On Tue, Jun 17, 2025 at 02:19:15PM -0500, Timothy Pearson wrote:
> ----- Original Message -----
> > From: "Bjorn Helgaas" <helgaas@xxxxxxxxxx>
> > To: "Timothy Pearson" <tpearson@xxxxxxxxxxxxxxxxxxxxx>
> > Cc: "linux-pci" <linux-pci@xxxxxxxxxxxxxxx>, "mahesh" <mahesh@xxxxxxxxxxxxx>, "Oliver" <oohall@xxxxxxxxx>, "Madhavan
> > Srinivasan" <maddy@xxxxxxxxxxxxx>, "Michael Ellerman" <mpe@xxxxxxxxxxxxxx>, "Lukas Wunner" <lukas@xxxxxxxxx>
> > Sent: Tuesday, June 17, 2025 1:59:38 PM
> > Subject: Re: [PATCH v10] PCI: Add pcie_link_is_active() function
>
> > On Tue, Jun 17, 2025 at 11:10:50AM -0500, Timothy Pearson wrote:
> >> ----- Original Message -----
> >> > From: "Bjorn Helgaas" <helgaas@xxxxxxxxxx>
> >> > To: "Timothy Pearson" <tpearson@xxxxxxxxxxxxxxxxxxxxx>
> >> > Cc: "linux-pci" <linux-pci@xxxxxxxxxxxxxxx>, "mahesh" <mahesh@xxxxxxxxxxxxx>,
> >> > "Oliver" <oohall@xxxxxxxxx>, "Madhavan
> >> > Srinivasan" <maddy@xxxxxxxxxxxxx>, "Michael Ellerman" <mpe@xxxxxxxxxxxxxx>,
> >> > "Lukas Wunner" <lukas@xxxxxxxxx>
> >> > Sent: Tuesday, June 17, 2025 11:04:28 AM
> >> > Subject: Re: [PATCH v10] PCI: Add pcie_link_is_active() function
> >>
> >> > On Tue, Jun 17, 2025 at 10:41:58AM -0500, Timothy Pearson wrote:
> >> >> Add pcie_link_is_active() function to check if the physical PCIe link is
> >> >> active, replacing duplicate code in multiple locations.
> >
> >> Will do in the future. There is some urgency on this overall
> >> patchset as we have had hotplug support broken for many years now,
> >> and it's causing continued problems on customer deployed machines.
> >> While we can continue to point customers at the patchset and have
> >> them compile their own kernels, patience from our customer side is
> >> wearing a bit thin for that. I will continue to try to push this
> >> forward (along with the associated ppc patch set) at a more
> >> reasonable pace.
> >
> > Apparently this got pulled out of a series where the urgency might
> > have been obvious. As a standalone patch, AFAICT it's a cleanup that
> > doesn't actually fix any problems.
>
> Correct -- my apologies for not making that clear. The original
> patch set for ppc (PowerNV) hotplug fixes was blocked on this
> pending patch by the ppc maintainers -- logicially it makes some
> sense, but I can't really submit a series that requires both groups
> of maintainers to approve.
Two paths:
- Best choice: since this is a cleanup that doesn't fix anything,
pull it out of the ppc series, fix the problem, and add the
cleanup to your TODO list for resolution in the future. AFAICT
this patch, at least v10, doesn't touch anything remotely
ppc-related, so I don't see the connection with PowerNV.
- Alternatively: send the patch to both sets of maintainers (ppc and
pci), and maintainer less affected can ack it and the other can
appy it.
