Apple M1 Microarchitecture Research by Dougall Johnson

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

ADDHN (vector, 8H)

Test 1: uops

Code:

  addhn v0.8b, v0.8h, v1.8h
  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
1004303723017025482510001000100039831303018303730372415328951000100020003037303711100110000373116112630100030383038303830383086
100430372308225482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037308411100110000073116112630100030383038303830383038
100430372306125484510001000100039831303054303730842417828951000100020003037303711100110000373116112630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372208225482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430852206125482510001000100039831303018303730372415328951000100020003037303711100110000373116112630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372306125482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038
100430372308225482510001000100039831303018303730372415328951000100020003037303711100110000073116112630100030383038303830383038

Test 2: Latency 1->2

Code:

  addhn v0.8b, v0.8h, v1.8h
  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)0308090b191e1f3f4e51inst 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)9fa1a7a8acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773130300183017930037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
1020430037225000000536295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000006071011611296340100001003008530038300383003830038
102043003722500000061295482510100100100001001000050042773131300183003730037282653287451010020010333200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773130300183003730037282763287451010020010000200200003003730037111020110099100100100001000030071011611296340100001003003830038300383003830230
1020430037225000012061295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000010071011611297790100001003003830038300383003830077
102043003722500000061295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038
102043003722500000061295482510100108100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000030071011611296340100001003003830038300383003830038
102043003722500000061295482510100100100001001000050042773130300183003730037282653287451010020010000200200003003730037111020110099100100100001000000071011611296340100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030809191e3f4e51inst 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006402164229630010000103003830038300383003830038
100243003722510006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
1002430037224000061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000034036402162229630010000103003830038300383003830038
1002430037225010126129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000036402162229701010000103003830038300383003830038
100243003722400006129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500006129548251001010100001010000604277313130018300373003728287328767100102010000202000030037300373110021109101010000100000006402162229630010000103003830038300383003830038
1002430037225000010329548251001010100001010000504277313130018300373017828287328767100102010000202000030037300371110021109101010000100000006402162229630010000103003830038300383003830038
100243003722500006129548251001010100001010000504282741130018300373003728287328767100102010000202000030037300371110021109101010000100002006402162329630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  addhn v0.8b, v1.8h, v0.8h
  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)03081e3f4e51inst 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)9fa6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300873008630038
102043008522500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000522427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830085
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722503612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037225006129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000337101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0318191e3f4e51inst 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)9fa0a6a8accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225000822954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
1002430037225000842954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
1002430084225000842954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
100243003722500010522954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
10024300372250001912954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
1002430037225000612954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
1002430037225000842954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038
10024300372250005412954825100101010000101000050427731303001803003730070282873287671001020100002020000300373003711100211091010100001000036403163329630010000103003830038300383003830038
10024300372250001702954825100101010000101000050427731303001803003730037282873287671001020100002020000300373003711100211091010100001000006403163329630010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  addhn v0.8b, v8.8h, v9.8h
  addhn v1.8b, v8.8h, v9.8h
  addhn v2.8b, v8.8h, v9.8h
  addhn v3.8b, v8.8h, v9.8h
  addhn v4.8b, v8.8h, v9.8h
  addhn v5.8b, v8.8h, v9.8h
  addhn v6.8b, v8.8h, v9.8h
  addhn v7.8b, v8.8h, v9.8h
  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)031e1f3f51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042004915000622580100100800001008000050064000002002002003920039997339997801002008000020016000020039200391180201100991001008000010000051102161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000012002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000002002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000012002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000002002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000012002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000012002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402009020040
8020420039150120412580100100800001008000050064000002002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040
802042003915000412580100100800001008000050064000002002002003920039997339997801002008000020016000020039200391180201100991001008000010040051101161120036800001002004020040200402004020040
8020420039150030412580100100800001008000050064000002002002003920039997339997801002008000020016000020039200391180201100991001008000010000051101161120036800001002004020040200402004020040

1000 unrolls and 10 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)9fa8accfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaebec? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420039150082258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000502000015160001762003600080000102004020040200402004020040
800242003915004025800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100050200006160001762003600080000102004020040200402004020040
800242003915004025800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100050200008160001762003600080000102004020040200402004020040
8002420039150070525800101080000108000050640000120020200392003999963100198001020800002016000020039200391180021109101080000100050200007160006172003600080000102004020040200402004020040
8002420039150040258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000502000061600017172003600080000102004020040200402004020040
800242003915001262580010108000010800005064000002002020039200399996310019800102080000201600002003920039118002110910108000010005020000171620017172003600080000102004020040200402004020040
8002420039150080258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000502000017160008172003600080000102004020040200402004020040
8002420039150040258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000502000017160008172003600080000102004020040200402004020040
8002420039150040258001010800001080000506400000200202003920039999631001980010208000020160000200392003911800211091010800001000502000081600017172003600080000102004020040200402004020040
80024200391500515258001010800001080000506400001200202003920039999631001980010208000020160000200392003911800211091010800001000502000081600017172003600080000102004020040200402004020040