Apple M1 Microarchitecture Research by Dougall Johnson

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

USUBW (vector, 2D)

Test 1: uops

Code:

  usubw v0.2d, v0.2d, 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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037160841687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
10042037160611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371612611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371602591687251000100010002646802018203720371572318951000100020002037203721100110000073216221787100020382038203820382038
10042037160611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
10042037160611687251000100010002646802018203720371572318951000100020002037203711100110000973216221787100020382038203820382038
10042037160611687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371601031687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038
100420371501561687251000100010002646802018203720371572318951000100020002037203711100110001073216221787100020382038203820382038
10042037160721687251000100010002646802018203720371572318951000100020002037203711100110000073216221787100020382038203820382038

Test 2: Latency 1->2

Code:

  usubw v0.2d, v0.2d, 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)030a0b191e3f4e51inst 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)9fa8accfd0d2d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102042003715620006119687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010010710002162219791100001002003820038200382003820038
102042003715520006119687251010010010000100100005002847680002001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820087
1020420037155200010319687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
1020420037155200126119687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
102042003715620006119687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
10204200371552001225719687251010010010000100100005002847680002001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
102042003715520006119687251010010010000100100005002847680002001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
102042003715500006119687251010010010000100100005002847680002001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
102042003715500006119687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038
102042003715520006119687251010010010000100100005002847680102001820037200371842231874510100200100002002000020037200371110201100991001001000010000710002162219791100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03191e3f4e51inst 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)9fa9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000686416551978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640416551978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640616661978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640516661978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768002001820224202281844431876710010201000020200002003720037211002110910101000010000640516551978510000102003820038200382003820038
1002420037150001031968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640416551978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640616561978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640516561978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640416551978510000102003820038200382003820038
100242003715000611968725100101010000101000050284768002001820037200371844431876710010201000020200002003720037111002110910101000010000640516551978510000102003820038200382003820038

Test 3: Latency 1->3

Code:

  usubw v0.2d, v1.2d, 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)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)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)9fa0a1a6a8a9acc2cfd5d6dbddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371560000000611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000150710116011197910100001002003820038200382003820038
10204200371550000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000210710116011197910100001002003820038200382003820038
1020420037155000000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000000000710116011197910100001002003820038200382003820038
10204200371550000000761968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000870710116011197910100001002003820038200382003820038
102042003715500000006411968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000210710116011197910100001002003820038200382003820038
1020420037156000000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000000090710116011197910100001002003820038200382003820038
10204200371560000000611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000000690710116011197910100001002003820038200382003820038
102042003715500000007261968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000000720710116011197910100001002003820038200382003820038
1020420037155000000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037111020110099100100100001000000090710116011197910100001002003820038200382003820038
1020420037156000000061196872510100100100001001000050028476800200182003720037184223187451010020010000200200002003720037211020110099100100100001000000000710116011197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030b0f1e3f4e51inst 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)72scheduler 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
1002420037155001510319687251001010100001010000502847680120018200372003718444031876710010201000020200002003720037111002110910101000010030640316331978510000102003820038200382003820038
10024200371550006119687251001010100001010000502847680120018200372003718444031876710010201000020200002003720037111002110910101000010027640316331978510000102003820038200382003820038
1002420037155000611968725100101010000101000050284768012001820037200371844403187671001020100002020000200372003711100211091010100001009640316331978510000102003820038200382003820038
10024200371560006791968725100101010000101000050284768012001820037200371844403187671001020100002020000200372003711100211091010100001009640316331978510000102003820038200382003820038
1002420037155000611968725100101010000101000050284768012001820037200371844403187671001020100002020000200372003711100211091010100001000640316331978510000102003820038200382003820038
10024200371560006119687251001010100001010000502847680120018200372003718444031876710010201000020200002003720037111002110910101000010060640316341978510000102003820038200382003820038
1002420037155000611968725100101010000101000050284768012001820037200371844403187671001020100002020000200372003711100211091010100001009640316331978510000102003820038200382003820038
1002420037155000611968725100101010000101000050284768012001820037200371844403187671001020100002020000200372003711100211091010100001000640316331978510000102003820038200842003820038
10024200371560006119687251001010100001010000502847680120018200372003718444031876710010201000020200002003720037111002110910101000010012640316331978510000102003820038200382003820038
10024200371550006119687251001010100001010000502847680120018200372003718444031876710010201000020200002003720037111002110910101000010018640316331978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  usubw v0.2d, v8.2d, v9.2s
  usubw v1.2d, v8.2d, v9.2s
  usubw v2.2d, v8.2d, v9.2s
  usubw v3.2d, v8.2d, v9.2s
  usubw v4.2d, v8.2d, v9.2s
  usubw v5.2d, v8.2d, v9.2s
  usubw v6.2d, v8.2d, v9.2s
  usubw v7.2d, v8.2d, 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)03080b18191e1f3a3f51inst 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)9fa0a1a6a7a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005915500000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511031633200350800001002003920039200392003920039
80204200381550000000184258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000001500511031633200350800001002003920039200392003920039
802042003815500000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511021632200350800001002003920039200392003920039
802042003815500000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000006511031633200350800001002003920039200392003920039
802042003815500000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000100511031633200350800001002003920039200392003920039
80204200381560000120040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000000060511031633200350800001002003920039200392003920039
8020420038155000000090625801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000000511031623200350800001002003920039200392003920039
802042003815500000008225801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000100511031632200350800001002003920039200392003920039
802042003815500000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000003511031632200350800001002003920039200392003920039
802042003815500000004025801001008000010080000500640000120019200382003899733999680100200800002001600002003820038118020110099100100800001000000103511031633200350800001002003920039200392003920039

1000 unrolls and 10 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)5f60696d6edispatch 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)9fa6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420047155000392580010108000010800005064000000200192003820038999631001880010208000020160000200382003811800211091010800001000135502011169212003580000102003920039200392003920039
8002420038155000392580010108000010800005064000011200192003820038999631001880010208000020160000200382003811800211091010800001000050209169212003580000102003920039200392003920039
80024200381550093925800101080000108000050640000002001920038200389996310018800102080000201600002003820038118002110910108000010000502021169212003580000102003920039200392003920039
8002420038155000392580010108000010800005064000000200192003820038999631001880010208000020160000200382003811800211091010800001000050209162192003580000102003920039200392003920039
8002420038156000392580010108000010800005064000000200192003820038999631001880010208000020160000200382003811800211091010800001000050209162192003580000102003920039200392003920039
800242003816100039258001010800001080000506400000020019200382003899963100188020820800002016000020038200381180021109101080000100005020211621212003580000102003920039200392003920039
80024200381550002102580010108000010800005064000000200192003820038999631001880010208000020160000200382003811800211091010800001000050209162192003580000102003920039200392003920039
8002420038156000514258001010800001080000506400000020019200382003899963100188001020800002016039020038200381180021109101080000100005020211621212003580000102003920039200392003920039
8002420038155000392580010108000010800005064000000200192003820038999631001880010208000020160000200382003811800211091010800001000050209162192003580000102003920039200392003920039
800242003815500039258001010800001080000506400000020019200382003899963100188001020800002016000020038200381180021109101080000100012502021169212003580000102003920039200392003920039