linux/kernel/power/swap.c

1512 lines
36 KiB
C
Raw Normal View History

/*
* linux/kernel/power/swap.c
*
* This file provides functions for reading the suspend image from
* and writing it to a swap partition.
*
* Copyright (C) 1998,2001-2005 Pavel Machek <pavel@ucw.cz>
* Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
* Copyright (C) 2010-2012 Bojan Smojver <bojan@rexursive.com>
*
* This file is released under the GPLv2.
*
*/
#include <linux/module.h>
#include <linux/file.h>
#include <linux/delay.h>
#include <linux/bitops.h>
#include <linux/genhd.h>
#include <linux/device.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/swap.h>
#include <linux/swapops.h>
#include <linux/pm.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <linux/lzo.h>
#include <linux/vmalloc.h>
#include <linux/cpumask.h>
#include <linux/atomic.h>
#include <linux/kthread.h>
#include <linux/crc32.h>
#include "power.h"
#define HIBERNATE_SIG "S1SUSPEND"
/*
* The swap map is a data structure used for keeping track of each page
* written to a swap partition. It consists of many swap_map_page
* structures that contain each an array of MAP_PAGE_ENTRIES swap entries.
* These structures are stored on the swap and linked together with the
* help of the .next_swap member.
*
* The swap map is created during suspend. The swap map pages are
* allocated and populated one at a time, so we only need one memory
* page to set up the entire structure.
*
* During resume we pick up all swap_map_page structures into a list.
*/
#define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(sector_t) - 1)
/*
* Number of free pages that are not high.
*/
static inline unsigned long low_free_pages(void)
{
return nr_free_pages() - nr_free_highpages();
}
/*
* Number of pages required to be kept free while writing the image. Always
* half of all available low pages before the writing starts.
*/
static inline unsigned long reqd_free_pages(void)
{
return low_free_pages() / 2;
}
struct swap_map_page {
sector_t entries[MAP_PAGE_ENTRIES];
sector_t next_swap;
};
struct swap_map_page_list {
struct swap_map_page *map;
struct swap_map_page_list *next;
};
/**
* The swap_map_handle structure is used for handling swap in
* a file-alike way
*/
struct swap_map_handle {
struct swap_map_page *cur;
struct swap_map_page_list *maps;
sector_t cur_swap;
sector_t first_sector;
unsigned int k;
unsigned long reqd_free_pages;
u32 crc32;
};
struct swsusp_header {
char reserved[PAGE_SIZE - 20 - sizeof(sector_t) - sizeof(int) -
sizeof(u32)];
u32 crc32;
sector_t image;
swsusp: introduce restore platform operations At least on some machines it is necessary to prepare the ACPI firmware for the restoration of the system memory state from the hibernation image if the "platform" mode of hibernation has been used. Namely, in that cases we need to disable the GPEs before replacing the "boot" kernel with the "frozen" kernel (cf. http://bugzilla.kernel.org/show_bug.cgi?id=7887). After the restore they will be re-enabled by hibernation_ops->finish(), but if the restore fails, they have to be re-enabled by the restore code explicitly. For this purpose we can introduce two additional hibernation operations, called pre_restore() and restore_cleanup() and call them from the restore code path. Still, they should be called if the "platform" mode of hibernation has been used, so we need to pass the information about the hibernation mode from the "frozen" kernel to the "boot" kernel in the image header. Apparently, we can't drop the disabling of GPEs before the restore because of Bug #7887 .  We also can't do it unconditionally, because the GPEs wouldn't have been enabled after a successful restore if the suspend had been done in the 'shutdown' or 'reboot' mode. In principle we could (and probably should) unconditionally disable the GPEs before each snapshot creation *and* before the restore, but then we'd have to unconditionally enable them after the snapshot creation as well as after the restore (or restore failure)   Still, for this purpose we'd need to modify acpi_enter_sleep_state_prep() and acpi_leave_sleep_state() and we'd have to introduce some mechanism synchronizing the disablind/enabling of the GPEs with the device drivers' .suspend()/.resume() routines and with disable_/enable_nonboot_cpus().  However, this would have affected the suspend (ie. s2ram) code as well as the hibernation, which I'd like to avoid in this patch series. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Nigel Cunningham <nigel@nigel.suspend2.net> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:47:30 +00:00
unsigned int flags; /* Flags to pass to the "boot" kernel */
char orig_sig[10];
char sig[10];
} __attribute__((packed));
static struct swsusp_header *swsusp_header;
/**
* The following functions are used for tracing the allocated
* swap pages, so that they can be freed in case of an error.
*/
struct swsusp_extent {
struct rb_node node;
unsigned long start;
unsigned long end;
};
static struct rb_root swsusp_extents = RB_ROOT;
static int swsusp_extents_insert(unsigned long swap_offset)
{
struct rb_node **new = &(swsusp_extents.rb_node);
struct rb_node *parent = NULL;
struct swsusp_extent *ext;
/* Figure out where to put the new node */
while (*new) {
ext = container_of(*new, struct swsusp_extent, node);
parent = *new;
if (swap_offset < ext->start) {
/* Try to merge */
if (swap_offset == ext->start - 1) {
ext->start--;
return 0;
}
new = &((*new)->rb_left);
} else if (swap_offset > ext->end) {
/* Try to merge */
if (swap_offset == ext->end + 1) {
ext->end++;
return 0;
}
new = &((*new)->rb_right);
} else {
/* It already is in the tree */
return -EINVAL;
}
}
/* Add the new node and rebalance the tree. */
ext = kzalloc(sizeof(struct swsusp_extent), GFP_KERNEL);
if (!ext)
return -ENOMEM;
ext->start = swap_offset;
ext->end = swap_offset;
rb_link_node(&ext->node, parent, new);
rb_insert_color(&ext->node, &swsusp_extents);
return 0;
}
/**
* alloc_swapdev_block - allocate a swap page and register that it has
* been allocated, so that it can be freed in case of an error.
*/
sector_t alloc_swapdev_block(int swap)
{
unsigned long offset;
offset = swp_offset(get_swap_page_of_type(swap));
if (offset) {
if (swsusp_extents_insert(offset))
swap_free(swp_entry(swap, offset));
else
return swapdev_block(swap, offset);
}
return 0;
}
/**
* free_all_swap_pages - free swap pages allocated for saving image data.
* It also frees the extents used to register which swap entries had been
* allocated.
