Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-8179040

Avoid Ticks::now calls when EventClassLoad is not enabled

    XMLWordPrintable

    Details

    • Subcomponent:
    • Resolved In Build:
      b22

      Description

      Startup profiling shows that we're spending quite a bit of time in Ticks::now during class load, but it appears as if this time is only used if the EventClassLoad is to be committed.

      Extracting a pre_class_load_event method to only take the Ticks::now if the event is enabled should be recorded has a measurable effect on a minimal program:

      Baseline run:
      chrt -f 99 perf stat -r 250 java Hello > /dev/null

       Performance counter stats for 'java Hello' (250 runs):

              104,616782 task-clock (msec) # 1,230 CPUs utilized ( +- 0,26% )
                     847 context-switches # 0,008 M/sec ( +- 0,17% )
                      54 cpu-migrations # 0,516 K/sec ( +- 1,65% )
                   3 160 page-faults # 0,030 M/sec ( +- 0,37% )
             377 267 792 cycles # 3,606 GHz ( +- 0,25% ) (80,38%)
             238 128 005 stalled-cycles-frontend # 63,12% frontend cycles idle ( +- 0,39% ) (85,54%)
             184 874 177 stalled-cycles-backend # 49,00% backend cycles idle ( +- 0,38% ) (69,21%)
             301 374 735 instructions # 0,80 insn per cycle
                                                        # 0,79 stalled cycles per insn ( +- 0,12% ) (84,57%)
              58 648 242 branches # 560,601 M/sec ( +- 0,08% ) (82,65%)
               2 198 462 branch-misses # 3,75% of all branches ( +- 0,19% ) (83,06%)

             0,085036213 seconds time elapsed ( +- 0,22% )

      Patched:
      chrt -f 99 perf stat -r 250 java Hello > /dev/null

       Performance counter stats for 'java Hello' (250 runs):

              100,586215 task-clock (msec) # 1,231 CPUs utilized ( +- 0,24% )
                     827 context-switches # 0,008 M/sec ( +- 0,14% )
                      60 cpu-migrations # 0,601 K/sec ( +- 1,87% )
                   3 173 page-faults # 0,032 M/sec ( +- 0,36% )
             364 094 344 cycles # 3,620 GHz ( +- 0,24% ) (81,98%)
             225 931 914 stalled-cycles-frontend # 62,05% frontend cycles idle ( +- 0,39% ) (86,06%)
             174 214 186 stalled-cycles-backend # 47,85% backend cycles idle ( +- 0,39% ) (65,68%)
             298 467 000 instructions # 0,82 insn per cycle
                                                        # 0,76 stalled cycles per insn ( +- 0,17% ) (83,39%)
              58 031 033 branches # 576,928 M/sec ( +- 0,09% ) (83,07%)
               2 193 340 branch-misses # 3,78% of all branches ( +- 0,19% ) (84,00%)

             0,081682242 seconds time elapsed ( +- 0,19% )

        Attachments

          Activity

            People

            Assignee:
            redestad Claes Redestad
            Reporter:
            redestad Claes Redestad
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Dates

              Created:
              Updated:
              Resolved: