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

Bug#719680: Driver mv643xx_eth_port requests probe deferral



On Thu, 2013-09-05 at 12:32 +0100, Ben Hutchings wrote:
> On Thu, 2013-09-05 at 08:22 +0100, Ian Campbell wrote:
> > On Thu, 2013-09-05 at 03:39 +0100, Ben Hutchings wrote:
> > > On Sun, 2013-09-01 at 16:07 +0100, Ian Campbell wrote:
> > > > On Sun, 2013-09-01 at 15:53 +0100, Ben Hutchings wrote:
> > > > > On Sun, 2013-09-01 at 15:19 +0100, Ian Campbell wrote:
> > > > > > On Sat, 2013-08-31 at 19:52 +0100, Ian Campbell wrote:
> > > > > > > On Sat, 2013-08-31 at 18:59 +0100, Ian Campbell wrote:
> > > > > > > > On Sat, 2013-08-31 at 11:29 +0100, Ian Campbell wrote:
> > > > > > > > > Could this be due to lack of CONFIG_MVMDIO on armel? I'm going to try
> > > > > > > > > that.
> > > > > > > > 
> > > > > > > > It appears to have done the trick on trunk (3.11-rc5).
> > > > > > > > 
> > > > > > > > The kernel doesn't find my disks though :-(
> > > > > > > 
> > > > > > > More than likely the lack of CONFIG_USB_EHCI_HCD_ORION I should think.
> > > > > > 
> > > > > > That does the trick but mkinitramfs also needs to be taught about the
> > > > > > new ehci-orion module which this produces.
> > > > > 
> > > > > Maybe it should be including drivers/usb/host/* (although that currently
> > > > > includes some really obscure drivers we probably shouldn't build at
> > > > > all).
> > > > 
> > > > That's an option I suppose, but it's a pretty long list, of which the
> > > > majority are, I expect, useless on any given platform I think. An
> > > > example of grep HCD on the kirkwood config is below (it won't even
> > > > include things whose dependencies aren't met as disabled).
> > > [...]
> > > 
> > > That prompted me to remove some of the sillier HCD drivers from the
> > > config in trunk.  It's also possible to pass a list of exclusions to
> > > copy_modules_dir, as we do for the hid and net directories.
> > 
> > My greps in linux/debian and kernel-wedge are completely failing to find
> > those lists. I can see the support for excluding entire udebs in kernel
> > wedge, but that is all I found.
> 
> I thought we were talking about initramfs-tools here...

oops, got udebs on the brain... sorry!

Ian.


Reply to: