linux/kernel/power/user.c
Rafael J. Wysocki 181e9bdef3 PM / Hibernate: Fix s2disk regression related to freezing workqueues
Commit 2aede851dd

  PM / Hibernate: Freeze kernel threads after preallocating memory

introduced a mechanism by which kernel threads were frozen after
the preallocation of hibernate image memory to avoid problems with
frozen kernel threads not responding to memory freeing requests.
However, it overlooked the s2disk code path in which the
SNAPSHOT_CREATE_IMAGE ioctl was run directly after SNAPSHOT_FREE,
which caused freeze_workqueues_begin() to BUG(), because it saw
that worqueues had been already frozen.

Although in principle this issue might be addressed by removing
the relevant BUG_ON() from freeze_workqueues_begin(), that would
reintroduce the very problem that commit 2aede851dd
attempted to avoid into that particular code path.  For this reason,
to fix the issue at hand, introduce thaw_kernel_threads() and make
the SNAPSHOT_FREE ioctl execute it.

Special thanks to Srivatsa S. Bhat for detailed analysis of the
problem.

Reported-and-tested-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: stable@kernel.org
2012-01-29 20:35:52 +01:00

477 lines
9.9 KiB
C

/*
* linux/kernel/power/user.c
*
* This file provides the user space interface for software suspend/resume.
*
* Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
*
* This file is released under the GPLv2.
*
*/
#include <linux/suspend.h>
#include <linux/syscalls.h>
#include <linux/reboot.h>
#include <linux/kmod.h>
#include <linux/string.h>
#include <linux/device.h>
#include <linux/miscdevice.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/swapops.h>
#include <linux/pm.h>
#include <linux/fs.h>
#include <linux/compat.h>
#include <linux/console.h>
#include <linux/cpu.h>
#include <linux/freezer.h>
#include <scsi/scsi_scan.h>
#include <asm/uaccess.h>
#include "power.h"
#define SNAPSHOT_MINOR 231
static struct snapshot_data {
struct snapshot_handle handle;
int swap;
int mode;
char frozen;
char ready;
char platform_support;
} snapshot_state;
atomic_t snapshot_device_available = ATOMIC_INIT(1);
static int snapshot_open(struct inode *inode, struct file *filp)
{
struct snapshot_data *data;
int error;
lock_system_sleep();
if (!atomic_add_unless(&snapshot_device_available, -1, 0)) {
error = -EBUSY;
goto Unlock;
}
if ((filp->f_flags & O_ACCMODE) == O_RDWR) {
atomic_inc(&snapshot_device_available);
error = -ENOSYS;
goto Unlock;
}
if(create_basic_memory_bitmaps()) {
atomic_inc(&snapshot_device_available);
error = -ENOMEM;
goto Unlock;
}
nonseekable_open(inode, filp);
data = &snapshot_state;
filp->private_data = data;
memset(&data->handle, 0, sizeof(struct snapshot_handle));
if ((filp->f_flags & O_ACCMODE) == O_RDONLY) {
/* Hibernating. The image device should be accessible. */
data->swap = swsusp_resume_device ?
swap_type_of(swsusp_resume_device, 0, NULL) : -1;
data->mode = O_RDONLY;
error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE);
if (error)
pm_notifier_call_chain(PM_POST_HIBERNATION);
} else {
/*
* Resuming. We may need to wait for the image device to
* appear.
*/
wait_for_device_probe();
scsi_complete_async_scans();
data->swap = -1;
data->mode = O_WRONLY;
error = pm_notifier_call_chain(PM_RESTORE_PREPARE);
if (error)
pm_notifier_call_chain(PM_POST_RESTORE);
}
if (error) {
free_basic_memory_bitmaps();
atomic_inc(&snapshot_device_available);
}
data->frozen = 0;
data->ready = 0;
data->platform_support = 0;
Unlock:
unlock_system_sleep();
return error;
}
