Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRSHRN (S)

Test 1: uops

Code:

  sqrshrn h0, s0, #3
  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)030b183f4e51inst 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
1004303723006125472510001000100039816003018303730372414328951000100010003037303711100110000073316332629100030383038303830383038
1004303722006125472510001000100039816003018303730372414328951000100010003037303711100110000073216332629100030383038303830383038
10043037220014725472510001000100039816013018303730372414328951000100010003037303711100110000073316332629100030383038303830383038
1004303722019225472510001000100039816003018303730372414328951000100010003037303711100110000073316332629100030383038303830383038
1004303723006125472510001000100039816003018303730372414328951000100010003037303711100110000073216222629100030383038303830383038
1004303723006125472510001000100039816003018303730372414328951000100010003037303711100110000073316332629100030383038303830383038
1004303722008225472510001000100039816003018303730372414328951000100010003037303711100110000073316332629100030383038303830383038
10043037230012625472510001000100039816003018303730372414328951000100010003037303711100110000373216332629100030383038303830383038
10043037230010025472510001000100039816013018303730372414328951000100010003037303711100110000073316332629100030383038303830383038
1004303723008425472510001000100039816003018303730372414328951000100010003037303711100110000073316332629100030383038303830383038

Test 2: Latency 1->2

Code:

  sqrshrn h0, s0, #3
  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)0308090b18191e1f3a3f4e5051inst 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)9fa0a1a6a7a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372320000000023429547025101001001000010010000500427716003001830037300372826432874510100200100002001016630037300371110201100991001001000010004000000071011611296330100001003003830038300383003830038
102043003723300001000229295472044101001001000010010000500427716003001830037300372826432874510100200103302001000030037300374110201100991001001000010000000000071011611296330100001003003830038300383003830038
10204300372330000000061295470251010010010000100100005004277160030018300373003728264182876110100200100002001000030037300371110201100991001001000010000000000071011611296330100001003003830038300383003830038
1020430037233000000006129547025101001001000010010000500427716003001830037300372826432874510100200100002001000030037300371110201100991001001000010000000000071011611296330100001003003830038300383013330038
1020430037232000000006129547025101001001000010010000500427716003001830037300372826532874510100200100002001000030037300371110201100991001001000010000000000071011611296330100001003003830038300383003830038
1020430037233000000006129547025101001001000010010000500427716003001830037300372826432874510100200100002001000030037300371110201100991001001000010000000000071011611296330100001003003830038300383003830086
10204300372330000000010329547025101001001000010010000602427716003001830037300372826432874510100200100002001000030037300371110201100991001001000010000000000071011611296330100001003003830038300383003830038
10204300372330000000072629547025101001001000010010000500427716003001830037300372826432874510100200100002001000030037300371110201100991001001000010002000000071011611296330100001003003830038300383013530038
10204300372330000000084295470621010010010000100100005004277160030018300373003728264328745101002001000020010000300373003711102011009910010010000100000000000710116112963312100001003003830038300383003830038
1020430037233000000018429547025101001001000010010000500427716003001830037300372826432874510100200100002001000030037300831110201100991001001000010000000000071021611296330100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e1f3a3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing ld/st (5a)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)9fa0a1a2a6a7a8a9acbcc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037232010000006129547251001010100001010000500427716013001830037300372828632876710010201000020100003003730037111002110910101000010000004006006402162229629010000103003830038300383003830038
100243003724200000000972954725100101010000101000050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000102210006402162229629010000103003830086300853003830038
100243003723200010000612954725100101010008111015082042798641300543008430084282908287861016320101622210168300843008521100211091010100001000030022738026612243329667210000103013330131301333013330086
10024301782330000000013322952943100301510008101015055042785121300183013130037282863287791001020100002210000301803003711100211091010100001000000002613046613322229629010000103003830085300383003830038
1002430037233000000011032954743100201210000101000050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000109006402162229667010000103013330134300383003830038
1002430037233000000007262954725100101010000101000050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000329006403162229629010000103003830038300383003830038
10024300372470000000014072954725100101010008101000050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000200006402162229629010000103003830038300383003830038
10024300372330000020700612954725100101010000101015050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000209006402162229629010000103003830038300853003830038
100243003723400000000612954725100101010000101045050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000203006402162229629010000103003830038300383003830038
100243003723300000000612954725100101010000101000050042771601300183003730037282863287671001020100002010000300373003711100211091010100001000000303006402162229629010000103003830038300383003830038

