Apple M1 Microarchitecture Research by Dougall Johnson

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

SQDMULL (by element, S)

Test 1: uops

Code:

  sqdmull s0, h0, v1.h[1]
  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)03191e3f4e51inst 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
1004303722008225482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
1004303722006125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
1004303722008225482510001000100039831303018303730372415328951000100020003037303711100110001073116112630100030383085303830383038
1004303723008225482510001000100039831313018303730372415328951000100020003037303711100110002073116112630100030383038303830383038
1004303722006125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
10043037230010525482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
1004303723008225482510001000100039831303018303730842415328951000100020003037303711100110000073116112630100030383038303830383038
1004303723006125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
1004303722006125482510081000100039831303018303730372415328951000100020003037303711100110000373116112630100030383038303830383038
10043037220075254825100010001000398313030183037303724153289510001000200030373037111001100003973116112630100030383038303830383038

Test 2: Latency 1->2

Code:

  sqdmull s0, h0, v1.h[1]
  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)03080a0b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372240003000140392954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038
102043003722400000000612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038
1020430037225000003391320612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038
102043003722400000000612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830087300383008630038
1020430037225000000001032954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383008530038
102043003722500000000612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000080521622296340100001003003830038300383003830038
102043003722400000000612954825101001071000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038
102043003722500000000612954825101001001000010010000500427731313001803003730037282653287451010020010000200200003003730037111020110099100100100001000000000071021622296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e1f3f4e51inst 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)9fa0a8a9acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500002101032954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006404163429630010000103003830038300383003830038
10024300372250000120612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000102103006407164429630010000103003830038300383003830038
10024300372250000001032954825100101010000101000050427731330018300373003728287328767100102210000202000030037300371110021109101010000100000006403164429630010000103003830085300383003830038
10024300372250000120612954825100101010000101000050427731330018300373003728287328786100102010326222097630130301322110021109101010000100000006403164429630010000103003830038300383003830038
100243003722501003006312954844100211010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100123006403243429630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006404164329630010000103003830038300383003830038
10024300372240000005362954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100003006403244429668010000103003830038300383003830038
10024300372250000007262954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006404333429630010000103003830038300383003830038
10024300372250000005362954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006403163429630010000103003830038300383003830038
1002430037225000000612953025100101010000101000050427731330018300373003728287628767100102010000202000030037300371110021109101010000100000006404164429630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  sqdmull s0, h1, v0.h[1]
  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)696b6d6edispatch 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)9fa0a6a8accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722596129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037224576129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
1020430037225156129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
102043003722466129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
10204300372252706129548251010010010000100100005004277313300180300373003728265728763101002001000020020000300373003711102011009910010010000100000007102161129634100001003003830038300383003830038
102043003722596129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129707100001003003830038300383003830038
10204300372254326129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
10204300372254776129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
10204300372252886129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038
10204300372482076129548251010010010000100100005004277313300180300373003728265328745101002001000020020000300373003711102011009910010010000100000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e1f3f4e51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000096061295482510010101000010100005042773131300540300843003728287112880410159221016220203323008530085211002110910101000010000000401506402242229630010000103003830038300383003830038
100243013122611000006129530251001010100001010000504277313030018030037300372828732876710010201000020200003003730037311002110910101000010400000565506403242229630010000103003830038300383003830038
100243003722500000009329548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
100243003722500000006129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010001010006402162229630010000103003830038300383003830038
10024300372250000030006129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000047106129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000045906129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
100243003722500000006129530251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372240000031806129548251001010100001010000504277313030018030037300372828732876710160201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038
10024300372250000025206129548251001010100001010000504277313030018030037300372828732876710010201000020200003003730037111002110910101000010000000006402162229630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  sqdmull s0, h8, v9.h[1]
  sqdmull s1, h8, v9.h[1]
  sqdmull s2, h8, v9.h[1]
  sqdmull s3, h8, v9.h[1]
  sqdmull s4, h8, v9.h[1]
  sqdmull s5, h8, v9.h[1]
  sqdmull s6, h8, v9.h[1]
  sqdmull s7, h8, v9.h[1]
  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)0309191e1f3f51inst 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)9fa0a6a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420058150000023125801001008000010080000500640000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000511021611200360800001002004020040200402004020040
802042003915000004125801001008000010080000500640000200202003920039999039997801002008000020016000020039200391180201100991001008000010000100511015211200360800001002004020040200402004020040
8020420039150002404125801001008000010080000500640000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000511011611200360800001002004020040200402004020040
802042003915000004125801001008000010080000500640000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000511011611200360800001002004020040200402004020040
802042003915000004125801001008000010080000500640000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000511011611200360800001002004020040200402004020040
802042003915000004125801001008000010080000500640000200202003920039997339997801002008000020016000020088200391180201100991001008000010000000511011611200360800001002004020040200402004020040
80204200391500015041258010010080000100800005006400002002020039200399973111002380100200800002001600002003920039118020110099100100800001007204730511012811200360800001002004020040200402004020040
80204200391501001084125801001008000010080000500640000200612003920039997379997801002008000020016000020039200391180201100991001008000010000000511011611200360800001002004020040200402004020040
80204200391500000323258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000005110116112003624800001002004020040200402004020040
802042003915000004125801001008000010080000500640000200202003920039997339997801002008000020016000020039200391180201100991001008000010000000511011611200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03081e3a3f51inst 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)9fa7a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481500240402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050204163320036080000102004020040200402004020114
8002420039155060402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050205164320036080000102004020040200402004020040
8002420039150000402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050203167820036080000102004020040200402004020040
8002420039150000402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050204163420036080000102004020040200402004020040
800242003915004500402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050204165320036080000102004020040200402004020040
8002420039150090402580010108000010800005064000012002020039200399996310019800102080000201600002003920039418002110910108000010000050204164320036080000102004020040200402004020040
8002420039150000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050203164420036080000102004020040200402004020040
8002420039150000402580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000050205163520036080000102004020040200402004020040
8002420039150000402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050203163320036080000102004020040200402004020040
80024200391500660402580010108000010800005064000012002020039200399996310019800102080000201600002003920039118002110910108000010000050204164320036080000102004020040200402004020040