Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230117181533.2350335-1-neelx@redhat.com> <2bdeb975-6d45-67bb-3017-f19df62fe7af@intel.com>
 <CACjP9X-hKf8g2UqitV8_G7WQW7u6Js5EsCNutsAMA4WD7YYSwA@mail.gmail.com> <42e74619-f2d0-1079-28b1-61e9e17ae953@intel.com>
In-Reply-To: <42e74619-f2d0-1079-28b1-61e9e17ae953@intel.com>
From:   Daniel Vacek <neelx@redhat.com>
Date:   Wed, 18 Jan 2023 23:11:23 +0100
Message-ID: <CACjP9X8SHZAd_+HSLJCxYxSRQuRmq3r48id13r17n2ehrec2YQ@mail.gmail.com>
Subject: Re: [Intel-wired-lan] [PATCH] ice/ptp: fix the PTP worker retrying
 indefinitely if the link went down
To:     Jacob Keller <jacob.e.keller@intel.com>
Cc:     Jesse Brandeburg <jesse.brandeburg@intel.com>,
        Tony Nguyen <anthony.l.nguyen@intel.com>,
        "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        Richard Cochran <richardcochran@gmail.com>,
        "Kolacinski, Karol" <karol.kolacinski@intel.com>,
        Siddaraju <siddaraju.dh@intel.com>,
        "Michalik, Michal" <michal.michalik@intel.com>,
        netdev@vger.kernel.org, intel-wired-lan@lists.osuosl.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Jan 18, 2023 at 9:59 PM Jacob Keller <jacob.e.keller@intel.com> wrote:
> On 1/18/2023 7:14 AM, Daniel Vacek wrote:
> > On Tue, Jan 17, 2023 at 7:47 PM Jacob Keller <jacob.e.keller@intel.com> wrote:
> >>
> >> At a minimum I think I would only set drop_ts but not not goto skip_ts_read.
> >
> > IIUC, that would still fail to clear the tx->in_use bit in case ice_read_phy_tstamp()
> > returns with error. It would only work for the other case where no error is
> > returned but rather the returned &raw_tstamp is invalid. I'll send a v2 of
> > this fix trying to address the goto concern.
> >
>
> Please re-send the patch with the goto concern addressed. (only set
> drop_ts, and still read the timestamp just in case and make sure that if
> the read function fails while drop_ts is set we still drop the ts).

Just in case it got lost I did send the patch here:

https://lore.kernel.org/intel-wired-lan/20230118161727.2485457-1-neelx@redhat.com/

But unfortunately I misplaced the --in-reply-to header with --reply-to one :-(
Hopefully it arrived to you safely.

> I believe that alleviates my concerns regarding the potential link down
> race with a completed timestamp above.
>
> We also should already be preventing requesting new timestamps while
> link is down so that should be sufficient to combine with this and cover
> the three flows:
>
> 1) request tx timestamp
> 2) link down
> 3) timestamp never occurs
>
> 1) request tx timestamp
> 2) timestamp occurs
> 3) link goes down while processing

I was thinking this is the case we got reported. But then again, I'm
not really experienced in this field.

--nX

> 1) link down
> 2) request tx timestamp rejected
>
> Thanks!
>
> -Jake