[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20230419101242.44f2d143@kernel.org>
Date: Wed, 19 Apr 2023 10:12:42 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Jesper Dangaard Brouer <jbrouer@...hat.com>
Cc: Lorenzo Bianconi <lorenzo@...nel.org>, brouer@...hat.com,
Eric Dumazet <edumazet@...gle.com>, netdev@...r.kernel.org,
hawk@...nel.org, ilias.apalodimas@...aro.org, davem@...emloft.net,
pabeni@...hat.com, bpf@...r.kernel.org,
lorenzo.bianconi@...hat.com, nbd@....name,
Toke Hoiland Jorgensen <toke@...hat.com>
Subject: Re: issue with inflight pages from page_pool
On Wed, 19 Apr 2023 17:36:20 +0200 Jesper Dangaard Brouer wrote:
> This reminds me that Jakub's recent defer patches returning pages
> 'directly' to the page_pool alloc-cache, will actually result in this
> kind of bug. This is because page_pool_destroy() assumes that pages
> cannot be returned to alloc-cache, as driver will have "disconnected" RX
> side. We need to address this bug separately.
Mm, I think you're right. The NAPI can get restarted with a different
page pool and the old page pool will still think it's in use.
Powered by blists - more mailing lists