Apple M1 Microarchitecture Research by Dougall Johnson

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

SQRSHL (scalar, H)

Test 1: uops

Code:

  sqrshl h0, h0, h1
  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)03070a3a3f4e51inst 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)a0a1a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372311273625482510001000100039831313018303730372415328951000100020003037303711100110000000077416442630100030383038303830383038
10043037221126825482510001000100039831313018303730372415328951000100020003037303711100110004010377416442630100030383038303830383038
10043037231128925482510001000100039831313018303730372415328951000100020003037303711100110000000377416442630100030383038303830383038
10043037231126825482510001000100039831313018303730372415328951000100020003037303711100110000000077416442630100030383038303830383038
10043037231126825482510001000100039831313018303730372415328951000100020003037303711100110000000077416442630100030383038303830383038
10043037221126825482510001000100039831313018303730372415328951000100020003037303711100110000410077416442630100030383038303830383038
10043037231126825482510001000100039831313018303730372415328951000100020003037303711100110000000077416442630100030383038303830383038
1004303723112682548251000100010003983131301830373037241532895100010002000303730371110011000000011177416442630100030383038303830863038
100430372311215425482510001000100039831313018303730372415328951000100020003037303711100110000000077416442630100030383038303830383038
10043037221126825482510001000100039831313018303730372415328951000100020003037303711100110000000077416442630100030383038303830383038

Test 2: Latency 1->2

Code:

  sqrshl h0, h0, h1
  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)0308090b18191e1f3a3f4e51inst 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)9d9fa0a1a6a7a8a9acc2c5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000000001892954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071041622296340100001003003830038300383003830038
1020430037224000000001702954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
1020430037225000000002792954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
102043003722400000000612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000001100000071021622296340100001003003830038300383003830038
10204300372240000000010282954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
102043003722500000000842954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
102043003722500000000842954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
102043003722500000000842954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
10204300372250000000014092954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038
102043003722400000000842954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000001000000000000071021622296340100001003003830038300383003830038

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)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)9fa0a1a6a7a8a9acc2c5cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000000186295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000168295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000168295482510010101000010100005042773130300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000145295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000166295482510010101000010100005042773130300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000189295482510010101000010100005042773131300183008230037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000189295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038
10024300372250000000210295482510010101001610100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830181
10024300372250000000168295482510010101000010100005042773130300183003730037282873287671001020100002020000300373003711100211091010100001000004030106402162229630010000103003830038300383003830038
10024300372250000000233295482510010101001610100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001000000000006402162229630010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  sqrshl h0, h1, h0
  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)031e3f4e5051inst 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)9fa0a1a8acc2cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722501242954830021251010010010000100100005004278670300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300843003830038
1020430037225061295480251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
1020430037225061295480251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
1020430037225061295480251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
10204300372250612954802510100100100001001000050042773133001830037300372826932874510100200114022322297830515304551111020110099100100100001000000007101161129634100001003003830038300383003830038
102043003722507872954802510100122100001001000050042773133001830037300372826513287451010020010000200200003003730037111020110099100100100001002023007101161229634100001003003830038300383003830038
10204300372251261295390251010010010000100100006494277313300183003730037282653287451026620010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
1020430037225061295480251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000010007101161129634100001003003830038300383003830038
10204300372251261295480251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038
10204300372251261295480251010010010000100100005004277313300183003730037282653287451010020010000200200003003730037111020110099100100100001000000007101161129634100001003003830038300383003830038

1000 unrolls and 10 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)9fa0a1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
1002430037225020729548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010000202000030037300371110021109101010000100000640216222963010000103003830038300383003830038
100243003722506129548251001010100001010000504277313030018300373003728287328767100102010172202000030037300371110021109101010000100000640216222963010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  sqrshl h0, h8, h9
  sqrshl h1, h8, h9
  sqrshl h2, h8, h9
  sqrshl h3, h8, h9
  sqrshl h4, h8, h9
  sqrshl h5, h8, h9
  sqrshl h6, h8, h9
  sqrshl h7, h8, h9
  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)0308090b18191e1f3a3f51inst 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
802042004815000000000940258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511051611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000100511011611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511021611200360800001002004020040200402004020040
802042003915000000000611258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500000000062508010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040
80204200391500000000041258010010080000100800005006400002002020039200399973399978010020080000200160000200392003911802011009910010080000100000000511011611200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030918191e1f3f51inst 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)6061696d6edispatch 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)9fa0a1a6a8a9acc5cfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481500000040258001010800001080000506400000020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000216112003680000102004020040200402004020040
80024200391500000040258001010800001080000506400001020020200392003999963100198001020800002016000020039200391180021109101080000102000000502000216212003680000102004020040200402004020040
80024200391500000040258001010800001080000506400000020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116112003680000102004020040200402004020040
80024200391500000040258001010800001080000506400000020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116222003680000102004020040200402004020040
80024200391500000040258001010800001080000506400001020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116112003680000102004020040200402004020040
8002420039150000001488258001010800001080000506400001020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116112003680000102004020040200402004020040
80024200391500000040258001010800001080000506400001020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116222003680000102004020040200402004020040
80024200391500000040258001010800001080000506400001020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000216222003680000102004020040200402004020040
80024200391500000040258001010800001080000506400000020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116112003680000102004020040200402004020040
800242003915000000124258001010800001080000506400001020020200392003999963100198001020800002016000020039200391180021109101080000100000000502000116112003680000102004020040200402004020040