/******************************************************************************
- * balloon.c
- *
* Xen balloon driver - enables returning/claiming memory to/from Xen.
*
* Copyright (c) 2003, B Dragovic
*/
#include <linux/kernel.h>
-#include <linux/module.h>
#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/mutex.h>
#include <linux/list.h>
-#include <linux/sysdev.h>
#include <linux/gfp.h>
#include <asm/page.h>
#include <asm/pgalloc.h>
#include <asm/pgtable.h>
-#include <asm/uaccess.h>
#include <asm/tlb.h>
#include <asm/e820.h>
#include <xen/xen.h>
#include <xen/interface/xen.h>
#include <xen/interface/memory.h>
-#include <xen/xenbus.h>
+#include <xen/balloon.h>
#include <xen/features.h>
#include <xen/page.h>
-#define PAGES2KB(_p) ((_p)<<(PAGE_SHIFT-10))
-
-#define BALLOON_CLASS_NAME "xen_memory"
+/*
+ * balloon_process() state:
+ *
+ * BP_DONE: done or nothing to do,
+ * BP_EAGAIN: error, go to sleep,
+ * BP_ECANCELED: error, balloon operation canceled.
+ */
-struct balloon_stats {
- /* We aim for 'current allocation' == 'target allocation'. */
- unsigned long current_pages;
- unsigned long target_pages;
- /*
- * Drivers may alter the memory reservation independently, but they
- * must inform the balloon driver so we avoid hitting the hard limit.
- */
- unsigned long driver_pages;
- /* Number of pages in high- and low-memory balloons. */
- unsigned long balloon_low;
- unsigned long balloon_high;
+enum bp_state {
+ BP_DONE,
+ BP_EAGAIN,
+ BP_ECANCELED
};
-static DEFINE_MUTEX(balloon_mutex);
-
-static struct sys_device balloon_sysdev;
-static int register_balloon(struct sys_device *sysdev);
+static DEFINE_MUTEX(balloon_mutex);
-static struct balloon_stats balloon_stats;
+struct balloon_stats balloon_stats;
+EXPORT_SYMBOL_GPL(balloon_stats);
/* We increase/decrease in batches which fit in a page */
static unsigned long frame_list[PAGE_SIZE / sizeof(unsigned long)];
/* Main work function, always executed in process context. */
static void balloon_process(struct work_struct *work);
-static DECLARE_WORK(balloon_worker, balloon_process);
-static struct timer_list balloon_timer;
+static DECLARE_DELAYED_WORK(balloon_worker, balloon_process);
/* When ballooning out (allocating memory to return to Xen) we don't really
want the kernel to try too hard since that can trigger the oom killer. */
}
/* balloon_retrieve: rescue a page from the balloon, if it is not empty. */
-static struct page *balloon_retrieve(void)
+static struct page *balloon_retrieve(bool prefer_highmem)
{
struct page *page;
if (list_empty(&ballooned_pages))
return NULL;
- page = list_entry(ballooned_pages.next, struct page, lru);
+ if (prefer_highmem)
+ page = list_entry(ballooned_pages.prev, struct page, lru);
+ else
+ page = list_entry(ballooned_pages.next, struct page, lru);
list_del(&page->lru);
if (PageHighMem(page)) {
return list_entry(next, struct page, lru);
}
-static void balloon_alarm(unsigned long unused)
+static enum bp_state update_schedule(enum bp_state state)
{
- schedule_work(&balloon_worker);
+ if (state == BP_DONE) {
+ balloon_stats.schedule_delay = 1;
+ balloon_stats.retry_count = 1;
+ return BP_DONE;
+ }
+
+ ++balloon_stats.retry_count;
+
+ if (balloon_stats.max_retry_count != RETRY_UNLIMITED &&
+ balloon_stats.retry_count > balloon_stats.max_retry_count) {
+ balloon_stats.schedule_delay = 1;
+ balloon_stats.retry_count = 1;
+ return BP_ECANCELED;
+ }
+
+ balloon_stats.schedule_delay <<= 1;
+
+ if (balloon_stats.schedule_delay > balloon_stats.max_schedule_delay)
+ balloon_stats.schedule_delay = balloon_stats.max_schedule_delay;
+
+ return BP_EAGAIN;
}
static unsigned long current_target(void)
return target;
}
-static int increase_reservation(unsigned long nr_pages)
+static enum bp_state increase_reservation(unsigned long nr_pages)
{
+ int rc;
