[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aGK85SAE9kDcVAMe@google.com>
Date: Mon, 30 Jun 2025 16:35:49 +0000
From: Jaegeuk Kim <jaegeuk@...nel.org>
To: wangzijie <wangzijie1@...or.com>
Cc: chao@...nel.org, linux-f2fs-devel@...ts.sourceforge.net,
linux-kernel@...r.kernel.org, bintian.wang@...or.com,
feng.han@...or.com, niuzhiguo84@...il.com
Subject: Re: [f2fs-dev] [PATCH v5 2/2] f2fs: don't allow unaligned truncation
to smaller/equal size on pinned file
I think we can clean up more like this?
https://lore.kernel.org/linux-f2fs-devel/20250630160839.1142073-1-jaegeuk@kernel.org/T/#u
On 06/30, wangzijie wrote:
> To prevent scattered pin block generation, don't allow non-section aligned truncation
> to smaller or equal size on pinned file. But for truncation to larger size, after
> commit 3fdd89b452c2("f2fs: prevent writing without fallocate() for pinned files"),
> we only support overwrite IO to pinned file, so we don't need to consider
> attr->ia_size > i_size case.
> In addition, xfstests #494 assumes truncation on active swapfile(pinned) will return
> ETXTBSY by setattr_prepare() -> inode_newsize_ok(), so we relocate this check after
> setattr_prepare().
>
> Signed-off-by: wangzijie <wangzijie1@...or.com>
> ---
> v5:
> - fix xfstests #494 fail
> ---
> fs/f2fs/file.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> index 78368b793..d4feea44b 100644
> --- a/fs/f2fs/file.c
> +++ b/fs/f2fs/file.c
> @@ -1070,6 +1070,23 @@ int f2fs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
> if (err)
> return err;
>
> + /*
> + * To prevent scattered pin block generation, we don't allow
> + * smaller/equal size unaligned truncation for pinned file.
> + * We only support overwrite IO to pinned file, so don't
> + * care about larger size truncation.
> + * We need to check this after setattr_prepare() because xfstests
> + * #494 assumes truncation on active swapfile(pinned) will
> + * return ETXTBSY by setattr_prepare() -> inode_newsize_ok().
> + */
> + if (attr->ia_valid & ATTR_SIZE) {
> + if (f2fs_is_pinned_file(inode) &&
> + attr->ia_size <= i_size_read(inode) &&
> + !IS_ALIGNED(attr->ia_size,
> + F2FS_BLK_TO_BYTES(CAP_BLKS_PER_SEC(sbi))))
> + return -EINVAL;
> + }
> +
> err = fscrypt_prepare_setattr(dentry, attr);
> if (err)
> return err;
> --
> 2.25.1
Powered by blists - more mailing lists