static int snapshot_release(struct inode *inode, struct file *filp)
{
struct snapshot_data *data;
lock_system_sleep();
swsusp_free();
free_basic_memory_bitmaps();
data = filp->private_data;
free_all_swap_pages(data->swap);
if (data->frozen) {
pm_restore_gfp_mask();
thaw_processes();
}
pm_notifier_call_chain(data->mode == O_RDONLY ?
PM_POST_HIBERNATION : PM_POST_RESTORE);
atomic_inc(&snapshot_device_available);
unlock_system_sleep();
return 0;
}
static ssize_t snapshot_read(struct file *filp, char __user *buf,
size_t count, loff_t *offp)
{
struct snapshot_data *data;
ssize_t res;
loff_t pg_offp = *offp & ~PAGE_MASK;
lock_system_sleep();
data = filp->private_data;
if (!data->ready) {
res = -ENODATA;
goto Unlock;
}
if (!pg_offp) { /* on page boundary? */
res = snapshot_read_next(&data->handle);
if (res <= 0)
goto Unlock;
} else {
res = PAGE_SIZE - pg_offp;
}
res = simple_read_from_buffer(buf, count, &pg_offp,
data_of(data->handle), res);
if (res > 0)
*offp += res;
Unlock:
unlock_system_sleep();
return res;
}
static ssize_t snapshot_write(struct file *filp, const char __user *buf,
size_t count, loff_t *offp)
{
struct snapshot_data *data;
ssize_t res;
loff_t pg_offp = *offp & ~PAGE_MASK;
lock_system_sleep();
data = filp->private_data;
if (!pg_offp) {
res = snapshot_write_next(&data->handle);
if (res <= 0)
goto unlock;
} else {
res = PAGE_SIZE - pg_offp;
}
res = simple_write_to_buffer(data_of(data->handle), res, &pg_offp,
buf, count);
if (res > 0)
*offp += res;
unlock:
unlock_system_sleep();
return res;
}
static long snapshot_ioctl(struct file *filp, unsigned int cmd,
unsigned long arg)
{
int error = 0;
struct snapshot_data *data;
loff_t size;
sector_t offset;
if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC)
return -ENOTTY;
if (_IOC_NR(cmd) > SNAPSHOT_IOC_MAXNR)
return -ENOTTY;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (!mutex_trylock(&pm_mutex))
return -EBUSY;
data = filp->private_data;
switch (cmd) {
case SNAPSHOT_FREEZE:
if (data->frozen)
break;
printk("Syncing filesystems ... ");
sys_sync();
printk("done.\n");
error = usermodehelper_disable();
if (error)
break;
error = freeze_processes();
if (error)
usermodehelper_enable();
else
data->frozen = 1;
break;
case SNAPSHOT_UNFREEZE:
if (!data->frozen || data->ready)
break;
pm_restore_gfp_mask();
thaw_processes();
usermodehelper_enable();
data->frozen = 0;
break;
case SNAPSHOT_CREATE_IMAGE:
if (data->mode != O_RDONLY || !data->frozen || data->ready) {
error = -EPERM;
break;
}
pm_restore_gfp_mask();
error = hibernation_snapshot(data->platform_support);
if (!error) {
error = put_user(in_suspend, (int __user *)arg);
if (!error && !freezer_test_done)
data->ready = 1;
if (freezer_test_done) {
freezer_test_done = false;
thaw_processes();
}
}
break;
case SNAPSHOT_ATOMIC_RESTORE:
snapshot_write_finalize(&data->handle);
if (data->mode != O_WRONLY || !data->frozen ||
!snapshot_image_loaded(&data->handle)) {
error = -EPERM;
break;
}
error = hibernation_restore(data->platform_support);
break;
case SNAPSHOT_FREE:
swsusp_free();
memset(&data->handle, 0, sizeof(struct snapshot_handle));
data->ready = 0;
/*
* It is necessary to thaw kernel threads here, because
* SNAPSHOT_CREATE_IMAGE may be invoked directly after
* SNAPSHOT_FREE. In that case, if kernel threads were not
* thawed, the preallocation of memory carried out by
* hibernation_snapshot() might run into problems (i.e. it
* might fail or even deadlock).
*/
thaw_kernel_threads();
break;
case SNAPSHOT_PREF_IMAGE_SIZE:
image_size = arg;
break;
case SNAPSHOT_GET_IMAGE_SIZE:
