Apple M1 Microarchitecture Research by Dougall Johnson

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

FCVT (H to D)

Test 1: uops

Code:

  fcvt d0, h0
  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)a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372212612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303723135612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037230612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
100430372260612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303723751452547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037220822547251000100010003981600301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037220612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037230612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037220612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303722126612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038

Test 2: Latency 1->2

Code:

  fcvt d0, h0
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308091e3a3f4e51inst 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)9fa1a8acc2branch mispredict (cb)cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000300071011611296330100001003003830038300383003830038
10204300372250000165295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037224000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038
1020430037225000061295472510100100100001001000050042771601300183003730037282643287451010020010000200100003003730037111020110099100100100001000000071011611296330100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090a0b18191e1f3a3f4e51inst 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)9fa0a1a6a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
1002430037260000000000612954725100101010000101000050427716013001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103007230038300383003830038
1002430037260000000000612954725100101010000101000050427716003001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103003830038300383003830038
1002430037260000000000612954725100101010000101000050427716013001830037300372828632876710010201000020100003003730037111002110910101000010000000006402172229629010000103003830038300383003830038
10024300372410000000001242954725100101010000101000050427716013001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103003830038300383003830038
10024300372410000000004082954725100101010000101000050427716003001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103003830038300383003830038
1002430037241000000000842954725100101010000101000050427716003001830037300372828632876710010201000020100003003730037111002110910101000010000000006402242229629010000103003830038300383003830038
1002430037241000000000612954725100101010000101000050427716013001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103003830038300383003830038
1002430037233000000000612954725100101010000101000050427716003001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229701010000103003830038300383003830038
10024300372320000000002542954725100101010000101000050427716013001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103003830038300383003830038
1002430037229000000000612954725100101010000101000050427716003001830037300372828632876710010201000020100003003730037111002110910101000010000000006402162229629010000103003830038300383003830038

Test 3: throughput

Count: 8

Code:

  fcvt d0, h8
  fcvt d1, h8
  fcvt d2, h8
  fcvt d3, h8
  fcvt d4, h8
  fcvt d5, h8
  fcvt d6, h8
  fcvt d7, h8
  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)03070a0b18191e3f51inst 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)9fa0a6a8a9acc5branch mispredict (cb)cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420058150110000158258010810080008100800205006401320200202003920039997714999080120200800322008003220039200391180201100991001008000010000003111511811611200360800001002004020040200402004020040
80204200391501100103042803121008000810080128500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000111511811621200360800001002004020040200402004020040
80204200391501100003025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000003111511811611200360800001002004020040200402004020040
802042003914911000062225801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000111511841611200361800001002004020040200402004020040
802042003915011000049725801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000100111511811611200360800001002004020040200402004020040
802042003915011000127225801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000040111511812611200750800001002004020040200402004020091
80204200391501100003044801081008000810080020500641008020020200992003999776999080120200800322008003220039200391180201100991001008000010000000111511811612200360800001002004020040200402004020040
802042003915011000013925801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010002000111511811611200360800001002004020040200402004020040
80204200391501100005325801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000111511811611200360800001002004020040200402004020040
802042003915011000013925801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010000000111511811612200360800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)0309181e1f3a3f51inst 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)9fa0a8acbranch mispredict (cb)cfd5d6daddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
800242005115020000211258001010800001080000506400001200202003920039999631001980010208000020800002003920039118002110910108000010000050201016810222003680000102004020040200402004020040
800242003915000000402580010108000010800005064000012002020039200399996310019800102080000208000020039200391180021109101080000100000502091669182003680000102004020040200402004020040
80024200391500000040258001010800001080000506400001200202003920039999631001980010208000020800002003920039118002110910108000010000050202116621212003680000102004020040200402004020040
8002420039150000004025800101080000108000050640000120020200392003999963100198001020800002080000200392003911800211091010800001000005020211679212003680000102004020040200402004020040
8002420039150001210804025800101080000108000050640000120020200392003999963100198001020800002080000200392003911800211091010800001000005020916821212003680000102004020040200402004020040
80024200391500000040258001010800001080000506400001200202003920039999631001980010208000020800002003920039118002110910108000010000050202116721212003680000102004020040200402004020040
8002420039150000004025800101080000108000050640000120020200392003999963100198001020800002080000200392003911800211091010800001007015020211682192003680000102004020040200402004020040
8002420039150000004025800101080000108000050640000120020200392003999963100198001020800002080000200392003911800211091010800001000005020916721212003680000102004020040200402004020040
80024200391500000051525800101080000108000050640000120020200392003999963100198001020800002080000200392003911800211091010800001000005020221679212003680000102004020040200402004020040
80024200391500000012625800101080000108000050640000120020200392003999963100198001020800002080000200392003911800211091010800001000005020916822222003680000102004020040200402004020040