public inbox for ecos-discuss@sourceware.org
 help / color / mirror / Atom feed
* [ECOS] a qustion about measure real-time interrupt latency
@ 2000-11-29 22:38 xuas
  2000-11-30  9:46 ` Jonathan Larmour
  0 siblings, 1 reply; 2+ messages in thread
From: xuas @ 2000-11-29 22:38 UTC (permalink / raw)
  To: ecos

I cant understand something about isr doing.
can someone help me ?

what is mean measure real-time interrupt latency?
what is aim?

 
the isr code is:
 
cyg_uint32 Cyg_RealTimeClock::isr(cyg_vector vector, CYG_ADDRWORD data)
{
#if defined(CYGVAR_KERNEL_COUNTERS_CLOCK_LATENCY) && defined(HAL_CLOCK_LATENCY)
    if (measure_clock_latency) {
        cyg_int32 delta;
        HAL_CLOCK_LATENCY(&delta);
        // Note: Ignore a latency of 0 when finding min_clock_latency.
        if (delta > 0) {
            // Valid delta measured
            total_clock_latency += delta;
            total_clock_interrupts++;
            if (min_clock_latency > delta) min_clock_latency = delta;
            if (max_clock_latency < delta) max_clock_latency = delta;
        }
    }
#endif

    CYG_INSTRUMENT_CLOCK( ISR, 0, 0);

    HAL_CLOCK_RESET( CYGNUM_HAL_INTERRUPT_RTC, CYGNUM_KERNEL_COUNTERS_RTC_PERIOD );

    Cyg_Interrupt::acknowledge_interrupt(CYGNUM_HAL_INTERRUPT_RTC);

#if defined(CYGVAR_KERNEL_COUNTERS_CLOCK_DSR_LATENCY)
    HAL_CLOCK_READ(&clock_dsr_start);
#endif    
    return Cyg_Interrupt::CALL_DSR|Cyg_Interrupt::HANDLED;
}

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [ECOS] a qustion about measure real-time interrupt latency
  2000-11-29 22:38 [ECOS] a qustion about measure real-time interrupt latency xuas
@ 2000-11-30  9:46 ` Jonathan Larmour
  0 siblings, 0 replies; 2+ messages in thread
From: Jonathan Larmour @ 2000-11-30  9:46 UTC (permalink / raw)
  To: xuas; +Cc: ecos

xuas wrote:
> 
> I cant understand something about isr doing.
> can someone help me ?
> 
> what is mean measure real-time interrupt latency?
> what is aim?

It's there purely to, um...,  measure the latency. It's purely for
information purposes that's all. The kernel tm_basic test takes advantage
of it for example... if you enable this option, it will report the
interrupt (and DSR if also enabled) latency for the clock interrupt and
tell you what it is, purely for your information.

Jifl
-- 
Red Hat, 35 Cambridge Place, Cambridge, UK. CB2 1NS  Tel: +44 (1223) 728762
"Plan to be spontaneous tomorrow."  ||  These opinions are all my own fault

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2000-11-30  9:46 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-11-29 22:38 [ECOS] a qustion about measure real-time interrupt latency xuas
2000-11-30  9:46 ` Jonathan Larmour

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).