Re: [PATCH v2 0/2] USB: EHCI: Allow users to override 80% max periodic bandwidth

[Sarah Sharp <sarah.a.sharp@linux.intel.com>]

On Fri, Jun 24, 2011 at 08:48:06PM +0400, Kirill Smelkov wrote:
> 
> Changes since v1:
> 
> 
>  - dropped RFC status as "this seems like the sort of feature somebody might
>    reasonably want to use -- if they know exactly what they're doing";
> 
>  - new preparatory patch (1/2) which moves already-in-there sysfs code into
>    ehci-sysfs.c;
> 
>  - moved uframe_periodic_max parameter from module option to sysfs attribute,
>    so that it can be set per controller and at runtime, added validity checks;
> 
>  - clarified a bit bandwith analysis for 96% max periodic setup as noticed by
>    Alan Stern;
> 
>  - clarified patch description saying that set in stone 80% max periodic is
>    specified by USB 2.0;

Have you tested this patch by maxing out this bandwidth on various
types of host controllers?  It's entirely possible that you'll run into
vendor-specific bugs if you try to pack the schedule with isochronous
transfers.  I don't think any hardware designer would seriously test or
validate their hardware with a schedule that is basically a violation of
the USB bus spec (more than 80% for periodic transfers).

But if Alan is fine with giving users a way to shoot themselves in the
foot, and it's disabled by default, then I don't particularly mind this
patch.

Sarah Sharp