Apple M1 Microarchitecture Research by Dougall Johnson

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

UMIN (vector, 2S)

Test 1: uops

Code:

  umin 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)696d6edispatch 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)a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715084168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203716061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715082168725100010001000264680201820372037157231895100010002000203720371110011000573116111787100020382038203820382038
1004203715661168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038
1004203715061168725100010001000264680201820372037157231895100010002000203720371110011000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  umin 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)03080b181e3a3f4e51inst 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)9fa0a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000000821968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000067101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000727101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000997101161119791100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476800200180200372003718422318745101002001000020020000200372003711102011009910010010000100001087101161119791100001002003820038200382003820038
102042003715000000611968725101001001000010010000500284768002001802003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476800200180200372003718422318745101002001000020020000200372003711102011009910010010000100001087101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000607101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000787101161119791100001002003820038200382003820038
1020420037150000006119687251010010010000100100005002847680020018020037200371842231874510100200100002002000020037200371110201100991001001000010000877101161119791100001002003820038200382003820038
10204200371500000061196872510100100100001001000050028476800200180200372003718422318745101002001000020020000200372003711102011009910010010000100001207101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03090b0e191e3f4e51inst 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)9fa0a1a6a8a9acc2c5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000200611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000500006402162219785010000102003820038200382003820038
100242003714900000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000100006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000100006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000203006402162219785010000102003820038200382003820038
10024200371500000061196872510010101000010100005028476800200182003720037184443187671001020100002020000200372003711100211091010100001000000207006402162219785010000102003820038200382003820038
1002420037150000006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100000096006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000100016402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000009006402162219785010000102003820038200382003820038
100242003715000000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000103006402162219785010000102003820038200382003820038
100242003715100000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  umin 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)03181e3f4e51inst 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)9fa8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010011807101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500361196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001001307101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001001607101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001001007101161119791100001002003820038200382003820038
10204200371500061196872510100100100121251015250028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000007101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184223187451010020010000200200002003720037111020110099100100100001002007101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001002007101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001002007101161119791100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000017101161119791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03081e1f3f4e51inst 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
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010200640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010260640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010100640216221978510000102003820038200382003820038
1002420037150000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371500006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000106000640216221978510000102003820038200382003820038
10024200371500007261968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010530640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  umin v0.2s, v8.2s, v9.2s
  umin v1.2s, v8.2s, v9.2s
  umin v2.2s, v8.2s, v9.2s
  umin v3.2s, v8.2s, v9.2s
  umin v4.2s, v8.2s, v9.2s
  umin v5.2s, v8.2s, v9.2s
  umin v6.2s, v8.2s, v9.2s
  umin 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)0318191e3f51inst 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
80204200591500004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001005157051102161120035800001002003920039200392003920039
80204200381500004025801001008000010080000500640000020019200382003899733999680100200801362001600002003820038118020110099100100800001004593051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010042147051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010035105051101161120035800001002003920039200392003920039
80204200381500004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001004845051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010010051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100521406051101161120035800001002003920039200392003920039
80204200381510004025801001008000010080000500640000020019200382003899733999680100200800002001600002003820038118020110099100100800001005090051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010026051101161120035800001002003920039200392003920039
8020420038150000402580100100800001008000050064000012001920038200389973399968010020080000200160000200382003811802011009910010080000100493051101161120035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0304090b18191e1f3f51inst 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)9fa0a1a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242004815000000003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100000030502018169420035080000102003920039200392003920039
80024200381500000000392580010108000010800005064000002006820038200389996310018800102080000201600002003820038118002110910108000010000000050205168420035080000102003920039200392003920039
800242003815000000003925800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100001000502051651320035080000102003920039200392003920039
8002420038150000000028525800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100001000502091613920035080000102003920039200392003920039
800242003815000000903925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100000000502081613520035080000102003920039200392003920039
800242003815000000003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100000000502010164820035080000102003920039200392003920039
80024200381500000000392580010108000010800005064000012001920038200389996310018800102080000201600002003820038118002110910108000010000000050203169520035080000102003920039200392003920039
800242003815000000003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100000000502041651320035080000102003920039200392003920039
800242003815000000006025800101080000108000050640000020019200382003899963100188001020800002016000020038200381180021109101080000100000000502051641220035080000102003920039200392003920039
8002420038150000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000000005020131612820035080000102003920039200392003920039