[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <AD2B545E-FC05-402D-80A4-FB542A6D3C06@vmware.com>
Date: Fri, 18 Jan 2019 14:05:46 +0000
From: Ajay Kaher <akaher@...are.com>
To: Bjorn Helgaas <helgaas@...nel.org>
CC: "kys@...rosoft.com" <kys@...rosoft.com>,
"haiyangz@...rosoft.com" <haiyangz@...rosoft.com>,
"devel@...uxdriverproject.org" <devel@...uxdriverproject.org>,
"linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Jork Loeser <jloeser@...rosoft.com>,
"longli@...rosoft.com" <longli@...rosoft.com>
Subject: Re: [PATCH 3/3] PCI: hv: Use vPCI protocol version 1.2 for v4.9
> On 17/01/19, 8:37 PM, "Bjorn Helgaas" <helgaas@...nel.org> wrote:
>
> > On Fri, Jan 18, 2019 at 02:17:18AM +0530, Ajay Kaher wrote:
> > Update the Hyper-V vPCI driver to use the Server-2016 version of the vPCI
> > protocol, fixing MSI creation and retargeting issues.
> >
> > Replaced hv_tmp_cpu_nr_to_vp_nr() with vmbus_cpu_number_to_vp_number()
> > to make this patch compatibale for linux v4.9.
>
> s/compatibale for/compatible with/
>
> This change (to make it compatible with v4.9) sounds like it should be
> in its own separate patch. I don't see any use of
> hv_tmp_cpu_nr_to_vp_nr() being removed, so the changelog doesn't quite
> make sense.
hv_tmp_cpu_nr_to_vp_nr() is not removed, v4.9 doesn't have this API,
and because of this failed to apply the original patch on v4.9.
That's the reason to use vmbus_cpu_number_to_vp_number()
instead of hv_tmp_cpu_nr_to_vp_nr().
I will change the explanation in changelog to:
Because v4.9 doesn't have hv_tmp_cpu_nr_to_vp_nr(),
so original patch from v4.14 fails to apply on v4.9.
To solve this, using vmbus_cpu_number_to_vp_number()
instead of hv_tmp_cpu_nr_to_vp_nr() in this patch.
Here, I would like maintainer of HyperV to review this change and
provide their feedback. Tested these patches on Azure Server with
VM of size L16s_v2 and it's working fine.
> > Signed-off-by: Jork Loeser <jloeser@...rosoft.com>
> > Signed-off-by: Bjorn Helgaas <bhelgaas@...gle.com>
>
> I did not sign off on this; please remove.
Sorry, I kept as it is from original patch. I will change this in next version of this patch.
> > Reviewed-by: K. Y. Srinivasan <kys@...rosoft.com>
> > Acked-by: K. Y. Srinivasan <kys@...rosoft.com>
> > Signed-off-by: Ajay Kaher <akaher@...are.com>
>
> > + * struct hv_msi_desc2 - 1.2 version of hv_msi_desc
> > + * @vector: IDT entry
> > + * @delivery_mode: As defined in Intel's Programmer's
> > + * Reference Manual, Volume 3, Chapter 8.
> > + * @vector_count: Number of contiguous entries in the
> > + * Interrupt Descriptor Table that are
> > + * occupied by this Message-Signaled
> > + * Interrupt. For "MSI", as first defined
> > + * in PCI 2.2, this can be between 1 and
> > + * 32. For "MSI-X," as first defined in PCI
> > + * 3.0, this must be 1, as each MSI-X table
> > + * entry would have its own descriptor.
>
> Please reflow these descriptions to take advantage of an 80 column
> width. They are currently wrapped to fit in 50 columns, which is
> unnecessarily short.
Thanks, I will change this in next version of this patch.
> > + default:
> > + /* As we only negotiate protocol versions known to this driver,
> > + * this path should never hit. However, this is it not a hot
> > + * path so we print a message to aid future updates.
> > + */
> > + dev_err(&hbus->hdev->device,
> > + "Unexpected vPCI protocol, update driver.");
> Include the actual protocol version in the message?
Thanks, I will try to change this in next version.
Powered by blists - more mailing lists