Re: kern/50249 (XHCI driver missing delay after address set)

[Robert Sprowson <webpages%sprow.co.uk@localhost>]

The following reply was made to PR kern/50249; it has been noted by GNATS.

From: Robert Sprowson <webpages%sprow.co.uk@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/50249 (XHCI driver missing delay after address set)
Date: Sun, 11 Oct 2015 21:11:52 +0100

 On 11 Oct, David Holland <dholland-bugs%netbsd.org@localhost> wrote:
 > The following reply was made to PR kern/50249; it has been noted by GNATS.
 >  
 > > I've been following the branch with interest, but assumed the
 > > trunk was still open to submissions since only the bold/brave would
 > > be using the experimental version. Hence adding the
 > > SET_ADDRESS_SETTLE to the trunk benefits those users.
 >
 >  FWIW, that sounds potentially worthwhile to me, but I'm not the one
 >  doing the work. However, in cases like this please send one change per
 >  patch; needing to disentangle two changes strongly decreases the
 >  likelihood of patches getting merged...
 
 The SET_ADDRESS_SETTLE aspect is the 2 lines added at 1578, everything else
 was clarifying the name of the ring cycle state (rcs) in the state struct. 
 
 Happy to split the patch, but I think a subsequent discussion with skrll by
 email to restructure the XHCI<->USB calling may have superceded the
 SET_ADDRESS_SETTLE part anyway, by using the delay in usb_subr.c and calling
 out to the XHCI HC only for the HC specific bits.
 
 Regards,
 Robert.