Re: NETGEAR FA310TX problems

["Dennis E. Arce" <dennis@bournetech.com>]

Marty-

I'm using a pentium 200 MMX for this application.

A changed boguscnt = 1000000 (1 zero more); and 10000000 (two extra 0s) and
put
printf("boguscnt= %d", boguscnt);
tulip_wait(3*TICKS_PER_SEC);

after while and boguscnt was= 0 all three times.  Still no change in the
result.

Also, I tried several other versions of netboot to no avail.

-Dennis

Marty Connor wrote:

> On 6/20/2000 7:58 AM Dennis E. Arce dennis@bournetech.com wrote:
> >I have the came Card, FA310TX REV-D2 with LC82169C on it.
> >Thanks.  Is there a netboot version you know works with it for the time
> >being?
>
> I don't know too much about netboot, but I'm sure someone else on the
> list will be able to help with that.
>
> I would be curious about your machine, however.  I mainly have P200s,
> which are slow compared to more modern machines.  I have a theory that
> there are residual timing loops based on (i=10000;i>0;i--) kind of stuff
> that may be causing problems on faster machines.  I will build a PIII 550
> machine to test this theory and to debug a bit better.
>
> I replaced most of that kind of loop with loops based on
> (timeout=currenttime() + 1*TICKS_PER_SECOND).  When looking at the code,
> it appears there are still some of the old "spin loops" in the logic for
> reading the MAC address on the LC82169C.  We could easily test this.
> Here's some of the code in question:
>
>   /* Hardware Address retrieval method for LC82C168 */
>   if (vendor == PCI_VENDOR_ID_LINKSYS && dev_id ==
> PCI_DEVICE_ID_DEC_TULIP) {
>     for (i = 0; i < 3; i++) {
>       int value, boguscnt = 100000;
>       outl(0x600 | i, ioaddr + 0x98);
>       do
>         value = inl(ioaddr + CSR9);
>       while (value < 0  && --boguscnt > 0);
>
> notice that "boguscnt = 100000".  Just for testing, try adding a zero to
> make it 1000000, which would give more time for retrieval of the data.
> We should probably replace this logic with the timeout logic as I've done
> elsewhere, but this could help debug the situation.  Also, knowing the
> value of boguscnt after the loop might tell us if the loops is
> terminating before the data is retrieved.  We could also put some printf
> statements in to look at the data.  I'm betting on a timing bug, however.
>
> It will take me a little time to get back up to speed and to assemble my
> testing facility, but perhaps this "virtual debugging" will be of some
> help.
>
> Regards,
>
> Marty
>
> P.S. Any HAM Radio operators on the list? 73's
>
> ---
>    Name: Martin D. Connor
> US Mail: Entity Cyber, Inc.; P.O. Box 391827; Cambridge, MA 02139; USA
>   Voice: (617) 491-6935, Fax: (617) 491-7046
>   Email: mdc@thinguin.org
>     Web: http://www.thinguin.org/

===========================================================================
This Mail was sent to netboot mailing list by:
"Dennis E. Arce" <dennis@bournetech.com>
To get help about this list, send a mail with 'help' as the only string in
it's body to majordomo@baghira.han.de. If you have problems with this list,
send a mail to netboot-owner@baghira.han.de.