*/
void free_all_swap_pages(int swap)
{
struct rb_node *node;
while ((node = swsusp_extents.rb_node)) {
struct swsusp_extent *ext;
unsigned long offset;
ext = container_of(node, struct swsusp_extent, node);
rb_erase(node, &swsusp_extents);
for (offset = ext->start; offset <= ext->end; offset++)
swap_free(swp_entry(swap, offset));
kfree(ext);
}
}
int swsusp_swap_in_use(void)
{
return (swsusp_extents.rb_node != NULL);
}
/*
* General things
*/
static unsigned short root_swap = 0xffff;
struct block_device *hib_resume_bdev;
/*
* Saving part
*/
static int mark_swapfiles(struct swap_map_handle *handle, unsigned int flags)
{
int error;
hib_bio_read_page(swsusp_resume_block, swsusp_header, NULL);
if (!memcmp("SWAP-SPACE",swsusp_header->sig, 10) ||
!memcmp("SWAPSPACE2",swsusp_header->sig, 10)) {
memcpy(swsusp_header->orig_sig,swsusp_header->sig, 10);
memcpy(swsusp_header->sig, HIBERNATE_SIG, 10);
swsusp_header->image = handle->first_sector;
swsusp: introduce restore platform operations At least on some machines it is necessary to prepare the ACPI firmware for the restoration of the system memory state from the hibernation image if the "platform" mode of hibernation has been used. Namely, in that cases we need to disable the GPEs before replacing the "boot" kernel with the "frozen" kernel (cf. http://bugzilla.kernel.org/show_bug.cgi?id=7887). After the restore they will be re-enabled by hibernation_ops->finish(), but if the restore fails, they have to be re-enabled by the restore code explicitly. For this purpose we can introduce two additional hibernation operations, called pre_restore() and restore_cleanup() and call them from the restore code path. Still, they should be called if the "platform" mode of hibernation has been used, so we need to pass the information about the hibernation mode from the "frozen" kernel to the "boot" kernel in the image header. Apparently, we can't drop the disabling of GPEs before the restore because of Bug #7887 .  We also can't do it unconditionally, because the GPEs wouldn't have been enabled after a successful restore if the suspend had been done in the 'shutdown' or 'reboot' mode. In principle we could (and probably should) unconditionally disable the GPEs before each snapshot creation *and* before the restore, but then we'd have to unconditionally enable them after the snapshot creation as well as after the restore (or restore failure)   Still, for this purpose we'd need to modify acpi_enter_sleep_state_prep() and acpi_leave_sleep_state() and we'd have to introduce some mechanism synchronizing the disablind/enabling of the GPEs with the device drivers' .suspend()/.resume() routines and with disable_/enable_nonboot_cpus().  However, this would have affected the suspend (ie. s2ram) code as well as the hibernation, which I'd like to avoid in this patch series. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Nigel Cunningham <nigel@nigel.suspend2.net> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:47:30 +00:00
swsusp_header->flags = flags;
if (flags & SF_CRC32_MODE)
swsusp_header->crc32 = handle->crc32;
error = hib_bio_write_page(swsusp_resume_block,
swsusp_header, NULL);
} else {
printk(KERN_ERR "PM: Swap header not found!\n");
error = -ENODEV;
}
return error;
}
/**
* swsusp_swap_check - check if the resume device is a swap device
* and get its index (if so)
*
* This is called before saving image
*/
static int swsusp_swap_check(void)
{
int res;
res = swap_type_of(swsusp_resume_device, swsusp_resume_block,
&hib_resume_bdev);
if (res < 0)
return res;
root_swap = res;
block: make blkdev_get/put() handle exclusive access Over time, block layer has accumulated a set of APIs dealing with bdev open, close, claim and release. * blkdev_get/put() are the primary open and close functions. * bd_claim/release() deal with exclusive open. * open/close_bdev_exclusive() are combination of open and claim and the other way around, respectively. * bd_link/unlink_disk_holder() to create and remove holder/slave symlinks. * open_by_devnum() wraps bdget() + blkdev_get(). The interface is a bit confusing and the decoupling of open and claim makes it impossible to properly guarantee exclusive access as in-kernel open + claim sequence can disturb the existing exclusive open even before the block layer knows the current open if for another exclusive access. Reorganize the interface such that, * blkdev_get() is extended to include exclusive access management. @holder argument is added and, if is @FMODE_EXCL specified, it will gain exclusive access atomically w.r.t. other exclusive accesses. * blkdev_put() is similarly extended. It now takes @mode argument and if @FMODE_EXCL is set, it releases an exclusive access. Also, when the last exclusive claim is released, the holder/slave symlinks are removed automatically. * bd_claim/release() and close_bdev_exclusive() are no longer necessary and either made static or removed. * bd_link_disk_holder() remains the same but bd_unlink_disk_holder() is no longer necessary and removed. * open_bdev_exclusive() becomes a simple wrapper around lookup_bdev() and blkdev_get(). It also has an unexpected extra bdev_read_only() test which probably should be moved into blkdev_get(). * open_by_devnum() is modified to take @holder argument and pass it to blkdev_get(). Most of bdev open/close operations are unified into blkdev_get/put() and most exclusive accesses are tested atomically at the open time (as it should). This cleans up code and removes some, both valid and invalid, but unnecessary all the same, corner cases. open_bdev_exclusive() and open_by_devnum() can use further cleanup - rename to blkdev_get_by_path() and blkdev_get_by_devt() and drop special features. Well, let's leave them for another day. Most conversions are straight-forward. drbd conversion is a bit more involved as there was some reordering, but the logic should stay the same. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: Neil Brown <neilb@suse.de> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Acked-by: Mike Snitzer <snitzer@redhat.com> Acked-by: Philipp Reisner <philipp.reisner@linbit.com> Cc: Peter Osterlund <petero2@telia.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Jan Kara <jack@suse.cz> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <joel.becker@oracle.com> Cc: Alex Elder <aelder@sgi.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: dm-devel@redhat.com Cc: drbd-dev@lists.linbit.com Cc: Leo Chen <leochen@broadcom.com> Cc: Scott Branden <sbranden@broadcom.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com> Cc: Joern Engel <joern@logfs.org> Cc: reiserfs-devel@vger.kernel.org Cc: Alexander Viro <viro@zeniv.linux.org.uk>
2010-11-13 10:55:17 +00:00
res = blkdev_get(hib_resume_bdev, FMODE_WRITE, NULL);
if (res)
return res;
res = set_blocksize(hib_resume_bdev, PAGE_SIZE);
if (res < 0)
blkdev_put(hib_resume_bdev, FMODE_WRITE);
return res;
}
/**
* write_page - Write one page to given swap location.
* @buf: Address we're writing.
* @offset: Offset of the swap page we're writing to.
