Re: linux-next: manual merge of the amdgpu tree with the pci tree

[Bjorn Helgaas]

On Mon, Dec 14, 2020 at 06:18:54PM -0500, Alex Deucher wrote:
> On Mon, Dec 14, 2020 at 6:16 PM Bjorn Helgaas <helgaas@xxxxxxxxxx> wrote:
> > On Tue, Dec 15, 2020 at 07:34:31AM +1100, Stephen Rothwell wrote:
> > > Hi all,
> > >
> > > On Tue, 8 Dec 2020 13:56:20 +1100 Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> wrote:
> > > >
> > > > Today's linux-next merge of the amdgpu tree got a conflict in:
> > > >
> > > >   drivers/pci/pcie/err.c
> > > >
> > > > between commits:
> > > >
> > > >   8f1bbfbc3596 ("PCI/ERR: Rename reset_link() to reset_subordinates()")
> > > >   0791721d8007 ("PCI/ERR: Use "bridge" for clarity in pcie_do_recovery()")
> > > >   05e9ae19ab83 ("PCI/ERR: Add pci_walk_bridge() to pcie_do_recovery()")
> > > >
> > > > from the pci tree and commit:
> > > >
> > > >   36a8901e900a ("PCI/ERR: Fix reset logic in pcie_do_recovery() call")
> > > >
> > > > from the amdgpu tree.
> > > >
> > > > I fixed it up (I think - see below) and can carry the fix as
> > > > necessary. This is now fixed as far as linux-next is concerned, but any
> > > > non trivial conflicts should be mentioned to your upstream maintainer
> > > > when your tree is submitted for merging.  You may also want to consider
> > > > cooperating with the maintainer of the conflicting tree to minimise any
> > > > particularly complex conflicts.
> >
> > Huh.  It's sub-optimal to change this core code via both the PCI and
> > the amdgpu tree, with no heads-up to me.
> >
> > 400b308d388a ("PCI/ERR: Fix reset logic in pcie_do_recovery() call")
> > (apparently in the amdgpu tree) doesn't have a Link: tag to the
> > posting of the patch, where there was quite a lot of useful discussion
> > that should be connected somehow.
> >
> > I think the posting was
> > https://lore.kernel.org/r/cbba08a5e9ca62778c8937f44eda2192a2045da7.1595617529.git.sathyanarayanan.kuppuswamy@xxxxxxxxxxxxxxx
> >
> > I had deferred merging this patch so I could merge Sean's RCEC work
> > first, and then resolve the conflict when merging *this* patch.
> 
> I don't plan to merge this upstream via my tree.  I was just carrying
> it in my drm-next branch because we have a number of users that depend
> on it for working DPC and a number of people use this branch for
> testing.

OK, thanks.  FWIW, it's currently marked "Changes Requested" in
patchwork, so it isn't really going anywhere right now:

https://patchwork.kernel.org/project/linux-pci/patch/cbba08a5e9ca62778c8937f44eda2192a2045da7.1595617529.git.sathyanarayanan.kuppuswamy@xxxxxxxxxxxxxxx/

I have merged Sean's series, so this would be a good time to try to
move this one forward.