[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d7bb9ebf-4294-13a5-294d-5b9dcc821a07@huawei.com>
Date: Sat, 6 May 2023 21:11:06 +0800
From: Yunsheng Lin <linyunsheng@...wei.com>
To: Jesper Dangaard Brouer <jbrouer@...hat.com>, Ilias Apalodimas
<ilias.apalodimas@...aro.org>, <netdev@...r.kernel.org>, Eric Dumazet
<eric.dumazet@...il.com>, <linux-mm@...ck.org>, Mel Gorman
<mgorman@...hsingularity.net>
CC: <brouer@...hat.com>, <lorenzo@...nel.org>,
Toke Høiland-Jørgensen <toke@...hat.com>,
<bpf@...r.kernel.org>, "David S. Miller" <davem@...emloft.net>, Jakub
Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, Andrew Morton
<akpm@...ux-foundation.org>, <willy@...radead.org>
Subject: Re: [PATCH RFC net-next/mm V3 1/2] page_pool: Remove workqueue in new
shutdown scheme
On 2023/5/5 8:54, Yunsheng Lin wrote:
> It is not exactly the kind of refcnt bias trick in my mind, I was thinking
> about using pool->pages_state_hold_cnt as refcnt bias and merge it to
> pool->pages_state_release_cnt as needed, maybe I need to try to implement
> that to see if it turn out to be what I want it to be.
>
I did try implementing the above idea, not sure is there anything missing
yet as I only do the compile testing.
View attachment "page_pool_remove_wq.patch" of type "text/plain" (9805 bytes)
Powered by blists - more mailing lists