linux/ipc/shm.c

1180 lines
27 KiB
C
Raw Normal View History

/*
* linux/ipc/shm.c
* Copyright (C) 1992, 1993 Krishna Balasubramanian
* Many improvements/fixes by Bruno Haible.
* Replaced `struct shm_desc' by `struct vm_area_struct', July 1994.
* Fixed the shm swap deallocation (shm_unuse()), August 1998 Andrea Arcangeli.
*
* /proc/sysvipc/shm support (c) 1999 Dragos Acostachioaie <dragos@iname.com>
* BIGMEM support, Andrea Arcangeli <andrea@suse.de>
* SMP thread shm, Jean-Luc Boyard <jean-luc.boyard@siemens.fr>
* HIGHMEM support, Ingo Molnar <mingo@redhat.com>
* Make shmmax, shmall, shmmni sysctl'able, Christoph Rohland <cr@sap.com>
* Shared /dev/zero support, Kanoj Sarcar <kanoj@sgi.com>
* Move the mm functionality over to mm/shmem.c, Christoph Rohland <cr@sap.com>
*
[PATCH] Rework of IPC auditing 1) The audit_ipc_perms() function has been split into two different functions: - audit_ipc_obj() - audit_ipc_set_perm() There's a key shift here... The audit_ipc_obj() collects the uid, gid, mode, and SElinux context label of the current ipc object. This audit_ipc_obj() hook is now found in several places. Most notably, it is hooked in ipcperms(), which is called in various places around the ipc code permforming a MAC check. Additionally there are several places where *checkid() is used to validate that an operation is being performed on a valid object while not necessarily having a nearby ipcperms() call. In these locations, audit_ipc_obj() is called to ensure that the information is captured by the audit system. The audit_set_new_perm() function is called any time the permissions on the ipc object changes. In this case, the NEW permissions are recorded (and note that an audit_ipc_obj() call exists just a few lines before each instance). 2) Support for an AUDIT_IPC_SET_PERM audit message type. This allows for separate auxiliary audit records for normal operations on an IPC object and permissions changes. Note that the same struct audit_aux_data_ipcctl is used and populated, however there are separate audit_log_format statements based on the type of the message. Finally, the AUDIT_IPC block of code in audit_free_aux() was extended to handle aux messages of this new type. No more mem leaks I hope ;-) Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-04-02 23:07:33 +02:00
* support for audit of ipc object properties and permission changes
* Dustin Kirkland <dustin.kirkland@us.ibm.com>
*
* namespaces support
* OpenVZ, SWsoft Inc.
* Pavel Emelianov <xemul@openvz.org>
*/
#include <linux/slab.h>
#include <linux/mm.h>
#include <linux/hugetlb.h>
#include <linux/shm.h>
#include <linux/init.h>
#include <linux/file.h>
#include <linux/mman.h>
#include <linux/shmem_fs.h>
#include <linux/security.h>
#include <linux/syscalls.h>
#include <linux/audit.h>
#include <linux/capability.h>
#include <linux/ptrace.h>
#include <linux/seq_file.h>
#include <linux/rwsem.h>
#include <linux/nsproxy.h>
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
#include <linux/mount.h>
#include <asm/uaccess.h>
#include "util.h"
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct shm_file_data {
int id;
struct ipc_namespace *ns;
struct file *file;
const struct vm_operations_struct *vm_ops;
};
#define shm_file_data(file) (*((struct shm_file_data **)&(file)->private_data))
static const struct file_operations shm_file_operations;
static struct vm_operations_struct shm_vm_ops;
static struct ipc_ids init_shm_ids;
#define shm_ids(ns) (*((ns)->ids[IPC_SHM_IDS]))
#define shm_unlock(shp) \
ipc_unlock(&(shp)->shm_perm)
#define shm_buildid(id, seq) ipc_buildid(id, seq)
static int newseg(struct ipc_namespace *, struct ipc_params *);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
static void shm_open(struct vm_area_struct *vma);
static void shm_close(struct vm_area_struct *vma);
static void shm_destroy (struct ipc_namespace *ns, struct shmid_kernel *shp);
#ifdef CONFIG_PROC_FS
static int sysvipc_shm_proc_show(struct seq_file *s, void *it);
#endif
static void __shm_init_ns(struct ipc_namespace *ns, struct ipc_ids *ids)
{
ns->ids[IPC_SHM_IDS] = ids;
ns->shm_ctlmax = SHMMAX;
ns->shm_ctlall = SHMALL;
ns->shm_ctlmni = SHMMNI;
ns->shm_tot = 0;
ipc_init_ids(ids);
}
/*
* Called with shm_ids.rw_mutex (writer) and the shp structure locked.
* Only shm_ids.rw_mutex remains locked on exit.
*/
static void do_shm_rmid(struct ipc_namespace *ns, struct shmid_kernel *shp)
{
if (shp->shm_nattch){
shp->shm_perm.mode |= SHM_DEST;
/* Do not find it any more */
shp->shm_perm.key = IPC_PRIVATE;
shm_unlock(shp);
} else
shm_destroy(ns, shp);
}
int shm_init_ns(struct ipc_namespace *ns)
{
struct ipc_ids *ids;
ids = kmalloc(sizeof(struct ipc_ids), GFP_KERNEL);
if (ids == NULL)
return -ENOMEM;
__shm_init_ns(ns, ids);
return 0;
}
void shm_exit_ns(struct ipc_namespace *ns)
{
struct shmid_kernel *shp;
int next_id;
int total, in_use;
down_write(&shm_ids(ns).rw_mutex);
in_use = shm_ids(ns).in_use;
for (total = 0, next_id = 0; total < in_use; next_id++) {
shp = idr_find(&shm_ids(ns).ipcs_idr, next_id);
if (shp == NULL)
continue;
ipc_lock_by_ptr(&shp->shm_perm);
do_shm_rmid(ns, shp);
total++;
}
up_write(&shm_ids(ns).rw_mutex);
kfree(ns->ids[IPC_SHM_IDS]);
ns->ids[IPC_SHM_IDS] = NULL;
}
void __init shm_init (void)
{
__shm_init_ns(&init_ipc_ns, &init_shm_ids);
ipc_init_proc_interface("sysvipc/shm",
" key shmid perms size cpid lpid nattch uid gid cuid cgid atime dtime ctime\n",
IPC_SHM_IDS, sysvipc_shm_proc_show);
}
/*
* shm_lock_(check_)down routines are called in the paths where the rw_mutex
* is held to protect access to the idr tree.
