Apple M1 Microarchitecture Research by Dougall Johnson

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

TBL (single register table, 8B)

Test 1: uops

Code:

  tbl v0.8b, { v0.16b }, 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)03191e3f4e51inst 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)a1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150061168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
100420371502761168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037150061168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037150061168725100010001000264680120182037203715723189510001000200020372037111001100001073116111787100020382038203820382038
10042037150061168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037150061168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037150061168725100010001000264680020182037203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037150061168725100010001000264680120182084203715723189510001000200020372037111001100000073116111787100020382038203820382038
10042037150061168725100010001000264680120182037203715723189510001000200020372037111001100021073116111787100020382038203820382038
10042037150061168725100010001000264680120182037203715723189510001000200020372037111001100000073116111787100020382038203820382038

Test 2: Latency 1->2

Code:

  tbl v0.8b, { v0.16b }, 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)0318191e1f3f4e51inst 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)9fa1accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204200371490042061196872510100100100001001000050028476801200182003720037184221118745101002001000020020000200372003711102011009910010010000100007101161219791100001002003820038200382003820038
10204200371550024022919687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150003006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715000006119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003731102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150000053619687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100037101161119791100001002003820038200382003820038
1020420037150003906119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
102042003715000444014519687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038
1020420037150005106119687251010010010000100100005002847680020018200372003718422318745101002001000020020000200372003711102011009910010010000100007101161119791100001002003820038200382003820038

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)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)9fa6acc5cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002420037150636119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100006404163319785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100016403163919796010000102003820038200382003820038
1002420037150756119687251001010100001010000502847680120018200372003718444318767100102010000202000020084200371110021109101010000100006403163319785010000102003820038200382003820038
100242003714906119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100006403163319785010000102003820038200382003820038
10024200371506576119687251001010100001010000502847680120018200372003718444318804100102010000202000020037200371110021109101010000100006403163319785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100006403163319785010000102003820038200382003820038
100242003715066119687251001010100001010000502847680120018200372003718444318767100102010000202000020037200371110021109101010000100006403160319785010000102003820038200382003820038
1002420037150516119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100006403163319785010000102003820038200382003820038
100242003715006119687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100006403163319785010000102003820038200382003820038
1002420037150126122919687251001010100001010000502847680020018200372003718444318767100102010000202000020037200371110021109101010000100006403163319785010000102003820038200382003820038

Test 3: Latency 1->3

Code:

  tbl v0.8b, { v1.16b }, 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)9fa1a6a8branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150144611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768002001820037200851842231874510100200100002002000020037200371110201100991001001000010000007101162119791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371503611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371500611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
102042003715018611968725101001001000010010000500284768012001820037200851842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
102042003715048611968725101001001000010010000500284768012001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038
10204200371503611968725101001001000010010000500284768002001820037200371842231874510100200100002002000020037200371110201100991001001000010000007101161119791100001002003820038200382003820038

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)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
100242003715039611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371845231876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
100242003715003401968725100101010000101000050284768012001820084200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371506611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371509611968725100101010000101000050284768012001820178200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010009640216221978510000102003820038200382003820038
100242003715007261968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038
10024200371500611968725100101010000101000050284768012001820037200371844431876710010201000020200002003720037111002110910101000010000640216221978510000102003820038200382003820038

Test 4: throughput

Count: 8

Code:

  tbl v0.8b, { v8.16b }, v9.8b
  tbl v1.8b, { v8.16b }, v9.8b
  tbl v2.8b, { v8.16b }, v9.8b
  tbl v3.8b, { v8.16b }, v9.8b
  tbl v4.8b, { v8.16b }, v9.8b
  tbl v5.8b, { v8.16b }, v9.8b
  tbl v6.8b, { v8.16b }, v9.8b
  tbl v7.8b, { v8.16b }, 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)0308090b18191e1f3a3f51inst 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042006015000000630040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511031622200350800001002003920039200392003920039
802042003815000000180040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039
802042003815000000000323258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039
80204200381500000000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039
80204200381500000000040258010010080000100800005006400000200192003820038998439996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039
80204200381500000000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000202000511021622200350800001002003920039200392003920039
80204200381500000000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000460511021622200350800001002003920039200392003920039
80204200381500000060040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039
802042003815000000000610258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039
80204200381500000000040258010010080000100800005006400000200192003820038997339996801002008000020016000020038200381180201100991001008000010000000000511021622200350800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03070818191e3a3f51inst 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? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200471500000003925800101080000108000050640000120019200382003899963100188001020800002016000020038200381180021109101080000100005020131641220035080000102003920039200392003920039
8002420038150000000392580010108000010800005064000002001920038200389996310018800102080000201600002003820038118002110910108000010000502081611420035080000102003920039200392003920039
800242003815000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050206165620035080000102003920039200392003920039
800242003815000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050207167620035080000102011020039200392003920039
800242003815000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050205164520035080000102003920039200392003920039
800242003815000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050204164620035080000102003920039200392003920039
800242003815000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050206165520035080000102003920039200392003920039
8002420038150000000921258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050206166620035080000102003920039200392003920039
8002420038150000000324258001010800001080000506400000200192003820242999631001880010208000020160000200382003811800211091010800001000050204164420035080000102003920039200392003920039
800242003815000000039258001010800001080000506400000200192003820038999631001880010208000020160000200382003811800211091010800001000050206166620035080000102003920039200392003920039