* @bio_chain: Link the next write BIO here
*/
static int write_page(void *buf, sector_t offset, struct bio **bio_chain)
{
void *src;
int ret;
if (!offset)
return -ENOSPC;
if (bio_chain) {
src = (void *)__get_free_page(__GFP_WAIT | __GFP_NOWARN |
__GFP_NORETRY);
if (src) {
copy_page(src, buf);
} else {
ret = hib_wait_on_bio_chain(bio_chain); /* Free pages */
if (ret)
return ret;
src = (void *)__get_free_page(__GFP_WAIT |
__GFP_NOWARN |
__GFP_NORETRY);
if (src) {
copy_page(src, buf);
} else {
WARN_ON_ONCE(1);
bio_chain = NULL; /* Go synchronous */
src = buf;
}
}
} else {
src = buf;
}
return hib_bio_write_page(offset, src, bio_chain);
}
static void release_swap_writer(struct swap_map_handle *handle)
{
if (handle->cur)
free_page((unsigned long)handle->cur);
handle->cur = NULL;
}
static int get_swap_writer(struct swap_map_handle *handle)
{
int ret;
ret = swsusp_swap_check();
if (ret) {
if (ret != -ENOSPC)
printk(KERN_ERR "PM: Cannot find swap device, try "
"swapon -a.\n");
return ret;
}
handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_KERNEL);
if (!handle->cur) {
ret = -ENOMEM;
goto err_close;
}
handle->cur_swap = alloc_swapdev_block(root_swap);
if (!handle->cur_swap) {
ret = -ENOSPC;
goto err_rel;
}
handle->k = 0;
handle->reqd_free_pages = reqd_free_pages();
handle->first_sector = handle->cur_swap;
return 0;
err_rel:
release_swap_writer(handle);
err_close:
swsusp_close(FMODE_WRITE);
return ret;
}
static int swap_write_page(struct swap_map_handle *handle, void *buf,
struct bio **bio_chain)
{
int error = 0;
sector_t offset;
if (!handle->cur)
return -EINVAL;
offset = alloc_swapdev_block(root_swap);
error = write_page(buf, offset, bio_chain);
if (error)
return error;
handle->cur->entries[handle->k++] = offset;
if (handle->k >= MAP_PAGE_ENTRIES) {
offset = alloc_swapdev_block(root_swap);
if (!offset)
return -ENOSPC;
handle->cur->next_swap = offset;
error = write_page(handle->cur, handle->cur_swap, bio_chain);
if (error)
goto out;
clear_page(handle->cur);
handle->cur_swap = offset;
handle->k = 0;
if (bio_chain && low_free_pages() <= handle->reqd_free_pages) {
error = hib_wait_on_bio_chain(bio_chain);
if (error)
goto out;
/*
* Recalculate the number of required free pages, to
* make sure we never take more than half.
*/
handle->reqd_free_pages = reqd_free_pages();
}
}
out:
return error;
}
static int flush_swap_writer(struct swap_map_handle *handle)
{
if (handle->cur && handle->cur_swap)
return write_page(handle->cur, handle->cur_swap, NULL);
else
return -EINVAL;
}
static int swap_writer_finish(struct swap_map_handle *handle,
unsigned int flags, int error)
{
if (!error) {
flush_swap_writer(handle);
printk(KERN_INFO "PM: S");
error = mark_swapfiles(handle, flags);
printk("|\n");
}
if (error)
free_all_swap_pages(root_swap);
release_swap_writer(handle);
swsusp_close(FMODE_WRITE);
return error;
}
/* We need to remember how much compressed data we need to read. */
#define LZO_HEADER sizeof(size_t)
/* Number of pages/bytes we'll compress at one time. */
#define LZO_UNC_PAGES 32
#define LZO_UNC_SIZE (LZO_UNC_PAGES * PAGE_SIZE)
/* Number of pages/bytes we need for compressed data (worst case). */
#define LZO_CMP_PAGES DIV_ROUND_UP(lzo1x_worst_compress(LZO_UNC_SIZE) + \
LZO_HEADER, PAGE_SIZE)
#define LZO_CMP_SIZE (LZO_CMP_PAGES * PAGE_SIZE)
/* Maximum number of threads for compression/decompression. */
#define LZO_THREADS 3
/* Minimum/maximum number of pages for read buffering. */
#define LZO_MIN_RD_PAGES 1024
#define LZO_MAX_RD_PAGES 8192
/**
* save_image - save the suspend image data
*/
static int save_image(struct swap_map_handle *handle,
struct snapshot_handle *snapshot,
unsigned int nr_to_write)
{
unsigned int m;
int ret;
int nr_pages;
int err2;
struct bio *bio;
struct timeval start;
struct timeval stop;
printk(KERN_INFO "PM: Saving image data pages (%u pages)...\n",
nr_to_write);
m = nr_to_write / 10;
if (!m)
m = 1;
nr_pages = 0;
bio = NULL;
do_gettimeofday(&start);
while (1) {
ret = snapshot_read_next(snapshot);
if (ret <= 0)
break;
ret = swap_write_page(handle, data_of(*snapshot), &bio);
if (ret)
break;
if (!(nr_pages % m))
printk(KERN_INFO "PM: Image saving progress: %3d%%\n",
nr_pages / m * 10);
nr_pages++;
}
err2 = hib_wait_on_bio_chain(&bio);
do_gettimeofday(&stop);
if (!ret)
ret = err2;
if (!ret)
printk(KERN_INFO "PM: Image saving done.\n");
swsusp_show_speed(&start, &stop, nr_to_write, "Wrote");
return ret;
}
/**
* Structure used for CRC32.
*/
struct crc_data {
struct task_struct *thr; /* thread */
atomic_t ready; /* ready to start flag */
atomic_t stop; /* ready to stop flag */
unsigned run_threads; /* nr current threads */
wait_queue_head_t go; /* start crc update */
wait_queue_head_t done; /* crc update done */
u32 *crc32; /* points to handle's crc32 */
size_t *unc_len[LZO_THREADS]; /* uncompressed lengths */
unsigned char *unc[LZO_THREADS]; /* uncompressed data */
};
/**
* CRC32 update function that runs in its own thread.
*/
static int crc32_threadfn(void *data)
{
struct crc_data *d = data;
unsigned i;
while (1) {
wait_event(d->go, atomic_read(&d->ready) ||
kthread_should_stop());
if (kthread_should_stop()) {
d->thr = NULL;
atomic_set(&d->stop, 1);
wake_up(&d->done);
break;
}
atomic_set(&d->ready, 0);
for (i = 0; i < d->run_threads; i++)
*d->crc32 = crc32_le(*d->crc32,
d->unc[i], *d->unc_len[i]);
atomic_set(&d->stop, 1);
wake_up(&d->done);
}
return 0;
}
/**
* Structure used for LZO data compression.
*/
struct cmp_data {
struct task_struct *thr; /* thread */
atomic_t ready; /* ready to start flag */
atomic_t stop; /* ready to stop flag */
int ret; /* return code */
wait_queue_head_t go; /* start compression */
wait_queue_head_t done; /* compression done */
size_t unc_len; /* uncompressed length */
size_t cmp_len; /* compressed length */
unsigned char unc[LZO_UNC_SIZE]; /* uncompressed buffer */
unsigned char cmp[LZO_CMP_SIZE]; /* compressed buffer */
unsigned char wrk[LZO1X_1_MEM_COMPRESS]; /* compression workspace */
};
/**
* Compression function that runs in its own thread.
