Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRSHL (vector, 4H)

Test 1: uops

Code:

  sqrshl v0.4h, v0.4h, v1.4h
  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)031e3f4e51inst 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
100430372206125482510001000100039831303018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
1004303723061254825100010001000398313030183037303724153289510001000200030373037111001100006073216222630100030383038303830383038
100430372206125482510001000100039831303018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372296125482510001000100039831303018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372206125482510001000100039831303018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
1004303723061254825100010001000398313030183037303724153289510001000200030373037111001100009373216222630100030383038303830383038
1004303723061254825100010001000398313030183037303724153289510001000200030373037111001100003673216222630100030383038303830383038
100430372206125482510001000100039831303018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000373216222630100030383038303830383038

Test 2: Latency 1->2

Code:

  sqrshl v0.4h, v0.4h, v1.4h
  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)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)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)9fa1a7a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500025129548251010010010000100100005004277313030018030037300372826532874510100200100002002000030037300371110201100991001001000010000107101161129634100001003003830038300383003830038
102043003722500012429548251010010010000100100005004277313030018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500046629548251010010010000100100005004277313130018030037300372826532874510100200100002002033430037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372240006129548251010010010000100100005004277313030018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372240006129548251010010010000100100005004277313030018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500012829548251010010010000100100005004277313030018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500016829548251010010010000100100005004277313030018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
10204300372250006129548251010010010000100100005004277313130018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500025329548251010010010000100100005004277313130018030037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722400000001032954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000000306402242229630010000103003830038300383003830038
100243003722500000001492954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000000026402162229630010000103003830038300383003830038
100243003722500001200612954825100101010000101000050427731313001830084300852828762876710010201000020200003003730037111002110910101000010000000559806402163429630010000103008530038300383003830038
100243013222500000001032954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000010006402162229630010000103003830038300383003830038
10024300372250000000612954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000010006402162229630010000103003830038300383003830038
100243003722400000001242954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000000306402162229630010000103003830038300383003830038
1002430037225000000061295482510010101000010100005042773131300543060430654283385629023112072611952242424230604306501211002110910101000010000010306402162229630010000103003830038300383003830038
100243003722400000007432954825100101010000101000050427731313001830037300372828732876710310201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
100243003722500000002122954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103022930417304543041830405
10024304132280088795704057052953019910083141006415110436542881691302703041630449283142328928113572210324222266830413303699110021109101010000100201002229828084822529918210000103022830038300383003830038

Test 3: Latency 1->3

Code:

  sqrshl v0.4h, v1.4h, 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)031e3f4e51inst 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)9fa0acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731303001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612953925101001001000010010000500427731313001803003730037282653287451010020210000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b181e3f4e51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0ec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250006129548251001010100001010000504277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372240006129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500017029548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250006129548251001010100001010000604277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722500072629548251001010100001010000504277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
100243003722400017229548251001010100001010000504277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038
10024300372250006129548251001010100001010000504277313130018030037300372828732876710010201000020200003003730037111002110910101000010006402162229630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  sqrshl v0.4h, v8.4h, v9.4h
  sqrshl v1.4h, v8.4h, v9.4h
  sqrshl v2.4h, v8.4h, v9.4h
  sqrshl v3.4h, v8.4h, v9.4h
  sqrshl v4.4h, v8.4h, v9.4h
  sqrshl v5.4h, v8.4h, v9.4h
  sqrshl v6.4h, v8.4h, v9.4h
  sqrshl v7.4h, v8.4h, v9.4h
  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)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)9fa1acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042004915003025801081008000810080020500640132020020200392003999776999080120200800322001600642003920039118020110099100100800001000011151180160020036800001002004020040200402004020040
802042003915003025801081008000810080020500640132120020200392003999776999080120200800322001600642003920039118020110099100100800001000011151180160020036800001002004020040200402004020040
802042003915003025801081008000810080020500640132020020200392003999776999080120200800322001600642003920039118020110099100100800001000011151180160020036800001002004020040200402004020040
80204200391505733025801081008000810080020500640132120020200392003999776999080120200800322001600642003920039118020110099100100800001000000051101161120036800001002004020040200402004020040
802042003915004125801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000051102161120036800001002004020040200402004020040
802042003915006225801001008000010080000500640000120020200392003999733999780100200800002001600002003920039118020110099100100800001000000051102161120036800001002004020040200402004020040
802042003915006225801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000051102161120156800001002004020040200402004020040
802042003915004125801001008000010080000500640000120020200392003999733999780218200800002001600002003920039118020110099100100800001000000051101161120036800001002004020040200402004020040
802042003915004125801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000051101161120036800001002004020040200402004020040
802042003915004125801001008000010080000500640000120020200392003999733999780100200800002001600002003920039118020110099100100800001000000051101161120036800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03181e3a3f51inst 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)9fa1a8cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481500924525800101080000108000050640000120020200392003999963100198001020800002016000020045200391180021109101080000102005024171613162003680000102004020040200402004020040
80024200391500028725800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100005024161616132003680000102004020040200402004020040
80024200391500024525800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100005078161613132003680000102004020040200402004020040
800242003915020213325800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100005024141614142003680000102004020040200402004020040
80024200391500024525800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100005024171615162003680000102004020040200402004020040
8002420039150002452580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000502481615162003680000102004020040200402004020040
80024200391500024525800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100005024171617162003680000102004020040200402004020040
8002420039150002452580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000502414161882003680000102004020040200402004020040
80024200391500024525800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100005024131615162003680000102004020040200402004020040
80024200391500024525800101080000108000050640000120020200392003999963100198001020801402016000020045200391180021109101080000100005024131616152003680000102004020040200402004020040