*/
static inline struct shmid_kernel *shm_lock_down(struct ipc_namespace *ns,
int id)
{
struct kern_ipc_perm *ipcp = ipc_lock_down(&shm_ids(ns), id);
return container_of(ipcp, struct shmid_kernel, shm_perm);
}
static inline struct shmid_kernel *shm_lock_check_down(
struct ipc_namespace *ns,
int id)
{
struct kern_ipc_perm *ipcp = ipc_lock_check_down(&shm_ids(ns), id);
return container_of(ipcp, struct shmid_kernel, shm_perm);
}
/*
* shm_lock_(check_) routines are called in the paths where the rw_mutex
* is not held.
*/
static inline struct shmid_kernel *shm_lock(struct ipc_namespace *ns, int id)
{
struct kern_ipc_perm *ipcp = ipc_lock(&shm_ids(ns), id);
return container_of(ipcp, struct shmid_kernel, shm_perm);
}
static inline struct shmid_kernel *shm_lock_check(struct ipc_namespace *ns,
int id)
{
struct kern_ipc_perm *ipcp = ipc_lock_check(&shm_ids(ns), id);
return container_of(ipcp, struct shmid_kernel, shm_perm);
}
static inline void shm_rmid(struct ipc_namespace *ns, struct shmid_kernel *s)
{
ipc_rmid(&shm_ids(ns), &s->shm_perm);
}
static inline int shm_addid(struct ipc_namespace *ns, struct shmid_kernel *shp)
{
return ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni);
}
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
/* This is called by fork, once for every shm attach. */
static void shm_open(struct vm_area_struct *vma)
{
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct file *file = vma->vm_file;
struct shm_file_data *sfd = shm_file_data(file);
struct shmid_kernel *shp;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
shp = shm_lock(sfd->ns, sfd->id);
BUG_ON(IS_ERR(shp));
shp->shm_atim = get_seconds();
shp->shm_lprid = task_tgid_vnr(current);
shp->shm_nattch++;
shm_unlock(shp);
}
/*
* shm_destroy - free the struct shmid_kernel
*
* @ns: namespace
* @shp: struct to free
*
* It has to be called with shp and shm_ids.rw_mutex (writer) locked,
* but returns with shp unlocked and freed.
*/
static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp)
{
ns->shm_tot -= (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT;
shm_rmid(ns, shp);
shm_unlock(shp);
if (!is_file_hugepages(shp->shm_file))
shmem_lock(shp->shm_file, 0, shp->mlock_user);
else
user_shm_unlock(shp->shm_file->f_path.dentry->d_inode->i_size,
shp->mlock_user);
fput (shp->shm_file);
security_shm_free(shp);
ipc_rcu_putref(shp);
}
/*
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
* remove the attach descriptor vma.
* free memory for segment if it is marked destroyed.
* The descriptor has already been removed from the current->mm->mmap list
* and will later be kfree()d.
*/
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
static void shm_close(struct vm_area_struct *vma)
{
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct file * file = vma->vm_file;
struct shm_file_data *sfd = shm_file_data(file);
struct shmid_kernel *shp;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct ipc_namespace *ns = sfd->ns;
down_write(&shm_ids(ns).rw_mutex);
/* remove from the list of attaches of the shm segment */
shp = shm_lock_down(ns, sfd->id);
BUG_ON(IS_ERR(shp));
shp->shm_lprid = task_tgid_vnr(current);
shp->shm_dtim = get_seconds();
shp->shm_nattch--;
if(shp->shm_nattch == 0 &&
shp->shm_perm.mode & SHM_DEST)
shm_destroy(ns, shp);
else
shm_unlock(shp);
up_write(&shm_ids(ns).rw_mutex);
}
static int shm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
{
struct file *file = vma->vm_file;
struct shm_file_data *sfd = shm_file_data(file);
return sfd->vm_ops->fault(vma, vmf);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
}
#ifdef CONFIG_NUMA
static int shm_set_policy(struct vm_area_struct *vma, struct mempolicy *new)
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
{
struct file *file = vma->vm_file;
struct shm_file_data *sfd = shm_file_data(file);
int err = 0;
if (sfd->vm_ops->set_policy)
err = sfd->vm_ops->set_policy(vma, new);
return err;
}
static struct mempolicy *shm_get_policy(struct vm_area_struct *vma,
unsigned long addr)
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
{
struct file *file = vma->vm_file;
struct shm_file_data *sfd = shm_file_data(file);
struct mempolicy *pol = NULL;
if (sfd->vm_ops->get_policy)
pol = sfd->vm_ops->get_policy(vma, addr);
else if (vma->vm_policy)
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
pol = vma->vm_policy;
else
pol = current->mempolicy;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
return pol;
}
#endif
static int shm_mmap(struct file * file, struct vm_area_struct * vma)
{
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct shm_file_data *sfd = shm_file_data(file);
int ret;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
ret = sfd->file->f_op->mmap(sfd->file, vma);
if (ret != 0)
return ret;
sfd->vm_ops = vma->vm_ops;
#ifdef CONFIG_MMU
mm: merge populate and nopage into fault (fixes nonlinear) Nonlinear mappings are (AFAIKS) simply a virtual memory concept that encodes the virtual address -> file offset differently from linear mappings. ->populate is a layering violation because the filesystem/pagecache code should need to know anything about the virtual memory mapping. The hitch here is that the ->nopage handler didn't pass down enough information (ie. pgoff). But it is more logical to pass pgoff rather than have the ->nopage function calculate it itself anyway (because that's a similar layering violation). Having the populate handler install the pte itself is likewise a nasty thing to be doing. This patch introduces a new fault handler that replaces ->nopage and ->populate and (later) ->nopfn. Most of the old mechanism is still in place so there is a lot of duplication and nice cleanups that can be removed if everyone switches over. The rationale for doing this in the first place is that nonlinear mappings are subject to the pagefault vs invalidate/truncate race too, and it seemed stupid to duplicate the synchronisation logic rather than just consolidate the two. After this patch, MAP_NONBLOCK no longer sets up ptes for pages present in pagecache. Seems like a fringe functionality anyway. NOPAGE_REFAULT is removed. This should be implemented with ->fault, and no users have hit mainline yet. [akpm@linux-foundation.org: cleanup] [randy.dunlap@oracle.com: doc. fixes for readahead] [akpm@linux-foundation.org: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Mark Fasheh <mark.fasheh@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 10:46:59 +02:00