*/
static int lzo_compress_threadfn(void *data)
{
struct cmp_data *d = data;
while (1) {
wait_event(d->go, atomic_read(&d->ready) ||
kthread_should_stop());
if (kthread_should_stop()) {
d->thr = NULL;
d->ret = -1;
atomic_set(&d->stop, 1);
wake_up(&d->done);
break;
}
atomic_set(&d->ready, 0);
d->ret = lzo1x_1_compress(d->unc, d->unc_len,
d->cmp + LZO_HEADER, &d->cmp_len,
d->wrk);
atomic_set(&d->stop, 1);
wake_up(&d->done);
}
return 0;
}
/**
* save_image_lzo - Save the suspend image data compressed with LZO.
* @handle: Swap mam handle to use for saving the image.
* @snapshot: Image to read data from.
* @nr_to_write: Number of pages to save.
*/
static int save_image_lzo(struct swap_map_handle *handle,
struct snapshot_handle *snapshot,
unsigned int nr_to_write)
{
unsigned int m;
int ret = 0;
int nr_pages;
int err2;
struct bio *bio;
struct timeval start;
struct timeval stop;
size_t off;
unsigned thr, run_threads, nr_threads;
unsigned char *page = NULL;
struct cmp_data *data = NULL;
struct crc_data *crc = NULL;
/*
* We'll limit the number of threads for compression to limit memory
* footprint.
*/
nr_threads = num_online_cpus() - 1;
nr_threads = clamp_val(nr_threads, 1, LZO_THREADS);
page = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH);
if (!page) {
printk(KERN_ERR "PM: Failed to allocate LZO page\n");
ret = -ENOMEM;
goto out_clean;
}
data = vmalloc(sizeof(*data) * nr_threads);
if (!data) {
printk(KERN_ERR "PM: Failed to allocate LZO data\n");
ret = -ENOMEM;
goto out_clean;
}
for (thr = 0; thr < nr_threads; thr++)
memset(&data[thr], 0, offsetof(struct cmp_data, go));
crc = kmalloc(sizeof(*crc), GFP_KERNEL);
if (!crc) {
printk(KERN_ERR "PM: Failed to allocate crc\n");
ret = -ENOMEM;
goto out_clean;
}
memset(crc, 0, offsetof(struct crc_data, go));
/*
* Start the compression threads.
*/
for (thr = 0; thr < nr_threads; thr++) {
init_waitqueue_head(&data[thr].go);
init_waitqueue_head(&data[thr].done);
data[thr].thr = kthread_run(lzo_compress_threadfn,
&data[thr],
"image_compress/%u", thr);
if (IS_ERR(data[thr].thr)) {
data[thr].thr = NULL;
printk(KERN_ERR
"PM: Cannot start compression threads\n");
ret = -ENOMEM;
goto out_clean;
}
}
/*
* Start the CRC32 thread.
*/
init_waitqueue_head(&crc->go);
init_waitqueue_head(&crc->done);
handle->crc32 = 0;
crc->crc32 = &handle->crc32;
for (thr = 0; thr < nr_threads; thr++) {
crc->unc[thr] = data[thr].unc;
crc->unc_len[thr] = &data[thr].unc_len;
}
crc->thr = kthread_run(crc32_threadfn, crc, "image_crc32");
if (IS_ERR(crc->thr)) {
crc->thr = NULL;
printk(KERN_ERR "PM: Cannot start CRC32 thread\n");
ret = -ENOMEM;
goto out_clean;
}
/*
* Adjust the number of required free pages after all allocations have
* been done. We don't want to run out of pages when writing.
*/
handle->reqd_free_pages = reqd_free_pages();
printk(KERN_INFO
"PM: Using %u thread(s) for compression.\n"
"PM: Compressing and saving image data (%u pages)...\n",
nr_threads, nr_to_write);
m = nr_to_write / 10;
if (!m)
m = 1;
nr_pages = 0;
bio = NULL;
do_gettimeofday(&start);
for (;;) {
for (thr = 0; thr < nr_threads; thr++) {
for (off = 0; off < LZO_UNC_SIZE; off += PAGE_SIZE) {
ret = snapshot_read_next(snapshot);
if (ret < 0)
goto out_finish;
if (!ret)
break;
memcpy(data[thr].unc + off,
data_of(*snapshot), PAGE_SIZE);
if (!(nr_pages % m))
printk(KERN_INFO
"PM: Image saving progress: "
"%3d%%\n",
nr_pages / m * 10);
nr_pages++;
}
if (!off)
break;
data[thr].unc_len = off;
atomic_set(&data[thr].ready, 1);
wake_up(&data[thr].go);
}
if (!thr)
break;
crc->run_threads = thr;
atomic_set(&crc->ready, 1);
wake_up(&crc->go);
for (run_threads = thr, thr = 0; thr < run_threads; thr++) {
wait_event(data[thr].done,
atomic_read(&data[thr].stop));
atomic_set(&data[thr].stop, 0);
ret = data[thr].ret;
if (ret < 0) {
printk(KERN_ERR "PM: LZO compression failed\n");
goto out_finish;
}
if (unlikely(!data[thr].cmp_len ||
data[thr].cmp_len >
lzo1x_worst_compress(data[thr].unc_len))) {
printk(KERN_ERR
"PM: Invalid LZO compressed length\n");
ret = -1;
goto out_finish;
}
*(size_t *)data[thr].cmp = data[thr].cmp_len;
/*
* Given we are writing one page at a time to disk, we
* copy that much from the buffer, although the last
* bit will likely be smaller than full page. This is
* OK - we saved the length of the compressed data, so
* any garbage at the end will be discarded when we
* read it.
*/
for (off = 0;
off < LZO_HEADER + data[thr].cmp_len;
off += PAGE_SIZE) {
memcpy(page, data[thr].cmp + off, PAGE_SIZE);
ret = swap_write_page(handle, page, &bio);
if (ret)
goto out_finish;
}
}
wait_event(crc->done, atomic_read(&crc->stop));
atomic_set(&crc->stop, 0);
}
out_finish:
err2 = hib_wait_on_bio_chain(&bio);
do_gettimeofday(&stop);
if (!ret)
ret = err2;
if (!ret)
printk(KERN_INFO "PM: Image saving done.\n");
swsusp_show_speed(&start, &stop, nr_to_write, "Wrote");
out_clean:
if (crc) {
if (crc->thr)
kthread_stop(crc->thr);
kfree(crc);
}
if (data) {
for (thr = 0; thr < nr_threads; thr++)
if (data[thr].thr)
kthread_stop(data[thr].thr);
vfree(data);
}
if (page) free_page((unsigned long)page);
return ret;
}
/**
* enough_swap - Make sure we have enough swap to save the image.
*
* Returns TRUE or FALSE after checking the total amount of swap
* space avaiable from the resume partition.
