linux/fs/ecryptfs/miscdev.c

528 lines
16 KiB
C
Raw Normal View History

eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
/**
* eCryptfs: Linux filesystem encryption layer
*
* Copyright (C) 2008 International Business Machines Corp.
* Author(s): Michael A. Halcrow <mhalcrow@us.ibm.com>
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License version
* 2 as published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
* 02111-1307, USA.
*/
#include <linux/fs.h>
#include <linux/hash.h>
#include <linux/random.h>
#include <linux/miscdevice.h>
#include <linux/poll.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
#include <linux/wait.h>
#include <linux/module.h>
#include "ecryptfs_kernel.h"
static atomic_t ecryptfs_num_miscdev_opens;
/**
* ecryptfs_miscdev_poll
* @file: dev file (ignored)
* @pt: dev poll table (ignored)
*
* Returns the poll mask
*/
static unsigned int
ecryptfs_miscdev_poll(struct file *file, poll_table *pt)
{
struct ecryptfs_daemon *daemon;
unsigned int mask = 0;
uid_t euid = current_euid();
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
int rc;
mutex_lock(&ecryptfs_daemon_hash_mux);
/* TODO: Just use file->private_data? */
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
rc = ecryptfs_find_daemon_by_euid(&daemon, euid, current_user_ns());
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
BUG_ON(rc || !daemon);
mutex_lock(&daemon->mux);
mutex_unlock(&ecryptfs_daemon_hash_mux);
if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) {
printk(KERN_WARNING "%s: Attempt to poll on zombified "
"daemon\n", __func__);
goto out_unlock_daemon;
}
if (daemon->flags & ECRYPTFS_DAEMON_IN_READ)
goto out_unlock_daemon;
if (daemon->flags & ECRYPTFS_DAEMON_IN_POLL)
goto out_unlock_daemon;
daemon->flags |= ECRYPTFS_DAEMON_IN_POLL;
mutex_unlock(&daemon->mux);
poll_wait(file, &daemon->wait, pt);
mutex_lock(&daemon->mux);
if (!list_empty(&daemon->msg_ctx_out_queue))
mask |= POLLIN | POLLRDNORM;
out_unlock_daemon:
daemon->flags &= ~ECRYPTFS_DAEMON_IN_POLL;
mutex_unlock(&daemon->mux);
return mask;
}
/**
* ecryptfs_miscdev_open
* @inode: inode of miscdev handle (ignored)
* @file: file for miscdev handle (ignored)
*
* Returns zero on success; non-zero otherwise
*/
static int
ecryptfs_miscdev_open(struct inode *inode, struct file *file)
{
struct ecryptfs_daemon *daemon = NULL;
uid_t euid = current_euid();
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
int rc;
mutex_lock(&ecryptfs_daemon_hash_mux);
rc = try_module_get(THIS_MODULE);
if (rc == 0) {
rc = -EIO;
printk(KERN_ERR "%s: Error attempting to increment module use "
"count; rc = [%d]\n", __func__, rc);
goto out_unlock_daemon_list;
}
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
rc = ecryptfs_find_daemon_by_euid(&daemon, euid, current_user_ns());
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
if (rc || !daemon) {
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
rc = ecryptfs_spawn_daemon(&daemon, euid, current_user_ns(),
task_pid(current));
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
if (rc) {
printk(KERN_ERR "%s: Error attempting to spawn daemon; "
"rc = [%d]\n", __func__, rc);
goto out_module_put_unlock_daemon_list;
}
}
mutex_lock(&daemon->mux);
if (daemon->pid != task_pid(current)) {
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
rc = -EINVAL;
printk(KERN_ERR "%s: pid [0x%p] has registered with euid [%d], "
"but pid [0x%p] has attempted to open the handle "
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
"instead\n", __func__, daemon->pid, daemon->euid,
task_pid(current));
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
goto out_unlock_daemon;
}
if (daemon->flags & ECRYPTFS_DAEMON_MISCDEV_OPEN) {
rc = -EBUSY;
printk(KERN_ERR "%s: Miscellaneous device handle may only be "
"opened once per daemon; pid [0x%p] already has this "
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
"handle open\n", __func__, daemon->pid);
goto out_unlock_daemon;
}
daemon->flags |= ECRYPTFS_DAEMON_MISCDEV_OPEN;
atomic_inc(&ecryptfs_num_miscdev_opens);
out_unlock_daemon:
mutex_unlock(&daemon->mux);
out_module_put_unlock_daemon_list:
if (rc)
module_put(THIS_MODULE);
out_unlock_daemon_list:
mutex_unlock(&ecryptfs_daemon_hash_mux);
return rc;
}
/**
* ecryptfs_miscdev_release
* @inode: inode of fs/ecryptfs/euid handle (ignored)
* @file: file for fs/ecryptfs/euid handle (ignored)
*
* This keeps the daemon registered until the daemon sends another
* ioctl to fs/ecryptfs/ctl or until the kernel module unregisters.