unsigned long pfn, i;
struct page *page;
- long rc;
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
page = balloon_first_page();
for (i = 0; i < nr_pages; i++) {
- BUG_ON(page == NULL);
+ if (!page) {
+ nr_pages = i;
+ break;
+ }
frame_list[i] = page_to_pfn(page);
page = balloon_next_page(page);
}
set_xen_guest_handle(reservation.extent_start, frame_list);
reservation.nr_extents = nr_pages;
rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
- if (rc < 0)
- goto out;
+ if (rc <= 0)
+ return BP_EAGAIN;
for (i = 0; i < rc; i++) {
- page = balloon_retrieve();
+ page = balloon_retrieve(false);
BUG_ON(page == NULL);
pfn = page_to_pfn(page);
balloon_stats.current_pages += rc;
- out:
- return rc < 0 ? rc : rc != nr_pages;
+ return BP_DONE;
}
-static int decrease_reservation(unsigned long nr_pages)
+static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp)
{
+ enum bp_state state = BP_DONE;
unsigned long pfn, i;
struct page *page;
- int need_sleep = 0;
int ret;
struct xen_memory_reservation reservation = {
.address_bits = 0,
nr_pages = ARRAY_SIZE(frame_list);
for (i = 0; i < nr_pages; i++) {
- if ((page = alloc_page(GFP_BALLOON)) == NULL) {
+ if ((page = alloc_page(gfp)) == NULL) {
nr_pages = i;
- need_sleep = 1;
+ state = BP_EAGAIN;
break;
}
balloon_stats.current_pages -= nr_pages;
- return need_sleep;
+ return state;
}
/*
*/
static void balloon_process(struct work_struct *work)
{
- int need_sleep = 0;
+ enum bp_state state = BP_DONE;
long credit;
mutex_lock(&balloon_mutex);
do {
credit = current_target() - balloon_stats.current_pages;
+
if (credit > 0)
- need_sleep = (increase_reservation(credit) != 0);
+ state = increase_reservation(credit);
+
if (credit < 0)
- need_sleep = (decrease_reservation(-credit) != 0);
+ state = decrease_reservation(-credit, GFP_BALLOON);
+
+ state = update_schedule(state);
#ifndef CONFIG_PREEMPT
if (need_resched())
schedule();
#endif
- } while ((credit != 0) && !need_sleep);
+ } while (credit && state == BP_DONE);
/* Schedule more work if there is some still to be done. */
- if (current_target() != balloon_stats.current_pages)
- mod_timer(&balloon_timer, jiffies + HZ);
+ if (state == BP_EAGAIN)
+ schedule_delayed_work(&balloon_worker, balloon_stats.schedule_delay * HZ);
mutex_unlock(&balloon_mutex);
}
/* Resets the Xen limit, sets new target, and kicks off processing. */
-static void balloon_set_new_target(unsigned long target)
+void balloon_set_new_target(unsigned long target)
{
/* No need for lock. Not read-modify-write updates. */
balloon_stats.target_pages = target;
- schedule_work(&balloon_worker);
+ schedule_delayed_work(&balloon_worker, 0);
}
+EXPORT_SYMBOL_GPL(balloon_set_new_target);
-static struct xenbus_watch target_watch =
-{
- .node = "memory/target"
-};
-
-/* React to a change in the target key */
-static void watch_target(struct xenbus_watch *watch,
- const char **vec, unsigned int len)
+/**
+ * alloc_xenballooned_pages - get pages that have been ballooned out
+ * @nr_pages: Number of pages to get
+ * @pages: pages returned
+ * @return 0 on success, error otherwise
+ */
+int alloc_xenballooned_pages(int nr_pages, struct page** pages)
{
- unsigned long long new_target;
- int err;
-
- err = xenbus_scanf(XBT_NIL, "memory", "target", "%llu", &new_target);
- if (err != 1) {
- /* This is ok (for domain0 at least) - so just return */
- return;
+ int pgno = 0;
+ struct page* page;
+ mutex_lock(&balloon_mutex);
+ while (pgno < nr_pages) {
+ page = balloon_retrieve(true);
+ if (page) {
+ pages[pgno++] = page;
+ } else {
+ enum bp_state st;
+ st = decrease_reservation(nr_pages - pgno, GFP_HIGHUSER);
+ if (st != BP_DONE)
+ goto out_undo;
+ }
}
-
- /* The given memory/target value is in KiB, so it needs converting to
- * pages. PAGE_SHIFT converts bytes to pages, hence PAGE_SHIFT - 10.