*/
static int enough_swap(unsigned int nr_pages, unsigned int flags)
{
unsigned int free_swap = count_swap_pages(root_swap, 1);
unsigned int required;
pr_debug("PM: Free swap pages: %u\n", free_swap);
required = PAGES_FOR_IO + nr_pages;
return free_swap > required;
}
/**
* swsusp_write - Write entire image and metadata.
swsusp: introduce restore platform operations At least on some machines it is necessary to prepare the ACPI firmware for the restoration of the system memory state from the hibernation image if the "platform" mode of hibernation has been used. Namely, in that cases we need to disable the GPEs before replacing the "boot" kernel with the "frozen" kernel (cf. http://bugzilla.kernel.org/show_bug.cgi?id=7887). After the restore they will be re-enabled by hibernation_ops->finish(), but if the restore fails, they have to be re-enabled by the restore code explicitly. For this purpose we can introduce two additional hibernation operations, called pre_restore() and restore_cleanup() and call them from the restore code path. Still, they should be called if the "platform" mode of hibernation has been used, so we need to pass the information about the hibernation mode from the "frozen" kernel to the "boot" kernel in the image header. Apparently, we can't drop the disabling of GPEs before the restore because of Bug #7887 .  We also can't do it unconditionally, because the GPEs wouldn't have been enabled after a successful restore if the suspend had been done in the 'shutdown' or 'reboot' mode. In principle we could (and probably should) unconditionally disable the GPEs before each snapshot creation *and* before the restore, but then we'd have to unconditionally enable them after the snapshot creation as well as after the restore (or restore failure)   Still, for this purpose we'd need to modify acpi_enter_sleep_state_prep() and acpi_leave_sleep_state() and we'd have to introduce some mechanism synchronizing the disablind/enabling of the GPEs with the device drivers' .suspend()/.resume() routines and with disable_/enable_nonboot_cpus().  However, this would have affected the suspend (ie. s2ram) code as well as the hibernation, which I'd like to avoid in this patch series. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Nigel Cunningham <nigel@nigel.suspend2.net> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:47:30 +00:00
* @flags: flags to pass to the "boot" kernel in the image header
*
* It is important _NOT_ to umount filesystems at this point. We want
* them synced (in case something goes wrong) but we DO not want to mark
* filesystem clean: it is not. (And it does not matter, if we resume
* correctly, we'll mark system clean, anyway.)
*/
swsusp: introduce restore platform operations At least on some machines it is necessary to prepare the ACPI firmware for the restoration of the system memory state from the hibernation image if the "platform" mode of hibernation has been used. Namely, in that cases we need to disable the GPEs before replacing the "boot" kernel with the "frozen" kernel (cf. http://bugzilla.kernel.org/show_bug.cgi?id=7887). After the restore they will be re-enabled by hibernation_ops->finish(), but if the restore fails, they have to be re-enabled by the restore code explicitly. For this purpose we can introduce two additional hibernation operations, called pre_restore() and restore_cleanup() and call them from the restore code path. Still, they should be called if the "platform" mode of hibernation has been used, so we need to pass the information about the hibernation mode from the "frozen" kernel to the "boot" kernel in the image header. Apparently, we can't drop the disabling of GPEs before the restore because of Bug #7887 .  We also can't do it unconditionally, because the GPEs wouldn't have been enabled after a successful restore if the suspend had been done in the 'shutdown' or 'reboot' mode. In principle we could (and probably should) unconditionally disable the GPEs before each snapshot creation *and* before the restore, but then we'd have to unconditionally enable them after the snapshot creation as well as after the restore (or restore failure)   Still, for this purpose we'd need to modify acpi_enter_sleep_state_prep() and acpi_leave_sleep_state() and we'd have to introduce some mechanism synchronizing the disablind/enabling of the GPEs with the device drivers' .suspend()/.resume() routines and with disable_/enable_nonboot_cpus().  However, this would have affected the suspend (ie. s2ram) code as well as the hibernation, which I'd like to avoid in this patch series. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Nigel Cunningham <nigel@nigel.suspend2.net> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:47:30 +00:00
int swsusp_write(unsigned int flags)
{
struct swap_map_handle handle;
struct snapshot_handle snapshot;
struct swsusp_info *header;
unsigned long pages;
int error;
pages = snapshot_get_image_size();
error = get_swap_writer(&handle);
if (error) {
printk(KERN_ERR "PM: Cannot get swap writer\n");
return error;
}
if (flags & SF_NOCOMPRESS_MODE) {
if (!enough_swap(pages, flags)) {
printk(KERN_ERR "PM: Not enough free swap\n");
error = -ENOSPC;
goto out_finish;
}
}
memset(&snapshot, 0, sizeof(struct snapshot_handle));
error = snapshot_read_next(&snapshot);
if (error < PAGE_SIZE) {
if (error >= 0)
error = -EFAULT;
goto out_finish;
}
header = (struct swsusp_info *)data_of(snapshot);
error = swap_write_page(&handle, header, NULL);
if (!error) {
error = (flags & SF_NOCOMPRESS_MODE) ?
save_image(&handle, &snapshot, pages - 1) :
save_image_lzo(&handle, &snapshot, pages - 1);
}
out_finish:
error = swap_writer_finish(&handle, flags, error);
return error;
}
/**
* The following functions allow us to read data using a swap map
* in a file-alike way
*/
static void release_swap_reader(struct swap_map_handle *handle)
{
struct swap_map_page_list *tmp;
while (handle->maps) {
if (handle->maps->map)
free_page((unsigned long)handle->maps->map);
tmp = handle->maps;
handle->maps = handle->maps->next;
kfree(tmp);
}
handle->cur = NULL;
}
static int get_swap_reader(struct swap_map_handle *handle,
unsigned int *flags_p)
{
int error;
struct swap_map_page_list *tmp, *last;
sector_t offset;
*flags_p = swsusp_header->flags;
if (!swsusp_header->image) /* how can this happen? */
return -EINVAL;
handle->cur = NULL;
last = handle->maps = NULL;
offset = swsusp_header->image;
while (offset) {
tmp = kmalloc(sizeof(*handle->maps), GFP_KERNEL);
if (!tmp) {
release_swap_reader(handle);
return -ENOMEM;
}
memset(tmp, 0, sizeof(*tmp));
if (!handle->maps)
handle->maps = tmp;
if (last)
last->next = tmp;
last = tmp;
tmp->map = (struct swap_map_page *)
__get_free_page(__GFP_WAIT | __GFP_HIGH);
if (!tmp->map) {
release_swap_reader(handle);
return -ENOMEM;
}
error = hib_bio_read_page(offset, tmp->map, NULL);
if (error) {
release_swap_reader(handle);
return error;
}
offset = tmp->map->next_swap;
}
handle->k = 0;
handle->cur = handle->maps->map;