BUG_ON(!sfd->vm_ops->fault);
#endif
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
vma->vm_ops = &shm_vm_ops;
shm_open(vma);
return ret;
}
static int shm_release(struct inode *ino, struct file *file)
{
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct shm_file_data *sfd = shm_file_data(file);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
put_ipc_ns(sfd->ns);
shm_file_data(file) = NULL;
kfree(sfd);
return 0;
}
static int shm_fsync(struct file *file, struct dentry *dentry, int datasync)
{
int (*fsync) (struct file *, struct dentry *, int datasync);
struct shm_file_data *sfd = shm_file_data(file);
int ret = -EINVAL;
fsync = sfd->file->f_op->fsync;
if (fsync)
ret = fsync(sfd->file, sfd->file->f_path.dentry, datasync);
return ret;
}
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
static unsigned long shm_get_unmapped_area(struct file *file,
unsigned long addr, unsigned long len, unsigned long pgoff,
unsigned long flags)
{
struct shm_file_data *sfd = shm_file_data(file);
return get_unmapped_area(sfd->file, addr, len, pgoff, flags);
}
int is_file_shm_hugepages(struct file *file)
{
int ret = 0;
if (file->f_op == &shm_file_operations) {
struct shm_file_data *sfd;
sfd = shm_file_data(file);
ret = is_file_hugepages(sfd->file);
}
return ret;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
}
static const struct file_operations shm_file_operations = {
.mmap = shm_mmap,
.fsync = shm_fsync,
.release = shm_release,
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
.get_unmapped_area = shm_get_unmapped_area,
};
static struct vm_operations_struct shm_vm_ops = {
.open = shm_open, /* callback for a new vm-area open */
.close = shm_close, /* callback for when the vm-area is released */
mm: merge populate and nopage into fault (fixes nonlinear) Nonlinear mappings are (AFAIKS) simply a virtual memory concept that encodes the virtual address -> file offset differently from linear mappings. ->populate is a layering violation because the filesystem/pagecache code should need to know anything about the virtual memory mapping. The hitch here is that the ->nopage handler didn't pass down enough information (ie. pgoff). But it is more logical to pass pgoff rather than have the ->nopage function calculate it itself anyway (because that's a similar layering violation). Having the populate handler install the pte itself is likewise a nasty thing to be doing. This patch introduces a new fault handler that replaces ->nopage and ->populate and (later) ->nopfn. Most of the old mechanism is still in place so there is a lot of duplication and nice cleanups that can be removed if everyone switches over. The rationale for doing this in the first place is that nonlinear mappings are subject to the pagefault vs invalidate/truncate race too, and it seemed stupid to duplicate the synchronisation logic rather than just consolidate the two. After this patch, MAP_NONBLOCK no longer sets up ptes for pages present in pagecache. Seems like a fringe functionality anyway. NOPAGE_REFAULT is removed. This should be implemented with ->fault, and no users have hit mainline yet. [akpm@linux-foundation.org: cleanup] [randy.dunlap@oracle.com: doc. fixes for readahead] [akpm@linux-foundation.org: build fix] Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Mark Fasheh <mark.fasheh@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 10:46:59 +02:00
.fault = shm_fault,
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
#if defined(CONFIG_NUMA)
.set_policy = shm_set_policy,
.get_policy = shm_get_policy,
#endif
};
/**
* newseg - Create a new shared memory segment
* @ns: namespace
* @params: ptr to the structure that contains key, size and shmflg
*
* Called with shm_ids.rw_mutex held as a writer.
*/
static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
{
key_t key = params->key;
int shmflg = params->flg;
size_t size = params->u.size;
int error;
struct shmid_kernel *shp;
int numpages = (size + PAGE_SIZE -1) >> PAGE_SHIFT;
struct file * file;
char name[13];
int id;
if (size < SHMMIN || size > ns->shm_ctlmax)
return -EINVAL;
if (ns->shm_tot + numpages > ns->shm_ctlall)
return -ENOSPC;
shp = ipc_rcu_alloc(sizeof(*shp));
if (!shp)
return -ENOMEM;
shp->shm_perm.key = key;
shp->shm_perm.mode = (shmflg & S_IRWXUGO);
shp->mlock_user = NULL;
shp->shm_perm.security = NULL;
error = security_shm_alloc(shp);
if (error) {
ipc_rcu_putref(shp);
return error;
}
sprintf (name, "SYSV%08x", key);
if (shmflg & SHM_HUGETLB) {
/* hugetlb_file_setup takes care of mlock user accounting */
file = hugetlb_file_setup(name, size);
shp->mlock_user = current->user;
} else {
int acctflag = VM_ACCOUNT;
/*
* Do not allow no accounting for OVERCOMMIT_NEVER, even
* if it's asked for.
*/
if ((shmflg & SHM_NORESERVE) &&
sysctl_overcommit_memory != OVERCOMMIT_NEVER)
acctflag = 0;
file = shmem_file_setup(name, size, acctflag);
}
error = PTR_ERR(file);
if (IS_ERR(file))
goto no_file;
id = shm_addid(ns, shp);
if (id < 0) {
error = id;
goto no_id;
}
shp->shm_cprid = task_tgid_vnr(current);
shp->shm_lprid = 0;
shp->shm_atim = shp->shm_dtim = 0;
shp->shm_ctim = get_seconds();
shp->shm_segsz = size;
shp->shm_nattch = 0;
shp->shm_perm.id = shm_buildid(id, shp->shm_perm.seq);
shp->shm_file = file;
/*
* shmid gets reported as "inode#" in /proc/pid/maps.
* proc-ps tools use this. Changing this will break them.
*/
file->f_dentry->d_inode->i_ino = shp->shm_perm.id;
ns->shm_tot += numpages;
error = shp->shm_perm.id;
shm_unlock(shp);
return error;
no_id:
fput(file);
no_file:
security_shm_free(shp);
ipc_rcu_putref(shp);
return error;
}
/*
* Called with shm_ids.rw_mutex and ipcp locked.
*/
static inline int shm_security(struct kern_ipc_perm *ipcp, int shmflg)
{
struct shmid_kernel *shp;
shp = container_of(ipcp, struct shmid_kernel, shm_perm);
return security_shm_associate(shp, shmflg);
}
/*
* Called with shm_ids.rw_mutex and ipcp locked.
