USB transfers in device drivers

[O'Connor, Daniel]

> On 3 May 2019, at 02:31, Hans Petter Selasky <hps at selasky.org> wrote:
> 
> On 2019-05-02 13:38, O'Connor, Daniel wrote:
>>> On 2 May 2019, at 20:33, Hans Petter Selasky <hps at selasky.org> wrote:
>>> 
>>> On 2019-05-02 12:44, O'Connor, Daniel wrote:
>>>>> On 2 May 2019, at 20:02, Hans Petter Selasky <hps at selasky.org> wrote:
>>>>> 
>>>>> On 2019-05-02 11:18, O'Connor, Daniel wrote:
>>>>>> OK, thanks.
>>>>>> To be honest I would much prefer to work out why this particular hardware & software seem to drop the ball for such a long time - 50msec without the kernel getting to schedule something (on a basically idle system) is quite perplexing to me.
>>>>> 
>>>>> Sounds like a lost IRQ issue. Did you try any of the EHCI quirks in hw.usb.ehci ?
>>>> No not, yet - thanks for the pointer!
>>> 
>>> The 50ms delay may also be due to a physical link data error and needed recovery through clear stall which is expensive.
>>> 
>> I see the same error on different hardware sets (both motherboard and USB device) so I don't think it's that.
> 
> If you can check if a USB BULK transfer is pending during this delay, then it might also be a firmware issue.

OK, I'll try and come up with a program to process usbdump output and show things like response latency.

--
Daniel O'Connor
"The nice thing about standards is that there
are so many of them to choose from."
 -- Andrew Tanenbaum