Hi!
I don't know what you trying to fix, but your patch dd72be99
"x86/pci/mrst: Use configuration mechanism 1 for 00:00.0, 00:02.0,
00:03.0" breaks Intel Medfield devices.
After it was applied (sorry, was on vacation, didn't noticed sooner)
lspci shows only 18 out of 34 devices. That means the so called fake
PCI devices are gone.
This is a bad approach and patch should be reverted. Bin, please,
explain what you are trying to fix and on what device. We will find
better solution.
P.S. Keep me in Cc list.
--
With Best Regards,
Andy Shevchenko
[+cc linux-pci and cc list of Bin's original posting]
On Mon, May 20, 2013 at 9:09 AM, Andy Shevchenko
<[email protected]> wrote:
> Hi!
>
> I don't know what you trying to fix, but your patch dd72be99
> "x86/pci/mrst: Use configuration mechanism 1 for 00:00.0, 00:02.0,
> 00:03.0" breaks Intel Medfield devices.
>
> After it was applied (sorry, was on vacation, didn't noticed sooner)
> lspci shows only 18 out of 34 devices. That means the so called fake
> PCI devices are gone.
>
> This is a bad approach and patch should be reverted. Bin, please,
> explain what you are trying to fix and on what device. We will find
> better solution.
I reverted dd72be99 in my for-linus branch. I guess I should have let
that sit in -next for a cycle before asking Linus to pull it. Sorry
about jumping the gun.
I'll wait for a patch tested by both Bin and Andy.
Bjorn