Re: [f2fs-dev] [PATCH] f2fs: don't call block truncation for aliased file

[Chao Yu <chao@kernel.org>]

On 2025/1/7 10:32, Jaegeuk Kim via Linux-f2fs-devel wrote:
> This patch should avoid the below warning which does not corrupt the metadata
> tho.
> 
> [   51.508120][  T253] F2FS-fs (dm-59): access invalid blkaddr:36
> [   51.508156][  T253]  __f2fs_is_valid_blkaddr+0x330/0x384
> [   51.508162][  T253]  f2fs_is_valid_blkaddr_raw+0x10/0x24
> [   51.508163][  T253]  f2fs_truncate_data_blocks_range+0x1ec/0x438
> [   51.508177][  T253]  f2fs_remove_inode_page+0x8c/0x148
> [   51.508194][  T253]  f2fs_evict_inode+0x230/0x76c
> 
> Fixes: 128d333f0dff ("f2fs: introduce device aliasing file")
> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
> ---
>   fs/f2fs/node.c | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
> index c04ee1a7ce57..06313e1eb884 100644
> --- a/fs/f2fs/node.c
> +++ b/fs/f2fs/node.c
> @@ -1274,6 +1274,12 @@ int f2fs_remove_inode_page(struct inode *inode)
>   		return err;
>   	}
>   
> +	/* The aliase file must be truncated successfully. */
> +	if (unlikely(IS_DEVICE_ALIASING(inode))) {
> +		f2fs_put_dnode(&dn);
> +		return 0;

Do we need to keep calling truncate_node() for device alias inode in 
f2fs_remove_inode_page()?

Thanks,

> +	}
> +
>   	/* remove potential inline_data blocks */
>   	if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
>   				S_ISLNK(inode->i_mode))