linux-kernel - Re: [PATCH v3 1/3] vfs: support caching symlink lengths in inodes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20241121-seilschaft-zeitig-7c8c3431bd00@brauner>
Date: Thu, 21 Nov 2024 11:12:52 +0100
From: Christian Brauner <brauner@...nel.org>
To: Mateusz Guzik <mjguzik@...il.com>
Cc: viro@...iv.linux.org.uk, jack@...e.cz, linux-kernel@...r.kernel.org, 
	linux-fsdevel@...r.kernel.org, hughd@...gle.com, linux-ext4@...r.kernel.org, tytso@....edu, 
	linux-mm@...ck.org
Subject: Re: [PATCH v3 1/3] vfs: support caching symlink lengths in inodes

On Wed, Nov 20, 2024 at 12:20:34PM +0100, Mateusz Guzik wrote:
> When utilized it dodges strlen() in vfs_readlink(), giving about 1.5%
> speed up when issuing readlink on /initrd.img on ext4.
> 
> Filesystems opt in by calling inode_set_cached_link() when creating an
> inode.
> 
> The size is stored in a new union utilizing the same space as i_devices,
> thus avoiding growing the struct or taking up any more space.
> 
> Churn-wise the current readlink_copy() helper is patched to accept the
> size instead of calculating it.
> 
> Signed-off-by: Mateusz Guzik <mjguzik@...il.com>
> ---
>  fs/namei.c                     | 34 +++++++++++++++++++---------------
>  fs/proc/namespaces.c           |  2 +-
>  include/linux/fs.h             | 15 +++++++++++++--
>  security/apparmor/apparmorfs.c |  2 +-
>  4 files changed, 34 insertions(+), 19 deletions(-)
> 
> diff --git a/fs/namei.c b/fs/namei.c
> index 9d30c7aa9aa6..e56c29a22d26 100644
> --- a/fs/namei.c
> +++ b/fs/namei.c
> @@ -5272,19 +5272,16 @@ SYSCALL_DEFINE2(rename, const char __user *, oldname, const char __user *, newna
>  				getname(newname), 0);
>  }
>  
> -int readlink_copy(char __user *buffer, int buflen, const char *link)
> +int readlink_copy(char __user *buffer, int buflen, const char *link, int linklen)
>  {
> -	int len = PTR_ERR(link);
> -	if (IS_ERR(link))
> -		goto out;
> +	int copylen;
>  
> -	len = strlen(link);
> -	if (len > (unsigned) buflen)
> -		len = buflen;
> -	if (copy_to_user(buffer, link, len))
> -		len = -EFAULT;
> -out:
> -	return len;
> +	copylen = linklen;
> +	if (unlikely(copylen > (unsigned) buflen))
> +		copylen = buflen;
> +	if (copy_to_user(buffer, link, copylen))
> +		copylen = -EFAULT;
> +	return copylen;
>  }
>  
>  /**
> @@ -5304,6 +5301,9 @@ int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen)
>  	const char *link;
>  	int res;
>  
> +	if (inode->i_opflags & IOP_CACHED_LINK)
> +		return readlink_copy(buffer, buflen, inode->i_link, inode->i_linklen);
> +
>  	if (unlikely(!(inode->i_opflags & IOP_DEFAULT_READLINK))) {
>  		if (unlikely(inode->i_op->readlink))
>  			return inode->i_op->readlink(dentry, buffer, buflen);
> @@ -5322,7 +5322,7 @@ int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen)
>  		if (IS_ERR(link))
>  			return PTR_ERR(link);
>  	}
> -	res = readlink_copy(buffer, buflen, link);
> +	res = readlink_copy(buffer, buflen, link, strlen(link));
>  	do_delayed_call(&done);
>  	return res;
>  }
> @@ -5391,10 +5391,14 @@ EXPORT_SYMBOL(page_put_link);
>  
>  int page_readlink(struct dentry *dentry, char __user *buffer, int buflen)
>  {
> +	const char *link;
> +	int res;
> +
>  	DEFINE_DELAYED_CALL(done);
> -	int res = readlink_copy(buffer, buflen,
> -				page_get_link(dentry, d_inode(dentry),
> -					      &done));
> +	link = page_get_link(dentry, d_inode(dentry), &done);
> +	res = PTR_ERR(link);
> +	if (!IS_ERR(link))
> +		res = readlink_copy(buffer, buflen, link, strlen(link));
>  	do_delayed_call(&done);
>  	return res;
>  }
> diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c
> index 8e159fc78c0a..c610224faf10 100644
> --- a/fs/proc/namespaces.c
> +++ b/fs/proc/namespaces.c
> @@ -83,7 +83,7 @@ static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int bufl
>  	if (ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) {
>  		res = ns_get_name(name, sizeof(name), task, ns_ops);
>  		if (res >= 0)
> -			res = readlink_copy(buffer, buflen, name);
> +			res = readlink_copy(buffer, buflen, name, strlen(name));
>  	}
>  	put_task_struct(task);
>  	return res;
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 7e29433c5ecc..2cc98de5af43 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -626,6 +626,7 @@ is_uncached_acl(struct posix_acl *acl)
>  #define IOP_XATTR	0x0008
>  #define IOP_DEFAULT_READLINK	0x0010
>  #define IOP_MGTIME	0x0020
> +#define IOP_CACHED_LINK	0x0040
>  
>  /*
>   * Keep mostly read-only and often accessed (especially for
> @@ -723,7 +724,10 @@ struct inode {
>  	};
>  	struct file_lock_context	*i_flctx;
>  	struct address_space	i_data;
> -	struct list_head	i_devices;
> +	union {
> +		struct list_head	i_devices;
> +		int			i_linklen;
> +	};

I think that i_devices should be moved into the union as it's really
only used with i_cdev but it's not that easily done because list_head
needs to be initialized. I roughly envisioned something like:

union {
        struct {
                struct cdev             *i_cdev;
                struct list_head        i_devices;
        };
        struct {
                char                    *i_link;
                unsigned int            i_link_len;
        };
        struct pipe_inode_info          *i_pipe;
        unsigned                        i_dir_seq;
};

But it's not important enough imho.