*
* Returns zero on success; non-zero otherwise
*/
static int
ecryptfs_miscdev_release(struct inode *inode, struct file *file)
{
struct ecryptfs_daemon *daemon = NULL;
uid_t euid = current_euid();
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
int rc;
mutex_lock(&ecryptfs_daemon_hash_mux);
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
rc = ecryptfs_find_daemon_by_euid(&daemon, euid, current_user_ns());
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
BUG_ON(rc || !daemon);
mutex_lock(&daemon->mux);
BUG_ON(daemon->pid != task_pid(current));
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
BUG_ON(!(daemon->flags & ECRYPTFS_DAEMON_MISCDEV_OPEN));
daemon->flags &= ~ECRYPTFS_DAEMON_MISCDEV_OPEN;
atomic_dec(&ecryptfs_num_miscdev_opens);
mutex_unlock(&daemon->mux);
rc = ecryptfs_exorcise_daemon(daemon);
if (rc) {
printk(KERN_CRIT "%s: Fatal error whilst attempting to "
"shut down daemon; rc = [%d]. Please report this "
"bug.\n", __func__, rc);
BUG();
}
module_put(THIS_MODULE);
mutex_unlock(&ecryptfs_daemon_hash_mux);
return rc;
}
/**
* ecryptfs_send_miscdev
* @data: Data to send to daemon; may be NULL
* @data_size: Amount of data to send to daemon
* @msg_ctx: Message context, which is used to handle the reply. If
* this is NULL, then we do not expect a reply.
* @msg_type: Type of message
* @msg_flags: Flags for message
* @daemon: eCryptfs daemon object
*
* Add msg_ctx to queue and then, if it exists, notify the blocked
* miscdevess about the data being available. Must be called with
* ecryptfs_daemon_hash_mux held.
*
* Returns zero on success; non-zero otherwise
*/
int ecryptfs_send_miscdev(char *data, size_t data_size,
struct ecryptfs_msg_ctx *msg_ctx, u8 msg_type,
u16 msg_flags, struct ecryptfs_daemon *daemon)
{
int rc = 0;
mutex_lock(&msg_ctx->mux);
msg_ctx->msg = kmalloc((sizeof(*msg_ctx->msg) + data_size),
GFP_KERNEL);
if (!msg_ctx->msg) {
rc = -ENOMEM;
printk(KERN_ERR "%s: Out of memory whilst attempting "
"to kmalloc(%zd, GFP_KERNEL)\n", __func__,
(sizeof(*msg_ctx->msg) + data_size));
goto out_unlock;
}
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
msg_ctx->msg->index = msg_ctx->index;
msg_ctx->msg->data_len = data_size;
msg_ctx->type = msg_type;
memcpy(msg_ctx->msg->data, data, data_size);
msg_ctx->msg_size = (sizeof(*msg_ctx->msg) + data_size);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
mutex_lock(&daemon->mux);
list_add_tail(&msg_ctx->daemon_out_list, &daemon->msg_ctx_out_queue);
daemon->num_queued_msg_ctx++;
wake_up_interruptible(&daemon->wait);
mutex_unlock(&daemon->mux);
out_unlock:
mutex_unlock(&msg_ctx->mux);
return rc;
}
/**
* ecryptfs_miscdev_read - format and send message from queue
* @file: fs/ecryptfs/euid miscdevfs handle (ignored)
* @buf: User buffer into which to copy the next message on the daemon queue
* @count: Amount of space available in @buf
* @ppos: Offset in file (ignored)
*
* Pulls the most recent message from the daemon queue, formats it for
* being sent via a miscdevfs handle, and copies it into @buf
*
* Returns the number of bytes copied into the user buffer
*/
static ssize_t
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
ecryptfs_miscdev_read(struct file *file, char __user *buf, size_t count,
loff_t *ppos)
{
struct ecryptfs_daemon *daemon;
struct ecryptfs_msg_ctx *msg_ctx;
size_t packet_length_size;
char packet_length[3];
size_t i;
size_t total_length;
uid_t euid = current_euid();
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
int rc;
mutex_lock(&ecryptfs_daemon_hash_mux);
/* TODO: Just use file->private_data? */
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
rc = ecryptfs_find_daemon_by_euid(&daemon, euid, current_user_ns());
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
BUG_ON(rc || !daemon);
mutex_lock(&daemon->mux);
if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) {
rc = 0;
mutex_unlock(&ecryptfs_daemon_hash_mux);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
printk(KERN_WARNING "%s: Attempt to read from zombified "
"daemon\n", __func__);
goto out_unlock_daemon;
}
if (daemon->flags & ECRYPTFS_DAEMON_IN_READ) {
rc = 0;
mutex_unlock(&ecryptfs_daemon_hash_mux);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
goto out_unlock_daemon;
}
