[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKF3qh1Bzj6LwBU0yzvLTxSO9RTU5GrJ7PRSwMBHdFJpbFzKEA@mail.gmail.com>
Date: Sat, 3 Oct 2020 13:46:04 +0800
From: Ethan Zhao <xerces.zhao@...il.com>
To: Sinan Kaya <okaya@...nel.org>
Cc: Ethan Zhao <haifeng.zhao@...el.com>,
Bjorn Helgaas <bhelgaas@...gle.com>, Oliver <oohall@...il.com>,
ruscur@...sell.cc, Lukas Wunner <lukas@...ner.de>,
Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
Stuart Hayes <stuart.w.hayes@...il.com>,
Alexandru Gagniuc <mr.nuke.me@...il.com>,
Mika Westerberg <mika.westerberg@...ux.intel.com>,
linux-pci <linux-pci@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
ashok.raj@...ux.intel.com,
Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@...el.com>
Subject: Re: [PATCH v6 4/5] PCI: only return true when dev io state is really changed
Sinan,
On Sat, Oct 3, 2020 at 12:08 AM Sinan Kaya <okaya@...nel.org> wrote:
>
> On 9/30/2020 3:05 AM, Ethan Zhao wrote:
> > When uncorrectable error happens, AER driver and DPC driver interrupt
> > handlers likely call
> >
> > pcie_do_recovery()
> > ->pci_walk_bus()
> > ->report_frozen_detected()
> >
> > with pci_channel_io_frozen the same time.
>
> We need some more data on this. If DPC is supported by HW, errors
> should be triggered by DPC not AER.
>
> If I remember right, there is a register that tells which AER errors
> should be handled by DPC.
When uncorrectable errors happen, non-fatal severity level, AER and
DPC could be triggered at the same time.
Thanks,
Ethan
>
Powered by blists - more mailing lists