*/
static inline int shm_more_checks(struct kern_ipc_perm *ipcp,
struct ipc_params *params)
{
struct shmid_kernel *shp;
shp = container_of(ipcp, struct shmid_kernel, shm_perm);
if (shp->shm_segsz < params->u.size)
return -EINVAL;
return 0;
}
asmlinkage long sys_shmget (key_t key, size_t size, int shmflg)
{
struct ipc_namespace *ns;
struct ipc_ops shm_ops;
struct ipc_params shm_params;
ns = current->nsproxy->ipc_ns;
shm_ops.getnew = newseg;
shm_ops.associate = shm_security;
shm_ops.more_checks = shm_more_checks;
shm_params.key = key;
shm_params.flg = shmflg;
shm_params.u.size = size;
return ipcget(ns, &shm_ids(ns), &shm_ops, &shm_params);
}
static inline unsigned long copy_shmid_to_user(void __user *buf, struct shmid64_ds *in, int version)
{
switch(version) {
case IPC_64:
return copy_to_user(buf, in, sizeof(*in));
case IPC_OLD:
{
struct shmid_ds out;
ipc64_perm_to_ipc_perm(&in->shm_perm, &out.shm_perm);
out.shm_segsz = in->shm_segsz;
out.shm_atime = in->shm_atime;
out.shm_dtime = in->shm_dtime;
out.shm_ctime = in->shm_ctime;
out.shm_cpid = in->shm_cpid;
out.shm_lpid = in->shm_lpid;
out.shm_nattch = in->shm_nattch;
return copy_to_user(buf, &out, sizeof(out));
}
default:
return -EINVAL;
}
}
struct shm_setbuf {
uid_t uid;
gid_t gid;
mode_t mode;
};
static inline unsigned long copy_shmid_from_user(struct shm_setbuf *out, void __user *buf, int version)
{
switch(version) {
case IPC_64:
{
struct shmid64_ds tbuf;
if (copy_from_user(&tbuf, buf, sizeof(tbuf)))
return -EFAULT;
out->uid = tbuf.shm_perm.uid;
out->gid = tbuf.shm_perm.gid;
out->mode = tbuf.shm_perm.mode;
return 0;
}
case IPC_OLD:
{
struct shmid_ds tbuf_old;
if (copy_from_user(&tbuf_old, buf, sizeof(tbuf_old)))
return -EFAULT;
out->uid = tbuf_old.shm_perm.uid;
out->gid = tbuf_old.shm_perm.gid;
out->mode = tbuf_old.shm_perm.mode;
return 0;
}
default:
return -EINVAL;
}
}
static inline unsigned long copy_shminfo_to_user(void __user *buf, struct shminfo64 *in, int version)
{
switch(version) {
case IPC_64:
return copy_to_user(buf, in, sizeof(*in));
case IPC_OLD:
{
struct shminfo out;
if(in->shmmax > INT_MAX)
out.shmmax = INT_MAX;
else
out.shmmax = (int)in->shmmax;
out.shmmin = in->shmmin;
out.shmmni = in->shmmni;
out.shmseg = in->shmseg;
out.shmall = in->shmall;
return copy_to_user(buf, &out, sizeof(out));
}
default:
return -EINVAL;
}
}
/*
* Called with shm_ids.rw_mutex held as a reader
*/
static void shm_get_stat(struct ipc_namespace *ns, unsigned long *rss,
unsigned long *swp)
{
int next_id;
int total, in_use;
*rss = 0;
*swp = 0;
in_use = shm_ids(ns).in_use;
for (total = 0, next_id = 0; total < in_use; next_id++) {
struct shmid_kernel *shp;
struct inode *inode;
shp = idr_find(&shm_ids(ns).ipcs_idr, next_id);
if (shp == NULL)
continue;
inode = shp->shm_file->f_path.dentry->d_inode;
if (is_file_hugepages(shp->shm_file)) {
struct address_space *mapping = inode->i_mapping;
*rss += (HPAGE_SIZE/PAGE_SIZE)*mapping->nrpages;
} else {
struct shmem_inode_info *info = SHMEM_I(inode);
spin_lock(&info->lock);
*rss += inode->i_mapping->nrpages;
*swp += info->swapped;
spin_unlock(&info->lock);
}
total++;
}
}
asmlinkage long sys_shmctl (int shmid, int cmd, struct shmid_ds __user *buf)
{
struct shm_setbuf setbuf;
struct shmid_kernel *shp;
int err, version;
struct ipc_namespace *ns;
if (cmd < 0 || shmid < 0) {
err = -EINVAL;
goto out;
}
version = ipc_parse_version(&cmd);
ns = current->nsproxy->ipc_ns;
switch (cmd) { /* replace with proc interface ? */
case IPC_INFO:
{
struct shminfo64 shminfo;
err = security_shm_shmctl(NULL, cmd);
if (err)
return err;
memset(&shminfo,0,sizeof(shminfo));
shminfo.shmmni = shminfo.shmseg = ns->shm_ctlmni;
shminfo.shmmax = ns->shm_ctlmax;
shminfo.shmall = ns->shm_ctlall;
shminfo.shmmin = SHMMIN;
if(copy_shminfo_to_user (buf, &shminfo, version))
return -EFAULT;
down_read(&shm_ids(ns).rw_mutex);
err = ipc_get_maxid(&shm_ids(ns));
up_read(&shm_ids(ns).rw_mutex);
if(err<0)
err = 0;
goto out;
}
case SHM_INFO:
{
struct shm_info shm_info;
err = security_shm_shmctl(NULL, cmd);
if (err)
return err;
memset(&shm_info,0,sizeof(shm_info));
down_read(&shm_ids(ns).rw_mutex);
shm_info.used_ids = shm_ids(ns).in_use;
shm_get_stat (ns, &shm_info.shm_rss, &shm_info.shm_swp);
shm_info.shm_tot = ns->shm_tot;
shm_info.swap_attempts = 0;
shm_info.swap_successes = 0;
err = ipc_get_maxid(&shm_ids(ns));
up_read(&shm_ids(ns).rw_mutex);
if(copy_to_user (buf, &shm_info, sizeof(shm_info))) {
err = -EFAULT;
goto out;
}
err = err < 0 ? 0 : err;
goto out;
}
case SHM_STAT:
case IPC_STAT:
{
struct shmid64_ds tbuf;
int result;
if (!buf) {
err = -EFAULT;
goto out;
}
if (cmd == SHM_STAT) {
shp = shm_lock(ns, shmid);
if (IS_ERR(shp)) {
err = PTR_ERR(shp);
goto out;
}
result = shp->shm_perm.id;
} else {
shp = shm_lock_check(ns, shmid);
if (IS_ERR(shp)) {
err = PTR_ERR(shp);
goto out;
}
result = 0;
}
err=-EACCES;
if (ipcperms (&shp->shm_perm, S_IRUGO))
goto out_unlock;
err = security_shm_shmctl(shp, cmd);
if (err)
goto out_unlock;
memset(&tbuf, 0, sizeof(tbuf));
kernel_to_ipc64_perm(&shp->shm_perm, &tbuf.shm_perm);
tbuf.shm_segsz = shp->shm_segsz;
tbuf.shm_atime = shp->shm_atim;
tbuf.shm_dtime = shp->shm_dtim;
tbuf.shm_ctime = shp->shm_ctim;
tbuf.shm_cpid = shp->shm_cprid;
tbuf.shm_lpid = shp->shm_lprid;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
tbuf.shm_nattch = shp->shm_nattch;
shm_unlock(shp);
if(copy_shmid_to_user (buf, &tbuf, version))
err = -EFAULT;
else
err = result;
goto out;
}
case SHM_LOCK:
case SHM_UNLOCK:
{
shp = shm_lock_check(ns, shmid);
if (IS_ERR(shp)) {
err = PTR_ERR(shp);
goto out;
}
