Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRDMULH (by element, 8H)

Test 1: uops

Code:

  sqrdmulh v0.8h, v0.8h, 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)03191e1f3f4e51inst 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
10043037220006125482510001000100039831313018303730372415328951000100020003037303711100110000373116112630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100020003037303711100110000973116112630100030383038303830383038
100430372300025125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
10043037230006125482510001000100039831313018303730372415328951000100020003037303711100110003073116112630100030383038303830383038
100430372300010625482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
10043037220006125482510001000100039831313018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
10043037280006125482510001000100039831313018303730372415328951000100020003037303711100110004073116112630100030383038303830383038
100430372300061254825100010001000398313130183037303724153289510001000200030373037111001100013673116112630100030383038303830383038
100430372302106125482510001000100039831313018303730372415328951000100020003037303711100110003973116112630100030383038303830383038

Test 2: Latency 1->2

Code:

  sqrdmulh v0.8h, v0.8h, 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)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)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)9fa0a1a6a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250145295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250440295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250401295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722501604295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250437295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250896295302510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250422295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372250447295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
10204300372240465295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722501047295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038

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)9fa0a1a6a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000822954825100101010016131000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100002006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328788100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103013430038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101000050427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000000612954825100101010000101014850427731330018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  sqrdmulh v0.8h, v1.8h, 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)0308181e3f4e51inst 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)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)9fa1a6a8a9acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000237295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000438295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
10204300372250001492954825101001001000010010000500427731303001803003730037282650328745101002001000020020000300373003711102011009910010010000100001030710116112963417100001003003830038300383003830038
1020430037225000425295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071021611296340100001003003830038300383003830038
1020430037225000890295482510100100100001001000050042773130300180300373003728265732874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
102043003722500061295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000200071011611296340100001003003830038300383003830038
1020430037225000409295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000386295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000399295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038
1020430037225000386295482510100100100001001000050042773130300180300373003728265032874510100200100002002000030037300371110201100991001001000010000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030818191e1f3f4e51inst 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)9fa6a7a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500090102229548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250000012829530251001010100001010000504277313130018300373003728287032878310010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372240000042629548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250000031329548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372320000046229548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250000036129548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250000097729548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250000052529548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372240000023129548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038
10024300372250000095729548251001010100001010000504277313130018300373003728287032876710010201000020200003003730037111002110910101000010000006402162229630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  sqrdmulh v0.8h, v8.8h, v9.h[1]
  sqrdmulh v1.8h, v8.8h, v9.h[1]
  sqrdmulh v2.8h, v8.8h, v9.h[1]
  sqrdmulh v3.8h, v8.8h, v9.h[1]
  sqrdmulh v4.8h, v8.8h, v9.h[1]
  sqrdmulh v5.8h, v8.8h, v9.h[1]
  sqrdmulh v6.8h, v8.8h, v9.h[1]
  sqrdmulh v7.8h, v8.8h, 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)030b1e1f3a3f51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042004815000001272580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051104162320036800001002004020040200402004020040
80204200391500000412580100100800001008000050064000002002020039200399973039997801002008000020016000020039200391180201100991001008000010000051102162320036800001002004020040200402004020040
80204200391500000412580100100800001008000050064000002002020039200399973039997801002008000020016000020039200391180201100991001008000010000051102163320036800001002004020040200402004020040
802042003915000003262580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051102162320036800001002004020040200402004020040
80204200391500000412580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051103163220036800001002004020040200402004020040
80204200391500000622580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051102162320036800001002004020040200402004020040
80204200391500000412580100100800001008000050064000002002020039200399973039997801002008000020016000020039200391180201100991001008000010000051103163320036800001002004020040200402004020040
802042003915500001252580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051101163220036800001002004020040200402004020040
80204200391500000412580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051103162320036800001002004020040200402004020040
80204200391500600622580100100800001008000050064000012002020039200399973039997801002008000020016000020039200391180201100991001008000010000051103163320036800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030818191e1f3a3f51inst 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)9fa6a8a9acc2cfd0d5d6ddinst fetch restart (de)dfe0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481550000009662580012128000012800006064000002002020039200399996310019800122080000201600002003920039118002110910108000010000005020011613020036080000102004020040200402004020040
80024200391500000009692580010108000012800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005020011611020036080000102004020040200402004020244
80024200391500000017322580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005022011611020036080000102004020040200402004020040
800242003915000000012132580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005020011611020036080000102004020040200402004020040
8002420039150010000402580012128000010800005064000002002020039200399996310019800122080000201600002003920039118002110910108000010000005020011611020036080000102004020040200402004020040
8002420039150000000612580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005020011611220036080000102004020040200402004020040
80024200391500000001742580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005020011611020036080000102004020040200402004020040
800242003915000000012232580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005020011621220036080000102004020040200402004020040
800242003915000000011112580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010000005022011611020036080000102004020040200402004020040
800242003915000000011732580010108000010800005064000012005820039200399996310019800102080000201600002003920039118002110910108000010000005020011611020036080000102004020040200402004020040