Messages in this thread | | | Date | Mon, 2 Aug 2010 19:28:20 +0200 | From | Daniel Mack <> | Subject | Re: [098/140] USB: ehci-mxc: bail out on transceiver problems |
| |
On Mon, Aug 02, 2010 at 10:10:50AM -0700, Greg KH wrote: > On Mon, Aug 02, 2010 at 02:45:14PM +0200, Wolfram Sang wrote: > > Daniel mentioned that this patch could cause a "regression" with boards > > which have hardware problems in form of a floating chip select > > (originating from the reference design!). For such boards, the probing > > might now fail, leaving the boards without USB, while it worked before > > (although it only worked because of ignoring the actual problem). > > > > I can't make my mind if it is better to fix the potential OOPS or to > > keep those boards working for older kernels. I just wanted to mention > > it, so this is issue won't be overlooked. > > Odd. Well, I'd rather match what the newer kernels do here, so I'll > keep it for now unless someone objects.
Yep, I was about to write the same thing. Let's see what it breaks and fix up the mess afterwards :)
Thanks, Daniel
| |