Apple M1 Microarchitecture Research by Dougall Johnson

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

SHA512SU0

Test 1: uops

Code:

  sha512su0 v0.2d, v1.2d
  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)03071e3f4e51inst 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)a0a1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203716006193725100010001000689840201820372037178831895100010002000203720371110011000000073116111969100020382038203820382038
1004203715006193725100010001000689840201820372037178831895100010002000203720371110011000000073116111969100020382038203820382038
1004203715006193725100010001000689841201820372037178831895100010002000203720371110011000000075216111969100020382038203820382038
1004203715006193725100010001000689841201820372037178831895100010002000203720371110011000000073116111969100020382038203820382038
1004203715106193725100010001000689841201820372037178731895100010002000203720371110011000000073116111969100020382038203820382038
1004203715006193725100010001000689841201820372037178731895100010002000203720371110011000000073116111969100020382038203820382038
1004203716006193725100010001000689840201820372037178831895100010002000203720371110011000000073116111969100020382038203820382038
1004203715006193725100010001000689841201820372037178731895100010002000203720371110011000000073116111969100020382038203820382038
1004203715006193725100010001000689841201820372037178831895100010002000203720371110011000000073116111969100020382038203820382038
1004203715008293725100010001000689840201820372037178831895100010002000203720371110011000000073116111969100020382038203820382038

Test 2: Latency 1->1

Code:

  sha512su0 v0.2d, v1.2d
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)033f4e51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420054200372003718638318745101002001000020020000200372003711102011009910010010000100107102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101482001000020020000200372003711102011009910010010000100107102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100107102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038
10204200371506199372510100100100001001000050070798420018200372003718638318745101002001000020020000200372003711102011009910010010000100007102162219967100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)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)9fa1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000000084993725100101010000101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100040006402162219969010000102003820038200382003820038
100242003715000000061993725100101010000101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100000006402162219969010000102003820038200382003820038
100242003715000000061993725100101010000111000050707984120018200372003718660318767100102010000202000020037200371110021109101010000100000306402162220005010000102003820038200382003820038
100242003715000000061993725100101010000101000050707984120018200372003718660318767100102010000202000020037200371110021109101010000100000006402162319969010000102003820038200382003820038
100242003715000000061993725100101010000101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100010006402162219969010000102003820038200382003820038
100242003715000000061993725100101010000101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100000306402162219969010000102003820038200382003820038
100242003715000000061993725100101010000101000050707984120018200372003718660318767100562010000202000020037200371110021109101010000100000306542162219969010000102003820038200382008420038
100242003715000100061993725100371010012101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100200006402162219969010000102003820038200382003820038
1002420037150000000103993725100101010000101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100000006402162219969010000102023120038200382003820038
100242003715000000061993725100101010000101000050707984020018200372003718660318767100102010000202000020037200371110021109101010000100000006402162219969010000102003820038200382003820038

Test 3: Latency 1->2

Code:

  sha512su0 v0.2d, v0.2d
  movi v0.16b, 1

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03181e3a3f4e51inst 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)9fa8acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715040061993625101261001000010010000500707921120018200372003718644718740101002001000820020016200372003711102011009910010010000100001117170160019978100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921020018200372003718644718741101002001000820020016200372003711102011009910010010000100001117180160019978100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921120018200372003718644718741101002001000820020016200372003711102011009910010010000100001117170160019979100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921020018200372003718644718741101002001000820020016200372003711102011009910010010000100001117170160019979100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921120018200372003718644718740101002001000820020016200372003711102011009910010010000100001117180160019979100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921020018200372003718644618741101002001000820020016200372003711102011009910010010000100001117180160019979100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921120018200372003718644618740101002001000820020016200372003711102011009910010010000100001117180160019978100001002003820038200382003820038
102042003715000061993625101001001000010010000500707921120018200372003718644618740101002001000820020016200372003711102011009910010010000100001117170160019979100001002003820038200382003820038
1020420037150000652993625101001001000010010000500707921020018200372003718644718741101002001000820020016200372003711102011009910010010000100001117170160019978100001002003820038200382003820038
102042003714900061993625101001001000010010000500707921020018200372003718644618740101002001000820020016200372003711102011009910010010000100001117170160019979100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150017099362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
1002420037150978599362510010101000010100005070792112001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
100242003715006199362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
100242003715006199362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
100242003715006199362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010002714640216221996810000102003820038200382003820038
100242003715008799362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
100242003715006199362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
100242003715006199362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
1002420037150017099362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038
1002420037150075899362510010101000010100005070792102001820037200371865931876710010201000020200002003720037111002110910101000010000640216221996810000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  movi v0.16b, 0
  sha512su0 v0.2d, v8.2d
  movi v1.16b, 0
  sha512su0 v1.2d, v8.2d
  movi v2.16b, 0
  sha512su0 v2.2d, v8.2d
  movi v3.16b, 0
  sha512su0 v3.2d, v8.2d
  movi v4.16b, 0
  sha512su0 v4.2d, v8.2d
  movi v5.16b, 0
  sha512su0 v5.2d, v8.2d
  movi v6.16b, 0
  sha512su0 v6.2d, v8.2d
  movi v7.16b, 0
  sha512su0 v7.2d, v8.2d
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 1.0005

