Apple M1 Microarchitecture Research by Dougall Johnson

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

SHSUB (vector, 8B)

Test 1: uops

Code:

  shsub v0.8b, 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)03091e3f4e51inst 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)a1a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203716006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203715006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038
1004203716006116872510001000100026468012018203720371572318951000100020002037203711100110000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  shsub v0.8b, 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)0304081e3f4e51inst 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)9fa8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371501006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101431119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500096119687251010012710000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150003278419687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745102632001000020220000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500036119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038

1000 unrolls and 10 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)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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371501500611968725100101010000101000050284768002006520037200371844431876710010201000020200002003720037111002110910101000010000640416221978510000102003820038200382003820038
1002420037150180611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715060611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820086
100242018015000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371502130611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150390611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371501206119687251001010100001010000502847680020018200372003718444121876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715090611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  shsub v0.8b, 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)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)9fa8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371506611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038
102042003715063611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038
1020420037150249431968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010007392162219791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020085200861110201100991001001000010007102162219791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038
1020420037150213611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038
1020420037150195611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010017102162219791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038
10204200371509611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010007102162219791100001002003820038200382003820038

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)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
1002420037186336119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640416341978510000102003820038200382003820038
100242003718606119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640316341978510000102003820038200382003820038
100242003717306119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640316431978510000102003820038200382003820038
1002420037173366119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000664416441978510000102003820038200382003820038
10024200371741386119687251001010100001010000502847680200182003720037184493187671001020100002020000200372003711100211091010100001000640416441978510000102003820038200382003820038
100242003717306119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640416441978510000102003820038200382003820038
1002420037161186119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640316341978510000102003820038200382003820038
100242003716006119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640416341978510000102003820038200382003820038
100242003716106119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640416431978510000102003820038200382003820038
100242003716106119687251001010100001010000502847680200182003720037184443187671001020100002020000200372003711100211091010100001000640416341978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  shsub v0.8b, v8.8b, v9.8b
  shsub v1.8b, v8.8b, v9.8b
  shsub v2.8b, v8.8b, v9.8b
  shsub v3.8b, v8.8b, v9.8b
  shsub v4.8b, v8.8b, v9.8b
  shsub v5.8b, v8.8b, v9.8b
  shsub v6.8b, v8.8b, v9.8b
  shsub v7.8b, 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)03071e3f51inst 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)9fa6a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420038150004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511021611200350800001002003920039200392003920039
8020420038150004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
8020420038150094025801001008000010080000620640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
8020420038150094025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
8020420038150004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
8020420038150064025801001008000010080000500640000201802003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
8020420038150004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511011611200350800001002003920039200392003920039
80204200381500040258010010080000100800005006400002001920038200389973399968010020080000200160000200382003811802011009910010080000100000005110116112003525800001002003920039200392003920039
8020420038150104025801001008000010080000500640000200232003820038997339996801002008000020016000020038200381180201100991001008000010000000511011711200350800001002003920039200392003920039
8020420038150004025801001008000010080000500640000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000511221733200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030407080a0b18191e1f3a3f51inst 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)9fa1a6a8acc2cfitlb miss (d4)d5d6daddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200481501000000000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020027160251220035080000102003920039200392003920039
800242003815010000000003925800101080000108000060640000020019200382003899961910018800102080000201600002003820038118002110910108000010001005020019160241220035080000102003920039200392003920039
80024200381501000000000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020021160271620035080000102003920039200392003920039
80024200381501000000600392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020021160272120035080000102003920039200392003920039
80024200381501000000000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020027160152720035080000102003920039200392003920039
80024200381501000000900392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020013160271520035080000102003920039200392003920039
800242003815010000001200392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020026160152720035080000102003920039200392003920039
80024200381501000000000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020027160152520035080000102003920039200392003920039
800242003815010000000009542580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020013160252720035080000102003920039200392003920039
80024200381501000000000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000005020026160272720035080000102003920039200392003920039