.. so, even if i blank the receive buffer, the NIC is copying some old
RX data into it, then the only information that's valid is actually
the link quality (ie, how much time the link has been silent or not.)
Which I guess is what the LINK_DETERIORIATION notification is.
I wish I knew how to tell
.. actually, this is even more screwed up. here's what I get in the RX
notification messages:
time=1396123498.839215
g 9 %�
� � � � � � $ � � D � � �
� � P � � � E
U � @ �
�
�
D \ D \ z � d { " h o s t _ i
n t " : 7 4 0 7 6 9 6 2 7 ,
" v e r s i o n " : [ 1
On 29 March 2014 12:40, Adrian Chadd wrote:
> So I exposed the iwi statistics messages but for some odd reason I
> actually get garbage statistics. The linux driver only looks at two
> fields (which are correctly filled out) but the rest of the fields
> aren't filled out.
>
> I'm not yet sure why.
So I exposed the iwi statistics messages but for some odd reason I
actually get garbage statistics. The linux driver only looks at two
fields (which are correctly filled out) but the rest of the fields
aren't filled out.
I'm not yet sure why.
-a
On 17 March 2014 18:46, Adrian Chadd wrote:
> Y
Yup, I found a 2100 and 2200. And yes it does dump out a bunch of
statistics stuff when the NIC is running.
I'll see about exposing the statistics messages via some API soon.
-a
On 14 March 2014 18:55, Adrian Chadd wrote:
> Nope; there's no code for handling the messages yet.
>
> I'll dig arou
Nope; there's no code for handling the messages yet.
I'll dig around to see if I can find a 2200 series NIC in a laptop
here or in my collection. I know I have 3945 NICs, but not the 2200..
-a
On 14 March 2014 18:06, Jim Long wrote:
> Thank you for your reply.
>
> This would be via syslog in /
Thank you for your reply.
This would be via syslog in /var/log/messages? Nothing that I see:
$ zgrep -il notification /var/log/messages*
$
On Fri, Mar 14, 2014 at 05:50:12PM -0700, Adrian Chadd wrote:
> It'd be stuff like this:
>
> case IWI_NOTIF_TYPE_CALIBRATION:
> case IWI
Hi,
right. The linux driver (ipw2x00) has a bunch more quality related
stuff that I don't think the iwi driver is decoding. But i bet a
little bit of C coding would get the statistics out into a useful
format that we can log and analyse.
-a
On 14 March 2014 17:50, Adrian Chadd wrote:
> It'd b
It'd be stuff like this:
case IWI_NOTIF_TYPE_CALIBRATION:
case IWI_NOTIF_TYPE_NOISE:
case IWI_NOTIF_TYPE_LINK_QUALITY:
DPRINTFN(5, ("Notification (%u)\n", notif->type));
break;
-a
On 14 March 2014 17:18, Adrian Chadd wrote:
> Hi,
>
> I'
Hi,
I'm not sure what kind of statistics or diagnostics iwi spits out.
It's likely worth reviewing the linux and freebsd drivers to see if it
does spit out any kind of statistics messages. That's a good starting
point.
Thanks!
-a
On 14 March 2014 16:00, Jim Long wrote:
> I have a long-standi
I have a long-standing problem that involves the iwi interface on
my Thinkpad T42 running 9.2-PRERELEASE circa 28 Aug 2013.
Some wifi connections I make show good signal strength, but poor
latency and/or packet loss to the WAP IP. I can't find other
wifi users who perceive the wifi performance as
11 matches
Mail list logo