retire (01)cycle (02)03080b18191e1f3a3f51inst 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)9fa7a8a9acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020480039599000000046525801001008000010080000500640000080167800388003859970659991801002008000820016001680038800381116020110099100100160000100000011110117016008003501600001008003980039800398003980039
1602048003859900000003825801001008000010080000500640000180019800388003859970659991801002008000820016001680038800381116020110099100100160000100000011110117016008003501600001008003980039800398003980039
160204800385990000120097525801001008000010080000500640000080019800388003859970659991801002008000820016001680038800381116020110099100100160000100000011110117016008003501600001008003980039800398003980039
1602048003860000000008025801001008000010080000500640195180019801828003859970659991801252008011620016008880134800386116020110099100100160000100000011110117016008003501600001008003980039800398003980039
1602048003859900000003825801001008000010080000500640000080019800388003859970659991801002008000820016001680038800381116020110099100100160000100050311110117016008003501600001008003980039800398003980039
16020480038600000000038258010010080023100800005006400001800198003880038599706599918010020080008200160016800388003811160201100991001001600001000210011110117016008003501600001008003980039800398003980039
1602048003860000000003825801001008000010080000500640000080019800388003859970659991801002008000820016001680038800381116020110099100100160000100000011110117016008003501600001008003980039800398003980039
1602048003859900000003825801001008000010080000500640000080019800388003859970659991801002008000820016001680038800381116020110099100100160000100000011110117016008003501600001008003980039800398003980039
1602048003860000001260074525801001008000010080050500640000080019800388008759961959978802002008001120016008880135800381116020110099100100160000100000022210127123118003501600001008003980039800398003980039
1602048003859900000007143801231008000010080000500640000180019800388003859961959979801002028001120016002280038800384116020110099100100160000100000022210128023118003501600001008003980039800398003980039

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 1.0005

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)5f60696b6d6edispatch 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)9fa1a8acc5cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1600248003959900055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022354341621123800351510160000108003980039800398003980039
1600248003859900055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022345331721132800351520160000108003980039800398003980039
1600248003859900055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022352331621133800351510160000108003980039800398003980097
1600248003859900055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022346321621133800351510160000108003980039800398003980039
1600248003860000055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022348321621132800351520160000108003980039800398003980039
16002480038599000196258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022348331621133800351510160000108003980039800398003980039
1600248003860000055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022347331621123800351510160000108003980039800398003980039
1600248003859900055258001010800231080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022349341621123800351510160000108003980039800398003980039
1600248003859900055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022348321621143800351510160000108003980039800398003980039
1600248003860000055258001010800001080000506400001180019080038800385998636001880010208000020160000800388003811160021109101016000010000010022348331622133800351510160000108003980039800398003980039

Test 5: throughput

Count: 16

Code:

  sha512su0 v0.2d, v16.2d
  sha512su0 v1.2d, v16.2d
  sha512su0 v2.2d, v16.2d
  sha512su0 v3.2d, v16.2d
  sha512su0 v4.2d, v16.2d
  sha512su0 v5.2d, v16.2d
  sha512su0 v6.2d, v16.2d
  sha512su0 v7.2d, v16.2d
  sha512su0 v8.2d, v16.2d
  sha512su0 v9.2d, v16.2d
  sha512su0 v10.2d, v16.2d
  sha512su0 v11.2d, v16.2d
  sha512su0 v12.2d, v16.2d
  sha512su0 v13.2d, v16.2d
  sha512su0 v14.2d, v16.2d
  sha512su0 v15.2d, v16.2d
  movi v16.16b, 17

