[iPAQ] Re: Documentation, battery, kernel (was: Latest versions)

Le sextidi 16 nivôse, an CCXIII, Paul Eggleton a écrit :
> See http://handhelds.org/moin/moin.cgi/BootLoader

Thanks. It does not answer all my question, but along with the source code
found on <URL: http://handhelds.org/cgi-bin/cvsweb.cgi/bootldr/ > I have
enough.

> They are still down at the moment. You can pick them up on http://gmane.org
> though - search for the name of each list.

I should have thought of gmane.org!

> sources can be retrieved and cross-compiled using it. See also
> http://handhelds.org/sources.html

This page is interesting, but all the "FTP/HTTP repository" links are
broken.

> CVS is the easiest place to get the kernel source with patches applied. I'm
> not sure if anyone is maintaining the patches, although I sent a big patch to
> someone recently and I understand he is working on it. (I would recommend
> just using CVS, it's easier).

Alas, the public CVS for handhelds is still down, as far as I can see. I
have tried a perl script to crawl the ViewCVS interface, but I somehow
missed some files. And I do not want to stress too much the ViewCVS server
while debugging it.

Does anyone know if the public read-only CVS will be back online soon?

Nonetheless, I could download some files I wanted to see in the source.

My main problem now is the USB link: it does not work at all, and NFS over
serial is uncomfortably slow when loading a glibc. The craddle and the
hardware are ok, since I used USB to backup the wince image. But when I plug
tie iPaq, I get:

usb 1-2: new full speed USB device using address 16
usb 1-2: device not accepting address 16, error -71
usb 1-2: new full speed USB device using address 17
usb 1-2: device not accepting address 17, error -71

(or -110 instead of on another host).

I tried unplugging/plugging several times, as stated in some old mail here,
but without success.

As far as I can see, the iPaq does not get any IRQ. According to the source
and /proc/interrupts, USB dialogs should trigger the 13, 20 and 21 IRQs, but
all these stay to 0. In fact, I could not see any IRQ reacting specifically
to the USB host configuration requests.

Does anyone have an idea about that problem? Maybe I should try and contact
Ward Willats.