[BENCHMARK] ctxbench for some recent kernels

From: Bill Davidsen (davidsen@tmr.com)
Date: Wed Dec 11 2002 - 07:07:01 EST

Next message: Nicolas ASPERT: "Re: 2.4.20 AGP for I845 wrong ?"
Previous message: chinnakka.b@bplmail.com: "Kernel Module"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Comparing the Redhat 7.3 kernel with some recent 2.5 kernels. All on a
dual Celeron 500 with an SMP kernel run with the "nosmp" option.

================================================================
Results by IPC type
================================================================

                                   loops/sec
SIGUSR1 low high average
  2.4.18-10nosmp-bl 62700 63254 63049
  2.5.41nosmp-bl 55048 55660 55418
  2.5.43nosmp-bl 50093 50250 50176
  2.5.47-ac6nosmp-bl 54200 54236 54215
  2.5.50nosmp-bl 51675 56403 54818

                                   loops/sec
message queue low high average
  2.4.18-10nosmp-bl 105275 105737 105513
  2.5.41nosmp-bl 108057 108413 108232
  2.5.43nosmp-bl 100698 101835 101325
  2.5.47-ac6nosmp-bl 98086 98223 98157
  2.5.50nosmp-bl 104965 105261 105105

                                   loops/sec
pipes low high average
  2.4.18-10nosmp-bl 86794 88484 87637
  2.5.41nosmp-bl 74026 78242 76734
  2.5.43nosmp-bl 80559 81233 80953
  2.5.47-ac6nosmp-bl 71277 78883 75415
  2.5.50nosmp-bl 81211 81529 81411

                                   loops/sec
semiphore low high average
  2.4.18-10nosmp-bl 114407 114644 114537
  2.5.41nosmp-bl 114459 114658 114535
  2.5.43nosmp-bl 113222 113423 113322
  2.5.47-ac6nosmp-bl 95986 96347 96179
  2.5.50nosmp-bl 110582 112044 111450

                                   loops/sec
spin+yield low high average
  2.4.18-10nosmp-bl 189489 190677 190102
  2.5.41nosmp-bl 250737 251116 250867
  2.5.43nosmp-bl 249474 249836 249607
  2.5.47-ac6nosmp-bl 245594 245905 245740
  2.5.50nosmp-bl 252055 252529 252361

                                   loops/sec
spinlock low high average
  2.4.18-10nosmp-bl 3 3 3
  2.5.41nosmp-bl 3 3 3
  2.5.43nosmp-bl 3 3 3
  2.5.47-ac6nosmp-bl 3 3 3
  2.5.50nosmp-bl 3 3 3

-- bill davidsen, CTO TMR Associates, Inc <davidsen@tmr.com> Having the feature freeze for Linux 2.5 on Hallow'een is appropriate, since using 2.5 kernels includes a lot of things jumping out of dark corners to scare you.

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/

Next message: Nicolas ASPERT: "Re: 2.4.20 AGP for I845 wrong ?"
Previous message: chinnakka.b@bplmail.com: "Kernel Module"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sun Dec 15 2002 - 22:00:20 EST