No subject

[<bogus@does.not.exist.com>]

the silicon itself at power-on. 'abcd' sounds a bit too convenient to be
what's in EEPROM/OTP; so maybe it's a default value in the silicon?

(All just conjecture here at this point.)


Adrian

On 10 April 2011 23:17, Mohammed Shafi <shafi.ath9k@gmail.com> wrote:

> On Sun, Apr 10, 2011 at 8:41 PM, Peter Stuge <peter@stuge.se> wrote:
> > Mohammed Shafi wrote:
> >> > Is this a serious proposal from Atheros, or just your attempt at
> >> > a quick fix?
> >>
> >> No! its purely a personal idea (am completely responsible for the
> >> mistake),and I will take a look at it carefully to fix this.
> >
> > Sorry, I didn't mean that you made a mistake, just that the
> > suggestion probably would not get us closer to the actual issue.
> >
> > Bus level issues are indeed difficult. :\
>
> thanks, i did not know that. thought simple as adding another device id.
> >
> >
> >> > A device having an unexpected PCI id means that something is really
> >> > wrong in the device or on the bus, and the solution is rarely to
> >> > pretend that it didn't happen.]
> >>
> >> Yeah I can see that, hoping that I may get a correct Device ID from
> >> the reporter. I dont think 'abcd' is a proper vendor id.
> >
> > Yes, it's easy to spot. The question is how we can find out *why*
> > this happened, so that this error case can be prevented.
>
> Yes sure.
> >
> >
> >> > Since this card should work fine in principle, maybe it's some issue
> >> > with missing, or wrong, firmware stored on the Linux system.
> >>
> >> AR9382 does not seems to have firmware
> >
> > Aha! That's only for the USB devices maybe. I don't know much detail
> > for these latest devices.
> >
> currently only  ath9k_htc needs firmware.
>
> >
> >> and you have any idea what might went wrong.
> >
> > Sorry, I don't understand what you mean here.
>
> Your suggestions about what might have went wrong, as you had already
> told it might be a bus level issue.
>
> >
> >
> >> Also why its detected as Ethernet Controller rather than
> >> Network controller.
> >
> > This string comes from the pciutils package and could easily have
> > changed. Better look at the numerical device class code, which is
> > what is read from hardware.
>
> thanks, I will look into that.
>
> >
> > But I expect that when one thing in config space (device id) is bogus
> > then the rest of config space is also quite possibly bogus, for the
> > same reason, whatever it is.
> >
> >
> > //Peter
> > _______________________________________________
> > ath9k-devel mailing list
> > ath9k-devel at lists.ath9k.org
> > https://lists.ath9k.org/mailman/listinfo/ath9k-devel
> >
> _______________________________________________
> ath9k-devel mailing list
> ath9k-devel at lists.ath9k.org
> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>

--20cf300256ecb39f7b04a09232d4
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Incorrect or misplaced EEPROM/OTP data, perhaps?<div><br></div><div>From wh=
at I gather, the PCI ID on earlier devices is loaded out of EEPROM by the s=
ilicon itself at power-on. &#39;abcd&#39; sounds a bit too convenient to be=
 what&#39;s in EEPROM/OTP; so maybe it&#39;s a default value in the silicon=
?</div>
<div><br></div><div>(All just conjecture here at this point.)</div><div><br=
></div><div><br></div><div>Adrian</div><div><br><div class=3D"gmail_quote">=
On 10 April 2011 23:17, Mohammed Shafi <span dir=3D"ltr">&lt;<a href=3D"mai=
lto:shafi.ath9k@gmail.com">shafi.ath9k at gmail.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex;"><div class=3D"im">On Sun, Apr 10, 2011 at 8=
:41 PM, Peter Stuge &lt;<a href=3D"mailto:peter@stuge.se">peter at stuge.se</a=
>&gt; wrote:<br>

