From: "Gaëtan Rivet" <gaetan.rivet@6wind.com>
To: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Cc: Shahaf Shuler <shahafs@mellanox.com>,
Mordechay Haimovsky <motih@mellanox.com>,
"dev@dpdk.org" <dev@dpdk.org>,
"stable@dpdk.org" <stable@dpdk.org>
Subject: Re: [dpdk-stable] [dpdk-dev] [PATCH v3] net/mlx4: fix dev rmv not detected after port stop
Date: Wed, 31 Jan 2018 10:54:13 +0100 [thread overview]
Message-ID: <20180131095413.oth3ved2lyoriixt@bidouze.vm.6wind.com> (raw)
In-Reply-To: <20180131091513.GS4256@6wind.com>
On Wed, Jan 31, 2018 at 10:15:13AM +0100, Adrien Mazarguil wrote:
> On Tue, Jan 30, 2018 at 08:37:06PM +0000, Shahaf Shuler wrote:
> > Tuesday, January 30, 2018 11:40 AM, Adrien Mazarguil:
> > > Unfortunately I didn't get a chance to review this patch before it was applied.
> > > I'm not sure a stopped port is supposed to report events (interrupts). Will
> > > applications expect them to occur at this point?
> >
> > Why not?
> >
> > Stopped port is still counted as attached. The fact the application stopped the packet receive on it doesn't mean it should not receive a sync events (such as the remove event).
> > async events, by definition, are not related to traffic being flows through the port.
>
> My comment is based on my understanding of rte_eth_dev_stop(), which is a
> device (or port) is completely stopped, in a suspended state and no
> interrupts shall occur, as a means for applications to temporarily not be
> bothered by them until restarted.
>
> Think about it that way: applications do not want to get interrupts
> immediately after the device is initialized, because they might not be ready
> to process them at this point. An explicit call to rte_eth_dev_start() tells
> the PMD when it's OK to do so. The converse is rte_eth_dev_stop().
>
> Stopping traffic can already be achieved by not polling from the application
> side, calling rte_eth_dev_[rt]x_queue_stop() and/or toggling RX/TX
> interrupts through rte_eth_dev_[rt]x_intr_enable(). rte_eth_dev_stop()
> provides lower-level device control.
>
> Perhaps documentation is not clear, however that's how LSC seems implemented
> in all PMDs; it gets disabled after rte_eth_dev_stop() and one should
> explicitly use rte_eth_link_get() to retrieve link status afterward. I think
> RMV should behave similarly with rte_eth_dev_is_removed(). Adapting
> fail-safe should be easier than modifying all the remaining PMDs.
>
I think fail-safe should follow the API, but for that the API should be
defined explicitly, instead of relying on common interpretation.
It makes sense for relatively harmless interrupts (LSC, RX) to be masked when stopped,
but I'm not sure this holds true for all possible interrupts (RMV), or
future ones.
If applications needs to ignore all interrupts at some point, it could
be interesting to extend the interrupt API to allow for enabling /
disabling them instead of relying on PMDs to register / unregister their
callbacks each time.
On that note, it could also be interesting for applications that won't
use interrupts, not to launch the interrupt thread during
rte_eal_init(). This is nonsense to force an additional thread to
operate without clear affinity. But I digress.
Interrupt API needs an overhaul, and its implementation as well.
> > > In my opinion it's not a fix, as in, it doesn't address an issue introduced by the
> > > mentioned patch whose behavior was correct.
> > >
> > > It's probably too late to change it now and it does address an issue seen with
> > > a use case involving this PMD, however I think the fail-safe PMD could as well
> > > poll using the recently-added rte_eth_dev_is_removed() when it's aware
> > > the underlying port is stopped instead of expecting interrupts.
> > >
> > > --
> > > Adrien Mazarguil
> > > 6WIND
--
Gaëtan Rivet
6WIND
next prev parent reply other threads:[~2018-01-31 9:54 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1516356548-15057-1-git-send-email-motih@mellanox.com>
2018-01-19 10:16 ` [dpdk-stable] [PATCH v2] net/mlx4: fix port start fail after device removal Moti Haimovsky
2018-01-28 13:54 ` Shahaf Shuler
2018-01-29 8:34 ` [dpdk-stable] [PATCH v3] net/mlx4: fix dev rmv not detected after port stop Moti Haimovsky
2018-01-29 10:54 ` Shahaf Shuler
2018-01-30 9:21 ` Thomas Monjalon
2018-01-30 9:39 ` Adrien Mazarguil
2018-01-30 20:37 ` Shahaf Shuler
2018-01-31 9:15 ` Adrien Mazarguil
2018-01-31 9:54 ` Gaëtan Rivet [this message]
2018-01-31 10:08 ` [dpdk-stable] [dpdk-dev] " Matan Azrad
2018-01-31 10:43 ` Adrien Mazarguil
2018-01-31 13:44 ` Matan Azrad
2018-01-31 14:31 ` Adrien Mazarguil
2018-01-31 17:07 ` Matan Azrad
2018-02-02 19:53 ` Adrien Mazarguil
2018-02-03 19:42 ` Matan Azrad
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180131095413.oth3ved2lyoriixt@bidouze.vm.6wind.com \
--to=gaetan.rivet@6wind.com \
--cc=adrien.mazarguil@6wind.com \
--cc=dev@dpdk.org \
--cc=motih@mellanox.com \
--cc=shahafs@mellanox.com \
--cc=stable@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).