return 0;
}
static int swap_read_page(struct swap_map_handle *handle, void *buf,
struct bio **bio_chain)
{
sector_t offset;
int error;
struct swap_map_page_list *tmp;
if (!handle->cur)
return -EINVAL;
offset = handle->cur->entries[handle->k];
if (!offset)
return -EFAULT;
error = hib_bio_read_page(offset, buf, bio_chain);
if (error)
return error;
if (++handle->k >= MAP_PAGE_ENTRIES) {
handle->k = 0;
free_page((unsigned long)handle->maps->map);
tmp = handle->maps;
handle->maps = handle->maps->next;
kfree(tmp);
if (!handle->maps)
release_swap_reader(handle);
else
handle->cur = handle->maps->map;
}
return error;
}
static int swap_reader_finish(struct swap_map_handle *handle)
{
release_swap_reader(handle);
return 0;
}
/**
* load_image - load the image using the swap map handle
* @handle and the snapshot handle @snapshot
* (assume there are @nr_pages pages to load)
*/
static int load_image(struct swap_map_handle *handle,
struct snapshot_handle *snapshot,
unsigned int nr_to_read)
{
unsigned int m;
int ret = 0;
struct timeval start;
struct timeval stop;
struct bio *bio;
int err2;
unsigned nr_pages;
printk(KERN_INFO "PM: Loading image data pages (%u pages)...\n",
nr_to_read);
m = nr_to_read / 10;
if (!m)
m = 1;
nr_pages = 0;
bio = NULL;
do_gettimeofday(&start);
for ( ; ; ) {
ret = snapshot_write_next(snapshot);
if (ret <= 0)
break;
ret = swap_read_page(handle, data_of(*snapshot), &bio);
if (ret)
break;
if (snapshot->sync_read)
ret = hib_wait_on_bio_chain(&bio);
if (ret)
break;
if (!(nr_pages % m))
printk(KERN_INFO "PM: Image loading progress: %3d%%\n",
nr_pages / m * 10);
nr_pages++;
}
err2 = hib_wait_on_bio_chain(&bio);
do_gettimeofday(&stop);
if (!ret)
ret = err2;
if (!ret) {
printk(KERN_INFO "PM: Image loading done.\n");
[PATCH] swsusp: Improve handling of highmem Currently swsusp saves the contents of highmem pages by copying them to the normal zone which is quite inefficient (eg. it requires two normal pages to be used for saving one highmem page). This may be improved by using highmem for saving the contents of saveable highmem pages. Namely, during the suspend phase of the suspend-resume cycle we try to allocate as many free highmem pages as there are saveable highmem pages. If there are not enough highmem image pages to store the contents of all of the saveable highmem pages, some of them will be stored in the "normal" memory. Next, we allocate as many free "normal" pages as needed to store the (remaining) image data. We use a memory bitmap to mark the allocated free pages (ie. highmem as well as "normal" image pages). Now, we use another memory bitmap to mark all of the saveable pages (highmem as well as "normal") and the contents of the saveable pages are copied into the image pages. Then, the second bitmap is used to save the pfns corresponding to the saveable pages and the first one is used to save their data. During the resume phase the pfns of the pages that were saveable during the suspend are loaded from the image and used to mark the "unsafe" page frames. Next, we try to allocate as many free highmem page frames as to load all of the image data that had been in the highmem before the suspend and we allocate so many free "normal" page frames that the total number of allocated free pages (highmem and "normal") is equal to the size of the image. While doing this we have to make sure that there will be some extra free "normal" and "safe" page frames for two lists of PBEs constructed later. Now, the image data are loaded, if possible, into their "original" page frames. The image data that cannot be written into their "original" page frames are loaded into "safe" page frames and their "original" kernel virtual addresses, as well as the addresses of the "safe" pages containing their copies, are stored in one of two lists of PBEs. One list of PBEs is for the copies of "normal" suspend pages (ie. "normal" pages that were saveable during the suspend) and it is used in the same way as previously (ie. by the architecture-dependent parts of swsusp). The other list of PBEs is for the copies of highmem suspend pages. The pages in this list are restored (in a reversible way) right before the arch-dependent code is called. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-07 04:34:18 +00:00
snapshot_write_finalize(snapshot);
if (!snapshot_image_loaded(snapshot))
ret = -ENODATA;
}
swsusp_show_speed(&start, &stop, nr_to_read, "Read");
return ret;
}
/**
* Structure used for LZO data decompression.
*/
struct dec_data {
struct task_struct *thr; /* thread */
atomic_t ready; /* ready to start flag */
atomic_t stop; /* ready to stop flag */
int ret; /* return code */
wait_queue_head_t go; /* start decompression */
wait_queue_head_t done; /* decompression done */
size_t unc_len; /* uncompressed length */
size_t cmp_len; /* compressed length */
unsigned char unc[LZO_UNC_SIZE]; /* uncompressed buffer */
unsigned char cmp[LZO_CMP_SIZE]; /* compressed buffer */
};
/**
* Deompression function that runs in its own thread.
*/
static int lzo_decompress_threadfn(void *data)
{
struct dec_data *d = data;
while (1) {
wait_event(d->go, atomic_read(&d->ready) ||
kthread_should_stop());
if (kthread_should_stop()) {
d->thr = NULL;
d->ret = -1;
atomic_set(&d->stop, 1);
wake_up(&d->done);
break;
}
atomic_set(&d->ready, 0);
d->unc_len = LZO_UNC_SIZE;
d->ret = lzo1x_decompress_safe(d->cmp + LZO_HEADER, d->cmp_len,
d->unc, &d->unc_len);
atomic_set(&d->stop, 1);
wake_up(&d->done);
}
return 0;
}
/**
* load_image_lzo - Load compressed image data and decompress them with LZO.
* @handle: Swap map handle to use for loading data.
* @snapshot: Image to copy uncompressed data into.
* @nr_to_read: Number of pages to load.
*/
static int load_image_lzo(struct swap_map_handle *handle,
struct snapshot_handle *snapshot,
unsigned int nr_to_read)
{
unsigned int m;
int ret = 0;
int eof = 0;
struct bio *bio;
struct timeval start;
struct timeval stop;
unsigned nr_pages;
size_t off;
unsigned i, thr, run_threads, nr_threads;
unsigned ring = 0, pg = 0, ring_size = 0,
have = 0, want, need, asked = 0;
unsigned long read_pages = 0;
unsigned char **page = NULL;
struct dec_data *data = NULL;
struct crc_data *crc = NULL;
/*
* We'll limit the number of threads for decompression to limit memory
* footprint.
*/
nr_threads = num_online_cpus() - 1;
nr_threads = clamp_val(nr_threads, 1, LZO_THREADS);
page = vmalloc(sizeof(*page) * LZO_MAX_RD_PAGES);
if (!page) {
printk(KERN_ERR "PM: Failed to allocate LZO page\n");
ret = -ENOMEM;
goto out_clean;
}
data = vmalloc(sizeof(*data) * nr_threads);
if (!data) {
printk(KERN_ERR "PM: Failed to allocate LZO data\n");
ret = -ENOMEM;
goto out_clean;
}
for (thr = 0; thr < nr_threads; thr++)
memset(&data[thr], 0, offsetof(struct dec_data, go));
crc = kmalloc(sizeof(*crc), GFP_KERNEL);
if (!crc) {
printk(KERN_ERR "PM: Failed to allocate crc\n");
ret = -ENOMEM;
goto out_clean;
}
memset(crc, 0, offsetof(struct crc_data, go));
/*
* Start the decompression threads.
