mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jeff Layton <jlayton@kernel.org>
To: Alexander Viro <viro@zeniv.linux.org.uk>,
	Christian Brauner	 <brauner@kernel.org>, Jan Kara <jack@suse.cz>,
	Steven Rostedt	 <rostedt@goodmis.org>,
	Masami Hiramatsu <mhiramat@kernel.org>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Chuck Lever <chuck.lever@oracle.com>,
	NeilBrown	 <neil@brown.name>,
	Olga Kornievskaia <okorniev@redhat.com>,
	Dai Ngo	 <Dai.Ngo@oracle.com>, Tom Talpey <tom@talpey.com>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	 linux-trace-kernel@vger.kernel.org, linux-nfs@vger.kernel.org
Subject: Re: [PATCH 2/2] vfs: fix delegated timestamp handling in setattr_copy()
Date: Tue, 22 Jul 2025 18:19:51 -0400	[thread overview]
Message-ID: <e28297a2abe8253c0aa590831b3857432bef60f7.camel@kernel.org> (raw)
In-Reply-To: <20250722-nfsd-testing-v1-2-31321c7fc97f@kernel.org>

On Tue, 2025-07-22 at 14:52 -0400, Jeff Layton wrote:
> There are a couple of problems with delegated timestamp updates via
> setattr:
> 
> 1/ the ia_ctime is always clobbered by notify_change(), so setting the
> ia_ctime to the same value as the ia_mtime in nfsd4_decode_fattr4()
> doesn't work.
> 
> 2/ while it does test the ctime's validity vs. the existing ctime and
> current_time(), the same is not done for the atime or mtime. The spec
> requires this.
> 
> Add a new setattr_copy_delegts() function that handles updating the
> timestamps whenever ATTR_DELEG is set. For both atime and mtime,
> validate and clamp the value to current_time(), and then set it. If the
> mtime gets updated, also update the ctime.
> 
> Fixes: 7f2c86cba3c5 ("fs: handle delegated timestamps in setattr_copy_mgtime")
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
>  fs/attr.c         | 52 ++++++++++++++++++++++++++++++++++++++--------------
>  fs/nfsd/nfs4xdr.c |  4 +---
>  2 files changed, 39 insertions(+), 17 deletions(-)
> 
> diff --git a/fs/attr.c b/fs/attr.c
> index 9caf63d20d03e86c535e9c8c91d49c2a34d34b7a..3e636943d26a36aeeed0ff8b428b6dd3e63f8dde 100644
> --- a/fs/attr.c
> +++ b/fs/attr.c
> @@ -287,14 +287,7 @@ static void setattr_copy_mgtime(struct inode *inode, const struct iattr *attr)
>  	struct timespec64 now;
>  
>  	if (ia_valid & ATTR_CTIME) {
> -		/*
> -		 * In the case of an update for a write delegation, we must respect
> -		 * the value in ia_ctime and not use the current time.
> -		 */
> -		if (ia_valid & ATTR_DELEG)
> -			now = inode_set_ctime_deleg(inode, attr->ia_ctime);
> -		else
> -			now = inode_set_ctime_current(inode);
> +		now = inode_set_ctime_current(inode);
>  	} else {
>  		/* If ATTR_CTIME isn't set, then ATTR_MTIME shouldn't be either. */
>  		WARN_ON_ONCE(ia_valid & ATTR_MTIME);
> @@ -312,6 +305,39 @@ static void setattr_copy_mgtime(struct inode *inode, const struct iattr *attr)
>  		inode_set_mtime_to_ts(inode, now);
>  }
>  
> +/*
> + * Skip update if new value is older than existing time. Clamp
> + * to current_time() if it's in the future.
> + */
> +static void setattr_copy_delegts(struct inode *inode, const struct iattr *attr)
> +{
> +	struct timespec64 now = current_time(inode);
> +	unsigned int ia_valid = attr->ia_valid;
> +
> +	if (ia_valid & ATTR_MTIME) {
> +		struct timespec64 cur = inode_get_mtime(inode);
> +
> +		if (timespec64_compare(&attr->ia_mtime, &cur) > 0) {
> +			if (timespec64_compare(&attr->ia_mtime, &now) > 0)
> +				inode_set_mtime_to_ts(inode, now);
> +			else
> +				inode_set_mtime_to_ts(inode, attr->ia_mtime);
> +			inode_set_ctime_deleg(inode, attr->ia_mtime);
> +		}
> +	}
> +
> +	if (ia_valid & ATTR_ATIME) {
> +		struct timespec64 cur = inode_get_atime(inode);
> +
> +		if (timespec64_compare(&attr->ia_atime, &cur) > 0) {
> +			if (timespec64_compare(&attr->ia_atime, &now) > 0)
> +				inode_set_atime_to_ts(inode, now);
> +			else
> +				inode_set_atime_to_ts(inode, attr->ia_atime);
> +		}
> +	}
> +}
> +
>  /**
>   * setattr_copy - copy simple metadata updates into the generic inode
>   * @idmap:	idmap of the mount the inode was found from
> @@ -352,6 +378,8 @@ void setattr_copy(struct mnt_idmap *idmap, struct inode *inode,
>  		inode->i_mode = mode;
>  	}
>  
> +	if (ia_valid & ATTR_DELEG)
> +		return setattr_copy_delegts(inode, attr);
>  	if (is_mgtime(inode))
>  		return setattr_copy_mgtime(inode, attr);
>  
> @@ -359,12 +387,8 @@ void setattr_copy(struct mnt_idmap *idmap, struct inode *inode,
>  		inode_set_atime_to_ts(inode, attr->ia_atime);
>  	if (ia_valid & ATTR_MTIME)
>  		inode_set_mtime_to_ts(inode, attr->ia_mtime);
> -	if (ia_valid & ATTR_CTIME) {
> -		if (ia_valid & ATTR_DELEG)
> -			inode_set_ctime_deleg(inode, attr->ia_ctime);
> -		else
> -			inode_set_ctime_to_ts(inode, attr->ia_ctime);
> -	}
> +	if (ia_valid & ATTR_CTIME)
> +		inode_set_ctime_to_ts(inode, attr->ia_ctime);
>  }
>  EXPORT_SYMBOL(setattr_copy);
>  
> diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
> index 8b68f74a8cf08c6aa1305a2a3093467656085e4a..e6899a3502332d686138abee2284c87fc7fbc0ae 100644
> --- a/fs/nfsd/nfs4xdr.c
> +++ b/fs/nfsd/nfs4xdr.c
> @@ -537,9 +537,7 @@ nfsd4_decode_fattr4(struct nfsd4_compoundargs *argp, u32 *bmval, u32 bmlen,
>  			return nfserr_bad_xdr;
>  		iattr->ia_mtime.tv_sec = modify.seconds;
>  		iattr->ia_mtime.tv_nsec = modify.nseconds;
> -		iattr->ia_ctime.tv_sec = modify.seconds;
> -		iattr->ia_ctime.tv_nsec = modify.seconds;
> -		iattr->ia_valid |= ATTR_CTIME | ATTR_MTIME | ATTR_MTIME_SET | ATTR_DELEG;
> +		iattr->ia_valid |= ATTR_MTIME | ATTR_MTIME_SET | ATTR_DELEG;
>  	}
>  
>  	/* request sanity: did attrlist4 contain the expected number of words? */