[PATCH] Rework of IPC auditing 1) The audit_ipc_perms() function has been split into two different functions: - audit_ipc_obj() - audit_ipc_set_perm() There's a key shift here... The audit_ipc_obj() collects the uid, gid, mode, and SElinux context label of the current ipc object. This audit_ipc_obj() hook is now found in several places. Most notably, it is hooked in ipcperms(), which is called in various places around the ipc code permforming a MAC check. Additionally there are several places where *checkid() is used to validate that an operation is being performed on a valid object while not necessarily having a nearby ipcperms() call. In these locations, audit_ipc_obj() is called to ensure that the information is captured by the audit system. The audit_set_new_perm() function is called any time the permissions on the ipc object changes. In this case, the NEW permissions are recorded (and note that an audit_ipc_obj() call exists just a few lines before each instance). 2) Support for an AUDIT_IPC_SET_PERM audit message type. This allows for separate auxiliary audit records for normal operations on an IPC object and permissions changes. Note that the same struct audit_aux_data_ipcctl is used and populated, however there are separate audit_log_format statements based on the type of the message. Finally, the AUDIT_IPC block of code in audit_free_aux() was extended to handle aux messages of this new type. No more mem leaks I hope ;-) Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-04-02 23:07:33 +02:00
err = audit_ipc_obj(&(shp->shm_perm));
if (err)
goto out_unlock;
if (!capable(CAP_IPC_LOCK)) {
err = -EPERM;
if (current->euid != shp->shm_perm.uid &&
current->euid != shp->shm_perm.cuid)
goto out_unlock;
if (cmd == SHM_LOCK &&
!current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur)
goto out_unlock;
}
err = security_shm_shmctl(shp, cmd);
if (err)
goto out_unlock;
if(cmd==SHM_LOCK) {
struct user_struct * user = current->user;
if (!is_file_hugepages(shp->shm_file)) {
err = shmem_lock(shp->shm_file, 1, user);
if (!err && !(shp->shm_perm.mode & SHM_LOCKED)){
shp->shm_perm.mode |= SHM_LOCKED;
shp->mlock_user = user;
}
}
} else if (!is_file_hugepages(shp->shm_file)) {
shmem_lock(shp->shm_file, 0, shp->mlock_user);
shp->shm_perm.mode &= ~SHM_LOCKED;
shp->mlock_user = NULL;
}
shm_unlock(shp);
goto out;
}
case IPC_RMID:
{
/*
* We cannot simply remove the file. The SVID states
* that the block remains until the last person
* detaches from it, then is deleted. A shmat() on
* an RMID segment is legal in older Linux and if
* we change it apps break...
*
* Instead we set a destroyed flag, and then blow
* the name away when the usage hits zero.
*/
down_write(&shm_ids(ns).rw_mutex);
shp = shm_lock_check_down(ns, shmid);
if (IS_ERR(shp)) {
err = PTR_ERR(shp);
goto out_up;
}
[PATCH] Rework of IPC auditing 1) The audit_ipc_perms() function has been split into two different functions: - audit_ipc_obj() - audit_ipc_set_perm() There's a key shift here... The audit_ipc_obj() collects the uid, gid, mode, and SElinux context label of the current ipc object. This audit_ipc_obj() hook is now found in several places. Most notably, it is hooked in ipcperms(), which is called in various places around the ipc code permforming a MAC check. Additionally there are several places where *checkid() is used to validate that an operation is being performed on a valid object while not necessarily having a nearby ipcperms() call. In these locations, audit_ipc_obj() is called to ensure that the information is captured by the audit system. The audit_set_new_perm() function is called any time the permissions on the ipc object changes. In this case, the NEW permissions are recorded (and note that an audit_ipc_obj() call exists just a few lines before each instance). 2) Support for an AUDIT_IPC_SET_PERM audit message type. This allows for separate auxiliary audit records for normal operations on an IPC object and permissions changes. Note that the same struct audit_aux_data_ipcctl is used and populated, however there are separate audit_log_format statements based on the type of the message. Finally, the AUDIT_IPC block of code in audit_free_aux() was extended to handle aux messages of this new type. No more mem leaks I hope ;-) Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-04-02 23:07:33 +02:00
err = audit_ipc_obj(&(shp->shm_perm));
if (err)
goto out_unlock_up;
if (current->euid != shp->shm_perm.uid &&
current->euid != shp->shm_perm.cuid &&
!capable(CAP_SYS_ADMIN)) {
err=-EPERM;
goto out_unlock_up;
}
err = security_shm_shmctl(shp, cmd);
if (err)
goto out_unlock_up;
do_shm_rmid(ns, shp);
up_write(&shm_ids(ns).rw_mutex);
goto out;
}
case IPC_SET:
{
if (!buf) {
err = -EFAULT;
goto out;
}
if (copy_shmid_from_user (&setbuf, buf, version)) {
err = -EFAULT;
goto out;
}
down_write(&shm_ids(ns).rw_mutex);
shp = shm_lock_check_down(ns, shmid);
if (IS_ERR(shp)) {
err = PTR_ERR(shp);
goto out_up;
}
[PATCH] Rework of IPC auditing 1) The audit_ipc_perms() function has been split into two different functions: - audit_ipc_obj() - audit_ipc_set_perm() There's a key shift here... The audit_ipc_obj() collects the uid, gid, mode, and SElinux context label of the current ipc object. This audit_ipc_obj() hook is now found in several places. Most notably, it is hooked in ipcperms(), which is called in various places around the ipc code permforming a MAC check. Additionally there are several places where *checkid() is used to validate that an operation is being performed on a valid object while not necessarily having a nearby ipcperms() call. In these locations, audit_ipc_obj() is called to ensure that the information is captured by the audit system. The audit_set_new_perm() function is called any time the permissions on the ipc object changes. In this case, the NEW permissions are recorded (and note that an audit_ipc_obj() call exists just a few lines before each instance). 2) Support for an AUDIT_IPC_SET_PERM audit message type. This allows for separate auxiliary audit records for normal operations on an IPC object and permissions changes. Note that the same struct audit_aux_data_ipcctl is used and populated, however there are separate audit_log_format statements based on the type of the message. Finally, the AUDIT_IPC block of code in audit_free_aux() was extended to handle aux messages of this new type. No more mem leaks I hope ;-) Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-04-02 23:07:33 +02:00
