Apple M1 Microarchitecture Research by Dougall Johnson

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

SMINV (vector, 16B)

Test 1: uops

Code:

  sminv b0, v0.16b
  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)030818191e1f3f4e51inst 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)a0a1a7a8a9acc2cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
100430372300000612547251000100010003981601301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
100430372300000612547251000100010003981601301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
100430372300060612547251000100010003981600301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
1004303722000006125472510001000100039816003018303730372414328951000100020003037303711100110000000045073216222629100030383038303830383038
100430372200000612547251000100010003981600301830373037241432895100010002000303730371110011000000006073216222629100030383038303830383038
100430372300000612547251000100010003981600301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
1004303723000120612547251000100010003981600301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
100430372300000612547251000100010003981600301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
100430372200000612547251000100010003981601301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038
100430372300060612547251000100010003981600301830373037241432895100010002000303730371110011000000000073216222629100030383038303830383038

Test 2: Latency 1->2

Code:

  sminv b0, v0.16b
  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)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)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)9fa8acc5cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001000307101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001000007101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730084111020110099100100100001001307101161129633100001003003830038300383003830038
102043003722500103295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001001307101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001001607101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300543003730037282643287451010020010000200200003003730037111020110099100100100001003007101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001001307101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001000607101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001001907101161129633100001003003830038300383003830038
10204300372250061295472510100100100001001000050042771601300183003730037282643287451010020010000200200003003730037111020110099100100100001002307101161129633100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)0308090b18191e1f3f4e51inst 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)9fa0a1a8a9acc2cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
100243003722400000006129547251001010100001010000504277160130018300373003728286328767100102010000202000030037300371110021109101010000100000306402162229629010000103003830038300383003830038
100243017922500000006129547251001010100001010000504277160130018300373003728286328767100102010000202000030037300371110021109101010000100010306402162229629010000103003830038300383003830038
1002430037225000000061295472510010101000010100005042771601300183003730037282863287671001020100002020000300373003711100211091010100001000220006402162229629010000103003830038300383003830038
100243003722500000006129547251001010100001010000504277160130018300373003728286328767100102010000202000030037300371110021109101010000100010306402162229629010000103003830038300383003830038
1002430037225000000061295472510010101000010100005042771601300183003730037282863287671001020100002020000300373003711100211091010100001000380006402162229629010000103003830038300383003830038
100243003722500000006129547251001010100001010000504277160130018300373003728286328767100102010000202000030037300371110021109101010000100010306402162229629010000103003830038300383003830038
100243003722500000006129547251001810100001010000504277160130018300373003728286328767100102010000202000030037300371110021109101010000100000606402162229629010000103003830038300383003830038
100243003722400000006129547251001010100001010000504277160030018300373003728286328767100102010000202000030037300371110021109101010000100000306402162229629010000103003830038300383003830038
1002430037225000000094329547251001010100001010000504277160130018300373003728286328767100102010000202000030037300371110021109101010000100010306402162229629010000103003830038300383003830038
1002430037225000000061295472510010101000010100005042771601300183003730037282863287671001020100002020000300373003711100211091010100001000101206402162229629010000103003830038300383003830038

Test 3: throughput

Count: 8

Code:

  sminv b0, v8.16b
  sminv b1, v8.16b
  sminv b2, v8.16b
  sminv b3, v8.16b
  sminv b4, v8.16b
  sminv b5, v8.16b
  sminv b6, v8.16b
  sminv b7, v8.16b
  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)030b181e3f51inst 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)9fa1a8acc2c5branch mispredict (cb)cdcfd5d6dde0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010000001115118216020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010000001115118016020036800001002004020040200402004020040
802042003915004030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010040001115118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010001001115118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010002001115118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010000001115118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010000001215118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010001001115118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010001301115118016020036800001002004020040200402004020040
802042003915000030258010810080008100800205006401320200202003920039997769990801202008003220016006420039200391180201100991001008000010001301115118016020036800001002004020040200402004020040

1000 unrolls and 10 iterations

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

retire (01)cycle (02)03081e3f51inst 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)696b6d6edispatch 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)9fa8acc5cficache miss (d3)d5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
80024200511500040258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001000050200151611112003680000102004020040200402004020040
80024200391500040258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001000050200121611122003680000102004020040200402004020040
800242003915000104258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001000050200121613112003680000102004020040200402004020040
80024201441500040258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001003050200121611112003680000102004020040200402004020040
8002420039150004025800101080000108000050640000200200200392003999963100198001020800002016000020039200391180021109101080000100005020013161292003680000102004020040200402004020040
80024200391500040258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001030050200121613132003680000102004020040200402004020040
800242003915000402580010108000010800005064000020020020039200399996310019800102080000201600002003920039118002110910108000010110050200131613122003680000102004020040200402004020040
800242003915000230258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001000050200131612122003680000102004020040200402004020040
80024200391500040258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001000050200111613112003680000102004020040200402004020040
80024200391500084258001010800001080000506400002002002003920039999631001980010208000020160000200392003911800211091010800001000050200111611132003680000102004020040200402004020040