[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240202162040.GA2087318@ZenIV>
Date: Fri, 2 Feb 2024 16:20:40 +0000
From: Al Viro <viro@...iv.linux.org.uk>
To: JonasZhou-oc <JonasZhou-oc@...oxin.com>
Cc: brauner@...nel.org, jack@...e.cz, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org, CobeChen@...oxin.com,
LouisQi@...oxin.com, JonasZhou@...oxin.com
Subject: Re: [PATCH] fs/address_space: move i_mmap_rwsem to mitigate a false
sharing with i_mmap.
On Fri, Feb 02, 2024 at 04:33:04PM +0800, JonasZhou-oc wrote:
> From: JonasZhou <JonasZhou@...oxin.com>
>
> In the struct address_space, there is a 32-byte gap between i_mmap
> and i_mmap_rwsem. Due to the alignment of struct address_space
> variables to 8 bytes, in certain situations, i_mmap and i_mmap_rwsem
> may end up in the same CACHE line.
>
> While running Unixbench/execl, we observe high false sharing issues
> when accessing i_mmap against i_mmap_rwsem. We move i_mmap_rwsem
> after i_private_list, ensuring a 64-byte gap between i_mmap and
> i_mmap_rwsem.
>
> For Intel Silver machines (2 sockets) using kernel v6.8 rc-2, the score
> of Unixbench/execl improves by ~3.94%, and the score of Unixbench/shell
> improves by ~3.26%.
Looks sane.
Powered by blists - more mailing lists