err = audit_ipc_obj(&(shp->shm_perm));
if (err)
goto out_unlock_up;
[PATCH] update of IPC audit record cleanup The following patch addresses most of the issues with the IPC_SET_PERM records as described in: https://www.redhat.com/archives/linux-audit/2006-May/msg00010.html and addresses the comments I received on the record field names. To summarize, I made the following changes: 1. Changed sys_msgctl() and semctl_down() so that an IPC_SET_PERM record is emitted in the failure case as well as the success case. This matches the behavior in sys_shmctl(). I could simplify the code in sys_msgctl() and semctl_down() slightly but it would mean that in some error cases we could get an IPC_SET_PERM record without an IPC record and that seemed odd. 2. No change to the IPC record type, given no feedback on the backward compatibility question. 3. Removed the qbytes field from the IPC record. It wasn't being set and when audit_ipc_obj() is called from ipcperms(), the information isn't available. If we want the information in the IPC record, more extensive changes will be necessary. Since it only applies to message queues and it isn't really permission related, it doesn't seem worth it. 4. Removed the obj field from the IPC_SET_PERM record. This means that the kern_ipc_perm argument is no longer needed. 5. Removed the spaces and renamed the IPC_SET_PERM field names. Replaced iuid and igid fields with ouid and ogid in the IPC record. I tested this with the lspp.22 kernel on an x86_64 box. I believe it applies cleanly on the latest kernel. -- ljk Signed-off-by: Linda Knippers <linda.knippers@hp.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-05-17 04:03:48 +02:00
err = audit_ipc_set_perm(0, setbuf.uid, setbuf.gid, setbuf.mode);
[PATCH] Rework of IPC auditing 1) The audit_ipc_perms() function has been split into two different functions: - audit_ipc_obj() - audit_ipc_set_perm() There's a key shift here... The audit_ipc_obj() collects the uid, gid, mode, and SElinux context label of the current ipc object. This audit_ipc_obj() hook is now found in several places. Most notably, it is hooked in ipcperms(), which is called in various places around the ipc code permforming a MAC check. Additionally there are several places where *checkid() is used to validate that an operation is being performed on a valid object while not necessarily having a nearby ipcperms() call. In these locations, audit_ipc_obj() is called to ensure that the information is captured by the audit system. The audit_set_new_perm() function is called any time the permissions on the ipc object changes. In this case, the NEW permissions are recorded (and note that an audit_ipc_obj() call exists just a few lines before each instance). 2) Support for an AUDIT_IPC_SET_PERM audit message type. This allows for separate auxiliary audit records for normal operations on an IPC object and permissions changes. Note that the same struct audit_aux_data_ipcctl is used and populated, however there are separate audit_log_format statements based on the type of the message. Finally, the AUDIT_IPC block of code in audit_free_aux() was extended to handle aux messages of this new type. No more mem leaks I hope ;-) Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-04-02 23:07:33 +02:00
if (err)
goto out_unlock_up;
err=-EPERM;
if (current->euid != shp->shm_perm.uid &&
current->euid != shp->shm_perm.cuid &&
!capable(CAP_SYS_ADMIN)) {
goto out_unlock_up;
}
err = security_shm_shmctl(shp, cmd);
if (err)
goto out_unlock_up;
shp->shm_perm.uid = setbuf.uid;
shp->shm_perm.gid = setbuf.gid;
shp->shm_perm.mode = (shp->shm_perm.mode & ~S_IRWXUGO)
| (setbuf.mode & S_IRWXUGO);
shp->shm_ctim = get_seconds();
break;
}
default:
err = -EINVAL;
goto out;
}
err = 0;
out_unlock_up:
shm_unlock(shp);
out_up:
up_write(&shm_ids(ns).rw_mutex);
goto out;
out_unlock:
shm_unlock(shp);
out:
return err;
}
/*
* Fix shmaddr, allocate descriptor, map shm, add attach descriptor to lists.
*
* NOTE! Despite the name, this is NOT a direct system call entrypoint. The
* "raddr" thing points to kernel space, and there has to be a wrapper around
* this.
*/
long do_shmat(int shmid, char __user *shmaddr, int shmflg, ulong *raddr)
{
struct shmid_kernel *shp;
unsigned long addr;
unsigned long size;
struct file * file;
int err;
unsigned long flags;
unsigned long prot;
int acc_mode;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
unsigned long user_addr;
struct ipc_namespace *ns;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
struct shm_file_data *sfd;
struct path path;
mode_t f_mode;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
err = -EINVAL;
if (shmid < 0)
goto out;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
else if ((addr = (ulong)shmaddr)) {
if (addr & (SHMLBA-1)) {
if (shmflg & SHM_RND)
addr &= ~(SHMLBA-1); /* round down */
else
#ifndef __ARCH_FORCE_SHMLBA
if (addr & ~PAGE_MASK)
#endif
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
goto out;
}
flags = MAP_SHARED | MAP_FIXED;
} else {
if ((shmflg & SHM_REMAP))
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
goto out;
flags = MAP_SHARED;
}
if (shmflg & SHM_RDONLY) {
prot = PROT_READ;
acc_mode = S_IRUGO;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
f_mode = FMODE_READ;
} else {
prot = PROT_READ | PROT_WRITE;
acc_mode = S_IRUGO | S_IWUGO;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
f_mode = FMODE_READ | FMODE_WRITE;
}
if (shmflg & SHM_EXEC) {
prot |= PROT_EXEC;
acc_mode |= S_IXUGO;
}
/*
* We cannot rely on the fs check since SYSV IPC does have an
* additional creator id...