Test 3: throughput

Count: 8

Code:

  sqrshrn h0, s8, #3
  sqrshrn h1, s8, #3
  sqrshrn h2, s8, #3
  sqrshrn h3, s8, #3
  sqrshrn h4, s8, #3
  sqrshrn h5, s8, #3
  sqrshrn h6, s8, #3
  sqrshrn h7, s8, #3
  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)0307080a0b18191e1f3f51inst 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)9fa0a1a6a7a8a9acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042006015510100000862580108100800081008002050064013202002020039200399977699908012020080032200800322003920039118020110099100100800001000000000111511811612200360800001002004020040200402004020040
802042003916110100100302580108100800081008002050064013202002020039200399977699908012020080032200800322003920039118020110099100100800001000000200111511811611200360800001002004020040200402004020040
8020420039156101000003025801081008000810080020500640132120020200392003999771199908012020080032200800322003920039118020110099100100800001000000000111511811611200360800001002017620040200912004020040
802042003915510100000302580108100800081008002050064013212002020039200399977699908012020080032200801372011120039118020110099100100800001000000100111511811611200360800001002004020040200402004020040
802042003915510100000302580108100800081008002050064013212002020039200399977699908012020080032200800322003920039118020110099100100800001000000000111511811611200360800001002004020040200402004020040
8020420039155101000006952580108100800081008002051164013212002020039200399977699908012020080032200800322003920039118020110099100100800001000000100111511811611200360800001002004020040200402004020040
802042003915510100000302580108100800081008002050064013212002020039200399977699908012020080032200800322003920039118020110099100100800001000000100111511811611200360800001002004020040200402004020040
802042003915610100000302580502100800081008002050064013212002020039200399977699908012020080032200800322003920039118020110099100100800001000000000111511811611200360800001002004020040200402004020040
802042003915510101000302580108100800081008002050064013212002020039200399977699908012020080032200800322003920039118020110099100100800001000200100111511811611200360800001002004020040200402004020040
802042003915510100090302580108100800081008002050064013202002020039200399977699908012020080032200800322003920104118020110099100100800001000000100111511811611200360800001002004020040200402004020094

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0307090b18191e1f3a3f51inst 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)5f60696b6d6edispatch 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)9fa0a1a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420039161000000004025800101080000108000050640000002002002003920039999631001980010208000020800002003920039118002110910108000010000008705020416442003680000102004020040200402004020040
8002420039160000009001672580010108000010800005064332400200200200392003999963100198001020800002080000200392003911800211091010800001020000005020416422008680000102004020040200402004020040
8002420039155000000004025800101080000108000050640000002002002003920039999631001980010208000020800002003920039118002110910108000010000007505020416422003680000102004020040200402004020040
800242003915500000000402580010108000010800005064000001200200200392003999968100198001020800002080000200392003911800211091010800001000000005020416422003680000102004020040200402004020040
8002420039155000000004025800101080000108000050640000002002002003920039999631001980010208000020800002003920039118002110910108000010000008105020416422003680000102004020040200402004020040
8002420039155000000004025800101080000108000050640000012002002003920039999631001980010208000020800002003920039118002110910108000010000008705020439242003680000102004020040200402004020040
8002420039156000000004025800101080000108000050640000012002002003920039999631001980010208000020800002003920039118002110910108000010000007805020516422003680000102004020040200402004020090
8002420039156000000004025800101080000108000050640000002002002003920039999631001980010208000020800002003920039118002110910108000010000008405020216242003680000102004020040200402004020040
800242003915500000001402580010108000010800005064000000200200200392003999963100198001020800002080000200392003911800211091010800001000000005020416422003680000102004020144200402004020040
80024200391550000000040258001010800001080000506400000120020020039200399996310019800102080000208000020039200391180021109101080000100000093505020416422003680000102004020040200402004020040