Apple M1 Microarchitecture Research by Dougall Johnson

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

UZP1 (vector, 2S)

Test 1: uops

Code:

  uzp1 v0.2s, 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)031e3f4e51inst 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
100420371636116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715336116872510001000100026468002018203720371572318951000100020002037203711100110000094116111787100020382038203820852038
100420371606116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715126116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371506116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
10042037161686116872510001000100026468002018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
100420371606116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  uzp1 v0.2s, 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): 2.0037

retire (01)cycle (02)03080b181e1f3f4e51inst 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
1020420037155000002511968725101001001000010010000500284768020055200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715600000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037160000005231968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715500000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100037101161119791100001002003820038200382003820038
102042003715500000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100397101161119791100001002003820038200382003820038
102042013415500000891968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715500000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119833100001002003820038200382003820038
102042003715500000611968725101001001000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037156000006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000247101161119791100001002003820038200382003820038
102042003715500000611968725101001211000010010000500284768020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030409181e1f3f4e51inst 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)9fa1a8a9acbranch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371501000061196872510010101001210100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500006061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371610000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371490000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640224221978510000102003820038200382003820038
10024200371500000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500009061196872510010101000010100005028476801200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038
10024200371500000084196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000001640216221978510000102003820038200382003820038
100242003715500000611968725100101010000101000050284768012001820037200371844426187671001020100002020000200372003711100211091010100001000000640216221978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  uzp1 v0.2s, 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): 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)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)9fa8acbranch mispredict (cb)cfd0d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037155006119687251010010010000100100005002847680002001820037200371842231874510100200100002002000020037200371110201100991001001000010067905071001161119791100001002003820038200382003820038
10204200371551010319687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010000071001161119791100001002003820038200382003820038
10204200371550084196872510100100100001001000050028476800020018200372003718422318745101002001000020020000200372003711102011009910010010000100390071001161119791100001002003820038200382003820038
1020420037155006119687251010010010000100100005002847680002001820037200371842231874510100200100002002000020037200371110201100991001001000010030071001161119791100001002003820038200382003820038
102042003715500494196872510100100100001001000050028476800020018200372003718422318745101002001000020020000200372003711102011009910010010000100300071001161119791100001002003820038200382003820038
10204200371550061196872510100100100001001000050028476800020018200372003718422318745101002001000020020000200372003711102011009910010010000100280071001161119791100001002003820038200382003820038
10204200371550061196872510100100100001001000050028476800020018200372003718422318745101002001000020020000200372003711102011009910010010000100270071001161119791100001002003820038200382003820038
10204200371550061196872510100100100001001000050028476800020018200372003718422318745101002001000020020000200372003711102011009910010010000100530071001161119791100001002003820038200382003820038
10204200371550061196872510100100100001001000050028476800020018200372003718422318745101002001000020020000200372003711102011009910010010000100276071051161119791100001002003820038200382003820038
102042003715500611968725101001001000010010000500284768000200182003720037184223187451010020010000200200002003720037111020110099100100100001004212071001161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030408090b18191e1f3f4e51inst 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)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715500000000891968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010030064021622197850010000102003820038200382003820038
1002420037155000000001031968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010030064021622197850010000102003820038200382003820038
10024200371550000000072619687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100300064021622197850010000102003820038200382003820038
10024200371550000000010319687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100270064021622197850010000102003820038200382003820038
1002420037155000000006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100150064024934200392010000102018020369203732036920180
10024203721560011736726762730196101731010119100481311064552856661020090203712022618467311884111080221016924200002003720037111002110910101000010341700074545723199654010000102027820074202752027820180
10024201321571111256605286119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100120064021622197850010000102003820038200382003820038
1002420037155000000006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100150064021622197850010000102003820038200382003820038
100242003715500000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010060064021622197850010000102003820038200382003820038
100242003715500000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010190064021622197850010000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  uzp1 v0.2s, v8.2s, v9.2s
  uzp1 v1.2s, v8.2s, v9.2s
  uzp1 v2.2s, v8.2s, v9.2s
  uzp1 v3.2s, v8.2s, v9.2s
  uzp1 v4.2s, v8.2s, v9.2s
  uzp1 v5.2s, v8.2s, v9.2s
  uzp1 v6.2s, v8.2s, v9.2s
  uzp1 v7.2s, 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)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)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)9fa0a1a6a8accfd2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420059156184025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511002162120035800001002003920039200392003920039
802042003815609625801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511001161120035800001002003920039200392003920039
802042003815666125801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000050511001161120035800001002003920039200392003920039
802042003815504025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511001161120035800001002003920039200392003920039
802042003815504025801001008000010080000500640000200192003820038997339996801002008000020016000020038200891180201100991001008000010000000511001161120035800001002003920039200392003920039
802042003815504025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000030511001161120035800001002003920039200392003920039
8020420038155060225801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511001161120035800001002003920039200392003920039
802042003815504025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010020000511001161120035800001002003920039200392003920039
802042003815504025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511001161120035800001002003920039200392003920039
802042003815604025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511001161120035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03181e1f3f51inst 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)9fa0a8a9accficache miss (d3)d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420039155000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000350200816532003580000102003920039200392003920039
8002420038155000812580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000050200516642003580000102003920042200392003920172
8002420038155000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000050200516352003580000102003920039200392003920039
8002420038155000952580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010030050200516532003580000102003920039200392003920039
8002420038155000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000050200616532003580000102003920039200392003920039
8002420038155000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000350200316352003580000102003920039200392003920039
8002420038155000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010032350200316462003580000102003920114200392003920039
80024200381550003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100430650200516532003580000102003920039200392003920086
8002420038155000812580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000050200716532003580000102003920039200392003920039
80024200381550003925800101080000108000050640772020019201032008999963100188001020800002016000020038200383180021109101080000100100650200316532003580000102003920039200392011320039