*/
ns = current->nsproxy->ipc_ns;
shp = shm_lock_check(ns, shmid);
if (IS_ERR(shp)) {
err = PTR_ERR(shp);
goto out;
}
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
err = -EACCES;
if (ipcperms(&shp->shm_perm, acc_mode))
goto out_unlock;
err = security_shm_shmat(shp, shmaddr, shmflg);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
if (err)
goto out_unlock;
path.dentry = dget(shp->shm_file->f_path.dentry);
r/o bind mounts: filesystem helpers for custom 'struct file's Why do we need r/o bind mounts? This feature allows a read-only view into a read-write filesystem. In the process of doing that, it also provides infrastructure for keeping track of the number of writers to any given mount. This has a number of uses. It allows chroots to have parts of filesystems writable. It will be useful for containers in the future because users may have root inside a container, but should not be allowed to write to somefilesystems. This also replaces patches that vserver has had out of the tree for several years. It allows security enhancement by making sure that parts of your filesystem read-only (such as when you don't trust your FTP server), when you don't want to have entire new filesystems mounted, or when you want atime selectively updated. I've been using the following script to test that the feature is working as desired. It takes a directory and makes a regular bind and a r/o bind mount of it. It then performs some normal filesystem operations on the three directories, including ones that are expected to fail, like creating a file on the r/o mount. This patch: Some filesystems forego the vfs and may_open() and create their own 'struct file's. This patch creates a couple of helper functions which can be used by these filesystems, and will provide a unified place which the r/o bind mount code may patch. Also, rename an existing, static-scope init_file() to a less generic name. Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 08:31:13 +02:00
path.mnt = shp->shm_file->f_path.mnt;
shp->shm_nattch++;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
size = i_size_read(path.dentry->d_inode);
shm_unlock(shp);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
err = -ENOMEM;
sfd = kzalloc(sizeof(*sfd), GFP_KERNEL);
if (!sfd)
r/o bind mounts: filesystem helpers for custom 'struct file's Why do we need r/o bind mounts? This feature allows a read-only view into a read-write filesystem. In the process of doing that, it also provides infrastructure for keeping track of the number of writers to any given mount. This has a number of uses. It allows chroots to have parts of filesystems writable. It will be useful for containers in the future because users may have root inside a container, but should not be allowed to write to somefilesystems. This also replaces patches that vserver has had out of the tree for several years. It allows security enhancement by making sure that parts of your filesystem read-only (such as when you don't trust your FTP server), when you don't want to have entire new filesystems mounted, or when you want atime selectively updated. I've been using the following script to test that the feature is working as desired. It takes a directory and makes a regular bind and a r/o bind mount of it. It then performs some normal filesystem operations on the three directories, including ones that are expected to fail, like creating a file on the r/o mount. This patch: Some filesystems forego the vfs and may_open() and create their own 'struct file's. This patch creates a couple of helper functions which can be used by these filesystems, and will provide a unified place which the r/o bind mount code may patch. Also, rename an existing, static-scope init_file() to a less generic name. Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 08:31:13 +02:00
goto out_put_dentry;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
err = -ENOMEM;
r/o bind mounts: filesystem helpers for custom 'struct file's Why do we need r/o bind mounts? This feature allows a read-only view into a read-write filesystem. In the process of doing that, it also provides infrastructure for keeping track of the number of writers to any given mount. This has a number of uses. It allows chroots to have parts of filesystems writable. It will be useful for containers in the future because users may have root inside a container, but should not be allowed to write to somefilesystems. This also replaces patches that vserver has had out of the tree for several years. It allows security enhancement by making sure that parts of your filesystem read-only (such as when you don't trust your FTP server), when you don't want to have entire new filesystems mounted, or when you want atime selectively updated. I've been using the following script to test that the feature is working as desired. It takes a directory and makes a regular bind and a r/o bind mount of it. It then performs some normal filesystem operations on the three directories, including ones that are expected to fail, like creating a file on the r/o mount. This patch: Some filesystems forego the vfs and may_open() and create their own 'struct file's. This patch creates a couple of helper functions which can be used by these filesystems, and will provide a unified place which the r/o bind mount code may patch. Also, rename an existing, static-scope init_file() to a less generic name. Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 08:31:13 +02:00
file = alloc_file(path.mnt, path.dentry, f_mode, &shm_file_operations);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
if (!file)
goto out_free;
file->private_data = sfd;
file->f_mapping = shp->shm_file->f_mapping;
sfd->id = shp->shm_perm.id;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
sfd->ns = get_ipc_ns(ns);
sfd->file = shp->shm_file;
sfd->vm_ops = NULL;
down_write(&current->mm->mmap_sem);
if (addr && !(shmflg & SHM_REMAP)) {
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
err = -EINVAL;
if (find_vma_intersection(current->mm, addr, addr + size))
goto invalid;
/*
* If shm segment goes below stack, make sure there is some
* space left for the stack to grow (at least 4 pages).
