Apple M1 Microarchitecture Research by Dougall Johnson

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

SQSUB (vector, 4H)

Test 1: uops

Code:

  sqsub 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)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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037230612548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372206125482510001008100039831330183037303724153289510001000200030373037111001100003073216222630100030383038303830383038
100430372301036254825100010001000398313301830373037241532895100010002000303730371110011000015373216222630100030383038303830383038
10043037233612548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037220612548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372301912548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
10043037230612548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038
100430372206125482510001000100039831330183037303724153289510001000200030373037111001100004873216222630100030383038303830383038
100430372203392548251000100010003983133018303730372415328951000100020003037303711100110000073216222630100030383038303830383038

Test 2: Latency 1->2

Code:

  sqsub 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)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)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)9fa0a6a7a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000002242954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037224000000842954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037225000600612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300793003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011621296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
1020430037225000000612954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038
10204300372250000002082954825101001001000010010000500427731330018300373003728265328745101002001000020020000300373003711102011009910010010000100000000071011611296340100001003003830038300383003830038

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)696d6edispatch stall (70)72scheduler 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)9fa0a1a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037224000000015929548251001010100001010000504277313300183003730037282870328767100102010000202000030037300372110021109101010000100000006403162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000094329548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037224000000012629548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722400000006129548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
10024300372250000000267329548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000016629548251001010100001010000504277313300183003730037282870328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  sqsub 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)030809181e1f3f4e51inst 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)9fa0a6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020360300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313030018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037224000008229548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225000006129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b18191e1f3f4e51inst 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)61696d6edispatch 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000006129548251001010100001010000504277313030018300373003728287828786100102010000202000030037300371110021109101010000100096402162229630010000103003830038300383003830038
1002430037225000008229548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038
1002430037224000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038
10024300372240001806129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006403162229630010000103003830038300383003830038
1002430037225000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006682162229630010000103003830038300383003830038
1002430037224000006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100006402162229630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  sqsub v0.4h, v8.4h, v9.4h
  sqsub v1.4h, v8.4h, v9.4h
  sqsub v2.4h, v8.4h, v9.4h
  sqsub v3.4h, v8.4h, v9.4h
  sqsub v4.4h, v8.4h, v9.4h
  sqsub v5.4h, v8.4h, v9.4h
  sqsub v6.4h, v8.4h, v9.4h
  sqsub 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)03191e1f3f51inst 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)9fa1a8acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005915003004152580108100800081008002050064013202002020039200399977699908012020080032200160064200392003911802011009910010080000100000111511801600200360800001002004020040200402004020040
80204200391500150302580108100800081008002050064013212002020039200399977699908012020080032200160064200392003911802011009910010080000100000111511801600200360800001002004020040200402004020040
80204200391500870302580108100800081008002050064013212002020039200399977699908012020080032200160064200392003911802011009910010080000100000111511801600200360800001002004020040200402004020040
80204200391500240302580108100800081008002050064013202002020039200399977699908012020080032200160064200392003911802011009910010080000100000111511801600200360800001002004020040200402004020040
80204200391500120412580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500150412580100100800001008000050064000002002020039200399973399978010020080000200160000200902003911802011009910010080000100000000511011621200360800001002004020040200402004020040
8020420039150000412580100100800001008000050064082802002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500240412580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
8020420039150000622580100100800001008000050064000002002020039200399973399978022120080000200160000200392003911802011009910010080000100000000511014211200360800001002004020040200402004020040
802042003915000041258010010080000100800005006400000200202003920039997339997801002008000020016000020093200391180201100991001008000010001468000511011611200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03040b18191e1f3a3f51inst 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)5f60696d6edispatch 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)9fa8a9accfd0d5d6daddinst fetch restart (de)e0? int output thing (e9)ea? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200491500000000402580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000050200111601111200360080000102004020040200402004020040
80024200391500000000612580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000350200121601313200360080000102004020040200402004020040
80024200391500000000402580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000050200121601112200360080000102004020040200402004020040
80024200391500000000402580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000050200101601210200360080000102004020040200402004020040
80024200391500000000402580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000050200131601313200360080000102004020040200402004020040
800242003915000000004025800101080000108000050640000112002020039200399996310019800102080000201600002003920039118002110910108000010000502007160812200360080000102004020040200402004020040
80024200391500000000402580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001000050200131601212200361080000102004020040200402004020040
80024200391500000300822580010108000010800005064000001200202003920039999631001980010208000020160000200392003911800211091010800001010350200121601310200360080000102004020040200402004020040
800242003915001068044400402580010108000010800005064166400200202019120347999631004580010208009920160000200392003941800211091010800001012051310144301820202403080000102035620350203492037220199
8002420343153116679261621134258001010800001080000506400000120142203492034110013251012680643208073820160616200392003911800211091010800001010332851090121601213200360080000102045620299205482019420586