160107160049
80024160048119905367993725800271080000108000050567798411600290160048160048149910315002880010208000020800001600481600481180021109101080000100000050201716161315997128000010160049160049160049160049160049
8002416004811990617993725800101080000108000050567798411600293160048160048149910315002880010208000020800001600481600481180021109101080000100000050201716171415997108000010160049160049160049160049160049
8002416004811980617993725800101080000108000050567798411600290160048160048149910315002880010208000020800001600481600481180021109101080000100000050201316121215997108000010160049160049160049160049160049
800241600481199082799372580010108000010800005056779841160029016004816004814991031500288001020800002080000160048160107118002110910108000010018300050431516151615997108000010160049160049160049160049160049
8002416004811990617993725800101080017108000050567798411600290160048160048149910315002880010208000020800001600481600481180021109101080000100000050201216111415997108000010160049160049160049160049160049
8002416004811980617993725800101080000108000050567798401600290160048160048149910315002880010208000020800001600481600481180021109101080000100000050201116131716004708000010160049160049160049160049160049
8002416004811990617993725800101080000108000050567798401600290160048160048149910315002880010208000020800001600481600481180021109101080000100000050201016121715997108000010160049160049160049160049160049
8002416004811990617993725800101080000108000050567798411600290160106160048149910315002880010208000020800001600481600481180021109101080000100000050201616141615997108000010160049160049160049160049160049