Apple M1 Microarchitecture Research by Dougall Johnson

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

SABDL (vector, 2D)

Test 1: uops

Code:

  sabdl v0.2d, v0.2s, v1.2s
  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)031e1f3f4e51inst 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)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10043037233061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037220061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
100430372300126254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037230061254825100010001000399670130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
100430372200156254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037220061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038
10043037230061254825100010001000398313130183037303724153289510001000200030373037111001100073216222630100030383038303830383038

Test 2: Latency 1->2

Code:

  sabdl v0.2d, v0.2s, v1.2s
  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)030b1e3f4e51inst 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)9fa6a8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722400612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510267200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037225007492954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
102043003722500612954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038
1020430037224003462954825101001001000010010000500427731303001830037300372826532874510100200100002002000030037300371110201100991001001000010000007101161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03191e3f4e51inst 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)9fa8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722500612954825100101010000101000050427731313001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225021612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722400612954825100101010000101000050427731303001830037300372828732876710010201016320200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
1002430037225007262954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038
100243003722500612954825100101010000101000050427731303001830037300372828732876710010201000020200003003730037111002110910101000010000640216222963010000103003830038300383003830038

Test 3: Latency 1->3

Code:

  sabdl v0.2d, v1.2s, v0.2s
  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)9fa0a1a6a8a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722506129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000006007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010710000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000020013507101161129634100001003003830038300383003830038
102043003722506129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000009007101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000004201207101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000000307101161129634100001003003830038300383003830038
1020430037225061295482510100100100001001000050042773131300183003730037282653287451010020010000200200003003730037111020110099100100100001000000018007101161129634100001003003830038300383003830038
10204300372250612954825101001001000010010000500427731313001830037300372826532874510100200100002002000030037300371110201100991001001000010000000007101161129634100001003003830038300383003830038
102043003722506129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000007207101161129634100001003003830038300383003830038
102043003722506129548251010010010000100100005004277313130018300373003728265328745101002001000020020000300373003711102011009910010010000100000650007102161129634100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)031e3a3f4e51inst 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)9fa8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037225006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000101000640216222963010000103003830038300383003830038
1002430037225006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100030640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001051090640216222963010000103003830038300383003830038
1002430037225006129548251001010100001010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000100000640116222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001034090640216222963010000103003830038300383003830038
10024300372240061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001031030640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001020660640216222963010000103003830038300383003830038
10024300372250061295482510010101000010100005042773131300183003730037282873287671001020100002020000300373003711100211091010100001052030640216222963010000103003830038300383003830038
1002430037225009429548251001010100081010000504277313130018300373003728287328767100102010000202000030037300371110021109101010000103000640216222963010000103003830038300383003830038
1002430037225006129548251001010100081010000504277313130018300373003728287328789100102010000202000030037300371110021109101010000104030640216222963010000103003830038300383003830038

Test 4: throughput

Count: 8

Code:

  sabdl v0.2d, v8.2s, v9.2s
  sabdl v1.2d, v8.2s, v9.2s
  sabdl v2.2d, v8.2s, v9.2s
  sabdl v3.2d, v8.2s, v9.2s
  sabdl v4.2d, v8.2s, v9.2s
  sabdl v5.2d, v8.2s, v9.2s
  sabdl v6.2d, v8.2s, v9.2s
  sabdl v7.2d, v8.2s, v9.2s
  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)03081e3f51inst 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
8020420039150003025801081008000810080020500640132020020200392003999776999080120200800322001600642003920039118020110099100100800001000011151182160020036800001002004020040200402004020040
8020420039150003025801081008000810080020500640132020020200392003999776999080120200800322001600642003920039118020110099100100800001002000051101161120036800001002004020040200402004020040
8020420039150004125801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001001300051101161120036800001002004020040200402004020040
8020420039150064125801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001004616200051101161120036800001002004020091200402004020040
8020420039150002312580100100800001008000050064000002002020039200399973399978010020080000200160000200392003911802011009910010080000100012900051101161120036800001002004020040200402004020040
8020420039150004125801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000051101161120036800001002004020040200402004020040
80204200391500041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010018000051101161120036800001002004020040200402004020040
80204200391500041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010025000051101161120036800001002004020040200402004020040
8020420039150004125801001008000010080000500640000020020200392003999733999780100200800002001600002003920039118020110099100100800001000000051101171120036800001002004020040200402004020040
80204200391500041258010010080000100800005006400000200202003920039997339997801002008000020016000020039200391180201100991001008000010002400051101161120036800001002004020040200402004020040

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)5f60696d6edispatch 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
800242004815004025800101080000108000050640000012002020039200399996310019800102080000201600002003920039118002110910108000010105020116112003680000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996310019800102080000201600002003920039118002110910108000010805020116112003680000102004020040200402004020040
80024200391500402580010108000010800005064000000200202003920039999631001980010208000020160000200392003911800211091010800001001355020116112003680000102004020040200402004020040
8002420039150040258001010800001080000506400000120020200392003999963100198001020800002016000020039200391180021109101080000100695020116112003680000102004020040200402004020040
800242003915004025800101080000108000050640000002002020039200399996310019800102080000201600002003920039118002110910108000010135020116112003680000102004020040200402004020040
8002420039150040258001010800001080000506400000120020200392003999963100198001020800002016000020039200391180021109101080000102905020116112003680000102004020040200402004020040
800242003915004025800101080000108000050640000012002020039200399996310019800102080000201600002003920039118002110910108000010235020116112003680000102004020040200402004020040
800242003915004025800101080000108000050640000012002020039200399996310019800102080000201600002003920039118002110910108000010005020116112003680000102004020040200402004020040
800242003915004025800101080000108000050640000012002020039200399996310019800102080000201600002003920039118002110910108000010235020116112003680000102004020040200402004020040
8002420039150040258001010800001080000506400000120020200392003999963100198001020800002016000020039200391180021109101080000100935020116112003680000102004020040200402004020040