linux-um.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Geoffrey Thorpe <geoff@geoffthorpe.net>
To: Hongbo Li <lihongbo22@huawei.com>,
	richard@nod.at, anton.ivanov@cambridgegreys.com,
	johannes@sipsolutions.net
Cc: linux-um@lists.infradead.org, linux-fsdevel@vger.kernel.org,
	brauner@kernel.org
Subject: Re: [PATCH] hostfs: Fix only passing host root in boot stage with new mount
Date: Sun, 12 Oct 2025 11:41:29 -0400	[thread overview]
Message-ID: <100e0c55-a016-491a-ac41-89e640851771@geoffthorpe.net> (raw)
In-Reply-To: <20251011092235.29880-1-lihongbo22@huawei.com>

(This time, responding in plain text so the list-server doesn't reject. 
oops.)

Yes, I've tested your patch and it has fixed the problem for me. Thanks


On 2025-10-11 05:22, Hongbo Li wrote:
> In the old mount proceedure, hostfs could only pass root directory during
> boot. This is because it constructed the root directory using the @root_ino
> event without any mount options. However, when using it with the new mount
> API, this step is no longer triggered. As a result, if users mounts without
> specifying any mount options, the @host_root_path remains uninitialized. To
> prevent this issue, the @host_root_path should be initialized at the time
> of allocation.
>
> Reported-by: Geoffrey Thorpe <geoff@geoffthorpe.net>
> Closes: https://lore.kernel.org/all/643333a0-f434-42fb-82ac-d25a0b56f3b7@geoffthorpe.net/
> Fixes: cd140ce9f611 ("hostfs: convert hostfs to use the new mount API")
> Signed-off-by: Hongbo Li <lihongbo22@huawei.com>
> ---
>   fs/hostfs/hostfs_kern.c | 29 ++++++++++++++++++-----------
>   1 file changed, 18 insertions(+), 11 deletions(-)
>
> diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
> index 1e1acf5775ab..86455eebbf6c 100644
> --- a/fs/hostfs/hostfs_kern.c
> +++ b/fs/hostfs/hostfs_kern.c
> @@ -979,7 +979,7 @@ static int hostfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
>   {
>   	struct hostfs_fs_info *fsi = fc->s_fs_info;
>   	struct fs_parse_result result;
> -	char *host_root;
> +	char *host_root, *tmp_root;
>   	int opt;
>   
>   	opt = fs_parse(fc, hostfs_param_specs, param, &result);
> @@ -990,11 +990,13 @@ static int hostfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
>   	case Opt_hostfs:
>   		host_root = param->string;
>   		if (!*host_root)
> -			host_root = "";
> -		fsi->host_root_path =
> -			kasprintf(GFP_KERNEL, "%s/%s", root_ino, host_root);
> -		if (fsi->host_root_path == NULL)
> +			break;
> +		tmp_root = kasprintf(GFP_KERNEL, "%s%s",
> +				     fsi->host_root_path, host_root);
> +		if (!tmp_root)
>   			return -ENOMEM;
> +		kfree(fsi->host_root_path);
> +		fsi->host_root_path = tmp_root;
>   		break;
>   	}
>   
> @@ -1004,17 +1006,17 @@ static int hostfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
>   static int hostfs_parse_monolithic(struct fs_context *fc, void *data)
>   {
>   	struct hostfs_fs_info *fsi = fc->s_fs_info;
> -	char *host_root = (char *)data;
> +	char *tmp_root, *host_root = (char *)data;
>   
>   	/* NULL is printed as '(null)' by printf(): avoid that. */
>   	if (host_root == NULL)
> -		host_root = "";
> +		return 0;
>   
> -	fsi->host_root_path =
> -		kasprintf(GFP_KERNEL, "%s/%s", root_ino, host_root);
> -	if (fsi->host_root_path == NULL)
> +	tmp_root = kasprintf(GFP_KERNEL, "%s%s", fsi->host_root_path, host_root);
> +	if (!tmp_root)
>   		return -ENOMEM;
> -
> +	kfree(fsi->host_root_path);
> +	fsi->host_root_path = tmp_root;
>   	return 0;
>   }
>   
> @@ -1049,6 +1051,11 @@ static int hostfs_init_fs_context(struct fs_context *fc)
>   	if (!fsi)
>   		return -ENOMEM;
>   
> +	fsi->host_root_path = kasprintf(GFP_KERNEL, "%s/", root_ino);
> +	if (!fsi->host_root_path) {
> +		kfree(fsi);
> +		return -ENOMEM;
> +	}
>   	fc->s_fs_info = fsi;
>   	fc->ops = &hostfs_context_ops;
>   	return 0;


  reply	other threads:[~2025-10-12 15:41 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-11  9:22 [PATCH] hostfs: Fix only passing host root in boot stage with new mount Hongbo Li
2025-10-12 15:41 ` Geoffrey Thorpe [this message]
2025-10-21 12:16 ` Christian Brauner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=100e0c55-a016-491a-ac41-89e640851771@geoffthorpe.net \
    --to=geoff@geoffthorpe.net \
    --cc=anton.ivanov@cambridgegreys.com \
    --cc=brauner@kernel.org \
    --cc=johannes@sipsolutions.net \
    --cc=lihongbo22@huawei.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-um@lists.infradead.org \
    --cc=richard@nod.at \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).