On Fri, 2015-11-13 at 14:36 +0530, Sudip Mukherjee wrote: > On Fri, Nov 13, 2015 at 01:07:05PM +0530, Sudip Mukherjee wrote: <snip> > > or another easy solution. after 8250_pci probes and binds to your device > you can unbind it and then you can use the out-of-tree driver. > > regards > sudip That would certainly seem like an easy solution, and was attempted as a workaround, but two things came up against it: 1) One of our customers that build DSVs indicated that though they had used "unbind" to successfully get the 8250 out and the Exar-provided driver in, they did not want to rely on the PCI location of the board never changing, or on having to correctly parse the location of the board each time. It was an uncertainty they were unwilling to accept. Now, it's quite probable that there is no uncertainty in either of those things (PCI location, or the ability to parse it), but it was still seen as an un-needed risk by them. 2) When I attempted to writeup the unbind workaround to put into the README, I had no success. Every time I unbound the board, and then insmoded the Exar driver, I got an exception flood (IRQ happened and no one cared, IRQ disabled). After putting about 3 weeks into this effort, I concluded that it had nothing to do with RTD boards themselves, and that Exar wasn't able to help resolve it, partly because they could not reproduce it on their end. Their mantra at this point is to remove the offending code from the 8250_pci and rebuild the kernel. After that, it was decided to shelve working on Exar's problem and see if I could find some other way so that we don't have to require our customers to do a kernel build just to use our boards...in other words, back to the pre-3.8 utopia. I decided to try a patch to exclude RTD-only boards from the 8250. Greg kindly offered just to put the Exar driver in instead. And now we're here. Sorry for the long story. Rob G. -- To unsubscribe from this list: send the line "unsubscribe linux-serial" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