*/
if (addr < current->mm->start_stack &&
addr > current->mm->start_stack - size - PAGE_SIZE * 5)
goto invalid;
}
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
user_addr = do_mmap (file, addr, size, prot, flags, 0);
*raddr = user_addr;
err = 0;
if (IS_ERR_VALUE(user_addr))
err = (long)user_addr;
invalid:
up_write(&current->mm->mmap_sem);
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
fput(file);
out_nattch:
down_write(&shm_ids(ns).rw_mutex);
shp = shm_lock_down(ns, shmid);
BUG_ON(IS_ERR(shp));
shp->shm_nattch--;
if(shp->shm_nattch == 0 &&
shp->shm_perm.mode & SHM_DEST)
shm_destroy(ns, shp);
else
shm_unlock(shp);
up_write(&shm_ids(ns).rw_mutex);
out:
return err;
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
out_unlock:
shm_unlock(shp);
goto out;
out_free:
kfree(sfd);
r/o bind mounts: filesystem helpers for custom 'struct file's Why do we need r/o bind mounts? This feature allows a read-only view into a read-write filesystem. In the process of doing that, it also provides infrastructure for keeping track of the number of writers to any given mount. This has a number of uses. It allows chroots to have parts of filesystems writable. It will be useful for containers in the future because users may have root inside a container, but should not be allowed to write to somefilesystems. This also replaces patches that vserver has had out of the tree for several years. It allows security enhancement by making sure that parts of your filesystem read-only (such as when you don't trust your FTP server), when you don't want to have entire new filesystems mounted, or when you want atime selectively updated. I've been using the following script to test that the feature is working as desired. It takes a directory and makes a regular bind and a r/o bind mount of it. It then performs some normal filesystem operations on the three directories, including ones that are expected to fail, like creating a file on the r/o mount. This patch: Some filesystems forego the vfs and may_open() and create their own 'struct file's. This patch creates a couple of helper functions which can be used by these filesystems, and will provide a unified place which the r/o bind mount code may patch. Also, rename an existing, static-scope init_file() to a less generic name. Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 08:31:13 +02:00
out_put_dentry:
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
dput(path.dentry);
goto out_nattch;
}
asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg)
{
unsigned long ret;
long err;
err = do_shmat(shmid, shmaddr, shmflg, &ret);
if (err)
return err;
force_successful_syscall_return();
return (long)ret;
}
/*
* detach and kill segment if marked destroyed.
* The work is done in shm_close.
*/
asmlinkage long sys_shmdt(char __user *shmaddr)
{
struct mm_struct *mm = current->mm;
struct vm_area_struct *vma, *next;
unsigned long addr = (unsigned long)shmaddr;
loff_t size = 0;
int retval = -EINVAL;
if (addr & ~PAGE_MASK)
return retval;
down_write(&mm->mmap_sem);
/*
* This function tries to be smart and unmap shm segments that
* were modified by partial mlock or munmap calls:
* - It first determines the size of the shm segment that should be
* unmapped: It searches for a vma that is backed by shm and that
* started at address shmaddr. It records it's size and then unmaps
* it.
* - Then it unmaps all shm vmas that started at shmaddr and that
* are within the initially determined size.
* Errors from do_munmap are ignored: the function only fails if
* it's called with invalid parameters or if it's called to unmap
* a part of a vma. Both calls in this function are for full vmas,
* the parameters are directly copied from the vma itself and always
* valid - therefore do_munmap cannot fail. (famous last words?)
*/
/*
* If it had been mremap()'d, the starting address would not
* match the usual checks anyway. So assume all vma's are
* above the starting address given.
*/
vma = find_vma(mm, addr);
while (vma) {
next = vma->vm_next;
/*
* Check if the starting address would match, i.e. it's
* a fragment created by mprotect() and/or munmap(), or it
* otherwise it starts at this address with no hassles.
*/
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
if ((vma->vm_ops == &shm_vm_ops) &&
(vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff) {
size = vma->vm_file->f_path.dentry->d_inode->i_size;
do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start);
/*
* We discovered the size of the shm segment, so
* break out of here and fall through to the next
* loop that uses the size information to stop
* searching for matching vma's.
*/
retval = 0;
vma = next;
break;
}
vma = next;
}
/*
* We need look no further than the maximum address a fragment
* could possibly have landed at. Also cast things to loff_t to
* prevent overflows and make comparisions vs. equal-width types.
*/
size = PAGE_ALIGN(size);
while (vma && (loff_t)(vma->vm_end - addr) <= size) {
next = vma->vm_next;
/* finding a matching vma now does not alter retval */
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
if ((vma->vm_ops == &shm_vm_ops) &&
(vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff)
do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start);
vma = next;
}
up_write(&mm->mmap_sem);
return retval;
}
#ifdef CONFIG_PROC_FS
static int sysvipc_shm_proc_show(struct seq_file *s, void *it)
{
struct shmid_kernel *shp = it;
char *format;
#define SMALL_STRING "%10d %10d %4o %10u %5u %5u %5d %5u %5u %5u %5u %10lu %10lu %10lu\n"
#define BIG_STRING "%10d %10d %4o %21u %5u %5u %5d %5u %5u %5u %5u %10lu %10lu %10lu\n"
if (sizeof(size_t) <= sizeof(int))
format = SMALL_STRING;
else
format = BIG_STRING;
return seq_printf(s, format,
shp->shm_perm.key,
shp->shm_perm.id,
shp->shm_perm.mode,
shp->shm_segsz,
shp->shm_cprid,
shp->shm_lprid,
[PATCH] shm: make sysv ipc shared memory use stacked files The current ipc shared memory code runs into several problems because it does not quite use files like the rest of the kernel. With the option of backing ipc shared memory with either hugetlbfs or ordinary shared memory the problems got worse. With the added support for ipc namespaces things behaved so unexpected that we now have several bad namespace reference counting bugs when using what appears at first glance to be a reasonable idiom. So to attack these problems and hopefully make the code more maintainable this patch simply uses the files provided by other parts of the kernel and builds it's own files out of them. The shm files are allocated in do_shmat and freed when their reference count drops to zero with their last unmap. The file and vm operations that we don't want to implement or we don't implement completely we just delegate to the operations of our backing file. This means that we now get an accurate shm_nattch count for we have a hugetlbfs inode for backing store, and the shm accounting of last attach and last detach time work as well. This means that getting a reference to the ipc namespace when we create the file and dropping the referenece in the release method is now safe and correct. This means we no longer need a special case for clearing VM_MAYWRITE as our file descriptor now only has write permissions when we have requested write access when calling shmat. Although VM_SHARED is now cleared as well which I believe is harmless and is mostly likely a minor bug fix. By using the same set of operations for both the hugetlb case and regular shared memory case shmdt is not simplified and made slightly more correct as now the test "vma->vm_ops == &shm_vm_ops" is 100% accurate in spotting all shared memory regions generated from sysvipc shared memory. Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-20 22:57:53 +01:00
shp->shm_nattch,
shp->shm_perm.uid,
shp->shm_perm.gid,
shp->shm_perm.cuid,
shp->shm_perm.cgid,
shp->shm_atim,
shp->shm_dtim,
shp->shm_ctim);
}
#endif