Received-SPF: pass (google.com: domain of linux-kernel+bounces-81243-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240105152129.196824-1-aleksandr.mikhalitsyn@canonical.com> <20240105152129.196824-3-aleksandr.mikhalitsyn@canonical.com>
In-Reply-To: <20240105152129.196824-3-aleksandr.mikhalitsyn@canonical.com>
From: Miklos Szeredi <miklos@szeredi.hu>
Date: Mon, 26 Feb 2024 12:01:07 +0100
Message-ID: <CAJfpegsttFdeZnahAFQS=jG_uaw6XMHFfw7WKgAhujLaNszcsw@mail.gmail.com>
Subject: Re: [PATCH v1 2/3] fuse: use GFP_KERNEL_ACCOUNT for allocations in fuse_dev_alloc
To: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Cc: mszeredi@redhat.com, Amir Goldstein <amir73il@gmail.com>, linux-fsdevel@vger.kernel.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"

On Fri, 5 Jan 2024 at 16:21, Alexander Mikhalitsyn
<aleksandr.mikhalitsyn@canonical.com> wrote:
>
> fuse_dev_alloc() is called from the process context and it makes
> sense to properly account allocated memory to the kmemcg as these
> allocations are for long living objects.

Are the rules about when to use __GFP_ACCOUNT and when not documented somewhere?

I notice that most filesystem objects are allocated with
__GFP_ACCOUNT, but struct super_block isn't.  Is there a reason for
that?

Thanks,
Miklos