[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZmoZDBPrrLY8l6nM@C02YVCJELVCG.dhcp.broadcom.net>
Date: Wed, 12 Jun 2024 17:54:20 -0400
From: Andy Gospodarek <andrew.gospodarek@...adcom.com>
To: David Wei <dw@...idwei.uk>
Cc: David Ahern <dsahern@...nel.org>,
Michael Chan <michael.chan@...adcom.com>,
Andy Gospodarek <andrew.gospodarek@...adcom.com>,
Adrian Alvarado <adrian.alvarado@...adcom.com>,
Somnath Kotur <somnath.kotur@...adcom.com>, netdev@...r.kernel.org,
Pavel Begunkov <asml.silence@...il.com>,
Jakub Kicinski <kuba@...nel.org>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>, Paolo Abeni <pabeni@...hat.com>
Subject: Re: [PATCH net-next v1 0/3] bnxt_en: implement netdev_queue_mgmt_ops
On Wed, Jun 12, 2024 at 11:19:47AM -0700, David Wei wrote:
> On 2024-06-12 08:52, David Ahern wrote:
> > On 6/10/24 9:41 PM, David Wei wrote:
> >>
> >> This patchset is orthogonal to header split and page pool memory
> >> providers. It implements netdev_queue_mgmt_ops which enables dynamically
> >
> > Ok, where is the validation that these queues must be configured for
> > header-data split to use non-kernel memory?
>
> Any validation would be done outside of this patchset, which only
> focuses on resetting an Rx queue. Reconfiguring page pools and HDS is
> orthogonal and unrelated to this patchset.
I can confirm this. These changes are just to keep the allocation
schemes the same as they are today, but add support for stopping/freeing
and starting/allocating rings individually and via the new API.
> The netdev core API consuming netdev_queue_mgmt_ops would be
> netdev_rx_queue_restart() added in [1].
>
> [1]: https://lore.kernel.org/lkml/20240607005127.3078656-2-almasrymina@google.com/
>
> Validation would be done at a layer above, i.e. whatever that calls
> netdev_rx_queue_restart(), as opposed to netdev_rx_queue_restart() or
> netdev_queue_mgmt_ops itself.
Powered by blists - more mailing lists