Comment 12 for bug 915985

2012/4/11 Robert Ayrapetyan <email address hidden>

> Seems I've missed confirmation mail somewhere, now registered.
>
>
> > have you patched your 2.6.1 or used trunk + the patch?
> I've used trunk + the patch.
>
>
> > I'd be interested in a trace without the patch, to
> > see if we're still on the overflow side.
>
> trunk (rev.3529) "with no patch" log attached.
>
> Btw I've checked all logs I've sent so far - all of them contain:
>
> 1.199490 libusb_get_report: Unknown error
> 1.199512 Can't retrieve Report 0c: Input/output error
>
> Seems you are talking about Dmitry's case (overflow).
>

you're right: your issues are the same, but BSD doesn't report EOVERFLOW
has it should.
while Dmitry is running on Linux, the patch should work there.

@Robert: could you please run the test again (with wathever version), using
debug level 5 (-DDDDD) and doing "export USB_DEBUG=3" before launching the
driver?

if there is no way to catch overflow on BSD, I fear I'll have to create a
driver option to bypass this...

cheers,
Arno

>
>
> On 04/11/12 16:35, Arnaud Quette wrote:
>
>>
>> 2012/4/11 Robert Ayrapetyan <<email address hidden>
>> <mailto:robert.ayrapetyan@**gmail.com <email address hidden>>>
>>
>>
>> Hi.
>>
>>
>> Hi Robert,
>>
>> please check your subscription.
>> I'm still told that you're not subscribed!
>>
>> Mine output after patch looks same (attached).
>>
>>
>> in fact, it's different.
>> It's no more an overflow error, but an I/O error!
>>
>> have you patched your 2.6.1 or used trunk + the patch?
>> in the latter case, I'd be interested in a trace without the patch, to
>> see if we're still on the overflow side.
>>
>> On Wed, Apr 11, 2012 at 12:55 AM, Arnaud Quette
>> <<email address hidden> <mailto:<email address hidden>>**> wrote:
>> > Hi Robert and Dmitry,
>> >
>> > I'm crossing LP with the nut mailing list, since this is the same
>> EOVERFLOW
>> > issue.
>> >
>> > @Dmitry: it comes out that my previous patch was missing the
>> libusb.c part
>> > :-/
>> >
>> > to both, the attached patch should fix your issue.
>> > please send compressed debug output to confirm the fix.
>> >
>> > I'll have to do some more testing tomorrow since it's "blind
>> coded" (Ie,
>> > just compiled, not tested with HW)
>> >
>> > cheers,
>> > Arnaud
>>
>>
>> cheers,
>> Arnaud
>> --
>> Linux / Unix Expert R&D - Eaton - http://powerquality.eaton.com
>> Network UPS Tools (NUT) Project Leader - http://www.networkupstools.**
>> org/ <http://www.networkupstools.org/>
>> Debian Developer - http://www.debian.org
>> Free Software Developer - http://arnaud.quette.free.fr/
>>
>>