>What's the hardware? We've seen interesting differences with chipsets
>and motheboards. Consider that the pinheaded PC architecture has
>to pass irqs via an emulated 8259A. sometimes a isa/pci interface,
>a pci/processor bus interface and back again!
It's a DELL PC at 450 MHz. That much speed should help.
>(of course, we are now working on a Moto MBX board that has an even
>stupider design, so I guess its not fair to blame everything on Intel
>and Gates).
>
>Also, your terminology is confusing. Are you timing the start of an
>interrupt handler or a periodic scheduled task?
I'm measuring the time between the Interrupt Handler and the Task that is
woken up by the Interrupt Handler.
>The microseconds add up. I'd like to see some measurements on a BX
>board.
I ran the same test on exactly the same hardware (dual boot) running LynxOS.
With LynxOS the time between IT Handler and RT_Task (that is a kernel thread
for LynxOS) is in the range 10-15 usecs. [20-25 for RTL].
My question is then, is there any possible optimization in the rt_sched
to get a better response time ?
>On Wed, Jan 20, 1999 at 09:01:50AM -0800, frasc wrote:
>>
>> Hi folks,
>>
>> We are currently doing some experiment with RTL.
>>
>> Basically what we have been trying to measure so far is the task
>> scheduler latency:
>>
>> Real Time Clock ---> IT wake_up ---> RT Task
>> | |
>> V V
>> Toggle Pin2 Toggle Pin3
>> parallel port parallel port
>>
>> Using a scope we measure an average 22 micro-seconds between IT and Task.
>> [ best case 19.6 usecs, worst case 25 usecs ]
>>
>> Did you guys ever been able to get faster response time ?
>>
>> Thanks in advance for any advice!
--
Frédéric.R.R.Roussel
[EMAIL PROTECTED]
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