&gt; Mohammed Shafi wrote:<br>
&gt;&gt; &gt; Is this a serious proposal from Atheros, or just your attempt=
 at<br>
&gt;&gt; &gt; a quick fix?<br>
&gt;&gt;<br>
&gt;&gt; No! its purely a personal idea (am completely responsible for the<=
br>
&gt;&gt; mistake),and I will take a look at it carefully to fix this.<br>
&gt;<br>
&gt; Sorry, I didn&#39;t mean that you made a mistake, just that the<br>
&gt; suggestion probably would not get us closer to the actual issue.<br>
&gt;<br>
&gt; Bus level issues are indeed difficult. :\<br>
<br>
</div>thanks, i did not know that. thought simple as adding another device =
id.<br>
<div class=3D"im">&gt;<br>
&gt;<br>
&gt;&gt; &gt; A device having an unexpected PCI id means that something is =
really<br>
&gt;&gt; &gt; wrong in the device or on the bus, and the solution is rarely=
 to<br>
&gt;&gt; &gt; pretend that it didn&#39;t happen.]<br>
&gt;&gt;<br>
&gt;&gt; Yeah I can see that, hoping that I may get a correct Device ID fro=
m<br>
&gt;&gt; the reporter. I dont think &#39;abcd&#39; is a proper vendor id.<b=
r>
&gt;<br>
&gt; Yes, it&#39;s easy to spot. The question is how we can find out *why*<=
br>
&gt; this happened, so that this error case can be prevented.<br>
<br>
</div>Yes sure.<br>
<div class=3D"im">&gt;<br>
&gt;<br>
&gt;&gt; &gt; Since this card should work fine in principle, maybe it&#39;s=
 some issue<br>
&gt;&gt; &gt; with missing, or wrong, firmware stored on the Linux system.<=
br>
&gt;&gt;<br>
&gt;&gt; AR9382 does not seems to have firmware<br>
&gt;<br>
&gt; Aha! That&#39;s only for the USB devices maybe. I don&#39;t know much =
detail<br>
&gt; for these latest devices.<br>
&gt;<br>
</div>currently only =A0ath9k_htc needs firmware.<br>
<div class=3D"im"><br>
&gt;<br>
&gt;&gt; and you have any idea what might went wrong.<br>
&gt;<br>
&gt; Sorry, I don&#39;t understand what you mean here.<br>
<br>
</div>Your suggestions about what might have went wrong, as you had already=
<br>
told it might be a bus level issue.<br>
<div class=3D"im"><br>
&gt;<br>
&gt;<br>
&gt;&gt; Also why its detected as Ethernet Controller rather than<br>
&gt;&gt; Network controller.<br>
&gt;<br>
&gt; This string comes from the pciutils package and could easily have<br>
&gt; changed. Better look at the numerical device class code, which is<br>
&gt; what is read from hardware.<br>
<br>
</div>thanks, I will look into that.<br>
<div><div></div><div class=3D"h5"><br>
&gt;<br>
&gt; But I expect that when one thing in config space (device id) is bogus<=
br>
&gt; then the rest of config space is also quite possibly bogus, for the<br=
>
&gt; same reason, whatever it is.<br>
&gt;<br>
&gt;<br>
&gt; //Peter<br>
&gt; _______________________________________________<br>
&gt; ath9k-devel mailing list<br>
&gt; <a href=3D"mailto:ath9k-devel@lists.ath9k.org">ath9k-devel at lists.ath9k=
.org</a><br>
&gt; <a href=3D"https://lists.ath9k.org/mailman/listinfo/ath9k-devel" targe=
t=3D"_blank">https://lists.ath9k.org/mailman/listinfo/ath9k-devel</a><br>
&gt;<br>
_______________________________________________<br>
ath9k-devel mailing list<br>
<a href=3D"mailto:ath9k-devel@lists.ath9k.org">ath9k-devel at lists.ath9k.org<=
/a><br>
<a href=3D"https://lists.ath9k.org/mailman/listinfo/ath9k-devel" target=3D"=
_blank">https://lists.ath9k.org/mailman/listinfo/ath9k-devel</a><br>
</div></div></blockquote></div><br></div>

--20cf300256ecb39f7b04a09232d4--