[ECOS] Ping inconsistent ecos

[ECOS] Ping inconsistent ecos

[prashanthu baragur]

Hi,

We are using ecos on IXP425 processor and ported intel npe microcode
to enable ethernet interface. Now i observer a wierd behavior with the
ICMP test packets.
When i tried to ping the target, following was the observation.

64 bytes from 192.168.250.235: icmp_seq=1832 ttl=64 time=14.2 ms
64 bytes from 192.168.250.235: icmp_seq=1833 ttl=64 time=212 ms
64 bytes from 192.168.250.235: icmp_seq=1834 ttl=64 time=47.3 ms
64 bytes from 192.168.250.235: icmp_seq=1835 ttl=64 time=1000 ms
64 bytes from 192.168.250.235: icmp_seq=1836 ttl=64 time=0.665 ms
64 bytes from 192.168.250.235: icmp_seq=1837 ttl=64 time=32.5 ms
64 bytes from 192.168.250.235: icmp_seq=1838 ttl=64 time=182 ms
64 bytes from 192.168.250.235: icmp_seq=1839 ttl=64 time=17.2 ms
64 bytes from 192.168.250.235: icmp_seq=1840 ttl=64 time=1000 ms

Min time taken is 0.665 ms and Max 1 second. I also tried with
cross(pear to pear) ethernet cable to isolate from the LAN. There was
no difference.
Due to this all ethernet based applications are becoming slow.

Any coments are appreciable.

Thanks
Prashanth

--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] Ping inconsistent ecos

[Andrew Lunn]

On Wed, Aug 03, 2005 at 04:23:00PM +0530, prashanthu baragur wrote:
> Hi,
> 
> We are using ecos on IXP425 processor and ported intel npe microcode
> to enable ethernet interface. Now i observer a wierd behavior with the
> ICMP test packets.
> When i tried to ping the target, following was the observation.
> 
> 64 bytes from 192.168.250.235: icmp_seq=1832 ttl=64 time=14.2 ms
> 64 bytes from 192.168.250.235: icmp_seq=1833 ttl=64 time=212 ms
> 64 bytes from 192.168.250.235: icmp_seq=1834 ttl=64 time=47.3 ms
> 64 bytes from 192.168.250.235: icmp_seq=1835 ttl=64 time=1000 ms
> 64 bytes from 192.168.250.235: icmp_seq=1836 ttl=64 time=0.665 ms
> 64 bytes from 192.168.250.235: icmp_seq=1837 ttl=64 time=32.5 ms
> 64 bytes from 192.168.250.235: icmp_seq=1838 ttl=64 time=182 ms
> 64 bytes from 192.168.250.235: icmp_seq=1839 ttl=64 time=17.2 ms
> 64 bytes from 192.168.250.235: icmp_seq=1840 ttl=64 time=1000 ms
> 
> Min time taken is 0.665 ms and Max 1 second. I also tried with
> cross(pear to pear) ethernet cable to isolate from the LAN. There was
> no difference.
> Due to this all ethernet based applications are becoming slow.

I _think_ the IXP435 code is polled, not interrupt drive. So this is
normal behavior. I _think_ there has been work done to make this
interrupt driven, so try googling.

        Andrew

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] Ping inconsistent ecos

[Gary Thomas]

On Wed, 2005-08-03 at 16:23 +0530, prashanthu baragur wrote:
> Hi,
> 
> We are using ecos on IXP425 processor and ported intel npe microcode
> to enable ethernet interface. Now i observer a wierd behavior with the
> ICMP test packets.
> When i tried to ping the target, following was the observation.
> 
> 64 bytes from 192.168.250.235: icmp_seq=1832 ttl=64 time=14.2 ms
> 64 bytes from 192.168.250.235: icmp_seq=1833 ttl=64 time=212 ms
> 64 bytes from 192.168.250.235: icmp_seq=1834 ttl=64 time=47.3 ms
> 64 bytes from 192.168.250.235: icmp_seq=1835 ttl=64 time=1000 ms
> 64 bytes from 192.168.250.235: icmp_seq=1836 ttl=64 time=0.665 ms
> 64 bytes from 192.168.250.235: icmp_seq=1837 ttl=64 time=32.5 ms
> 64 bytes from 192.168.250.235: icmp_seq=1838 ttl=64 time=182 ms
> 64 bytes from 192.168.250.235: icmp_seq=1839 ttl=64 time=17.2 ms
> 64 bytes from 192.168.250.235: icmp_seq=1840 ttl=64 time=1000 ms
> 
> Min time taken is 0.665 ms and Max 1 second. I also tried with
> cross(pear to pear) ethernet cable to isolate from the LAN. There was
        peer to peer

> no difference.
> Due to this all ethernet based applications are becoming slow.
> 
> Any coments are appreciable.

Are you running an eCos application, or is this just a ping
to RedBoot?  If the latter, this is quite normal behaviour
as the RedBoot networking is purely polled and the time between
polls varies [due to other factors]

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss