| Message ID | 1445338426-4126-2-git-send-email-daniel.baluta@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, Oct 20, 2015 at 01:53:42PM +0300, Daniel Baluta wrote: > We don't want to hardcode default groups at subsystem > creation time. We export: > * configfs_register_group > * configfs_unregister_group > to allow drivers to programatically create/destroy groups > later, after module init time. > > This is needed for IIO configfs support. This looks like a good start, and maybe we should merge it as is given that merge window is about to end. But I have two requests, one trivial and one not so: a) please provide kerneldoc comments for the new functions b) just pass the name and item type and dynamically allocate the actua group. With that we can avoid a lot of boilerplate code and could also replace the horrible static sized defaults_groups array in current users with your new function. -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, Oct 22, 2015 at 11:40 AM, Christoph Hellwig <hch@lst.de> wrote: > On Tue, Oct 20, 2015 at 01:53:42PM +0300, Daniel Baluta wrote: >> We don't want to hardcode default groups at subsystem >> creation time. We export: >> * configfs_register_group >> * configfs_unregister_group >> to allow drivers to programatically create/destroy groups >> later, after module init time. >> >> This is needed for IIO configfs support. > > This looks like a good start, and maybe we should merge it as is given > that merge window is about to end. But I have two requests, one trivial > and one not so: > > a) please provide kerneldoc comments for the new functions > b) just pass the name and item type and dynamically allocate the > actua group. With that we can avoid a lot of boilerplate code and > could also replace the horrible static sized defaults_groups array > in current users with your new function. Will try to see how quick I can do this. Eventually we can introduce the new function + use it in IIO and fix current users later. Thanks Christoph! Daniel. -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index c81ce7f..1f5337c 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -1636,6 +1636,56 @@ const struct file_operations configfs_dir_operations = { .iterate = configfs_readdir, }; +int configfs_register_group(struct config_group *parent_group, + struct config_group *group) +{ + struct configfs_subsystem *subsys = parent_group->cg_subsys; + struct dentry *parent; + int ret; + + mutex_lock(&subsys->su_mutex); + link_group(parent_group, group); + mutex_unlock(&subsys->su_mutex); + + parent = parent_group->cg_item.ci_dentry; + + mutex_lock_nested(&d_inode(parent)->i_mutex, I_MUTEX_PARENT); + ret = create_default_group(parent_group, group); + if (!ret) { + spin_lock(&configfs_dirent_lock); + configfs_dir_set_ready(group->cg_item.ci_dentry->d_fsdata); + spin_unlock(&configfs_dirent_lock); + } + mutex_unlock(&d_inode(parent)->i_mutex); + return ret; +} +EXPORT_SYMBOL(configfs_register_group); + +void configfs_unregister_group(struct config_group *group) +{ + struct configfs_subsystem *subsys = group->cg_subsys; + struct dentry *dentry = group->cg_item.ci_dentry; + struct dentry *parent = group->cg_item.ci_parent->ci_dentry; + + mutex_lock_nested(&d_inode(parent)->i_mutex, I_MUTEX_PARENT); + spin_lock(&configfs_dirent_lock); + configfs_detach_prep(dentry, NULL); + spin_unlock(&configfs_dirent_lock); + + configfs_detach_group(&group->cg_item); + d_inode(dentry)->i_flags |= S_DEAD; + dont_mount(dentry); + d_delete(dentry); + mutex_unlock(&d_inode(parent)->i_mutex); + + dput(dentry); + + mutex_lock(&subsys->su_mutex); + unlink_group(group); + mutex_unlock(&subsys->su_mutex); +} +EXPORT_SYMBOL(configfs_unregister_group); + int configfs_register_subsystem(struct configfs_subsystem *subsys) { int err; diff --git a/include/linux/configfs.h b/include/linux/configfs.h index 63a36e8..c40b8143 100644 --- a/include/linux/configfs.h +++ b/include/linux/configfs.h @@ -252,6 +252,10 @@ static inline struct configfs_subsystem *to_configfs_subsystem(struct config_gro int configfs_register_subsystem(struct configfs_subsystem *subsys); void configfs_unregister_subsystem(struct configfs_subsystem *subsys); +int configfs_register_group(struct config_group *parent_group, + struct config_group *group); +void configfs_unregister_group(struct config_group *group); + /* These functions can sleep and can alloc with GFP_KERNEL */ /* WARNING: These cannot be called underneath configfs callbacks!! */ int configfs_depend_item(struct configfs_subsystem *subsys, struct config_item *target);
We don't want to hardcode default groups at subsystem creation time. We export: * configfs_register_group * configfs_unregister_group to allow drivers to programatically create/destroy groups later, after module init time. This is needed for IIO configfs support. Suggested-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Daniel Baluta <daniel.baluta@intel.com> --- fs/configfs/dir.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/configfs.h | 4 ++++ 2 files changed, 54 insertions(+)