Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20200117162444.v2.1.I9c7e72144ef639cc135ea33ef332852a6b33730f@changeid>
 <CACK8Z6Ft95qj4e_fsA32r_bcz2SsHOW1xxqZJt3_DBAJw=NMGA@mail.gmail.com>
 <CAE=gft6fKQWExW-=xjZGzXs30XohfpA5SKggvL2WtYXAHmzMew@mail.gmail.com>
 <87y2tytv5i.fsf@nanos.tec.linutronix.de> <87eevqkpgn.fsf@nanos.tec.linutronix.de>
 <CAE=gft6YiM5S1A7iJYJTd5zmaAa8=nhLE3B94JtWa+XW-qVSqQ@mail.gmail.com>
In-Reply-To: <CAE=gft6YiM5S1A7iJYJTd5zmaAa8=nhLE3B94JtWa+XW-qVSqQ@mail.gmail.com>
From:   Evan Green <evgreen@chromium.org>
Date:   Thu, 23 Jan 2020 14:59:25 -0800
Message-ID: <CAE=gft5xta4XCJtctWe=R3w=kVr598JCbk9VSRue04nzKAk3CQ@mail.gmail.com>
Subject: Re: [PATCH v2] PCI/MSI: Avoid torn updates to MSI pairs
To:     Thomas Gleixner <tglx@linutronix.de>
Cc:     Rajat Jain <rajatja@google.com>,
        Bjorn Helgaas <bhelgaas@google.com>,
        linux-pci <linux-pci@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jan 23, 2020 at 12:59 PM Evan Green <evgreen@chromium.org> wrote:
>
> On Thu, Jan 23, 2020 at 10:17 AM Thomas Gleixner <tglx@linutronix.de> wrote:
> >
> > Evan,
> >
> > Thomas Gleixner <tglx@linutronix.de> writes:
> > > This is not yet debugged fully and as this is happening on MSI-X I'm not
> > > really convinced yet that your 'torn write' theory holds.
> >
> > can you please apply the debug patch below and run your test. When the
> > failure happens, stop the tracer and collect the trace.
> >
> > Another question. Did you ever try to change the affinity of that
> > interrupt without hotplug rapidly while the device makes traffic? If
> > not, it would be interesting whether this leads to a failure as well.
>
> Thanks for the patch. Looks pretty familiar :)
> I ran into issues where trace_printks on offlined cores seem to
> disappear. I even made sure the cores were back online when I
> collected the trace. So your logs might not be useful. Known issue
> with the tracer?
>
> I figured I'd share my own debug chicken scratch, in case you could
> glean anything from it. The LOG entries print out timestamps (divide
> by 1000000) that you can match up back to earlier in the log (ie so
> the last XHCI MSI change occurred at 74.032501, the last interrupt
> came in at 74.032405). Forgive the mess.
>
> I also tried changing the affinity rapidly without CPU hotplug, but
> didn't see the issue, at least not in the few minutes I waited
> (normally repros easily within 1 minute). An interesting datapoint.

One additional datapoint. The intel guys suggested enabling
CONFIG_IRQ_REMAP, which does seem to eliminate the issue for me. I'm
still hoping there's a smaller fix so I don't have to add all that in.
-Evan