Apple M1 Microarchitecture Research by Dougall Johnson

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

SADDL (vector, 8H)

Test 1: uops

Code:

  saddl v0.8h, v0.8b, v1.8b
  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)033f4e51inst 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)a1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100420371561168725100010001000264680120182037203715723189510001000200020372037111001100000373116111787100020382038203820382038
100420371561168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
100420371661168725100010001000264680120182037203715723189510001000200020372037111001100000373116111787100020382038203820382038
100420371561168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
100420371661168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
100420371661168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
100420371561168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037156116872510001000100026468012018203720371572318951000100020002037203711100110006022073116111787100020382038203820382038
100420371561168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
100420371561168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  saddl v0.8h, v0.8b, v1.8b
  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)031e1f3f4e51inst 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)9fa0a1a6a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150006119687251010010010000100100005002847680120018020037200371842231874510100200100002002000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680120018020037200371842231874510100200100002002000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000070007101161119791100001002003820038200382003820038
1020420037150006119687431011810010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000000007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000010007101161119791100001002003820038200382003820038
1020420037150006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010002010007101161119791100001002003820038201812003820038
1020420037150006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000030007101161119791100001002003820038200382003820038
1020420037150008219687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000000007351161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)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)9fa8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001012000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216421978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768020018200372003718444318767100102010000202000020037200371110021109101010000100000640216221978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  saddl v0.8h, v1.8b, v0.8b
  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)030f1e3f4e51inst 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)9fa8acc5cdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010003007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007412162219791100001002003820038200382003820038
102042003715003298196872510100100100001001000050028476802005420037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010010007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102162219791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476802001820037200371842231874510100200100002002000020037200371110201100991001001000010000007102165219791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

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)9fc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500266196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001006448161881978510000102003820038200382003820038
1002420037150026619687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100644111612111978510000102003820038200382003820038
10024200371500266196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001006447168101978510000102003820038200852003820038
1002420037150926619687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100644111610101978510000102003820038200382003820038
1002420037150026619687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100644101610101978510000102003820038200382003820038
10024200371500273119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100644101610101978510000102003820038200382003820038
10024200371500266196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001006445161051978510000102003820038200382003820038
100242003715002661968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010064411161081978510000102003820038200382003820038
100242003715002661968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010064411161181978510000102003820038200382003820038
10024200371500254119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100644101612101978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  saddl v0.8h, v8.8b, v9.8b
  saddl v1.8h, v8.8b, v9.8b
  saddl v2.8h, v8.8b, v9.8b
  saddl v3.8h, v8.8b, v9.8b
  saddl v4.8h, v8.8b, v9.8b
  saddl v5.8h, v8.8b, v9.8b
  saddl v6.8h, v8.8b, v9.8b
  saddl v7.8h, v8.8b, v9.8b
  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)030b18191e1f3f51inst 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)9fa0a6a8a9accdcfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005815100078040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051105164320035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051103163420035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051104163420035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051103164320035800001002003920039200392003920039
80204200381500003040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051103164420035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051104164320035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051103164420035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051104164420035800001002003920039200392003920039
80204200381500000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051104164320035800001002003920039200392003920039
8020420038150000474040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000051103163420035800001002003920039200392003920039

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)9fa8accficache miss (d3)d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420048150039258001010800001080000506400001200192003820038999631001880010208000020160000200382003811800211091010800001000502001161120035180000102003920039200392003920039
800242003815003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000101468502001161120035080000102003920039200392003920039
8002420038150159392580010108000010800005064000002011820038200381000431001880010208000020160000200382003821800211091010800001000502003161120035080000102003920039200392003920039
80024200381500146258001010800001080000506400001200192003820038999661001880010208000020160000200382003811800211091010800001000502001162120035080000102003920039200392003920039
8002420038150039258010410800001280097506400001200192003820038999631001880010208000020160590200382003811800211091010800001000502001161120035080000102003920039200392003920039
8002420038150039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000502001161120035080000102003920039200392003920039
80024200381507539258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000502001161120035080000102003920039200392003920039
8002420038150255271258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001020502001161120035080000102003920039200392003920039
8002420038150039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000502001161120035080000102003920039200392003920039
80024200381502739258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000502001161120035080000102003920039200392003920039