[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <978f0f1e4ab0c8f46820af305f5efb00cf94ffd2@linux.dev>
Date: Thu, 24 Apr 2025 09:19:28 +0000
From: "Jiayuan Chen" <jiayuan.chen@...ux.dev>
To: "Cong Wang" <xiyou.wangcong@...il.com>
Cc: john.fastabend@...il.com, jakub@...udflare.com, netdev@...r.kernel.org,
bpf@...r.kernel.org
Subject: Re:
April 24, 2025 at 08:59, "Jiayuan Chen" <jiayuan.chen@...ux.dev> wrote:
>
> April 24, 2025 at 08:40, "Cong Wang" <xiyou.wangcong@...il.com> wrote:
>
> >
> > netdev@...r.kernel.org, bpf@...r.kernel.org
> >
> > Bcc:
> >
> >
> > Subject: test_sockmap failures on the latest bpf-next
> >
> > Reply-To:
> >
> >
> >
> > Hi all,
> >
> >
> >
> > The latest bpf-next failed on test_sockmap tests, I got the following
> >
> > failures (including 1 kernel warning). It is 100% reproducible here.
> >
> > I don't have time to look into them, a quick glance at the changelog
> >
> > shows quite some changes from Jiayuan. So please take a look, Jiayuan.
> >
> > Meanwhile, please let me know if you need more information from me.
> >
> > Thanks!
> >
> >
> >
> > --------------->
> >
>
> Thanks, I'm working on it.
>
After resetting my commit to 0bb2f7a1ad1f, which is before my changes, the warning still exists.
The warning originates from test_txmsg_redir_wait_sndmem(), which performs
'KTLS + sockmap with redir EGRESS and limited receive buffer'.
The memory charge/uncharge logic is problematic, I need some time to investigate and fix it.
Thanks.
Powered by blists - more mailing lists