Create a helper function for filesystems do the checks required for
casefold directories and strict enconding.
Suggested-by: Gabriel Krisman Bertazi <gabriel@krisman.be>
Signed-off-by: André Almeida <andrealmeid@igalia.com>
---
fs/unicode/utf8-core.c | 26 ++++++++++++++++++++++++++
include/linux/unicode.h | 2 ++
2 files changed, 28 insertions(+)
diff --git a/fs/unicode/utf8-core.c b/fs/unicode/utf8-core.c
index 0400824ef493..4966e175ed71 100644
--- a/fs/unicode/utf8-core.c
+++ b/fs/unicode/utf8-core.c
@@ -214,3 +214,29 @@ void utf8_unload(struct unicode_map *um)
}
EXPORT_SYMBOL(utf8_unload);
+/**
+ * utf8_check_strict_name - Check if a given name is suitable for a directory
+ *
+ * This functions checks if the proposed filename is suitable for the parent
+ * directory. That means that only valid UTF-8 filenames will be accepted for
+ * casefold directories from filesystems created with the strict enconding flags.
+ * That also means that any name will be accepted for directories that doesn't
+ * have casefold enabled, or aren't being strict with the enconding.
+ *
+ * @inode: inode of the directory where the new file will be created
+ * @d_name: name of the new file
+ *
+ * Returns:
+ * * True if the filename is suitable for this directory. It can be true if a
+ * given name is not suitable for a strict enconding directory, but the
+ * directory being used isn't strict
+ * * False if the filename isn't suitable for this directory. This only happens
+ * when a directory is casefolded and is strict about its encoding.
+ */
+bool utf8_check_strict_name(struct inode *dir, struct qstr *d_name)
+{
+ return !(IS_CASEFOLDED(dir) && dir->i_sb->s_encoding &&
+ sb_has_strict_encoding(dir->i_sb) &&
+ utf8_validate(dir->i_sb->s_encoding, d_name));
+}
+EXPORT_SYMBOL(utf8_check_strict_name);
diff --git a/include/linux/unicode.h b/include/linux/unicode.h
index 4d39e6e11a95..fb56fb5e686c 100644
--- a/include/linux/unicode.h
+++ b/include/linux/unicode.h
@@ -76,4 +76,6 @@ int utf8_casefold_hash(const struct unicode_map *um, const void *salt,
struct unicode_map *utf8_load(unsigned int version);
void utf8_unload(struct unicode_map *um);
+bool utf8_check_strict_name(struct inode *dir, struct qstr *d_name);
+
#endif /* _LINUX_UNICODE_H */
--
2.46.0
André Almeida <andrealmeid@igalia.com> writes:
> Create a helper function for filesystems do the checks required for
> casefold directories and strict enconding.
>
> Suggested-by: Gabriel Krisman Bertazi <gabriel@krisman.be>
> Signed-off-by: André Almeida <andrealmeid@igalia.com>
> ---
> fs/unicode/utf8-core.c | 26 ++++++++++++++++++++++++++
> include/linux/unicode.h | 2 ++
> 2 files changed, 28 insertions(+)
>
> diff --git a/fs/unicode/utf8-core.c b/fs/unicode/utf8-core.c
> index 0400824ef493..4966e175ed71 100644
> --- a/fs/unicode/utf8-core.c
> +++ b/fs/unicode/utf8-core.c
I don't think this belongs in fs/unicode. it is filesystem semantics whether
they don't allow invalid utf8 names and, while fs/unicode provides
utf8_validate to verify if a string is valid, it has no business looking
into superblock and inode flags.
It would be better placed as a libfs helper.
> @@ -214,3 +214,29 @@ void utf8_unload(struct unicode_map *um)
> }
> EXPORT_SYMBOL(utf8_unload);
>
> +/**
> + * utf8_check_strict_name - Check if a given name is suitable for a directory
To follow the namespace in libfs, we could call it
generic_ci_validate_strict_name
> + *
> + * This functions checks if the proposed filename is suitable for the parent
suitable => valid
> + * directory. That means that only valid UTF-8 filenames will be accepted for
> + * casefold directories from filesystems created with the strict enconding flags.
enconding flags => encoding flag
> + * That also means that any name will be accepted for directories that doesn't
> + * have casefold enabled, or aren't being strict with the enconding.
encoding
> + *
> + * @inode: inode of the directory where the new file will be created
> + * @d_name: name of the new file
d_name means 'dentry name'. just 'name' is enough here since it doesn't
matter if the qstr is coming from the dentry.
> + *
> + * Returns:
> + * * True if the filename is suitable for this directory. It can be true if a
> + * given name is not suitable for a strict enconding directory, but the
> + * directory being used isn't strict
> + * * False if the filename isn't suitable for this directory. This only happens
> + * when a directory is casefolded and is strict about its encoding.
> + */
> +bool utf8_check_strict_name(struct inode *dir, struct qstr *d_name)
> +{
> + return !(IS_CASEFOLDED(dir) && dir->i_sb->s_encoding &&
> + sb_has_strict_encoding(dir->i_sb) &&
> + utf8_validate(dir->i_sb->s_encoding, d_name));
> +}
Now that it is a helper, it could now be unfolded to something more
readable:
if (!IS_CASEFOLDED(dir) || !sb_has_strict_encoding(dir->i_sb)))
return true;
/* Should never happen. Unless the filesystem is corrupt. */
if (WARN_ON_ONCE(!dir->i_sb->s_encoding))
return true;
return utf8_validate(...)
