Apple M1 Microarchitecture Research by Dougall Johnson

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

UMAX (vector, 8H)

Test 1: uops

Code:

  umax v0.8h, v0.8h, v1.8h
  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)a8acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203716061168725100010001000264680120182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203715082168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
10042037151561168725100010001000264680120182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203716061168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
10042037150215168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203715061168725100010001000264680020182037203715723189510001000200020372037111001100000073216221787100020382038203820382038
1004203715129251168725100010001000264680120182037203715723189510001000200020372037111001100000073216221787100020382038203820382038

Test 2: Latency 1->2

Code:

  umax v0.8h, v0.8h, v1.8h
  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)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)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)9fa8acc5cfd0d5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500082196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
102042003715000891196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
10204201321500061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
102042003715000145196872510100100100001001000050028476801200182003720037184220318760101002001000020020000200372003711102011009910010010000100002710011611197910100001002003820038200382003820038
10204200371500061196872510100100100001001000050028476800200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038
102042003715000119196872510100100100001001000050028476801200182003720037184220318745101002001000020020000200372003711102011009910010010000100000710011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)030708090b18191e1f3a3f4e51inst 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)9fa0a1a6a7a8a9acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150000000000147196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006403162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000003006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038
100242003715000000000061196872510010101000010100005028476800200180200372003718444318767100102010000202000020037200371110021109101010000100000000006402162219785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  umax v0.8h, v1.8h, v0.8h
  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)03191e1f3f4e51inst 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)9fa6a7a8a9accfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371500906119687251010010010000100100005002847680200182003720132184223187451010020010000200200002003720037111020110099100100100001000000071031611197910100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680200182003720037184229187451027720010000200200002003720037111020110099100100100001002120071011611197910100001002003820038200382003820038
102042003715000072619687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071013211198250100001002003820038200382003820038
102042003715020018919687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500006119687251010010010024100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382018320038
102042003715001206119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000000071011611197910100001002003820038200382003820038
10204200371500006119687251010010010000100100005002847680200182003720037184223187451010020010000200200002003720037111020110099100100100001000002971011611197910100001002003820038200382003820038

1000 unrolls and 10 iterations

Result (median cycles for code): 2.0037

retire (01)cycle (02)03091e3a3f4e51inst 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)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024200371500006119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640316461978510000102003820038200382003820038
10024200371500006119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640516561978510000102003820038200382003820038
10024200371550906119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640516651978510000102003820038200382003820038
10024200371500006119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010022014640516661978510000102003820038200382003820038
10024200371500006119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640516551978510000102003820038200382003820038
100242003715000015619687251001010100001010000502847680020054200372003718444031876710010201000020200002003720037111002110910101000010000640516551978510000102003820038200382003820038
10024200371500006119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640616561978510000102003820038200382003820038
10024200371500006119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640516551978510000102003820038200382003820038
100242003715004806119687251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010050640516651978510000102003820038200382003820038
100242003715001206119676251001010100001010000502847680020018200372003718444031876710010201000020200002003720037111002110910101000010000640416551978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  umax v0.8h, v8.8h, v9.8h
  umax v1.8h, v8.8h, v9.8h
  umax v2.8h, v8.8h, v9.8h
  umax v3.8h, v8.8h, v9.8h
  umax v4.8h, v8.8h, v9.8h
  umax v5.8h, v8.8h, v9.8h
  umax v6.8h, v8.8h, v9.8h
  umax v7.8h, v8.8h, v9.8h
  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)03040a1e3f51inst 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)9fa6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420049151106340258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051102161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150006640258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
802042003815000040258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150005440258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039
8020420038150005740258010010080000100800005006400001200192003820038997339996801002008000020016000020038200381180201100991001008000010000051101161120035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)030b1e3f51inst 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)9fa8accfd2d5d6dbddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420040150093925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021602220035080000102003920039200392003920039
800242003815002553925800101080000108000050640000002001920038200389996310018800102080000201600002003820038118002110910108000010005020031602220035080000102003920039200392003920039
80024200381501183925800101080000108009950640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021602320035080000102003920039200392003920039
800242003815002103925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021602320035080000102003920039200392003920039
80024200381500273925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021602220035080000102003920039200392003920039
800242003815003813925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021602220035080000102003920039200392003920039
8002420038150003925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021602220035080000102003920039200392003920039
8002420038150093925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020041602220035080000102003920039200392003920039
80024200381500123925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021603320035080000102003920039200392003920039
80024200381500123925800101080000108000050640000012001920038200389996310018800102080000201600002003820038118002110910108000010005020021604220035080000102003920039200392003920039