lkml.org 
[lkml]   [2011]   [Oct]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [Patch] Increase USBFS Bulk Transfer size
From
On Fri, Oct 14, 2011 at 5:42 AM, Markus Rechberger
<mrechberger@gmail.com> wrote:
> On Fri, Oct 14, 2011 at 4:47 AM, Markus Rechberger
> <mrechberger@gmail.com> wrote:
>> On Fri, Oct 14, 2011 at 12:19 AM, James Courtier-Dutton
>> <james.dutton@gmail.com> wrote:
>>> Why don't you try a bulk size of 12032 instead of 24064 and not try 12288 as
>>> you appear to be doing in the logs. Post the logs for that.
>>
>> I tried that earlier already of course it fails. If I could pick a
>> smaller transfer size I would be happy since
>> the device would work with all 2.6.x kernels out of the box and I
>> wouldn't have to waste my time with it.
>> Unfortunately it requires the 24064 bytes.
>> The more flexible device A which is mentioned here confirms that there
>> can be some impact on the
>> bulk transfer size.
>> However to learn about this it's needed to look at the bottom line of
>> USB on the physical layer.
>>
>> And I disagree with Alan Cox it's not about being a crappy device or
>> not, it's more like about something
>> that is not well understood here. Most people are familiar with
>> Software only here and not with the physical
>> USB bottom Layer, otherwise the fact that the devices can have an
>> impact on this wouldn't be such a surprise.
>>
>
> however to not say that I'm unwilling to do that and that is the
> reason for not accepting this patch
> http://www.sundtek.de/support/notworking2.log
>
> even if the value is exposed to sysfs, it still requires the static
> value in the kernel. The current value
> used in the patch is based on what is in the HW specs of device A
> which has the flexible bulk transfer setting.
> The inflexible device which uses 24064 bytes works with all other
> Operating systems by using that value
> and gives exactly the same results with other transfer sizes than that.
>

I didn't check this before the half of it is of course not a multiple
of 512 so the
logfile only shows up 11776 of course.

24064 is the smallest common multiple of 188 and 512.

BR,
Markus


\
 
 \ /
  Last update: 2011-10-14 05:51    [from the cache]
©2003-2011 Jasper Spaans