Skip to content

Instantly share code, notes, and snippets.

@toddlipcon
Created March 30, 2018 20:16
Show Gist options
  • Save toddlipcon/ad2e68cfcdf9e7ca1957dfe3e981fbf9 to your computer and use it in GitHub Desktop.
Save toddlipcon/ad2e68cfcdf9e7ca1957dfe3e981fbf9 to your computer and use it in GitHub Desktop.
r130 clang4
--------
Compressed 42553599 bytes into 10314046 bytes ==> 24.24%
Performance counter stats for './lz4-r130.clang-4 -f akey /dev/null' (10 runs):
88.483908 task-clock (msec) # 0.995 CPUs utilized ( +- 0.96% )
3 context-switches # 0.035 K/sec ( +- 3.23% )
0 cpu-migrations # 0.001 K/sec ( +-100.00% )
1,456 page-faults # 0.016 M/sec ( +- 0.01% )
247,030,123 cycles # 2.792 GHz ( +- 0.96% )
336,605,820 instructions # 1.36 insn per cycle ( +- 0.00% )
44,257,876 branches # 500.180 M/sec ( +- 0.00% )
2,364,846 branch-misses # 5.34% of all branches ( +- 0.02% )
0.088926232 seconds time elapsed ( +- 0.95% )
Performance counter stats for './lz4-r130.clang-6 -f akey /dev/null' (30 runs):
88.200872 task-clock (msec) # 0.995 CPUs utilized ( +- 0.58% )
3 context-switches # 0.036 K/sec ( +- 2.32% )
0 cpu-migrations # 0.004 K/sec ( +- 24.41% )
1,457 page-faults # 0.017 M/sec ( +- 0.01% )
246,259,940 cycles # 2.792 GHz ( +- 0.57% )
352,360,917 instructions # 1.43 insn per cycle ( +- 0.00% )
43,259,714 branches # 490.468 M/sec ( +- 0.00% )
2,331,903 branch-misses # 5.39% of all branches ( +- 0.01% )
0.088627625 seconds time elapsed ( +- 0.58% )
r130 gcc4:
Performance counter stats for './lz4-r130.gcc-4.8.2 -f akey /dev/null' (10 runs):
83.113943 task-clock (msec) # 0.996 CPUs utilized ( +- 1.27% )
3 context-switches # 0.036 K/sec
0 cpu-migrations # 0.002 K/sec ( +- 66.67% )
1,458 page-faults # 0.018 M/sec ( +- 0.01% )
232,091,061 cycles # 2.792 GHz ( +- 1.25% )
313,027,421 instructions # 1.35 insn per cycle ( +- 0.00% )
45,470,519 branches # 547.087 M/sec ( +- 0.00% )
2,322,913 branch-misses # 5.11% of all branches ( +- 0.02% )
0.083449061 seconds time elapsed ( +- 1.27% )
r130 gcc7:
Performance counter stats for './lz4-r130.gcc-7 -f akey /dev/null' (10 runs):
83.250287 task-clock (msec) # 0.995 CPUs utilized ( +- 0.90% )
3 context-switches # 0.037 K/sec ( +- 3.23% )
0 cpu-migrations # 0.001 K/sec ( +-100.00% )
1,457 page-faults # 0.018 M/sec ( +- 0.01% )
232,281,948 cycles # 2.790 GHz ( +- 0.89% )
321,944,699 instructions # 1.39 insn per cycle ( +- 0.00% )
44,455,097 branches # 533.993 M/sec ( +- 0.00% )
2,331,653 branch-misses # 5.24% of all branches ( +- 0.04% )
0.083658848 seconds time elapsed ( +- 0.90% )
trunk clang4:
88.154990 task-clock (msec) # 0.995 CPUs utilized ( +- 1.02% )
3 context-switches # 0.035 K/sec ( +- 3.23% )
0 cpu-migrations # 0.002 K/sec ( +- 66.67% )
1,462 page-faults # 0.017 M/sec ( +- 0.01% )
245,937,896 cycles # 2.790 GHz ( +- 1.01% )
341,814,940 instructions # 1.39 insn per cycle ( +- 0.00% )
45,331,269 branches # 514.222 M/sec ( +- 0.00% )
2,318,045 branch-misses # 5.11% of all branches ( +- 0.01% )
