RE: [PATCH v4] PCI: hv: Fix a timing issue which causes kdump to fail occasionally

[Wei Hu <weh@microsoft.com>]

> -----Original Message-----
> From: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
> Sent: Monday, July 27, 2020 7:19 PM
> To: Wei Hu <weh@microsoft.com>
> Cc: KY Srinivasan <kys@microsoft.com>; Haiyang Zhang
> <haiyangz@microsoft.com>; Stephen Hemminger <sthemmin@microsoft.com>;
> wei.liu@kernel.org; robh@kernel.org; bhelgaas@google.com; linux-
> hyperv@vger.kernel.org; linux-pci@vger.kernel.org; linux-
> kernel@vger.kernel.org; Dexuan Cui <decui@microsoft.com>; Michael Kelley
> <mikelley@microsoft.com>
> Subject: Re: [PATCH v4] PCI: hv: Fix a timing issue which causes kdump to fail
> occasionally
> 
> On Mon, Jul 27, 2020 at 03:17:31PM +0800, Wei Hu wrote:
> > Kdump could fail sometime on Hyper-V guest over Accelerated Network
> > interface. This is because the retry in hv_pci_enter_d0() releases
> > child device strurctures in hv_pci_bus_exit(). Although there is a
> > second asynchronous device relations message sending from the host, if
> > this message arrives guest after hv_send_resource_allocated() is
> > called, the retry would fail.
> >
> > Fix the problem by moving retry to hv_pci_probe() and starting retry
> > from hv_pci_query_relations() call.  This will cause a device
> > relations message to arrive guest synchronously.  The guest would be
> > able to rebuild the child device structures before calling
> > hv_send_resource_allocated().
> >
> > This problem only happens on Accelerated Network or SRIOV devices as
> > only such devices currently are attached under vmbus PCI bridge.
> >
> > Fixes: c81992e7f4aa ("PCI: hv: Retry PCI bus D0 entry on invalid
> > device state")
> > Signed-off-by: Wei Hu <weh@microsoft.com>
> > Reviewed-by: Michael Kelley <mikelley@microsoft.com>
> > ---
> >     v2: Adding Fixes tag according to Michael Kelley's review comment.
> >     v3: Fix couple typos and reword commit message to make it clearer.
> >     Thanks the comments from Bjorn Helgaas.
> >     v4: Adding more  problem descritpions in the commit message
> >     and code upon request from Lorenze Pieralisi.
> >
> >  drivers/pci/controller/pci-hyperv.c | 71
> > +++++++++++++++--------------
> >  1 file changed, 37 insertions(+), 34 deletions(-)
> >
> > diff --git a/drivers/pci/controller/pci-hyperv.c
> > b/drivers/pci/controller/pci-hyperv.c
> 
> I edited commit log and a comment in the code to fix a typo and pushed out to
> pci/hv.
> 
> Thanks,
> Lorenzo
> 
Thanks Lorenzo. Appreciate your helps!

Wei