if (!data->ready) {
error = -ENODATA;
break;
}
size = snapshot_get_image_size();
size <<= PAGE_SHIFT;
error = put_user(size, (loff_t __user *)arg);
break;
case SNAPSHOT_AVAIL_SWAP_SIZE:
size = count_swap_pages(data->swap, 1);
size <<= PAGE_SHIFT;
error = put_user(size, (loff_t __user *)arg);
break;
case SNAPSHOT_ALLOC_SWAP_PAGE:
if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
error = -ENODEV;
break;
}
offset = alloc_swapdev_block(data->swap);
if (offset) {
offset <<= PAGE_SHIFT;
error = put_user(offset, (loff_t __user *)arg);
} else {
error = -ENOSPC;
}
break;
case SNAPSHOT_FREE_SWAP_PAGES:
if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
error = -ENODEV;
break;
}
free_all_swap_pages(data->swap);
break;
case SNAPSHOT_S2RAM:
if (!data->frozen) {
error = -EPERM;
break;
}
/*
* Tasks are frozen and the notifiers have been called with
* PM_HIBERNATION_PREPARE
*/
error = suspend_devices_and_enter(PM_SUSPEND_MEM);
data->ready = 0;
break;
case SNAPSHOT_PLATFORM_SUPPORT:
data->platform_support = !!arg;
break;
case SNAPSHOT_POWER_OFF:
if (data->platform_support)
error = hibernation_platform_enter();
break;
case SNAPSHOT_SET_SWAP_AREA:
if (swsusp_swap_in_use()) {
error = -EPERM;
} else {
struct resume_swap_area swap_area;
dev_t swdev;
error = copy_from_user(&swap_area, (void __user *)arg,
sizeof(struct resume_swap_area));
if (error) {
error = -EFAULT;
break;
}
/*
* User space encodes device types as two-byte values,
* so we need to recode them
*/
swdev = new_decode_dev(swap_area.dev);
if (swdev) {
offset = swap_area.offset;
data->swap = swap_type_of(swdev, offset, NULL);
if (data->swap < 0)
error = -ENODEV;
} else {
data->swap = -1;
error = -EINVAL;
}
}
break;
default:
error = -ENOTTY;
}
mutex_unlock(&pm_mutex);
return error;
}
#ifdef CONFIG_COMPAT
struct compat_resume_swap_area {
compat_loff_t offset;
u32 dev;
} __packed;
static long
snapshot_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
BUILD_BUG_ON(sizeof(loff_t) != sizeof(compat_loff_t));
switch (cmd) {
case SNAPSHOT_GET_IMAGE_SIZE:
case SNAPSHOT_AVAIL_SWAP_SIZE:
case SNAPSHOT_ALLOC_SWAP_PAGE: {
compat_loff_t __user *uoffset = compat_ptr(arg);
loff_t offset;
mm_segment_t old_fs;
int err;
old_fs = get_fs();
set_fs(KERNEL_DS);
err = snapshot_ioctl(file, cmd, (unsigned long) &offset);
set_fs(old_fs);
if (!err && put_user(offset, uoffset))
err = -EFAULT;
return err;
}
case SNAPSHOT_CREATE_IMAGE:
return snapshot_ioctl(file, cmd,
(unsigned long) compat_ptr(arg));
case SNAPSHOT_SET_SWAP_AREA: {
struct compat_resume_swap_area __user *u_swap_area =
compat_ptr(arg);
struct resume_swap_area swap_area;
mm_segment_t old_fs;
int err;
err = get_user(swap_area.offset, &u_swap_area->offset);
err |= get_user(swap_area.dev, &u_swap_area->dev);
if (err)
return -EFAULT;
old_fs = get_fs();
set_fs(KERNEL_DS);
err = snapshot_ioctl(file, SNAPSHOT_SET_SWAP_AREA,
(unsigned long) &swap_area);
set_fs(old_fs);
return err;
}
default:
return snapshot_ioctl(file, cmd, arg);
}
}
#endif /* CONFIG_COMPAT */
static const struct file_operations snapshot_fops = {
.open = snapshot_open,
.release = snapshot_release,
.read = snapshot_read,
.write = snapshot_write,
.llseek = no_llseek,
.unlocked_ioctl = snapshot_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = snapshot_compat_ioctl,
#endif
};
static struct miscdevice snapshot_device = {
.minor = SNAPSHOT_MINOR,
.name = "snapshot",
.fops = &snapshot_fops,
};
static int __init snapshot_device_init(void)
{
return misc_register(&snapshot_device);
};
device_initcall(snapshot_device_init);