Apple M1 Microarchitecture Research by Dougall Johnson

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

DUP (element, vector, 8H)

Test 1: uops

Code:

  dup v0.8h, v0.h[1]
  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)031e3a3f4e51inst 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
1004203715026616862510001000100026452102018203720371571318951000100010002037203711100110000077416541786100020382038203820382038
1004203715028916862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
1004203715026616862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
1004203715026616862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
1004203715028716862510001000100026452102018203720841571318951000100010002037203711100110000077416441786100020382038203820382038
1004203716026616862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
10042037150261216862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
1004203715026616862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
1004203715026616862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038
1004203715026616862510001000100026452102018203720371571318951000100010002037203711100110000077416441786100020382038203820382038

Test 2: Latency 1->2

Code:

  dup v0.8h, v0.h[1]
  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)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)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)9fa0a1a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000004206119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
102042003715000003306119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
102042003715000001206119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
102042003715000000061196862510100100100001001000050028475212001820037200371842131874510268200100002001000020037200371110201100991001001000010000801571021622197910100001002003820038200382003820038
102042003715000001506119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
102042003715000002406119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
102042003715000001806119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
10204200371500000006119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038
102042003715000001806119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000100071021622197910100001002003820038200382003820038
10204200371500000006119686251010010010000100100005002847521200182003720037184213187451010020010000200100002003720037111020110099100100100001000000071021622197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03080b18191e1f3a3f4e51inst 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)696b6d6edispatch 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500000720061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006405162219786010000102003820038200382003820038
1002420037150000000061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
1002420037150000000061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
10024200371500000270061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
1002420037150000400061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
1002420037150000000061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
1002420037150000000061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000010006402162219786010000102003820038200382003820038
100242003715000004530061196862510010101000010100005028475212001832003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
1002420037150000000061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038
10024200371500000150061196862510010101000010100005028475212001802003720037184433187671001020100002010000200372003711100211091010100001000000006402162219786010000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  dup v0.8h, v8.h[1]
  dup v1.8h, v8.h[1]
  dup v2.8h, v8.h[1]
  dup v3.8h, v8.h[1]
  dup v4.8h, v8.h[1]
  dup v5.8h, v8.h[1]
  dup v6.8h, v8.h[1]
  dup v7.8h, v8.h[1]
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)03080b18191e3f51inst 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)9fa1a7a8a9acc5branch mispredict (cb)cdcfd5d6e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420059150000026729258010810080008100800205006401320200192003820038997769989801202008003220080032200382003811802011009910010080000100022240111511821620035800001002003920039200392003920039
8020420038150000002925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
8020420038150000002925801081008000810080020500640132120019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
8020420038150000002925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
8020420038150000002925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
802042003815000002732925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010020100111511801620035800001002003920039200392003920039
8020420038150000062925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
8020420038150000002925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
8020420038150000002925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039
80204200381500000212925801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000000111511801620035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03191e1f3f51inst 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)9fa8a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0eaeb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420050150000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502071635200350080000102003920039200392003920039
8002420038150000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502061635200350080000102003920039200392003920039
8002420038150000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502051653200350080000102003920039200392003920039
8002420038150000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502051654200350080000102003920039200392003920039
8002420038155000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502051635200350080000102003920039200392003920039
8002420038150000392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502051635200350080000102003920039200392003920039
80024200381500180392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502051653200350080000102003920039200392003920039
8002420038150000392580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000100000502031653200350080000102003920039200392003920039
8002420038150000392580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000100000502051636200350080000102003920039200392003920039
8002420038150090392580010108000010800005064000002001920038200389996310018800102080000208000020038200381180021109101080000100000502051655200350080000102003920039200392003920039