From: Stefano Brivio <sbrivio@redhat.com>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: passt-dev@passt.top
Subject: Re: [PATCH 0/6] RFC: Clean up tap-side event handling
Date: Thu, 5 Sep 2024 13:33:05 +0200 [thread overview]
Message-ID: <20240905133305.1d6bcbc4@elisabeth> (raw)
In-Reply-To: <20240905103238.69431d8f@elisabeth>
On Thu, 5 Sep 2024 10:32:38 +0200
Stefano Brivio <sbrivio@redhat.com> wrote:
> On Thu, 5 Sep 2024 10:35:14 +1000
> David Gibson <david@gibson.dropbear.id.au> wrote:
>
> > On Wed, Sep 04, 2024 at 07:19:22PM +0200, Stefano Brivio wrote:
> > > On Wed, 4 Sep 2024 13:17:53 +1000
> > > David Gibson <david@gibson.dropbear.id.au> wrote:
> > >
> > > > On Tue, Sep 03, 2024 at 09:25:54PM +0200, Stefano Brivio wrote:
> > > > > On Tue, 3 Sep 2024 22:02:29 +1000
> > > > > David Gibson <david@gibson.dropbear.id.au> wrote:
> > > > >
> > > > > > This is a draft patch working towards adding EPOLLOUT handling to the
> > > > > > tap code, which could then be used to "unstick" flows which have
> > > > > > unsent data from the socket side. For now that's just a stub, but
> > > > > > makes what I think are some worthwhile cleanups to the tap side event
> > > > > > handling in the meantime.
> > > > >
> > > > > Except for the issue in 3/6 and nits elsewhere, it all makes sense and
> > > > > tap-side EPOLLOUT handling is definitely going to be an improvement.
> > > > >
> > > > > I wonder if it's the right moment for this kind of series, though, in
> > > > > terms of future bisections, as long as we're grappling with
> > > > > https://github.com/containers/podman/issues/23686 and
> > > > > https://bugs.passt.top/show_bug.cgi?id=94. Assuming, of course, that
> > > > > this series doesn't fix anything.
> > > >
> > > > I don't think this series will fix anything as it stands. It is,
> > > > indirectly, aimed at addressing bug 94. I'm struggling to figure out
> > > > what to do with bug 94, because I find it almost impossible to reason
> > > > about the current event masks in TCP.
> > >
> > > I don't see at the moment anything indicating TCP issues other than the
> > > one you addressed with your tentative debug patch at:
> > >
> > > https://passt.top/passt/commit/?h=podman23686&id=026fb71d1dde60135d95741552906fd5320384bc
> > >
> > > Given that, with that patch, we had at least another report of event
> > > storms, this time on UDP, that is, the one from:
> > >
> > > https://github.com/containers/podman/issues/23686#issuecomment-2324945010
> > >
> > > I shared this other one on top:
> > >
> > > https://passt.top/passt/commit/?h=podman23686&id=0c6c20dee5c24bd324834a99f409ad43c50812ae
> >
> > Ah, nice.
> >
> > > > I'd really like to simplify
> > > > them so it's clearer what's correct and not and I think the most
> > > > obvious path to doing so is using EPOLLET all the time. That requires
> > > > some sort of kick when the tap is ready to accept more data, hence
> > > > this series as a prerequisite.
> > >
> > > Sure, it's going to be simpler and more robust, but on the other hand
> > > we wouldn't notice these kind of issues.
> >
> > Uh.. I'm confused. In what way would we not notice issues, other than
> > the issues not existing which.. would be good, right?
>
> Right now, we have a condition where we fail to handle EPOLLRDHUP
> before an outbound connection is established, see
> https://github.com/containers/podman/issues/23686#issuecomment-233023742,
Sorry, it's:
https://github.com/containers/podman/issues/23686#issuecomment-2330237424
> and we end up in a tight event processing loop.
>
> I guess what we're missing in tcp_sock_handler() is clear (we should
> orderly close the connection), but the tight loop didn't happen on
> 2024_06_24.1ee2eca (I'm bisecting right now) and we don't know why it
> didn't.
>
> If we set EPOLLET, we won't see that anymore, because the EPOLLRDHUP
> event is reported just once, but that doesn't mean we solved this.
--
Stefano
prev parent reply other threads:[~2024-09-05 11:33 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-03 12:02 [PATCH 0/6] RFC: Clean up tap-side event handling David Gibson
2024-09-03 12:02 ` [PATCH 1/6] tap: Split out handling of EPOLLIN events David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:17 ` David Gibson
2024-09-03 12:02 ` [PATCH 2/6] tap: Improve handling of EINTR in tap_passt_input() David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:30 ` David Gibson
2024-09-03 12:02 ` [PATCH 3/6] tap: Restructure in tap_pasta_input() David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:33 ` David Gibson
2024-09-03 12:02 ` [PATCH 4/6] tap: Don't risk truncating frames on full buffer " David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:33 ` David Gibson
2024-09-03 12:02 ` [PATCH 5/6] tap: Re-introduce EPOLLET for tap connections David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:36 ` David Gibson
2024-09-03 12:02 ` [PATCH 6/6] tap: Stub EPOLLOUT handling David Gibson
2024-09-03 19:25 ` [PATCH 0/6] RFC: Clean up tap-side event handling Stefano Brivio
2024-09-04 3:17 ` David Gibson
2024-09-04 17:19 ` Stefano Brivio
2024-09-05 0:35 ` David Gibson
2024-09-05 8:32 ` Stefano Brivio
2024-09-05 11:33 ` Stefano Brivio [this message]
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=20240905133305.1d6bcbc4@elisabeth \
--to=sbrivio@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=passt-dev@passt.top \
/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.
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
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 IMAP folder(s).