/* This daemon will not go away so long as this flag is set */
daemon->flags |= ECRYPTFS_DAEMON_IN_READ;
mutex_unlock(&ecryptfs_daemon_hash_mux);
check_list:
if (list_empty(&daemon->msg_ctx_out_queue)) {
mutex_unlock(&daemon->mux);
rc = wait_event_interruptible(
daemon->wait, !list_empty(&daemon->msg_ctx_out_queue));
mutex_lock(&daemon->mux);
if (rc < 0) {
rc = 0;
goto out_unlock_daemon;
}
}
if (daemon->flags & ECRYPTFS_DAEMON_ZOMBIE) {
rc = 0;
goto out_unlock_daemon;
}
if (list_empty(&daemon->msg_ctx_out_queue)) {
/* Something else jumped in since the
* wait_event_interruptable() and removed the
* message from the queue; try again */
goto check_list;
}
BUG_ON(euid != daemon->euid);
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
BUG_ON(current_user_ns() != daemon->user_ns);
BUG_ON(task_pid(current) != daemon->pid);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
msg_ctx = list_first_entry(&daemon->msg_ctx_out_queue,
struct ecryptfs_msg_ctx, daemon_out_list);
BUG_ON(!msg_ctx);
mutex_lock(&msg_ctx->mux);
if (msg_ctx->msg) {
rc = ecryptfs_write_packet_length(packet_length,
msg_ctx->msg_size,
&packet_length_size);
if (rc) {
rc = 0;
printk(KERN_WARNING "%s: Error writing packet length; "
"rc = [%d]\n", __func__, rc);
goto out_unlock_msg_ctx;
}
} else {
packet_length_size = 0;
msg_ctx->msg_size = 0;
}
/* miscdevfs packet format:
* Octet 0: Type
* Octets 1-4: network byte order msg_ctx->counter
* Octets 5-N0: Size of struct ecryptfs_message to follow
* Octets N0-N1: struct ecryptfs_message (including data)
*
* Octets 5-N1 not written if the packet type does not
* include a message */
total_length = (1 + 4 + packet_length_size + msg_ctx->msg_size);
if (count < total_length) {
rc = 0;
printk(KERN_WARNING "%s: Only given user buffer of "
"size [%zd], but we need [%zd] to read the "
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
"pending message\n", __func__, count, total_length);
goto out_unlock_msg_ctx;
}
rc = -EFAULT;
if (put_user(msg_ctx->type, buf))
goto out_unlock_msg_ctx;
if (put_user(cpu_to_be32(msg_ctx->counter), (__be32 __user *)(buf + 1)))
goto out_unlock_msg_ctx;
i = 5;
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
if (msg_ctx->msg) {
if (copy_to_user(&buf[i], packet_length, packet_length_size))
goto out_unlock_msg_ctx;
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
i += packet_length_size;
if (copy_to_user(&buf[i], msg_ctx->msg, msg_ctx->msg_size))
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
goto out_unlock_msg_ctx;
i += msg_ctx->msg_size;
}
rc = i;
list_del(&msg_ctx->daemon_out_list);
kfree(msg_ctx->msg);
msg_ctx->msg = NULL;
/* We do not expect a reply from the userspace daemon for any
* message type other than ECRYPTFS_MSG_REQUEST */
if (msg_ctx->type != ECRYPTFS_MSG_REQUEST)
ecryptfs_msg_ctx_alloc_to_free(msg_ctx);
out_unlock_msg_ctx:
mutex_unlock(&msg_ctx->mux);
out_unlock_daemon:
daemon->flags &= ~ECRYPTFS_DAEMON_IN_READ;
mutex_unlock(&daemon->mux);
return rc;
}
/**
* ecryptfs_miscdev_response - miscdevess response to message previously sent to daemon
* @data: Bytes comprising struct ecryptfs_message
* @data_size: sizeof(struct ecryptfs_message) + data len
* @euid: Effective user id of miscdevess sending the miscdev response
* @user_ns: The namespace in which @euid applies
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
* @pid: Miscdevess id of miscdevess sending the miscdev response
* @seq: Sequence number for miscdev response packet
*
* Returns zero on success; non-zero otherwise
*/
static int ecryptfs_miscdev_response(char *data, size_t data_size,
uid_t euid, struct user_namespace *user_ns,
struct pid *pid, u32 seq)
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
{
struct ecryptfs_message *msg = (struct ecryptfs_message *)data;
int rc;
if ((sizeof(*msg) + msg->data_len) != data_size) {
printk(KERN_WARNING "%s: (sizeof(*msg) + msg->data_len) = "
"[%zd]; data_size = [%zd]. Invalid packet.\n", __func__,
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
(sizeof(*msg) + msg->data_len), data_size);
rc = -EINVAL;
goto out;
}
rc = ecryptfs_process_response(msg, euid, user_ns, pid, seq);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
if (rc)
printk(KERN_ERR
"Error processing response message; rc = [%d]\n", rc);
