Re: [PATCH v10 0/4] PCI: Restrict 64-bit prefetchable bridge windows to 64-bit resources

[Wei Yang <weiyang@linux.vnet.ibm.com>]

Bjorn,

I tried to apply it on the lastest upstrea, but seems not work.

Which tree should I apply to? This one ?
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git

On Mon, May 19, 2014 at 09:45:45PM -0600, Bjorn Helgaas wrote:
>I'd like to merge these for v3.16.  The first is Yinghai's v9 patch
>unchanged except to resolve minor merge conflicts, write a changelog, and
>fold in the mem64_mask removal posted previously as [1].
>
>The rest are cleanups and comment changes to try to make the code more
>readable.
>
>This does fix a real bug -- in some cases we can't assign space to large
>64-bit BARs even though there is plenty of 64-bit space available.  But I
>am concerned that we are giving up performance in some cases (32-bit
>prefetchable resources below a bridge with a 64-bit prefetchable window
>will no longer be prefetchable).  With some additional work, it should be
>possible to get that performance back.
>
>Comments welcome.
>
>[1] http://lkml.kernel.org/r/CAE9FiQUKYMPvXLsk84o-gH-j0-wiJGL0Wt1aTW2o4FPSm0PvbQ@mail.gmail.com
>
>---
>
>Bjorn Helgaas (3):
>      PCI: Change pbus_size_mem() return values to be more conventional
>      PCI: Simplify __pci_assign_resource() coding style
>      PCI: Add resource allocation comments
>
>Yinghai Lu (1):
>      PCI: Restrict 64-bit prefetchable bridge windows to 64-bit resources
>
>
> drivers/pci/setup-bus.c |  207 +++++++++++++++++++++++++++++++++--------------
> drivers/pci/setup-res.c |   41 +++++++--
> 2 files changed, 176 insertions(+), 72 deletions(-)

-- 
Richard Yang
Help you, Help me