I looked at those results and noticed that they were not using the
POSIX RT features the normal linux provides. Also, the numbers look
suspect anyway, it looks like a lot of swapping might be taking place
during these tests.
POSIX-RT should give you decent 10ms level scheduling is user-land.
On 4 Mar, Thomas Fruehbeck wrote:
>
> That's a choke!
>
> Please have a look into 'Linux Device Drivers' by Alessandro Rubini, if you doubt it.
> Anyway I would advise you to read this book, before believing in test results like
> the one you refer to.
>
> Thomas
>
>
>> > Where did You take the information, that plain Linux may block interrupts
>> > for 10's of milliseconds?
>>
>>
>> In:
>>
>> http://www.zentropix.com/realtime/testdata/si.html
>>
>> there is a study which shows that the interrupt response of Std-Linux
>> can be ut to 27.5 ms at full load.
--
Tyson D Sawyer Real World Interface, Inc.
Senior Systems Engineer Robots for the Real World
[EMAIL PROTECTED] http://www.rwii.com/
--- [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/
