[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Zr+KQ4/iiIqmkUSk@gmail.com>
Date: Fri, 16 Aug 2024 10:20:03 -0700
From: Breno Leitao <leitao@...ian.org>
To: 黄朝阳 (Zhaoyang Huang) <zhaoyang.huang@...soc.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>,
Matthew Wilcox <willy@...radead.org>,
Suren Baghdasaryan <surenb@...gle.com>,
Minchan Kim <minchan@...nel.org>,
"linux-mm@...ck.org" <linux-mm@...ck.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Zhaoyang Huang <huangzhaoyang@...il.com>,
王科 (Ke Wang) <Ke.Wang@...soc.com>,
"usamaarif642@...il.com" <usamaarif642@...il.com>,
"riel@...riel.com" <riel@...riel.com>,
"hannes@...xchg.org" <hannes@...xchg.org>,
"nphamcs@...il.com" <nphamcs@...il.com>
Subject: Re: reply: [PATCHv5] mm: skip CMA pages when they are not available
Hello Zhaoyang,
On Tue, Aug 13, 2024 at 09:58:32AM +0000, 黄朝阳 (Zhaoyang Huang) wrote:
> >I've been discussing this with colleagues and we're speculating that the high
> >contention might be linked to the fact that CMA regions are now being skipped.
> >This could potentially extend the duration of the
> >isolate_lru_folios() 'while' loop, resulting in increased pressure on the lock.
> >However, I want to emphasize that I'm not an expert in this area and I am
> >simply sharing the data I collected.
> Could you please try below patch which could be helpful
>
> https://lore.kernel.org/linux-mm/CAOUHufa7OBtNHKMhfu8wOOE4f0w3b0_2KzzV7-hrc9rVL8e=iw@mail.gmail.com/
Yes, my colleague Usama have tried it, and it solved the problem. Thanks
for the heads-up, it was very useful.
--breno
Powered by blists - more mailing lists