*/
for (thr = 0; thr < nr_threads; thr++) {
init_waitqueue_head(&data[thr].go);
init_waitqueue_head(&data[thr].done);
data[thr].thr = kthread_run(lzo_decompress_threadfn,
&data[thr],
"image_decompress/%u", thr);
if (IS_ERR(data[thr].thr)) {
data[thr].thr = NULL;
printk(KERN_ERR
"PM: Cannot start decompression threads\n");
ret = -ENOMEM;
goto out_clean;
}
}
/*
* Start the CRC32 thread.
*/
init_waitqueue_head(&crc->go);
init_waitqueue_head(&crc->done);
handle->crc32 = 0;
crc->crc32 = &handle->crc32;
for (thr = 0; thr < nr_threads; thr++) {
crc->unc[thr] = data[thr].unc;
crc->unc_len[thr] = &data[thr].unc_len;
}
crc->thr = kthread_run(crc32_threadfn, crc, "image_crc32");
if (IS_ERR(crc->thr)) {
crc->thr = NULL;
printk(KERN_ERR "PM: Cannot start CRC32 thread\n");
ret = -ENOMEM;
goto out_clean;
}
/*
* Set the number of pages for read buffering.
* This is complete guesswork, because we'll only know the real
* picture once prepare_image() is called, which is much later on
* during the image load phase. We'll assume the worst case and
* say that none of the image pages are from high memory.
*/
if (low_free_pages() > snapshot_get_image_size())
read_pages = (low_free_pages() - snapshot_get_image_size()) / 2;
read_pages = clamp_val(read_pages, LZO_MIN_RD_PAGES, LZO_MAX_RD_PAGES);
for (i = 0; i < read_pages; i++) {
page[i] = (void *)__get_free_page(i < LZO_CMP_PAGES ?
__GFP_WAIT | __GFP_HIGH :
__GFP_WAIT | __GFP_NOWARN |
__GFP_NORETRY);
if (!page[i]) {
if (i < LZO_CMP_PAGES) {
ring_size = i;
printk(KERN_ERR
"PM: Failed to allocate LZO pages\n");
ret = -ENOMEM;
goto out_clean;
} else {
break;
}
}
}
want = ring_size = i;
printk(KERN_INFO
"PM: Using %u thread(s) for decompression.\n"
"PM: Loading and decompressing image data (%u pages)...\n",
nr_threads, nr_to_read);
m = nr_to_read / 10;
if (!m)
m = 1;
nr_pages = 0;
bio = NULL;
do_gettimeofday(&start);
ret = snapshot_write_next(snapshot);
if (ret <= 0)
goto out_finish;
for(;;) {
for (i = 0; !eof && i < want; i++) {
ret = swap_read_page(handle, page[ring], &bio);
if (ret) {
/*
* On real read error, finish. On end of data,
* set EOF flag and just exit the read loop.
*/
if (handle->cur &&
handle->cur->entries[handle->k]) {
goto out_finish;
} else {
eof = 1;
break;
}
}
if (++ring >= ring_size)
ring = 0;
}
asked += i;
want -= i;
/*
* We are out of data, wait for some more.
*/
if (!have) {
if (!asked)
break;
ret = hib_wait_on_bio_chain(&bio);
if (ret)
goto out_finish;
have += asked;
asked = 0;
if (eof)
eof = 2;
}
if (crc->run_threads) {
wait_event(crc->done, atomic_read(&crc->stop));
atomic_set(&crc->stop, 0);
crc->run_threads = 0;
}
for (thr = 0; have && thr < nr_threads; thr++) {
data[thr].cmp_len = *(size_t *)page[pg];
if (unlikely(!data[thr].cmp_len ||
data[thr].cmp_len >
lzo1x_worst_compress(LZO_UNC_SIZE))) {
printk(KERN_ERR
"PM: Invalid LZO compressed length\n");
ret = -1;
goto out_finish;
}
need = DIV_ROUND_UP(data[thr].cmp_len + LZO_HEADER,
PAGE_SIZE);
if (need > have) {
if (eof > 1) {
ret = -1;
goto out_finish;
}
break;
}
for (off = 0;
off < LZO_HEADER + data[thr].cmp_len;
off += PAGE_SIZE) {
memcpy(data[thr].cmp + off,
page[pg], PAGE_SIZE);
have--;
want++;
if (++pg >= ring_size)
pg = 0;
}
atomic_set(&data[thr].ready, 1);
wake_up(&data[thr].go);
}
/*
* Wait for more data while we are decompressing.
*/
if (have < LZO_CMP_PAGES && asked) {
ret = hib_wait_on_bio_chain(&bio);
if (ret)
goto out_finish;
have += asked;
asked = 0;
if (eof)
eof = 2;
}
for (run_threads = thr, thr = 0; thr < run_threads; thr++) {
wait_event(data[thr].done,
atomic_read(&data[thr].stop));
atomic_set(&data[thr].stop, 0);
ret = data[thr].ret;
if (ret < 0) {
printk(KERN_ERR
"PM: LZO decompression failed\n");
goto out_finish;
}
if (unlikely(!data[thr].unc_len ||
data[thr].unc_len > LZO_UNC_SIZE ||
data[thr].unc_len & (PAGE_SIZE - 1))) {
printk(KERN_ERR
"PM: Invalid LZO uncompressed length\n");
ret = -1;
goto out_finish;
}
for (off = 0;
off < data[thr].unc_len; off += PAGE_SIZE) {
memcpy(data_of(*snapshot),
data[thr].unc + off, PAGE_SIZE);
if (!(nr_pages % m))
printk(KERN_INFO
"PM: Image loading progress: "
"%3d%%\n",
nr_pages / m * 10);
nr_pages++;
ret = snapshot_write_next(snapshot);
if (ret <= 0) {
crc->run_threads = thr + 1;
atomic_set(&crc->ready, 1);
wake_up(&crc->go);
goto out_finish;
}
}
}
crc->run_threads = thr;
atomic_set(&crc->ready, 1);
wake_up(&crc->go);
}
out_finish:
if (crc->run_threads) {
wait_event(crc->done, atomic_read(&crc->stop));
atomic_set(&crc->stop, 0);
}
do_gettimeofday(&stop);
if (!ret) {
printk(KERN_INFO "PM: Image loading done.\n");
snapshot_write_finalize(snapshot);
if (!snapshot_image_loaded(snapshot))
ret = -ENODATA;
if (!ret) {
if (swsusp_header->flags & SF_CRC32_MODE) {
if(handle->crc32 != swsusp_header->crc32) {
printk(KERN_ERR
"PM: Invalid image CRC32!\n");
ret = -ENODATA;
}
}
}
}
swsusp_show_speed(&start, &stop, nr_to_read, "Read");
out_clean:
for (i = 0; i < ring_size; i++)
free_page((unsigned long)page[i]);
if (crc) {
if (crc->thr)
kthread_stop(crc->thr);
kfree(crc);
}
if (data) {
for (thr = 0; thr < nr_threads; thr++)
if (data[thr].thr)
kthread_stop(data[thr].thr);
vfree(data);
}
if (page) vfree(page);
return ret;
}
swsusp: introduce restore platform operations At least on some machines it is necessary to prepare the ACPI firmware for the restoration of the system memory state from the hibernation image if the "platform" mode of hibernation has been used. Namely, in that cases we need to disable the GPEs before replacing the "boot" kernel with the "frozen" kernel (cf. http://bugzilla.kernel.org/show_bug.cgi?id=7887). After the restore they will be re-enabled by hibernation_ops->finish(), but if the restore fails, they have to be re-enabled by the restore code explicitly. For this purpose we can introduce two additional hibernation operations, called pre_restore() and restore_cleanup() and call them from the restore code path. Still, they should be called if the "platform" mode of hibernation has been used, so we need to pass the information about the hibernation mode from the "frozen" kernel to the "boot" kernel in the image header. Apparently, we can't drop the disabling of GPEs before the restore because of Bug #7887 .  We also can't do it unconditionally, because the GPEs wouldn't have been enabled after a successful restore if the suspend had been done in the 'shutdown' or 'reboot' mode. In principle we could (and probably should) unconditionally disable the GPEs before each snapshot creation *and* before the restore, but then we'd have to unconditionally enable them after the snapshot creation as well as after the restore (or restore failure)   Still, for this purpose we'd need to modify acpi_enter_sleep_state_prep() and acpi_leave_sleep_state() and we'd have to introduce some mechanism synchronizing the disablind/enabling of the GPEs with the device drivers' .suspend()/.resume() routines and with disable_/enable_nonboot_cpus().  However, this would have affected the suspend (ie. s2ram) code as well as the hibernation, which I'd like to avoid in this patch series. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Nigel Cunningham <nigel@nigel.suspend2.net> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:47:30 +00:00
/**
* swsusp_read - read the hibernation image.
