linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Luís Henriques" <lhenriques@suse.de>
To: Sweet Tea Dorminy <sweettea-kernel@dorminy.me>
Cc: Chris Mason <clm@fb.com>, Josef Bacik <josef@toxicpanda.com>,
	David Sterba <dsterba@suse.com>,
	"Theodore Y . Ts'o" <tytso@mit.edu>,
	Jaegeuk Kim <jaegeuk@kernel.org>,
	kernel-team@meta.com, linux-btrfs@vger.kernel.org,
	linux-fscrypt@vger.kernel.org, Eric Biggers <ebiggers@kernel.org>
Subject: Re: [PATCH v3 13/16] fscrypt: allow multiple extents to reference one info
Date: Thu, 10 Aug 2023 16:51:31 +0100	[thread overview]
Message-ID: <87edkagakc.fsf@suse.de> (raw)
In-Reply-To: <2fc070a3990716077dee122740f21abcea8121a8.1691505882.git.sweettea-kernel@dorminy.me> (Sweet Tea Dorminy's message of "Tue, 8 Aug 2023 13:08:30 -0400")

Sweet Tea Dorminy <sweettea-kernel@dorminy.me> writes:

> btrfs occasionally splits in-memory extents while holding a mutex. This
> means we can't just copy the info, since setting up a new inlinecrypt
> key requires taking a semaphore. Thus adding a mechanism to split
> extents and merely take a new reference on the info is necessary.
>
> Signed-off-by: Sweet Tea Dorminy <sweettea-kernel@dorminy.me>
> ---
>  fs/crypto/fscrypt_private.h |  5 +++++
>  fs/crypto/keysetup.c        | 18 +++++++++++++++++-
>  include/linux/fscrypt.h     |  2 ++
>  3 files changed, 24 insertions(+), 1 deletion(-)
>
> diff --git a/fs/crypto/fscrypt_private.h b/fs/crypto/fscrypt_private.h
> index cd29c71b4349..03be2c136c0e 100644
> --- a/fs/crypto/fscrypt_private.h
> +++ b/fs/crypto/fscrypt_private.h
> @@ -287,6 +287,11 @@ struct fscrypt_info {
>  
>  	/* Hashed inode number.  Only set for IV_INO_LBLK_32 */
>  	u32 ci_hashed_ino;
> +
> +	/* Reference count. Normally 1, unless a extent info is shared by
> +	 * several virtual extents.
> +	 */
> +	refcount_t refs;
>  };
>  
>  typedef enum {
> diff --git a/fs/crypto/keysetup.c b/fs/crypto/keysetup.c
> index 8d50716bdf11..12c3851b7cd6 100644
> --- a/fs/crypto/keysetup.c
> +++ b/fs/crypto/keysetup.c
> @@ -598,7 +598,7 @@ static void put_crypt_info(struct fscrypt_info *ci)
>  {
>  	struct fscrypt_master_key *mk;
>  
> -	if (!ci)
> +	if (!ci || !refcount_dec_and_test(&ci->refs))
>  		return;
>  
>  	if (ci->ci_enc_key) {
> @@ -686,6 +686,7 @@ fscrypt_setup_encryption_info(struct inode *inode,
>  	crypt_info->ci_inode = inode;
>  	crypt_info->ci_sb = inode->i_sb;
>  	crypt_info->ci_policy = *policy;
> +	refcount_set(&crypt_info->refs, 1);
>  	memcpy(crypt_info->ci_nonce, nonce, FSCRYPT_FILE_NONCE_SIZE);
>  
>  	mode = select_encryption_mode(&crypt_info->ci_policy, inode);
> @@ -1046,6 +1047,21 @@ int fscrypt_load_extent_info(struct inode *inode, void *buf, size_t len,
>  }
>  EXPORT_SYMBOL_GPL(fscrypt_load_extent_info);
>  
> +/**
> + * fscrypt_get_extent_info_ref() - mark a second extent using the same info
> + * @info: the info to be used by another extent
> + *
> + * Sometimes, an existing extent must be split into multiple extents in memory.
> + * In such a case, this function allows multiple extents to use the same extent
> + * info without allocating or taking any lock, which is necessary in certain IO
> + * paths.
> + */
> +void fscrypt_get_extent_info_ref(struct fscrypt_info *info)
> +{
> +	if (info)
> +		refcount_inc(&info->refs);
> +}
> +

There's an EXPORT_SYMBOL_GPL() missing here, right?