> +EXPORT_SYMBOL(utf8_check_strict_name);
> diff --git a/include/linux/unicode.h b/include/linux/unicode.h
> index 4d39e6e11a95..fb56fb5e686c 100644
> --- a/include/linux/unicode.h
> +++ b/include/linux/unicode.h
> @@ -76,4 +76,6 @@ int utf8_casefold_hash(const struct unicode_map *um, const void *salt,
> struct unicode_map *utf8_load(unsigned int version);
> void utf8_unload(struct unicode_map *um);
>
> +bool utf8_check_strict_name(struct inode *dir, struct qstr *d_name);
> +
> #endif /* _LINUX_UNICODE_H */
--
Gabriel Krisman Bertazi
I'd suggest using the one-line summary: unicode: create the helper function utf8_check_strict_name() so that it's a bit more descriptive. On Mon, Sep 02, 2024 at 07:55:04PM -0300, André Almeida wrote: > +/** > + * utf8_check_strict_name - Check if a given name is suitable for a directory > + * > + * This functions checks if the proposed filename is suitable for the parent > + * directory. That means that only valid UTF-8 filenames will be accepted for > + * casefold directories from filesystems created with the strict enconding flags. > + * That also means that any name will be accepted for directories that doesn't > + * have casefold enabled, or aren't being strict with the enconding. I also suggest wrapping with a fill column of 72 characters, instead of 80. - Ted
Hi André,
kernel test robot noticed the following build errors:
[auto build test ERROR on akpm-mm/mm-everything]
[also build test ERROR on tytso-ext4/dev brauner-vfs/vfs.all linus/master v6.11-rc6 next-20240903]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Andr-Almeida/unicode-Fix-utf8_load-error-path/20240903-070149
base: https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything
patch link: https://lore.kernel.org/r/20240902225511.757831-3-andrealmeid%40igalia.com
patch subject: [PATCH v2 2/8] unicode: Create utf8_check_strict_name
config: powerpc64-randconfig-r073-20240903 (https://download.01.org/0day-ci/archive/20240903/202409031655.gO1eC1AL-lkp@intel.com/config)
compiler: clang version 20.0.0git (https://github.com/llvm/llvm-project dc19b59ea2502193c0e7bc16bb7d711c8053edcf)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240903/202409031655.gO1eC1AL-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202409031655.gO1eC1AL-lkp@intel.com/
All errors (new ones prefixed by >>):
>> fs/unicode/utf8-core.c:238:11: error: call to undeclared function 'IS_CASEFOLDED'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
238 | return !(IS_CASEFOLDED(dir) && dir->i_sb->s_encoding &&
| ^
>> fs/unicode/utf8-core.c:238:36: error: incomplete definition of type 'struct inode'
238 | return !(IS_CASEFOLDED(dir) && dir->i_sb->s_encoding &&
| ~~~^
include/linux/uprobes.h:21:8: note: forward declaration of 'struct inode'
21 | struct inode;
| ^
>> fs/unicode/utf8-core.c:239:9: error: call to undeclared function 'sb_has_strict_encoding'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
239 | sb_has_strict_encoding(dir->i_sb) &&
| ^
fs/unicode/utf8-core.c:239:35: error: incomplete definition of type 'struct inode'
239 | sb_has_strict_encoding(dir->i_sb) &&
| ~~~^
include/linux/uprobes.h:21:8: note: forward declaration of 'struct inode'
21 | struct inode;
| ^
fs/unicode/utf8-core.c:240:26: error: incomplete definition of type 'struct inode'
240 | utf8_validate(dir->i_sb->s_encoding, d_name));
| ~~~^
include/linux/uprobes.h:21:8: note: forward declaration of 'struct inode'
21 | struct inode;
| ^
5 errors generated.
vim +/IS_CASEFOLDED +238 fs/unicode/utf8-core.c
216
217 /**
218 * utf8_check_strict_name - Check if a given name is suitable for a directory
219 *
220 * This functions checks if the proposed filename is suitable for the parent
221 * directory. That means that only valid UTF-8 filenames will be accepted for
222 * casefold directories from filesystems created with the strict enconding flags.
223 * That also means that any name will be accepted for directories that doesn't
224 * have casefold enabled, or aren't being strict with the enconding.
225 *
226 * @inode: inode of the directory where the new file will be created
227 * @d_name: name of the new file
228 *
229 * Returns:
230 * * True if the filename is suitable for this directory. It can be true if a
231 * given name is not suitable for a strict enconding directory, but the
232 * directory being used isn't strict
233 * * False if the filename isn't suitable for this directory. This only happens
234 * when a directory is casefolded and is strict about its encoding.
235 */
236 bool utf8_check_strict_name(struct inode *dir, struct qstr *d_name)
237 {
> 238 return !(IS_CASEFOLDED(dir) && dir->i_sb->s_encoding &&
> 239 sb_has_strict_encoding(dir->i_sb) &&
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
© 2016 - 2025 Red Hat, Inc.