both better and worse interrupt latency with the "realtime" patces

From: Bo Brantén (bosse_at_acc.umu.se)
Date: 11/13/04

  • Next message: Andries Brouwer: "[PATH] __init in i386/kernel/cpu/nexgen.c" 
    Date:	Sat, 13 Nov 2004 20:50:54 +0100 (MET)
To: linux-kernel@vger.kernel.org

    Hello,

    I have measured the interrupt latency on a few different kernels.

    The setup is a function generator connected to ACK on the printerport and
    an line added to the interrupt handler that toggles an data pin on it.
    Both signals is viewed on an oscilloscope with storage. (The computer was
    an old Athlon 800MHz but the results should be considered relative to each
    other)

    Linux 2.4.27:
         most often 15us.
         often 10us or 15us
         time distribution decreases with time, didn't see any longer
         than 250us

    Linux 2.6.9:
         most often 15us.
         else almost always 10us to 30us
         didn't see any longer than 75us
    This is an improvement over the 2.4.x kernel

    realtime-preempt-2.6.10-rc1-mm2-V0.7.1 by Ingo Molnar:
         often up to 50us
         every few seconds one interrupt takes 400us to 600us

    MontaVistas 4 patces from 08 Oct againts Linux-2.6.9-rc3:
         most often 10us, 15us or 25us
         almost always below 30us
         but every few seconds an interrupt takes 600us.
    Except for the sporadic long latency this patch would be the best, it
    collects almost all interrupts below 30us.

    Bo Branten

    -
    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: Andries Brouwer: "[PATH] __init in i386/kernel/cpu/nexgen.c"

    Relevant Pages

    • RE: Linux v2.6.13-rc3
      ... Many people don't run -rc kernels and even more people ... the S3 interrupt resume issue. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: "transmit timed out" from kernel 2.6.19 on?
      ... Are these kernels that you've compiled yourself, ... recently about dynamic interrupt problems, ... is the removal of the interrupt table. ... LOC: 21047 21049 ...
      (comp.os.linux.networking)
    • Re: Attempted summary of "RT patch acceptance" thread
      ... > missing are noticeable only during measurements with preempt-RT config ... drivers you want and still get the same interrupt latency. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [BUG] long freezes on thinkpad t60
      ... think also tried!NOHZ kernels and older kernels, ... i think the only thing that eventually got Miklos' laptop out of the ... the second column is the number of times the profiling interrupt has hit ... Note the many zero entries - this means that for instructions that are ...
      (Linux-Kernel)
    • Re: [PATCH] local_irq_disable removal
      ... >> did one test which was just interrupt latency. ... > the jury is still out on the accuracy of those numbers. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)