Apple M1 Microarchitecture Research by Dougall Johnson

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

CMGE (zero, D)

Test 1: uops

Code:

  cmge d0, d0, #0
  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)0308090b18191e1f3f4e51inst 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)a0a1a6a8a9accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004203715000000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073216111786100020382038203820382038
1004203716000000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
1004203715000000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
10042037150000000105168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
1004203715000000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
10042037150000000183168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
1004203715000000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
1004203715000000061168625100010001000264521120182037203715713189510001000100020372037111001100000003073116111786100020382038203820382038
1004203716100000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073116111786100020382038203820382038
1004203715000000061168625100010001000264521120182037203715713189510001000100020372037111001100000000073124111786100020382038203820382038

Test 2: Latency 1->2

Code:

  cmge d0, d0, #0
  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)030708090b18193a3f4e51inst 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)9fa8acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150100000012419686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000000010319686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715000000006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000000121419686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
102042003715000000006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038
1020420037150000000012619686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000171031711197910100001002003820038200382008520038
1020420037150000000033719686251012510010000125100006262847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100200071011611197910100001002003820038200382003820038
102042003715000000008219686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100300071011631197910100001002003820038200382003820038
1020420037150000000051619686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071001611197910100001002003820038200382003820038
1020420037150000000012619686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000071011611197910100001002003820038200382003820038

1000 unrolls and 10 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)9fa0a1a6a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715000000001031968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
1002420037150000000010311968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010020000006402162219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
10024200371490000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402242219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038
10024200371500000000611968625100101010000101000050284752112001820037200371844331876710010201000020100002003720037111002110910101000010000000006402162219786010000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  cmge d0, d8, #0
  cmge d1, d8, #0
  cmge d2, d8, #0
  cmge d3, d8, #0
  cmge d4, d8, #0
  cmge d5, d8, #0
  cmge d6, d8, #0
  cmge d7, d8, #0
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

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

retire (01)cycle (02)031e3f51inst 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)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)9fa8acc5branch mispredict (cb)cdcfd6dde0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042005915001362580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
802042003815005992580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816120035800001002003920039200392003920039
80204200381500292580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
80204200381500292580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
802042003815004022580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
80204200381500292580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
80204200381500292580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
80204200381500502580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
80204200381500922580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039
802042003815001382580108100800081008002050064013212001902003820038997706998980120200800322008003220038200381180201100991001008000010000111511816020035800001002003920039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)031e3a3f51inst 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)9fa8acbranch mispredict (cb)cfd5d6dbddinst fetch restart (de)e0eb? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420040150001042580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011621120035080000102003920039200392003920039
800242003815000392580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011611120035080000102003920039200392003920039
8002420038150001022580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
800242003815000392580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
800242003815000392580010108000010800005064000002001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
800242003815000392580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
800242003815000392580010108000010800005064000002001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
800242003815000392580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
800242003815000392580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039
8002420038150001042580010108000010800005064000012001902003820038999631001880010208000020800002003820038118002110910108000010000502011601120035080000102003920039200392003920039