0.088566811 seconds time elapsed ( +- 1.01% )
trunk gcc4:
Performance counter stats for './lz4-trunk.gcc-4.8 -f akey /dev/null' (10 runs):
85.017470 task-clock (msec) # 0.995 CPUs utilized ( +- 0.84% )
3 context-switches # 0.036 K/sec ( +- 3.23% )
0 cpu-migrations # 0.000 K/sec
1,464 page-faults # 0.017 M/sec ( +- 0.01% )
237,050,136 cycles # 2.788 GHz ( +- 0.84% )
334,249,423 instructions # 1.41 insn per cycle ( +- 0.00% )
48,633,537 branches # 572.042 M/sec ( +- 0.00% )
2,342,490 branch-misses # 4.82% of all branches ( +- 0.02% )
0.085453846 seconds time elapsed ( +- 0.84% )
trunk gcc7:
Performance counter stats for './lz4-trunk.gcc-7 -f akey /dev/null' (10 runs):
81.850551 task-clock (msec) # 0.995 CPUs utilized ( +- 1.23% )
3 context-switches # 0.038 K/sec ( +- 3.23% )
0 cpu-migrations # 0.005 K/sec ( +- 40.82% )
1,467 page-faults # 0.018 M/sec ( +- 0.01% )
228,526,074 cycles # 2.792 GHz ( +- 1.23% )
316,521,269 instructions # 1.39 insn per cycle ( +- 0.00% )
42,744,409 branches # 522.225 M/sec ( +- 0.00% )
2,353,068 branch-misses # 5.50% of all branches ( +- 0.04% )
0.082268223 seconds time elapsed ( +- 1.23% )
trunk gcc7 with PGO:
Performance counter stats for './lz4 akey /dev/null' (20 runs):
80.234094 task-clock (msec) # 0.995 CPUs utilized ( +- 0.77% )
3 context-switches # 0.039 K/sec ( +- 2.22% )
0 cpu-migrations # 0.003 K/sec ( +- 39.74% )
1,450 page-faults # 0.018 M/sec ( +- 0.01% )
224,046,952 cycles # 2.792 GHz ( +- 0.76% )
288,348,111 instructions # 1.29 insn per cycle ( +- 0.00% )
39,210,357 branches # 488.699 M/sec ( +- 0.00% )
2,488,151 branch-misses # 6.35% of all branches ( +- 0.03% )
0.080614767 seconds time elapsed ( +- 0.78% )
decompression
--------------------
Performance counter stats for './lz4-trunk.gcc-4.8 -d akey.lz4 /dev/null' (30 runs):
37.009628 task-clock (msec) # 0.991 CPUs utilized ( +- 0.35% )
3 context-switches # 0.081 K/sec
0 cpu-migrations # 0.001 K/sec ( +-100.00% )
1,491 page-faults # 0.040 M/sec ( +- 0.01% )
103,266,553 cycles # 2.790 GHz ( +- 0.34% )
234,435,868 instructions # 2.27 insn per cycle ( +- 0.00% )
31,212,133 branches # 843.352 M/sec ( +- 0.00% )
913,618 branch-misses # 2.93% of all branches ( +- 0.03% )
0.037336149 seconds time elapsed ( +- 0.35% )
Performance counter stats for './lz4-trunk.gcc-7 -d akey.lz4 /dev/null' (30 runs):
34.903987 task-clock (msec) # 0.990 CPUs utilized ( +- 0.50% )
3 context-switches # 0.089 K/sec ( +- 2.37% )
0 cpu-migrations # 0.000 K/sec
1,491 page-faults # 0.043 M/sec ( +- 0.01% )
97,353,408 cycles # 2.789 GHz ( +- 0.50% )
219,831,395 instructions # 2.26 insn per cycle ( +- 0.00% )
30,220,646 branches # 865.822 M/sec ( +- 0.00% )
924,034 branch-misses # 3.06% of all branches ( +- 0.06% )
0.035266196 seconds time elapsed ( +- 0.51% )
gcc7 trunk with PGO:
Performance counter stats for './lz4 -d akey.lz4 /dev/null' (20 runs):
33.410912 task-clock (msec) # 0.990 CPUs utilized ( +- 1.13% )
3 context-switches # 0.090 K/sec
0 cpu-migrations # 0.006 K/sec ( +- 45.88% )
