Re: [PATCH 2/2] usb-uas: update MAINTAINERS entry

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Jan 25, 2013 at 03:19:22PM +0100, Gerd Hoffmann wrote:
>   Hi,
> 
> >> diff --git a/MAINTAINERS b/MAINTAINERS
> >> index 8ae709e..c5b37de 100644
> >> --- a/MAINTAINERS
> >> +++ b/MAINTAINERS
> >> @@ -7911,9 +7911,10 @@ F:	drivers/net/wireless/ath/ar5523/
> >>  USB ATTACHED SCSI
> >>  M:	Matthew Wilcox <willy@xxxxxxxxxxxxxxx>
> >>  M:	Sarah Sharp <sarah.a.sharp@xxxxxxxxxxxxxxx>
> >> +M:	Gerd Hoffmann <kraxel@xxxxxxxxxx>
> > 
> > Should Matthew be removed from this?
> 
> Dunno, Sarah?

Matthew has said he doesn't have time to work on UAS, so I would suggest
removing him.

> > Also, any word on when I can remove the CONFIG_BROKEN marking on this
> > driver?
> 
> With the patches in -next uas itself should be reasonable solid.
> 
> Problem is that uas is pretty much the only device using streams,
> so uas will be the one who triggers any stream bugs in xhci.
> I have no idea how solid xhci streams support is at the moment.

The xHCI streams support isn't well tested, because the UAS devices I
had were so buggy that I couldn't fully test them, and the UAS driver
didn't (doesn't still?) properly support cancellation or device reset.
So the first SCSI command that the device didn't like basically hung the
driver.

Also, I do know there are a couple of streams work-arounds that need to
be created for the Intel Panther Point xHCI host controller, so I would
suggest testing with an NEC/Renesas xHCI host instead.  I will get
around to implementing them, but other bugs have taken a higher
priority.

> Sarah, is there some way to avoid using streams?  The UAS specs seems to
> imply using streams is mandatory when connected to a USB-3 port, is that
> correct?  Is there some way to force usb3 devices into usb2 mode even
> when plugged into a usb3 port?  I'd like to have a no_streams module
> option if possible ...

USB 3.0 UAS devices are really supposed to be using streams.  You might
as well just use Bulk-only-Transport (BoT) if you're not using streams.

As for forcing a USB 3.0 port to USB 2.0 speeds, I think you could write
a userspace program to use usbfs to disable the USB 3.0 port.  I've
never tried it though.

Sarah Sharp
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux Media]     [Linux Input]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Old Linux USB Devel Archive]

  Powered by Linux