On Mon, Feb 22, 2021 at 7:29 PM Eric Dumazet wrote:
>
> On Mon, Feb 22, 2021 at 12:10 PM Honglei Wang wrote:
>
> > Really? I just came to the latest 5.11 including Linus' tree and
> > net-next, seems it's still here at line
> > 725 of inet_hashtables.c.. Do I miss something?
>
> 5.11 is old alrea
On Mon, Feb 22, 2021 at 3:47 PM Eric Dumazet wrote:
>
> On Sat, Feb 20, 2021 at 12:04 PM Honglei Wang wrote:
> >
> > We are getting port for connect() from even ports firstly now. This
> > makes bind() users have more available slots at odd part. But there is a
> > problem here when the even port
On Sun, Feb 21, 2021 at 6:06 AM Jeremy Sowden wrote:
>
> On 2021-02-20, at 13:44:02 -0800, David Miller wrote:
> > From: Honglei Wang
> > Date: Sat, 20 Feb 2021 19:03:56 +0800
> >
> > > We are getting port for connect() from even ports firstly now. This
> > > makes bind() users have more availabl
On 2021-02-20, at 13:44:02 -0800, David Miller wrote:
> From: Honglei Wang
> Date: Sat, 20 Feb 2021 19:03:56 +0800
>
> > We are getting port for connect() from even ports firstly now. This
> > makes bind() users have more available slots at odd part. But there is a
> > problem here when the even p
From: Honglei Wang
Date: Sat, 20 Feb 2021 19:03:56 +0800
> We are getting port for connect() from even ports firstly now. This
> makes bind() users have more available slots at odd part. But there is a
> problem here when the even ports are used up. This happens when there
> is a flood of short l
We are getting port for connect() from even ports firstly now. This
makes bind() users have more available slots at odd part. But there is a
problem here when the even ports are used up. This happens when there
is a flood of short life cycle connections. In this scenario, it starts
getting ports fr