out:
return rc;
}
/**
* ecryptfs_miscdev_write - handle write to daemon miscdev handle
* @file: File for misc dev handle (ignored)
* @buf: Buffer containing user data
* @count: Amount of data in @buf
* @ppos: Pointer to offset in file (ignored)
*
* miscdevfs packet format:
* Octet 0: Type
* Octets 1-4: network byte order msg_ctx->counter (0's for non-response)
* Octets 5-N0: Size of struct ecryptfs_message to follow
* Octets N0-N1: struct ecryptfs_message (including data)
*
* Returns the number of bytes read from @buf
*/
static ssize_t
ecryptfs_miscdev_write(struct file *file, const char __user *buf,
size_t count, loff_t *ppos)
{
__be32 counter_nbo;
u32 seq;
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
size_t packet_size, packet_size_length, i;
ssize_t sz = 0;
char *data;
uid_t euid = current_euid();
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
int rc;
if (count == 0)
goto out;
data = memdup_user(buf, count);
if (IS_ERR(data)) {
printk(KERN_ERR "%s: memdup_user returned error [%ld]\n",
__func__, PTR_ERR(data));
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
goto out;
}
sz = count;
i = 0;
switch (data[i++]) {
case ECRYPTFS_MSG_RESPONSE:
if (count < (1 + 4 + 1 + sizeof(struct ecryptfs_message))) {
printk(KERN_WARNING "%s: Minimum acceptable packet "
"size is [%zd], but amount of data written is "
"only [%zd]. Discarding response packet.\n",
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
__func__,
(1 + 4 + 1 + sizeof(struct ecryptfs_message)),
count);
goto out_free;
}
memcpy(&counter_nbo, &data[i], 4);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
seq = be32_to_cpu(counter_nbo);
i += 4;
rc = ecryptfs_parse_packet_length(&data[i], &packet_size,
&packet_size_length);
if (rc) {
printk(KERN_WARNING "%s: Error parsing packet length; "
"rc = [%d]\n", __func__, rc);
goto out_free;
}
i += packet_size_length;
if ((1 + 4 + packet_size_length + packet_size) != count) {
printk(KERN_WARNING "%s: (1 + packet_size_length([%zd])"
" + packet_size([%zd]))([%zd]) != "
"count([%zd]). Invalid packet format.\n",
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
__func__, packet_size_length, packet_size,
(1 + packet_size_length + packet_size), count);
goto out_free;
}
rc = ecryptfs_miscdev_response(&data[i], packet_size,
User namespaces: set of cleanups (v2) The user_ns is moved from nsproxy to user_struct, so that a struct cred by itself is sufficient to determine access (which it otherwise would not be). Corresponding ecryptfs fixes (by David Howells) are here as well. Fix refcounting. The following rules now apply: 1. The task pins the user struct. 2. The user struct pins its user namespace. 3. The user namespace pins the struct user which created it. User namespaces are cloned during copy_creds(). Unsharing a new user_ns is no longer possible. (We could re-add that, but it'll cause code duplication and doesn't seem useful if PAM doesn't need to clone user namespaces). When a user namespace is created, its first user (uid 0) gets empty keyrings and a clean group_info. This incorporates a previous patch by David Howells. Here is his original patch description: >I suggest adding the attached incremental patch. It makes the following >changes: > > (1) Provides a current_user_ns() macro to wrap accesses to current's user > namespace. > > (2) Fixes eCryptFS. > > (3) Renames create_new_userns() to create_user_ns() to be more consistent > with the other associated functions and because the 'new' in the name is > superfluous. > > (4) Moves the argument and permission checks made for CLONE_NEWUSER to the > beginning of do_fork() so that they're done prior to making any attempts > at allocation. > > (5) Calls create_user_ns() after prepare_creds(), and gives it the new creds > to fill in rather than have it return the new root user. I don't imagine > the new root user being used for anything other than filling in a cred > struct. > > This also permits me to get rid of a get_uid() and a free_uid(), as the > reference the creds were holding on the old user_struct can just be > transferred to the new namespace's creator pointer. > > (6) Makes create_user_ns() reset the UIDs and GIDs of the creds under > preparation rather than doing it in copy_creds(). > >David >Signed-off-by: David Howells <dhowells@redhat.com> Changelog: Oct 20: integrate dhowells comments 1. leave thread_keyring alone 2. use current_user_ns() in set_user() Signed-off-by: Serge Hallyn <serue@us.ibm.com>
