| Message ID | 20240402154842.508032-8-eugen.hristev@collabora.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Cache insensitive cleanup for ext4/f2fs | expand |
On Tue, Apr 02, 2024 at 06:48:40PM +0300, Eugen Hristev via Linux-f2fs-devel wrote: > If the volume is in strict mode, generi c_ci_compare can report a broken > encoding name. This will not trigger on a bad lookup, which is caught > earlier, only if the actual disk name is bad. > > Suggested-by: Gabriel Krisman Bertazi <krisman@suse.de> > Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com> > --- > fs/f2fs/dir.c | 15 ++++++++++----- > 1 file changed, 10 insertions(+), 5 deletions(-) > > diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c > index 88b0045d0c4f..64286d80dd30 100644 > --- a/fs/f2fs/dir.c > +++ b/fs/f2fs/dir.c > @@ -192,11 +192,16 @@ static inline int f2fs_match_name(const struct inode *dir, > struct fscrypt_name f; > > #if IS_ENABLED(CONFIG_UNICODE) > - if (fname->cf_name.name) > - return generic_ci_match(dir, fname->usr_fname, > - &fname->cf_name, > - de_name, de_name_len); > - > + if (fname->cf_name.name) { > + int ret = generic_ci_match(dir, fname->usr_fname, > + &fname->cf_name, > + de_name, de_name_len); > + if (ret == -EINVAL) > + f2fs_warn(F2FS_SB(dir->i_sb), > + "Directory contains filename that is invalid UTF-8"); > + Shouldn't this use f2fs_warn_ratelimited? - Eric
On 4/3/24 07:25, Eric Biggers wrote: > On Tue, Apr 02, 2024 at 06:48:40PM +0300, Eugen Hristev via Linux-f2fs-devel wrote: >> If the volume is in strict mode, generi c_ci_compare can report a broken >> encoding name. This will not trigger on a bad lookup, which is caught >> earlier, only if the actual disk name is bad. >> >> Suggested-by: Gabriel Krisman Bertazi <krisman@suse.de> >> Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com> >> --- >> fs/f2fs/dir.c | 15 ++++++++++----- >> 1 file changed, 10 insertions(+), 5 deletions(-) >> >> diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c >> index 88b0045d0c4f..64286d80dd30 100644 >> --- a/fs/f2fs/dir.c >> +++ b/fs/f2fs/dir.c >> @@ -192,11 +192,16 @@ static inline int f2fs_match_name(const struct inode *dir, >> struct fscrypt_name f; >> >> #if IS_ENABLED(CONFIG_UNICODE) >> - if (fname->cf_name.name) >> - return generic_ci_match(dir, fname->usr_fname, >> - &fname->cf_name, >> - de_name, de_name_len); >> - >> + if (fname->cf_name.name) { >> + int ret = generic_ci_match(dir, fname->usr_fname, >> + &fname->cf_name, >> + de_name, de_name_len); >> + if (ret == -EINVAL) >> + f2fs_warn(F2FS_SB(dir->i_sb), >> + "Directory contains filename that is invalid UTF-8"); >> + > > Shouldn't this use f2fs_warn_ratelimited? f2fs_warn_ratelimited appears to be very new in the kernel, Krisman do you think you can rebase your for-next on top of latest such that this function is available ? I am basing the series on your for-next branch. Thanks > > - Eric > _______________________________________________ > Kernel mailing list -- kernel@mailman.collabora.com > To unsubscribe send an email to kernel-leave@mailman.collabora.com > This list is managed by https://mailman.collabora.com
Eugen Hristev <eugen.hristev@collabora.com> writes: > On 4/3/24 07:25, Eric Biggers wrote: >> On Tue, Apr 02, 2024 at 06:48:40PM +0300, Eugen Hristev via Linux-f2fs-devel wrote: >>> If the volume is in strict mode, generi c_ci_compare can report a broken >>> encoding name. This will not trigger on a bad lookup, which is caught >>> earlier, only if the actual disk name is bad. >>> >>> Suggested-by: Gabriel Krisman Bertazi <krisman@suse.de> >>> Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com> >>> --- >>> fs/f2fs/dir.c | 15 ++++++++++----- >>> 1 file changed, 10 insertions(+), 5 deletions(-) >>> >>> diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c >>> index 88b0045d0c4f..64286d80dd30 100644 >>> --- a/fs/f2fs/dir.c >>> +++ b/fs/f2fs/dir.c >>> @@ -192,11 +192,16 @@ static inline int f2fs_match_name(const struct inode *dir, >>> struct fscrypt_name f; >>> >>> #if IS_ENABLED(CONFIG_UNICODE) >>> - if (fname->cf_name.name) >>> - return generic_ci_match(dir, fname->usr_fname, >>> - &fname->cf_name, >>> - de_name, de_name_len); >>> - >>> + if (fname->cf_name.name) { >>> + int ret = generic_ci_match(dir, fname->usr_fname, >>> + &fname->cf_name, >>> + de_name, de_name_len); >>> + if (ret == -EINVAL) >>> + f2fs_warn(F2FS_SB(dir->i_sb), >>> + "Directory contains filename that is invalid UTF-8"); >>> + >> >> Shouldn't this use f2fs_warn_ratelimited? > > f2fs_warn_ratelimited appears to be very new in the kernel, > > Krisman do you think you can rebase your for-next on top of latest such that this > function is available ? I am basing the series on your for-next > branch. I try to make unicode/for-next a non-rebase branch, and I don't want to pollute the tree with an unecessary backmerge. Instead, why not base your work on a more recent branch, since it has no dependencies on anything from unicode/for-next?
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index 88b0045d0c4f..64286d80dd30 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -192,11 +192,16 @@ static inline int f2fs_match_name(const struct inode *dir, struct fscrypt_name f; #if IS_ENABLED(CONFIG_UNICODE) - if (fname->cf_name.name) - return generic_ci_match(dir, fname->usr_fname, - &fname->cf_name, - de_name, de_name_len); - + if (fname->cf_name.name) { + int ret = generic_ci_match(dir, fname->usr_fname, + &fname->cf_name, + de_name, de_name_len); + if (ret == -EINVAL) + f2fs_warn(F2FS_SB(dir->i_sb), + "Directory contains filename that is invalid UTF-8"); + + return ret; + } #endif f.usr_fname = fname->usr_fname; f.disk_name = fname->disk_name;
If the volume is in strict mode, generi c_ci_compare can report a broken encoding name. This will not trigger on a bad lookup, which is caught earlier, only if the actual disk name is bad. Suggested-by: Gabriel Krisman Bertazi <krisman@suse.de> Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com> --- fs/f2fs/dir.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-)