Re: [PATCH v2 1/6] fuse: limit the length of ITER_KVEC dio by max_pages

From: Hou Tao <houtao@huaweicloud.com>
To: Miklos Szeredi <miklos@szeredi.hu>
Cc: linux-fsdevel@vger.kernel.org, Vivek Goyal <vgoyal@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	Bernd Schubert <bernd.schubert@fastmail.fm>,
	"Michael S . Tsirkin" <mst@redhat.com>,
	Matthew Wilcox <willy@infradead.org>,
	Benjamin Coddington <bcodding@redhat.com>,
	linux-kernel@vger.kernel.org, virtualization@lists.linux.dev,
	houtao1@huawei.com
Subject: Re: [PATCH v2 1/6] fuse: limit the length of ITER_KVEC dio by max_pages
Date: Sat, 9 Mar 2024 12:26:52 +0800	[thread overview]
Message-ID: <8f21c92f-5456-7a2e-59af-1a02d8a10c24@huaweicloud.com> (raw)
In-Reply-To: <CAJfpegtMhkKG-Hk5vQ5gi6bSqwb=eMG9_TzcW7b08AtXBmnQXQ@mail.gmail.com>

Hi,

On 3/1/2024 9:42 PM, Miklos Szeredi wrote:
> On Wed, 28 Feb 2024 at 15:40, Hou Tao <houtao@huaweicloud.com> wrote:
>
>> So instead of limiting both the values of max_read and max_write in
>> kernel, capping the maximal length of kvec iter IO by using max_pages in
>> fuse_direct_io() just like it does for ubuf/iovec iter IO. Now the max
>> value for max_pages is 256, so on host with 4KB page size, the maximal
>> size passed to kmalloc() in copy_args_to_argbuf() is about 1MB+40B. The
>> allocation of 2MB of physically contiguous memory will still incur
>> significant stress on the memory subsystem, but the warning is fixed.
>> Additionally, the requirement for huge physically contiguous memory will
>> be removed in the following patch.
> So the issue will be fixed properly by following patches?
>
> In that case this patch could be omitted, right?

Sorry for the late reply. Being busy with off-site workshop these days.

No, this patch is still necessary and it is used to limit the number of
scatterlist used for fuse request and reply in virtio-fs. If the length
of out_args[0].size is not limited, the number of scatterlist used to
map the fuse request may be greater than the queue size of virtio-queue
and the fuse request may hang forever.

>
> Thanks,
> Miklos