[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <10a38fe0-d609-0e25-4a79-a5d23a83600b@bytedance.com>
Date: Thu, 16 Dec 2021 23:39:02 +0800
From: Qi Zheng <zhengqi.arch@...edance.com>
To: Greg KH <gregkh@...uxfoundation.org>
Cc: akpm@...ux-foundation.org, rafael@...nel.org, linux-mm@...ck.org,
linux-kernel@...r.kernel.org, songmuchun@...edance.com
Subject: Re: [PATCH 2/2] mm: reimplement si_mem_available()
On 12/16/21 9:17 PM, Greg KH wrote:
> On Thu, Dec 16, 2021 at 08:46:55PM +0800, Qi Zheng wrote:
>> Reimplement si_mem_available() by reusing si_mem_available_node().
>
> That says what you did, but not why. Why are you doing this?
Oh, sorry, I will add that.
>
> And what does it have to do with patch 1/2?
With [PATCH 1/2], we can simply sum the MemAvailable of each node
to get the gobal MemAvailable. Essentially this is a cleanup, no
functional changes.
Thanks,
Qi
>
> thanks,
>
> greg k-h
>
--
Thanks,
Qi
Powered by blists - more mailing lists