Cheers,
-- 
Luís

> 
>  /**
>   * fscrypt_put_encryption_info() - free most of an inode's fscrypt data
>   * @inode: an inode being evicted
> diff --git a/include/linux/fscrypt.h b/include/linux/fscrypt.h
> index 4ba624beea91..b67054a2c965 100644
> --- a/include/linux/fscrypt.h
> +++ b/include/linux/fscrypt.h
> @@ -370,6 +370,8 @@ void fscrypt_free_extent_info(struct fscrypt_info **info_ptr);
>  int fscrypt_load_extent_info(struct inode *inode, void *buf, size_t len,
>  			     struct fscrypt_info **info_ptr);
>  
> +void fscrypt_get_extent_info_ref(struct fscrypt_info *info);
> +
>  /* fname.c */
>  int fscrypt_fname_encrypt(const struct inode *inode, const struct qstr *iname,
>  			  u8 *out, unsigned int olen);
> -- 
>
> 2.41.0
>

  reply	other threads:[~2023-08-10 15:51 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-08 17:08 [PATCH v3 00/16] fscrypt: add extent encryption Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 01/16] fscrypt: factor helper for locking master key Sweet Tea Dorminy
2023-08-09 17:53   ` Josef Bacik
2023-08-08 17:08 ` [PATCH v3 02/16] fscrypt: factor getting info for a specific block Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 03/16] fscrypt: adjust effective lblks based on extents Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 04/16] fscrypt: add a super_block pointer to fscrypt_info Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 05/16] fscrypt: setup leaf inodes for extent encryption Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 06/16] fscrypt: allow infos to be owned by extents Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 07/16] fscrypt: use an optional ino equivalent for per-extent infos Sweet Tea Dorminy
2023-08-09 18:51   ` Josef Bacik
2023-08-08 17:08 ` [PATCH v3 08/16] fscrypt: move function call warning of busy inodes Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 09/16] fscrypt: revamp key removal for extent encryption Sweet Tea Dorminy
2023-08-12 22:54   ` Eric Biggers
2023-08-08 17:08 ` [PATCH v3 10/16] fscrypt: add creation/usage/freeing of per-extent infos Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 11/16] fscrypt: allow load/save of extent contexts Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 12/16] fscrypt: save session key credentials for extent infos Sweet Tea Dorminy
2023-08-12 22:34   ` Eric Biggers
2023-08-08 17:08 ` [PATCH v3 13/16] fscrypt: allow multiple extents to reference one info Sweet Tea Dorminy
2023-08-10 15:51   ` Luís Henriques [this message]
2023-08-11 16:39     ` Sweet Tea Dorminy
2023-08-08 17:08 ` [PATCH v3 14/16] fscrypt: cache list of inlinecrypt devices Sweet Tea Dorminy
2023-08-12 22:41   ` Eric Biggers
2023-08-08 17:08 ` [PATCH v3 15/16] fscrypt: allow asynchronous info freeing Sweet Tea Dorminy
2023-08-12 22:43   ` Eric Biggers
2023-08-08 17:08 ` [PATCH v3 16/16] fscrypt: update documentation for per-extent keys Sweet Tea Dorminy
2023-08-09 19:52 ` [PATCH v3 00/16] fscrypt: add extent encryption Josef Bacik
2023-08-10  4:55 ` Eric Biggers
2023-08-10  9:52   ` Neal Gompa
2023-08-10 14:11   ` Sweet Tea Dorminy
2023-08-10 17:17     ` Eric Biggers
2023-08-12 22:15 ` Eric Biggers
2023-08-15 15:12   ` Josef Bacik
2023-08-16  3:37     ` Eric Biggers
2023-08-16 14:42       ` Josef Bacik

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=87edkagakc.fsf@suse.de \
    --to=lhenriques@suse.de \
    --cc=clm@fb.com \
    --cc=dsterba@suse.com \
    --cc=ebiggers@kernel.org \
    --cc=jaegeuk@kernel.org \
    --cc=josef@toxicpanda.com \
    --cc=kernel-team@meta.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fscrypt@vger.kernel.org \
    --cc=sweettea-kernel@dorminy.me \
    --cc=tytso@mit.edu \
    /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 NNTP newsgroup(s).