2008-10-15 21:38:45 +00:00
euid, current_user_ns(),
task_pid(current), seq);
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
if (rc)
printk(KERN_WARNING "%s: Failed to deliver miscdev "
"response to requesting operation; rc = [%d]\n",
__func__, rc);
break;
case ECRYPTFS_MSG_HELO:
case ECRYPTFS_MSG_QUIT:
break;
default:
ecryptfs_printk(KERN_WARNING, "Dropping miscdev "
"message of unrecognized type [%d]\n",
data[0]);
break;
}
out_free:
kfree(data);
out:
return sz;
}
static const struct file_operations ecryptfs_miscdev_fops = {
.open = ecryptfs_miscdev_open,
.poll = ecryptfs_miscdev_poll,
.read = ecryptfs_miscdev_read,
.write = ecryptfs_miscdev_write,
.release = ecryptfs_miscdev_release,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 16:52:59 +00:00
.llseek = noop_llseek,
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
};
static struct miscdevice ecryptfs_miscdev = {
.minor = MISC_DYNAMIC_MINOR,
.name = "ecryptfs",
.fops = &ecryptfs_miscdev_fops
};
/**
* ecryptfs_init_ecryptfs_miscdev
*
* Messages sent to the userspace daemon from the kernel are placed on
* a queue associated with the daemon. The next read against the
* miscdev handle by that daemon will return the oldest message placed
* on the message queue for the daemon.
*
* Returns zero on success; non-zero otherwise
*/
int __init ecryptfs_init_ecryptfs_miscdev(void)
eCryptfs: introduce device handle for userspace daemon communications A regular device file was my real preference from the get-go, but I went with netlink at the time because I thought it would be less complex for managing send queues (i.e., just do a unicast and move on). It turns out that we do not really get that much complexity reduction with netlink, and netlink is more heavyweight than a device handle. In addition, the netlink interface to eCryptfs has been broken since 2.6.24. I am assuming this is a bug in how eCryptfs uses netlink, since the other in-kernel users of netlink do not seem to be having any problems. I have had one report of a user successfully using eCryptfs with netlink on 2.6.24, but for my own systems, when starting the userspace daemon, the initial helo message sent to the eCryptfs kernel module results in an oops right off the bat. I spent some time looking at it, but I have not yet found the cause. The netlink interface breaking gave me the motivation to just finish my patch to migrate to a regular device handle. If I cannot find out soon why the netlink interface in eCryptfs broke, I am likely to just send a patch to disable it in 2.6.24 and 2.6.25. I would like the device handle to be the preferred means of communicating with the userspace daemon from 2.6.26 on forward. This patch: Functions to facilitate reading and writing to the eCryptfs miscellaneous device handle. This will replace the netlink interface as the preferred mechanism for communicating with the userspace eCryptfs daemon. Each user has his own daemon, which registers itself by opening the eCryptfs device handle. Only one daemon per euid may be registered at any given time. The eCryptfs module sends a message to a daemon by adding its message to the daemon's outgoing message queue. The daemon reads the device handle to get the oldest message off the queue. Incoming messages from the userspace daemon are immediately handled. If the message is a response, then the corresponding process that is blocked waiting for the response is awakened. Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 07:59:50 +00:00
{
int rc;
atomic_set(&ecryptfs_num_miscdev_opens, 0);
rc = misc_register(&ecryptfs_miscdev);
if (rc)
printk(KERN_ERR "%s: Failed to register miscellaneous device "
"for communications with userspace daemons; rc = [%d]\n",
__func__, rc);
return rc;
}
/**
* ecryptfs_destroy_ecryptfs_miscdev
*
* All of the daemons must be exorcised prior to calling this
* function.
*/
void ecryptfs_destroy_ecryptfs_miscdev(void)
{
BUG_ON(atomic_read(&ecryptfs_num_miscdev_opens) != 0);
misc_deregister(&ecryptfs_miscdev);
}