1,477 page-faults # 0.044 M/sec ( +- 0.01% )
93,262,330 cycles # 2.791 GHz ( +- 1.12% )
205,037,732 instructions # 2.20 insn per cycle ( +- 0.00% )
25,935,884 branches # 776.270 M/sec ( +- 0.00% )
927,920 branch-misses # 3.58% of all branches ( +- 0.06% )
0.033735029 seconds time elapsed ( +- 1.16% )
Performance counter stats for './lz4-trunk.clang-4 -d akey.lz4 /dev/null' (30 runs):
43.541295 task-clock (msec) # 0.992 CPUs utilized ( +- 0.52% )
3 context-switches # 0.073 K/sec ( +- 2.76% )
0 cpu-migrations # 0.002 K/sec ( +- 69.48% )
1,490 page-faults # 0.034 M/sec ( +- 0.00% )
121,465,440 cycles # 2.790 GHz ( +- 0.51% )
260,222,242 instructions # 2.14 insn per cycle ( +- 0.00% )
37,736,746 branches # 866.689 M/sec ( +- 0.00% )
908,653 branch-misses # 2.41% of all branches ( +- 0.06% )
0.043903767 seconds time elapsed ( +- 0.54% )
Performance counter stats for './lz4-r130.clang-4 -f -d akey.lz4 /dev/null' (30 runs):
51.739862 task-clock (msec) # 0.994 CPUs utilized ( +- 0.48% )
3 context-switches # 0.059 K/sec ( +- 1.51% )
0 cpu-migrations # 0.003 K/sec ( +- 41.52% )
1,481 page-faults # 0.029 M/sec ( +- 0.00% )
144,464,236 cycles # 2.792 GHz ( +- 0.47% )
337,867,166 instructions # 2.34 insn per cycle ( +- 0.00% )
47,822,849 branches # 924.294 M/sec ( +- 0.00% )
943,924 branch-misses # 1.97% of all branches ( +- 0.03% )
Performance counter stats for './lz4-r130.clang-6 -f -d akey.lz4 /dev/null' (30 runs):
52.219312 task-clock (msec) # 0.993 CPUs utilized ( +- 0.65% )
3 context-switches # 0.059 K/sec ( +- 1.80% )
0 cpu-migrations # 0.001 K/sec ( +- 69.48% )
1,481 page-faults # 0.028 M/sec ( +- 0.00% )
145,720,888 cycles # 2.791 GHz ( +- 0.64% )
347,017,966 instructions # 2.38 insn per cycle ( +- 0.00% )
47,308,313 branches # 905.954 M/sec ( +- 0.00% )
944,695 branch-misses # 2.00% of all branches ( +- 0.04% )
0.052583632 seconds time elapsed ( +- 0.66% )
Performance counter stats for './lz4-r130.gcc-4.8.2 -f -d akey.lz4 /dev/null' (30 runs):
36.580186 task-clock (msec) # 0.989 CPUs utilized ( +- 1.04% )
3 context-switches # 0.085 K/sec ( +- 1.80% )
0 cpu-migrations # 0.009 K/sec ( +- 26.26% )
1,481 page-faults # 0.040 M/sec ( +- 0.01% )
102,057,400 cycles # 2.790 GHz ( +- 1.03% )
223,966,272 instructions # 2.19 insn per cycle ( +- 0.01% )
35,014,769 branches # 957.206 M/sec ( +- 0.00% )
958,027 branch-misses # 2.74% of all branches ( +- 0.07% )
0.036986305 seconds time elapsed ( +- 1.04% )
Performance counter stats for './lz4-r130.gcc-7 -f -d akey.lz4 /dev/null' (30 runs):
37.844639 task-clock (msec) # 0.989 CPUs utilized ( +- 0.97% )
3 context-switches # 0.079 K/sec
0 cpu-migrations # 0.007 K/sec ( +- 30.79% )
1,480 page-faults # 0.039 M/sec ( +- 0.00% )
105,558,406 cycles # 2.789 GHz ( +- 0.95% )
214,306,245 instructions # 2.03 insn per cycle ( +- 0.00% )
36,005,944 branches # 951.415 M/sec ( +- 0.00% )
948,146 branch-misses # 2.63% of all branches ( +- 0.01% )
0.038247922 seconds time elapsed ( +- 0.97% )
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment