From: Gabriel Krisman Bertazi <krisman@suse.de>
To: Eugen Hristev <eugen.hristev@collabora.com>
Cc: tytso@mit.edu, adilger.kernel@dilger.ca,
linux-ext4@vger.kernel.org, jaegeuk@kernel.org,
chao@kernel.org, linux-f2fs-devel@lists.sourceforge.net,
linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
kernel@collabora.com, viro@zeniv.linux.org.uk,
brauner@kernel.org, jack@suse.cz,
Gabriel Krisman Bertazi <krisman@collabora.com>,
Eric Biggers <ebiggers@google.com>
Subject: Re: [PATCH v12 2/8] f2fs: Simplify the handling of cached insensitive names
Date: Tue, 27 Feb 2024 18:11:53 -0500 [thread overview]
Message-ID: <87cysh5wp2.fsf@mailhost.krisman.be> (raw)
In-Reply-To: <20240220085235.71132-3-eugen.hristev@collabora.com> (Eugen Hristev's message of "Tue, 20 Feb 2024 10:52:29 +0200")
Eugen Hristev <eugen.hristev@collabora.com> writes:
> From: Gabriel Krisman Bertazi <krisman@collabora.com>
>
> Keeping it as qstr avoids the unnecessary conversion in f2fs_match
>
> Reviewed-by: Eric Biggers <ebiggers@google.com>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
> [eugen.hristev@collabora.com: port to 6.8-rc3]
> Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
> ---
> fs/f2fs/dir.c | 53 ++++++++++++++++++++++++++--------------------
> fs/f2fs/f2fs.h | 17 ++++++++++++++-
> fs/f2fs/recovery.c | 5 +----
> 3 files changed, 47 insertions(+), 28 deletions(-)
>
> diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
> index 3f20d94e12f9..f5b65cf36393 100644
> --- a/fs/f2fs/dir.c
> +++ b/fs/f2fs/dir.c
> @@ -42,35 +42,49 @@ static unsigned int bucket_blocks(unsigned int level)
> return 4;
> }
>
> +#if IS_ENABLED(CONFIG_UNICODE)
> /* If @dir is casefolded, initialize @fname->cf_name from @fname->usr_fname. */
> int f2fs_init_casefolded_name(const struct inode *dir,
> struct f2fs_filename *fname)
> {
> -#if IS_ENABLED(CONFIG_UNICODE)
> struct super_block *sb = dir->i_sb;
> + unsigned char *buf;
> + int len;
>
> if (IS_CASEFOLDED(dir) &&
> !is_dot_dotdot(fname->usr_fname->name, fname->usr_fname->len)) {
> - fname->cf_name.name = f2fs_kmem_cache_alloc(f2fs_cf_name_slab,
> - GFP_NOFS, false, F2FS_SB(sb));
> - if (!fname->cf_name.name)
> + buf = f2fs_kmem_cache_alloc(f2fs_cf_name_slab,
> + GFP_NOFS, false, F2FS_SB(sb));
> + if (!buf)
> return -ENOMEM;
> - fname->cf_name.len = utf8_casefold(sb->s_encoding,
> - fname->usr_fname,
> - fname->cf_name.name,
> - F2FS_NAME_LEN);
> - if ((int)fname->cf_name.len <= 0) {
> - kmem_cache_free(f2fs_cf_name_slab, fname->cf_name.name);
> - fname->cf_name.name = NULL;
> +
> + len = utf8_casefold(sb->s_encoding, fname->usr_fname,
> + buf, F2FS_NAME_LEN);
> + if (len <= 0) {
> + kmem_cache_free(f2fs_cf_name_slab, buf);
> if (sb_has_strict_encoding(sb))
> return -EINVAL;
> /* fall back to treating name as opaque byte sequence */
> + return 0;
> }
> + fname->cf_name.name = buf;
> + fname->cf_name.len = len;
> }
> -#endif
> +
> return 0;
> }
>
> +void f2fs_free_casefolded_name(struct f2fs_filename *fname)
> +{
> + unsigned char *buf = (unsigned char *)fname->cf_name.name;
> +
> + if (buf) {
> + kmem_cache_free(f2fs_cf_name_slab, buf);
> + fname->cf_name.name = NULL;
> + }
> +}
If we use kfree here, we can drop the buf !=NULL check.
> +#endif /* CONFIG_UNICODE */
> +
> static int __f2fs_setup_filename(const struct inode *dir,
> const struct fscrypt_name *crypt_name,
> struct f2fs_filename *fname)
> @@ -142,12 +156,7 @@ void f2fs_free_filename(struct f2fs_filename *fname)
> kfree(fname->crypto_buf.name);
> fname->crypto_buf.name = NULL;
> #endif
> -#if IS_ENABLED(CONFIG_UNICODE)
> - if (fname->cf_name.name) {
> - kmem_cache_free(f2fs_cf_name_slab, fname->cf_name.name);
> - fname->cf_name.name = NULL;
> - }
> -#endif
> + f2fs_free_casefolded_name(fname);
> }
>
> static unsigned long dir_block_index(unsigned int level,
> @@ -235,11 +244,9 @@ static inline int f2fs_match_name(const struct inode *dir,
> struct fscrypt_name f;
>
> #if IS_ENABLED(CONFIG_UNICODE)
> - if (fname->cf_name.name) {
> - struct qstr cf = FSTR_TO_QSTR(&fname->cf_name);
> -
> - return f2fs_match_ci_name(dir, &cf, de_name, de_name_len);
> - }
> + if (fname->cf_name.name)
> + return f2fs_match_ci_name(dir, &fname->cf_name,
> + de_name, de_name_len);
> #endif
> f.usr_fname = fname->usr_fname;
> f.disk_name = fname->disk_name;
> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> index 84c9fead3ad4..2ff8e52642ec 100644
> --- a/fs/f2fs/f2fs.h
> +++ b/fs/f2fs/f2fs.h
> @@ -530,7 +530,7 @@ struct f2fs_filename {
> * internal operation where usr_fname is also NULL. In all these cases
> * we fall back to treating the name as an opaque byte sequence.
> */
> - struct fscrypt_str cf_name;
> + struct qstr cf_name;
> #endif
> };
>
> @@ -3533,8 +3533,23 @@ int f2fs_get_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
> /*
> * dir.c
> */
> +unsigned char f2fs_get_de_type(struct f2fs_dir_entry *de);
This is not part of the original patch and doesn't make sense here. It
seems to be included by a bad rebase?
--
Gabriel Krisman Bertazi
WARNING: multiple messages have this Message-ID (diff)
From: Gabriel Krisman Bertazi <krisman@suse.de>
To: Eugen Hristev <eugen.hristev@collabora.com>
Cc: brauner@kernel.org, kernel@collabora.com, tytso@mit.edu,
Eric Biggers <ebiggers@google.com>,
jack@suse.cz, linux-kernel@vger.kernel.org,
linux-f2fs-devel@lists.sourceforge.net, adilger.kernel@dilger.ca,
viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org,
jaegeuk@kernel.org, linux-ext4@vger.kernel.org,
Gabriel Krisman Bertazi <krisman@collabora.com>
Subject: Re: [f2fs-dev] [PATCH v12 2/8] f2fs: Simplify the handling of cached insensitive names
Date: Tue, 27 Feb 2024 18:11:53 -0500 [thread overview]
Message-ID: <87cysh5wp2.fsf@mailhost.krisman.be> (raw)
In-Reply-To: <20240220085235.71132-3-eugen.hristev@collabora.com> (Eugen Hristev's message of "Tue, 20 Feb 2024 10:52:29 +0200")
Eugen Hristev <eugen.hristev@collabora.com> writes:
> From: Gabriel Krisman Bertazi <krisman@collabora.com>
>
> Keeping it as qstr avoids the unnecessary conversion in f2fs_match
>
> Reviewed-by: Eric Biggers <ebiggers@google.com>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
> [eugen.hristev@collabora.com: port to 6.8-rc3]
> Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
> ---
> fs/f2fs/dir.c | 53 ++++++++++++++++++++++++++--------------------
> fs/f2fs/f2fs.h | 17 ++++++++++++++-
> fs/f2fs/recovery.c | 5 +----
> 3 files changed, 47 insertions(+), 28 deletions(-)
>
> diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
> index 3f20d94e12f9..f5b65cf36393 100644
> --- a/fs/f2fs/dir.c
> +++ b/fs/f2fs/dir.c
> @@ -42,35 +42,49 @@ static unsigned int bucket_blocks(unsigned int level)
> return 4;
> }
>
> +#if IS_ENABLED(CONFIG_UNICODE)
> /* If @dir is casefolded, initialize @fname->cf_name from @fname->usr_fname. */
> int f2fs_init_casefolded_name(const struct inode *dir,
> struct f2fs_filename *fname)
> {
> -#if IS_ENABLED(CONFIG_UNICODE)
> struct super_block *sb = dir->i_sb;
> + unsigned char *buf;
> + int len;
>
> if (IS_CASEFOLDED(dir) &&
> !is_dot_dotdot(fname->usr_fname->name, fname->usr_fname->len)) {
> - fname->cf_name.name = f2fs_kmem_cache_alloc(f2fs_cf_name_slab,
> - GFP_NOFS, false, F2FS_SB(sb));
> - if (!fname->cf_name.name)
> + buf = f2fs_kmem_cache_alloc(f2fs_cf_name_slab,
> + GFP_NOFS, false, F2FS_SB(sb));
> + if (!buf)
> return -ENOMEM;
> - fname->cf_name.len = utf8_casefold(sb->s_encoding,
> - fname->usr_fname,
> - fname->cf_name.name,
> - F2FS_NAME_LEN);
> - if ((int)fname->cf_name.len <= 0) {
> - kmem_cache_free(f2fs_cf_name_slab, fname->cf_name.name);
> - fname->cf_name.name = NULL;
> +
> + len = utf8_casefold(sb->s_encoding, fname->usr_fname,
> + buf, F2FS_NAME_LEN);
> + if (len <= 0) {
> + kmem_cache_free(f2fs_cf_name_slab, buf);
> if (sb_has_strict_encoding(sb))
> return -EINVAL;
> /* fall back to treating name as opaque byte sequence */
> + return 0;
> }
> + fname->cf_name.name = buf;
> + fname->cf_name.len = len;
> }
> -#endif
> +
> return 0;
> }
>
> +void f2fs_free_casefolded_name(struct f2fs_filename *fname)
> +{
> + unsigned char *buf = (unsigned char *)fname->cf_name.name;
> +
> + if (buf) {
> + kmem_cache_free(f2fs_cf_name_slab, buf);
> + fname->cf_name.name = NULL;
> + }
> +}
If we use kfree here, we can drop the buf !=NULL check.
> +#endif /* CONFIG_UNICODE */
> +
> static int __f2fs_setup_filename(const struct inode *dir,
> const struct fscrypt_name *crypt_name,
> struct f2fs_filename *fname)
> @@ -142,12 +156,7 @@ void f2fs_free_filename(struct f2fs_filename *fname)
> kfree(fname->crypto_buf.name);
> fname->crypto_buf.name = NULL;
> #endif
> -#if IS_ENABLED(CONFIG_UNICODE)
> - if (fname->cf_name.name) {
> - kmem_cache_free(f2fs_cf_name_slab, fname->cf_name.name);
> - fname->cf_name.name = NULL;
> - }
> -#endif
> + f2fs_free_casefolded_name(fname);
> }
>
> static unsigned long dir_block_index(unsigned int level,
> @@ -235,11 +244,9 @@ static inline int f2fs_match_name(const struct inode *dir,
> struct fscrypt_name f;
>
> #if IS_ENABLED(CONFIG_UNICODE)
> - if (fname->cf_name.name) {
> - struct qstr cf = FSTR_TO_QSTR(&fname->cf_name);
> -
> - return f2fs_match_ci_name(dir, &cf, de_name, de_name_len);
> - }
> + if (fname->cf_name.name)
> + return f2fs_match_ci_name(dir, &fname->cf_name,
> + de_name, de_name_len);
> #endif
> f.usr_fname = fname->usr_fname;
> f.disk_name = fname->disk_name;
> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> index 84c9fead3ad4..2ff8e52642ec 100644
> --- a/fs/f2fs/f2fs.h
> +++ b/fs/f2fs/f2fs.h
> @@ -530,7 +530,7 @@ struct f2fs_filename {
> * internal operation where usr_fname is also NULL. In all these cases
> * we fall back to treating the name as an opaque byte sequence.
> */
> - struct fscrypt_str cf_name;
> + struct qstr cf_name;
> #endif
> };
>
> @@ -3533,8 +3533,23 @@ int f2fs_get_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
> /*
> * dir.c
> */
> +unsigned char f2fs_get_de_type(struct f2fs_dir_entry *de);
This is not part of the original patch and doesn't make sense here. It
seems to be included by a bad rebase?
--
Gabriel Krisman Bertazi
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
next prev parent reply other threads:[~2024-02-27 23:12 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-20 8:52 [PATCH v12 0/8] Cache insensitive cleanup for ext4/f2fs Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-20 8:52 ` [PATCH v12 1/8] ext4: Simplify the handling of cached insensitive names Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-20 8:52 ` [PATCH v12 2/8] f2fs: " Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-27 23:11 ` Gabriel Krisman Bertazi [this message]
2024-02-27 23:11 ` Gabriel Krisman Bertazi
2024-02-20 8:52 ` [PATCH v12 3/8] libfs: Introduce case-insensitive string comparison helper Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-27 23:32 ` Gabriel Krisman Bertazi
2024-02-27 23:32 ` [f2fs-dev] " Gabriel Krisman Bertazi
2024-02-20 8:52 ` [PATCH v12 4/8] ext4: Reuse generic_ci_match for ci comparisons Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-20 8:52 ` [PATCH v12 5/8] f2fs: " Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-20 8:52 ` [PATCH v12 6/8] ext4: Log error when lookup of encoded dentry fails Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-27 23:38 ` Gabriel Krisman Bertazi
2024-02-27 23:38 ` [f2fs-dev] " Gabriel Krisman Bertazi
2024-02-20 8:52 ` [PATCH v12 7/8] ext4: Move CONFIG_UNICODE defguards into the code flow Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-20 8:52 ` [PATCH v12 8/8] f2fs: " Eugen Hristev
2024-02-20 8:52 ` [f2fs-dev] " Eugen Hristev via Linux-f2fs-devel
2024-02-27 23:48 ` [PATCH v12 0/8] Cache insensitive cleanup for ext4/f2fs Gabriel Krisman Bertazi
2024-02-27 23:48 ` [f2fs-dev] " Gabriel Krisman Bertazi
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=87cysh5wp2.fsf@mailhost.krisman.be \
--to=krisman@suse.de \
--cc=adilger.kernel@dilger.ca \
--cc=brauner@kernel.org \
--cc=chao@kernel.org \
--cc=ebiggers@google.com \
--cc=eugen.hristev@collabora.com \
--cc=jack@suse.cz \
--cc=jaegeuk@kernel.org \
--cc=kernel@collabora.com \
--cc=krisman@collabora.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tytso@mit.edu \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.