Based on Trond's comments in this thread, I think I'm going to have to
respin this:

https://lore.kernel.org/linux-nfs/bfa20f4a81e0c2d5df8525476fb29af156f4f5f1.camel@kernel.org/

I'll post a v2 in the near future.

Cheers,
-- 
Jeff Layton <jlayton@kernel.org>

  reply	other threads:[~2025-07-22 22:19 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-22 18:52 [PATCH 0/2] vfs: fix delegated timestamp updates Jeff Layton
2025-07-22 18:52 ` [PATCH 1/2] vfs: add tracepoints in inode_set_ctime_deleg Jeff Layton
2025-07-23  1:10   ` Steven Rostedt
2025-07-24 13:49   ` Chuck Lever
2025-07-22 18:52 ` [PATCH 2/2] vfs: fix delegated timestamp handling in setattr_copy() Jeff Layton
2025-07-22 22:19   ` Jeff Layton [this message]
2025-07-23 11:51     ` 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=e28297a2abe8253c0aa590831b3857432bef60f7.camel@kernel.org \
    --to=jlayton@kernel.org \
    --cc=Dai.Ngo@oracle.com \
    --cc=brauner@kernel.org \
    --cc=chuck.lever@oracle.com \
    --cc=jack@suse.cz \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mhiramat@kernel.org \
    --cc=neil@brown.name \
    --cc=okorniev@redhat.com \
    --cc=rostedt@goodmis.org \
    --cc=tom@talpey.com \
    --cc=viro@zeniv.linux.org.uk \
    /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

all inboxes | Powered by JetHome®