[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <161583677541.3715498.6118778324185171839.stgit@firesoul>
Date: Mon, 15 Mar 2021 20:33:24 +0100
From: Jesper Dangaard Brouer <brouer@...hat.com>
To: Mel Gorman <mgorman@...hsingularity.net>, linux-mm@...ck.org
Cc: Jesper Dangaard Brouer <brouer@...hat.com>, chuck.lever@...cle.com,
Alexander Duyck <alexander.duyck@...il.com>,
netdev@...r.kernel.org, linux-nfs@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH mel-git] Followup: Update [PATCH 7/7] in Mel's series
This patch is against Mel's git-tree:
git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux.git
Using branch: mm-bulk-rebase-v4r2 but replacing the last patch related to
the page_pool using __alloc_pages_bulk().
https://git.kernel.org/pub/scm/linux/kernel/git/mel/linux.git/log/?h=mm-bulk-rebase-v4r2
While implementing suggestions by Alexander Duyck, I realised that I could
simplify the code further, and simply take the last page from the
pool->alloc.cache given this avoids special casing the last page.
I re-ran performance tests and the improvement have been reduced to 13% from
18% before, but I don't think the rewrite of the specific patch have
anything to do with this.
Notes on tests:
https://github.com/xdp-project/xdp-project/blob/master/areas/mem/page_pool06_alloc_pages_bulk.org#test-on-mel-git-tree
---
Jesper Dangaard Brouer (1):
net: page_pool: use alloc_pages_bulk in refill code path
net/core/page_pool.c | 73 ++++++++++++++++++++++++++++++++------------------
1 file changed, 47 insertions(+), 26 deletions(-)
--
Powered by blists - more mailing lists