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 1.0002

retire (01)cycle (02)0318191e1f3f51inst 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)9fa6a8a9acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16020416003911990008838251601001001600001001600005001280000016001916003816008613997106139992160100200160008200320016160038160038111602011009910010016000010000001111011721621160035160000100160039160039160039160039160039
1602041600381199000038251601001001600001001600005001280000116001916003816003813997106139992160100200160008200320016160038160038111602011009910010016000010000031111011721622160035160000100160039160039160039160039160039
16020416003811990000418381601001001600001001600005001280000116001916003816003813999406139992160100200160008200320016160038160038111602011009910010016000010000001111011721621160035160000100160039160039160039160039160039
1602041600381199010038251601001001600001001600005001280000116001916003816003813997106139992160100200160008200320016160038160038111602011009910010016000010000001111013121622160035160000100160039160039160039160039160039
16020416003811980000608251601001001600001001600005001280000116001916003816003813997106139992160100200160008200320016160038160038111602011009910010016000010000001111011721621160035160000100160039160039160039160039160039
1602041600381199000038251601001001600001001600005001280000016001916003816003813997106139992160100200160008200320016160038160038111602011009910010016000010000001111011721612160035160000100160039160039160039160039160039
16020416003811990000755251601001001600001001600005001280000116001916003816003813997106139992160100200160008200320016160085160038111602011009910010016000010001001111011711612160035160000100160039160039160039160039160039
160204160038119900001986251601001001600001001600005001280000016001916003816003813997106139992160100200160008200320016160038160038111602021009910010016000010000001111011721612160035160000100160039160039160039160039160039
160204160038119900270382516010010016000010016000050012800000160019160038160038139971016139992160100200160008200320016160038160038111602011009910010016000010022210981111013121612160035160000100160039160039160039160039160039
16020416003811990012038251601001001600921001600005001280585116001916003816003813997106139992160100200160008200320016160038160038111602011009910010016000010002001111011721621160035160000100160039160039160086160039160039

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 1.0002

retire (01)cycle (02)0318191e1f3f51inst 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)5f6061696d6edispatch 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)9fa8acc5cfd0d2icache miss (d3)d5d6d9dadbddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
16002416003811990000552516001010160016101600005012800001101600191600381600381399863140018160010201600002032000016003816003811160021109101016000010032701002231113161111021160082201616000010160039160087160039160039160039
160024160038119800007202516001010160000101600005012800001101600191600381600381399863140018160010201600002032000016003816003811160021109101016000010031501002221112116212921160033201616000010160039160039160039160039160039
1600241600381198000061251600101016000010160000501280000010160019160038160038139986314001816001020160000203200001600381600381116002110910101600001023480100243119164221019160033403216000010160039160039160039160039160039
1600241600381199000061251600101016000010160000501280000110160019160038160038139986314001816001020160000203200001600381600381116002110910101600001003810100243112116211921160033203216000010160039160039160039160039160039
1600241600381200000061251600101016000010160000501280000010160019160038160038139986314001816001020160000203200001600381600381116002110910101600001003990100226212116442721160033201616000010160039160039160039160039160039
1600241600381199000055251600101016000010160000501280000010160019160038160038139986314001816001020160000203200001600381600381116002110910101600001003450100223118162112217160033401616000010160039160039160039160039160039
160024160038119900005525160010101600001016000050128000011016001916003816003813998631400181600102016000020320000160038160038111600211091010160000100211100223112216211228160033201716000010160039160039160039160039160039
16002416003811990000720251600101016000010160000501280000110160067160038160038139986314001816001020160000203200001600381600381116002110910101600001002760100223122216211719160033201616000010160039160039160039160039160039
16002416003811990015005525160010101600001016000050128000011016001916003816003813998631400181600102016000020320000160038160038111600211091010160000100210100243222116212919160033401616000010160039160039160039160039160085
160024160038120000005302516001010160000101600005012800000101600191600381600381399863140018160010201600002032008616003816008611160021109101016000010014101002431121162122510160033201616000010160039160039160039160039160039