Apple M1 Microarchitecture Research by Dougall Johnson

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

UABDL2 (vector, 2D)

Test 1: uops

Code:

  uabdl2 v0.2d, v0.4s, v1.4s
  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)030708090b18191e1f3f4e51inst 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)a0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004303723000000006125482510001000100039831303018303730372415328951000100020003037303711100110000000000073116132627100030383038303830383038
1004303722000000006125482510001000100039831303018303730372414328951000100020003037303711100110000000000073116132630100030383038303830383038
10043037230000000016225482510001000100039831303018303730372415328951000100020003037303711100110000000000073116122627100030383038303830383038
1004303723100000006125482510001000100039831303018303730372415328951000100020003037303711100110000000000073116222630100030383038303830383038
1004303722000000006125482510001000100039831313018303730372415328951000100020003037303711100110000000000073116112630100030383038303830383038
1004303723000000006125482510001000100039831303018303730372415328951000100020003037303711100110000000000075116122627100030383038303830383038
1004303723100000006125482510001000100039831303018303730372414328951000100020003037303711100110000000000073216112627100030383038303830383038
10043037230000008406125482510001000100039831303018303730372414328951000100020003037303711100110000000000073116112630100030383038303830383038
10043037230000016606125482510001000100039831313018303730372415328951000100020003037303711100110000000000073116112630100030383038303830383038
10043037220000006006125482510001000100039831313018303730372415328951000100020003037303711100110000000000073116112630100030383038303830383038

Test 2: Latency 1->2

Code:

  uabdl2 v0.2d, v0.4s, v1.4s
  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)0318191e3a3f4e51inst 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250000612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296347100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011621296340100001003003830038300383008530038
102043003722400007262954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500002512954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
10204300372250000612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038
102043003722500005362954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000071011611296340100001003003830038300383003830038

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)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)9fa0a1a6a7a8acc2c5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500000006129548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000000006404163429630010000103003830038300383003830038
100243003722500000008429548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006403164329630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006643164429630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006404164429630010000103003830038300383003830038
100243003722500006006129548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006404164429630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006404164429630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287032878610010201000020200003003730037211002110910101000010000000006404164429630010000103003830038300383003830038
100243008422500000006129548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000000006403164429630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006403163429630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018300373003728287032876710010201000020200003003730037111002110910101000010000000006404163429630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  uabdl2 v0.2d, v1.4s, v0.4s
  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)03080b1e3f4e51inst 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)9fa0a1a6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722400061295482510100100100071001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000037101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129705100001003003830038300383003830038
102043008522500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030f1e3f4e51inst 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)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)9fa8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500648295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500420295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500327295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500402295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500145295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500145295482510010101000010100005042773131300183003730037282870328767101612010000202000030037300372110021109101010000100640216222963010000103003830038300383003830038
1002430037225018212295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500482295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722500124295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038
100243003722400189295482510010101000010100005042773131300183003730037282870328767100102010000202000030037300371110021109101010000100640216222963010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  uabdl2 v0.2d, v8.4s, v9.4s
  uabdl2 v1.2d, v8.4s, v9.4s
  uabdl2 v2.2d, v8.4s, v9.4s
  uabdl2 v3.2d, v8.4s, v9.4s
  uabdl2 v4.2d, v8.4s, v9.4s
  uabdl2 v5.2d, v8.4s, v9.4s
  uabdl2 v6.2d, v8.4s, v9.4s
  uabdl2 v7.2d, v8.4s, v9.4s
  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)030b18191e1f3f51inst 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
80204200591500000041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000000511021611200360800001002004020040200402004020040
802042003915000000136258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000003000511011611200360800001002004020040200402009220040
80204200391500000041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000000511011611200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000005110116112003618800001002004020040200402004020040
80204200391500000041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000001000511011611200360800001002004020040200402004020040
80204200391500000041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000001030511011611200360800001002004020040200402004020040
802042003915000000412580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100020000005110116112003614800001002004020040200402004020040
8020420039150000004125801001248000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000104730511011611200360800001002004020040200402004020040
80204200391500000041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000000511011611200360800001002004020040200402004020040
80204200391500000041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000000511022711200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03090b181e3f51inst 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)9fa1a6a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200491500000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050203163520036080000102004020040200402004020040
80024200391500000402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010001050207164720036080000102004020040200402004020040
80024200391500000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050205167520036080000102004020040200402004020040
80024200391500000402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050205165320036080000102004020040200402004020040
80024200391500000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050205165320036080000102004020040200402004020040
80024200391500000402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050203163520036080000102004020040200402004020040
80024200391501000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050203163520036080000102004020040200402004020040
80024200391500000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050203164520036080000102004020040200402004020040
8002420039150000011632580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050204163520036080000102004020040200402004020040
80024200391500000862580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050205163520036080000102004020040200402004020040