[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Sony vaio, UMASS0, etc
- To: Dennis Edwards <dennis@tesla.cx>,Theo de Raadt <deraadt@cvs.openbsd.org>
- Subject: Re: Sony vaio, UMASS0, etc
- From: Dragos Ruiu <dr@kyx.net>
- Date: Tue, 1 Oct 2002 03:32:06 +0000
- Cc: Adrian Close <adrian@close.wattle.id.au>, misc@openbsd.org
- Organization: all terrain ninjas
- References: <Your message of "Sat, 28 Sep 2002 03:21:51 -0000." <200209280321.51256.dr@kyx.net> <200209280321.51256.dr@kyx.net> <5.1.0.14.2.20020928063717.02936900@nikola.tesla.cx>
Yep, it's the umass one. And I should read more carefully.
Because now as I recall, the last time I snagged data off a
stick I had put the stick in at boot.
I'll go run a quick test right now to eliminate doubt.
Sigh, yep I can read data off it after insertion at boot
and I do _not_ get insertion/extraction notification.
I'm chasing another USB error with that laptop right now too,
cause it isn't happy with with a flash disk that is ok
under wintendo so since i'm getting jiggy with the
usb sniffer and enabling driver diagnostic modes
anyway.... I did some tracing.
BTW It sucks when your usb sniffer makes your memory
stick reader completely inoperable under Windows even
after a reboot. Sigh... fortunately a new version of usb-sniff
off sourceforge seems to work better.
In case anyone cares I have a set of diagnostic traces
from a kernel compiled with full on gory usb and umass diagnostics
with the stick in at boot and with the stick out. In both cases I inserted
and removed the stick which seemed to produce no traffic. I also
have a win32 usb-sniff of stick access under windows if anyone
wants but it doesn't look too enlightening - as it contains no
notification traffic that I see.
Neither under OBSD nor Windows does inserting or removing the
memory stick produce a single peep over USB so whatever
notification the stick device gives of inserts and ejects it doesn't
look like it goes over usb. But windows somehow seems to know
(whether by IRQ or system message queue or whatever) that a stick
has been inserted. Now maybe the win32 usb sniffer and the obsd
kernel diagnostics do not pick up the messages, but I'm now looking
for some windows message queue sniffer so I can try to identify
just what the notification mechanism is... cause i want to be able
to switch memory sticks too!
So in any case it doesn't look like the fault lies in umass with mem
stick notifications on the VAIO. However the fact that umass does
not like my usb flash disk is another matter. Just in case anyone
cares about this I've also have win32 sniffs and obsd diagnostics
from my flash disk. Email me if they are of any use to you.
But it doesn't sound like anyone else is really working on fixing either
of these issues and I am left to my own devices. It looks like fixing
VAIO mem sticks isn't quite straightforward so I'm not holding my
breath for it soon :(. The usb flash disk OTOH should be fixable
more easily I hope.... If anyone else cares - mail me and
I'll keep you posted if I make any progress.
sigh,
--dr
On September 28, 2002 12:42 pm, Dennis Edwards wrote:
> I'm not aware of any variation whatsoever in the z505le/ls range.
>
>
> Dragos? Is yours the USB mass storage one?
> I explained mine was umass when this thread started...
>
> At 04:28 AM 9/28/2002, Theo de Raadt wrote:
> > > On September 27, 2002 07:21 am, Mark Hoff wrote:
> > > > Well.. that's what I wanted to hear... thanks.
> > > >
> > > >
> > > > Do you have the same model Vaio as I do? z505le (or ls, same thing
> >
> > really).
> >
> > > > If I boot without the stick in, I get a sense error... if I insert
> > > > the stick, I get no indications of any kind that it's been detected.
> > > > If I insert and remove the stick, there is also no indication or
> > > > change.
> > >
> > > Hum, z505ls, memory sticks work fine. just pre-3.2ish -current.
> >
> >There are at least 3 varients of the Sonly memory stick interface.
> >
> >The first type sat on the 2nd IDE bus of the pciide interface.
> >The second type sat on the PCI bus as a unique device (no docs).
> >The third type is a umass usb device.
> >
> >Depends what laptop you have. Can't even use the z505xx names, since some
> >models changed a lot without changing xx names.
>
> ----------------------------------------
> Potestatem obscuri lateris nescis.
--
dr@kyx.net pgp: http://dragos.com/kyxpgp
Advance CanSecWest/03 registration available: http://cansecwest.com
"The question of whether computers can think is like the question
of whether submarines can swim." --Edsger Wybe Dijkstra 1930-2002