Apple M1 Microarchitecture Research by Dougall Johnson

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

FCMGT (vector, zero, 8H)

Test 1: uops

Code:

  fcmgt v0.8h, v0.8h, #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)03191e1f3f4e51inst 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)a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037160006116862510001000100026452102018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037160006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
1004203715012061168625100010001000264521120182037203715713189510001000100020372037111001100002473116111786100020382038203820382038
10042037160006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037160006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037150006116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038
10042037150306116862510001000100026452112018203720371571318951000100010002037203711100110000073116111786100020382038203820382038

Test 2: Latency 1->2

Code:

  fcmgt v0.8h, v0.8h, #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)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)9fa0a6a7a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020420037150126119675251010010010000126100005002848785020018200372003718424318762101002061000020010165200372003711102011009910010010000100000220507101251119824100001002003820038200382003820038
1020420037150126119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000107101161119791100001002003820038200382003820038
1020420037150012419686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
1020420037150010319686251010010010000100100005002847521020018200372003718421318745101002001000020010000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
10204200371501261196862510100100100001001000050028475211200182003720037184213187451010020010000200100002003720037111020110099100100100001000003007101161119791100001002008620038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000007101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000107101161119791100001002003820038200382003820038
102042003715006119686251010010010000100100005002847521120018200372003718421318745101002001000020010000200372003711102011009910010010000100000107101161119791100001002003820038200382003820038

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)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)9fa6a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100242003715006119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010203640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010003640216221978610000102003820038200382003820038
1002420037150061196862510010101000010100005028475211200180200372003718443318767100102010000201000020037200371110021109101010000100015640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521120018020037200371844331876710010201000020100002003720037111002110910101000010003640216221978610000102003820038200382003820038
1002420037150516119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010003640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521120018020037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038
100242003715006119686251001010100001010000502847521020018020037200371844331876710010201000020100002003720037111002110910101000010000640216221978610000102003820038200382003820038

Test 3: throughput

Count: 8

Code:

  fcmgt v0.8h, v8.8h, #0
  fcmgt v1.8h, v8.8h, #0
  fcmgt v2.8h, v8.8h, #0
  fcmgt v3.8h, v8.8h, #0
  fcmgt v4.8h, v8.8h, #0
  fcmgt v5.8h, v8.8h, #0
  fcmgt v6.8h, v8.8h, #0
  fcmgt v7.8h, v8.8h, #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)03191e1f3f51inst 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)9fa7a8a9acc5branch mispredict (cb)cdcfd5d6dde0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042004815000010225801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000001115118116020035800001002003920039200392003920039
80204200381500305225801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000001115118016020035800001002003920039200392003920039
80204200381500009425801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010003061115118016120035800001002003920039200392003920039
802042003815000011525801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000001115118016020035800001002003920039200392003920039
80204200381500009225801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000001115118016020035800001002003920039200392003920039
802042003815000051025801081008000810080020500640132020019200382003899776998980120200800322008003220038200381180201100991001008000010000231115118016020035800001002003920039200392003920039
80204200381500002925801081008000810080020500640132120019200382003899776998980120200800322008023120099200381180201100991001008000010000001115118016020035800001002003920039200392003920039
8020420038150013202925801081008000810080020500640132120019200382003899776998980120200800322008003220038200381180201100991001008000010000001115118016020035800001002003920039200392003920039
802042003815001202925801081008000810080020500640132120019200382003899776998980120200800322008003220038200381180201100991001008000010000001115118016020035800001002003920039200392003920039
8020420038150000712580108100800081008002050064013212001920038200389977699898012020080032200800322009020038118020110099100100800001002202501115118025020035800001002015020039200392003920039

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03041e1f3f51inst 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)9fa6a8accfd5d6d9ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200511501003925800101080000108000050640000020019200382003899963100188001020800002080000200382003811800211091010800001000050202716016252003580000102003920039200392003920039
800242003815000010925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000050202516013232003580000102003920039200392003920039
800242003815000051425800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000050202516015262003580000102003920039200392003920039
80024200381500006452580010108000010800005064000012001920038200389996310018800102080000208000020038200381180021109101080000100094350202416025252003580000102003920039200392003920039
80024200381500003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001001350202816028272003580000102003920039200392003920039
80024200381500003925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000050202816027262003580000102003920039200392003920039
800242003815000069725800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001000050201816023202003580000102003920039200392003920039
800242003815000039258001010800001080000506400001200192003820038999613100188001020800002080000200382013721800211091010800001030050202816027272003580000102018920039200392003920039
800242003815002403925800101080000108000050640000120019200382003899963100188001020800002080000200382003811800211091010800001001050202716026262003580000102003920039200392003920039
800242003815001203925800101080000108000050640000120019200382003899963100188001020800982080293201392003811800211091010800001001050202716015262003580000102003920039200392003920039