* @flags_p: flags passed by the "frozen" kernel in the image header should
* be written into this memory location
swsusp: introduce restore platform operations At least on some machines it is necessary to prepare the ACPI firmware for the restoration of the system memory state from the hibernation image if the "platform" mode of hibernation has been used. Namely, in that cases we need to disable the GPEs before replacing the "boot" kernel with the "frozen" kernel (cf. http://bugzilla.kernel.org/show_bug.cgi?id=7887). After the restore they will be re-enabled by hibernation_ops->finish(), but if the restore fails, they have to be re-enabled by the restore code explicitly. For this purpose we can introduce two additional hibernation operations, called pre_restore() and restore_cleanup() and call them from the restore code path. Still, they should be called if the "platform" mode of hibernation has been used, so we need to pass the information about the hibernation mode from the "frozen" kernel to the "boot" kernel in the image header. Apparently, we can't drop the disabling of GPEs before the restore because of Bug #7887 .  We also can't do it unconditionally, because the GPEs wouldn't have been enabled after a successful restore if the suspend had been done in the 'shutdown' or 'reboot' mode. In principle we could (and probably should) unconditionally disable the GPEs before each snapshot creation *and* before the restore, but then we'd have to unconditionally enable them after the snapshot creation as well as after the restore (or restore failure)   Still, for this purpose we'd need to modify acpi_enter_sleep_state_prep() and acpi_leave_sleep_state() and we'd have to introduce some mechanism synchronizing the disablind/enabling of the GPEs with the device drivers' .suspend()/.resume() routines and with disable_/enable_nonboot_cpus().  However, this would have affected the suspend (ie. s2ram) code as well as the hibernation, which I'd like to avoid in this patch series. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Nigel Cunningham <nigel@nigel.suspend2.net> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 08:47:30 +00:00
*/
int swsusp_read(unsigned int *flags_p)
{
int error;
struct swap_map_handle handle;
struct snapshot_handle snapshot;
struct swsusp_info *header;
memset(&snapshot, 0, sizeof(struct snapshot_handle));
error = snapshot_write_next(&snapshot);
if (error < PAGE_SIZE)
return error < 0 ? error : -EFAULT;
header = (struct swsusp_info *)data_of(snapshot);
error = get_swap_reader(&handle, flags_p);
if (error)
goto end;
if (!error)
error = swap_read_page(&handle, header, NULL);
if (!error) {
error = (*flags_p & SF_NOCOMPRESS_MODE) ?
load_image(&handle, &snapshot, header->pages - 1) :
load_image_lzo(&handle, &snapshot, header->pages - 1);
}
swap_reader_finish(&handle);
end:
if (!error)
pr_debug("PM: Image successfully loaded\n");
else
pr_debug("PM: Error %d resuming\n", error);
return error;
}
/**
* swsusp_check - Check for swsusp signature in the resume device
*/
int swsusp_check(void)
{
int error;
hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device,
FMODE_READ, NULL);
if (!IS_ERR(hib_resume_bdev)) {
set_blocksize(hib_resume_bdev, PAGE_SIZE);
clear_page(swsusp_header);
error = hib_bio_read_page(swsusp_resume_block,
swsusp_header, NULL);
if (error)
goto put;
if (!memcmp(HIBERNATE_SIG, swsusp_header->sig, 10)) {
memcpy(swsusp_header->sig, swsusp_header->orig_sig, 10);
/* Reset swap signature now */
error = hib_bio_write_page(swsusp_resume_block,
swsusp_header, NULL);
} else {
error = -EINVAL;
}
put:
if (error)
blkdev_put(hib_resume_bdev, FMODE_READ);
else
pr_debug("PM: Image signature found, resuming\n");
} else {
error = PTR_ERR(hib_resume_bdev);
}
if (error)
pr_debug("PM: Image not found (code %d)\n", error);
return error;
}
/**
* swsusp_close - close swap device.
*/
void swsusp_close(fmode_t mode)
{
if (IS_ERR(hib_resume_bdev)) {
pr_debug("PM: Image device not initialised\n");
return;
}
blkdev_put(hib_resume_bdev, mode);
}
/**
* swsusp_unmark - Unmark swsusp signature in the resume device
*/
#ifdef CONFIG_SUSPEND
int swsusp_unmark(void)
{
int error;
hib_bio_read_page(swsusp_resume_block, swsusp_header, NULL);
if (!memcmp(HIBERNATE_SIG,swsusp_header->sig, 10)) {
memcpy(swsusp_header->sig,swsusp_header->orig_sig, 10);
error = hib_bio_write_page(swsusp_resume_block,
swsusp_header, NULL);
} else {
printk(KERN_ERR "PM: Cannot find swsusp signature!\n");
error = -ENODEV;
}
/*
* We just returned from suspend, we don't need the image any more.
*/
free_all_swap_pages(root_swap);
return error;
}
#endif
static int swsusp_header_init(void)
{
swsusp_header = (struct swsusp_header*) __get_free_page(GFP_KERNEL);
if (!swsusp_header)
panic("Could not allocate memory for swsusp_header\n");
return 0;
}
core_initcall(swsusp_header_init);