From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751445AbbCKHBF (ORCPT ); Wed, 11 Mar 2015 03:01:05 -0400 Received: from mail-qg0-f41.google.com ([209.85.192.41]:39576 "EHLO mail-qg0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750851AbbCKHBB convert rfc822-to-8bit (ORCPT ); Wed, 11 Mar 2015 03:01:01 -0400 References: <20150310171133.23081.49616.stgit@buzz> <20150310172205.23081.67263.stgit@buzz> Mime-Version: 1.0 (1.0) In-Reply-To: <20150310172205.23081.67263.stgit@buzz> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8BIT Message-Id: <3403886C-8932-4CB0-8B1D-F468A5B199BA@dilger.ca> Cc: "linux-fsdevel@vger.kernel.org" , Dave Chinner , Jan Kara , "linux-ext4@vger.kernel.org" , "Theodore Ts'o" , Dmitry Monakhov , Andy Lutomirski , "linux-kernel@vger.kernel.org" , Li Xi X-Mailer: iPhone Mail (12B466) From: Andreas Dilger Subject: Re: [PATCH RFC v2 1/6] fs: vfs ioctls for managing project id Date: Wed, 11 Mar 2015 03:00:59 -0400 To: Konstantin Khlebnikov Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org How do your patches relate to the project quota patches from Li Xi? Cheers, Andreas > On Mar 10, 2015, at 13:22, Konstantin Khlebnikov wrote: > > This patch adds generic vfs interface for project id and > related super-block methods. > > Two new ioctls: > int ioctl(fd, FS_IOC_GETPROJECT, unsigned *project); > int ioctl(fd, FS_IOC_SETPROJECT, unsigned *project); > > Signed-off-by: Konstantin Khlebnikov > --- > Documentation/filesystems/Locking | 4 +++ > Documentation/filesystems/vfs.txt | 8 +++++ > fs/compat_ioctl.c | 2 + > fs/ioctl.c | 58 +++++++++++++++++++++++++++++++++++++ > include/linux/fs.h | 2 + > include/uapi/linux/fs.h | 3 ++ > 6 files changed, 77 insertions(+) > > diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking > index f91926f2f482..7a01e2b606d0 100644 > --- a/Documentation/filesystems/Locking > +++ b/Documentation/filesystems/Locking > @@ -126,6 +126,8 @@ prototypes: > ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t); > ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); > int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t); > + kprojid_t (*get_project) (struct inode *); > + int (*set_project) (struct inode *, kprojid_t); > > locking rules: > All may block [not true, see below] > @@ -147,6 +149,8 @@ show_options: no (namespace_sem) > quota_read: no (see below) > quota_write: no (see below) > bdev_try_to_free_page: no (see below) > +get_project no > +set_project no (i_mutex) > > ->statfs() has s_umount (shared) when called by ustat(2) (native or > compat), but that's an accident of bad API; s_umount is used to pin > diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt > index 966b22829f3b..aeff3c54021b 100644 > --- a/Documentation/filesystems/vfs.txt > +++ b/Documentation/filesystems/vfs.txt > @@ -230,6 +230,8 @@ struct super_operations { > ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); > int (*nr_cached_objects)(struct super_block *); > void (*free_cached_objects)(struct super_block *, int); > + kprojid_t (*get_project)(struct inode *); > + int (*set_project)(struct inode *, kprojid_t); > }; > > All methods are called without any locks being held, unless otherwise > @@ -319,6 +321,12 @@ or bottom half). > implementations will cause holdoff problems due to large scan batch > sizes. > > + get_project: called by the ioctl and quota code to get project id of an inode. > + Method should return INVALID_PROJID if feature isn't supported. > + > + set_project: called by the ioctl to set project id for an inode. > + This is called with i_mutex held. > + > Whoever sets up the inode is responsible for filling in the "i_op" field. This > is a pointer to a "struct inode_operations" which describes the methods that > can be performed on individual inodes. > diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c > index afec6450450f..9a24e4038377 100644 > --- a/fs/compat_ioctl.c > +++ b/fs/compat_ioctl.c > @@ -889,6 +889,8 @@ COMPATIBLE_IOCTL(FIOASYNC) > COMPATIBLE_IOCTL(FIONBIO) > COMPATIBLE_IOCTL(FIONREAD) /* This is also TIOCINQ */ > COMPATIBLE_IOCTL(FS_IOC_FIEMAP) > +COMPATIBLE_IOCTL(FS_IOC_GETPROJECT) > +COMPATIBLE_IOCTL(FS_IOC_SETPROJECT) > /* 0x00 */ > COMPATIBLE_IOCTL(FIBMAP) > COMPATIBLE_IOCTL(FIGETBSZ) > diff --git a/fs/ioctl.c b/fs/ioctl.c > index 5d01d2638ca5..d351576d95c8 100644 > --- a/fs/ioctl.c > +++ b/fs/ioctl.c > @@ -9,6 +9,7 @@ > #include > #include > #include > +#include > #include > #include > #include > @@ -545,6 +546,57 @@ static int ioctl_fsthaw(struct file *filp) > return thaw_super(sb); > } > > +static int ioctl_getproject(struct file *filp, projid_t __user *argp) > +{ > + struct user_namespace *ns = current_user_ns(); > + struct inode *inode = file_inode(filp); > + struct super_block *sb = inode->i_sb; > + kprojid_t kprojid; > + projid_t projid; > + > + if (!sb->s_op->get_project) > + return -EOPNOTSUPP; > + kprojid = sb->s_op->get_project(inode); > + if (!projid_valid(kprojid)) > + return -EOPNOTSUPP; > + projid = from_kprojid(ns, kprojid); > + if (projid == (projid_t)-1) > + return -EACCES; > + return put_user(projid, argp); > +} > + > +static int ioctl_setproject(struct file *filp, projid_t __user *argp) > +{ > + struct user_namespace *ns = current_user_ns(); > + struct inode *inode = file_inode(filp); > + struct super_block *sb = inode->i_sb; > + kprojid_t kprojid; > + projid_t projid; > + int ret; > + > + if (!sb->s_op->set_project) > + return -EOPNOTSUPP; > + if (ns != &init_user_ns) > + return -EPERM; > + ret = get_user(projid, argp); > + if (ret) > + return ret; > + kprojid = make_kprojid(ns, projid); > + if (!projid_valid(kprojid)) > + return -EACCES; > + ret = mnt_want_write_file(filp); > + if (ret) > + return ret; > + mutex_lock(&inode->i_mutex); > + if (inode_owner_or_capable(inode)) > + ret = sb->s_op->set_project(inode, kprojid); > + else > + ret = -EPERM; > + mutex_unlock(&inode->i_mutex); > + mnt_drop_write_file(filp); > + return ret; > +} > + > /* > * When you add any new common ioctls to the switches above and below > * please update compat_sys_ioctl() too. > @@ -597,6 +649,12 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, > case FS_IOC_FIEMAP: > return ioctl_fiemap(filp, arg); > > + case FS_IOC_GETPROJECT: > + return ioctl_getproject(filp, argp); > + > + case FS_IOC_SETPROJECT: > + return ioctl_setproject(filp, argp); > + > case FIGETBSZ: > return put_user(inode->i_sb->s_blocksize, argp); > > diff --git a/include/linux/fs.h b/include/linux/fs.h > index b4d71b5e1ff2..42156801739e 100644 > --- a/include/linux/fs.h > +++ b/include/linux/fs.h > @@ -1655,6 +1655,8 @@ struct super_operations { > struct shrink_control *); > long (*free_cached_objects)(struct super_block *, > struct shrink_control *); > + kprojid_t (*get_project)(struct inode *); > + int (*set_project)(struct inode *, kprojid_t); > }; > > /* > diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h > index 9b964a5920af..144426326e15 100644 > --- a/include/uapi/linux/fs.h > +++ b/include/uapi/linux/fs.h > @@ -170,6 +170,9 @@ struct inodes_stat_t { > #define FS_IOC32_GETVERSION _IOR('v', 1, int) > #define FS_IOC32_SETVERSION _IOW('v', 2, int) > > +#define FS_IOC_GETPROJECT _IOR('f', 20, unsigned) > +#define FS_IOC_SETPROJECT _IOW('f', 21, unsigned) > + > /* > * Inode flags (FS_IOC_GETFLAGS / FS_IOC_SETFLAGS) > */ > > -- > 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