- */
- balloon_set_new_target(new_target >> (PAGE_SHIFT - 10));
+ mutex_unlock(&balloon_mutex);
+ return 0;
+ out_undo:
+ while (pgno)
+ balloon_append(pages[--pgno]);
+ /* Free the memory back to the kernel soon */
+ schedule_delayed_work(&balloon_worker, 0);
+ mutex_unlock(&balloon_mutex);
+ return -ENOMEM;
}
+EXPORT_SYMBOL(alloc_xenballooned_pages);
-static int balloon_init_watcher(struct notifier_block *notifier,
- unsigned long event,
- void *data)
+/**
+ * free_xenballooned_pages - return pages retrieved with get_ballooned_pages
+ * @nr_pages: Number of pages
+ * @pages: pages to return
+ */
+void free_xenballooned_pages(int nr_pages, struct page** pages)
{
- int err;
+ int i;
- err = register_xenbus_watch(&target_watch);
- if (err)
- printk(KERN_ERR "Failed to set balloon watcher\n");
+ mutex_lock(&balloon_mutex);
- return NOTIFY_DONE;
-}
+ for (i = 0; i < nr_pages; i++) {
+ if (pages[i])
+ balloon_append(pages[i]);
+ }
+
+ /* The balloon may be too large now. Shrink it if needed. */
+ if (current_target() != balloon_stats.current_pages)
+ schedule_delayed_work(&balloon_worker, 0);
-static struct notifier_block xenstore_notifier;
+ mutex_unlock(&balloon_mutex);
+}
+EXPORT_SYMBOL(free_xenballooned_pages);
static int __init balloon_init(void)
{
if (!xen_domain())
return -ENODEV;
- pr_info("xen_balloon: Initialising balloon driver.\n");
+ pr_info("xen/balloon: Initialising balloon driver.\n");
if (xen_pv_domain())
nr_pages = xen_start_info->nr_pages;
balloon_stats.target_pages = balloon_stats.current_pages;
balloon_stats.balloon_low = 0;
balloon_stats.balloon_high = 0;
- balloon_stats.driver_pages = 0UL;
-
- init_timer(&balloon_timer);
- balloon_timer.data = 0;
- balloon_timer.function = balloon_alarm;
- register_balloon(&balloon_sysdev);
+ balloon_stats.schedule_delay = 1;
+ balloon_stats.max_schedule_delay = 32;
+ balloon_stats.retry_count = 1;
+ balloon_stats.max_retry_count = RETRY_UNLIMITED;
/*
* Initialise the balloon with excess memory space. We need
__balloon_append(page);
}
- target_watch.callback = watch_target;
- xenstore_notifier.notifier_call = balloon_init_watcher;
-
- register_xenstore_notifier(&xenstore_notifier);
-
return 0;
}
subsys_initcall(balloon_init);
-static void balloon_exit(void)
-{
- /* XXX - release balloon here */
- return;
-}
-
-module_exit(balloon_exit);
-
-#define BALLOON_SHOW(name, format, args...) \
- static ssize_t show_##name(struct sys_device *dev, \
- struct sysdev_attribute *attr, \
- char *buf) \
- { \
- return sprintf(buf, format, ##args); \
- } \
- static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL)
-
-BALLOON_SHOW(current_kb, "%lu\n", PAGES2KB(balloon_stats.current_pages));
-BALLOON_SHOW(low_kb, "%lu\n", PAGES2KB(balloon_stats.balloon_low));
-BALLOON_SHOW(high_kb, "%lu\n", PAGES2KB(balloon_stats.balloon_high));
-BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(balloon_stats.driver_pages));
-
-static ssize_t show_target_kb(struct sys_device *dev, struct sysdev_attribute *attr,
- char *buf)
-{
- return sprintf(buf, "%lu\n", PAGES2KB(balloon_stats.target_pages));
-}
-
-static ssize_t store_target_kb(struct sys_device *dev,
- struct sysdev_attribute *attr,
- const char *buf,
- size_t count)
-{
- char *endchar;
- unsigned long long target_bytes;
-
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
-
- target_bytes = simple_strtoull(buf, &endchar, 0) * 1024;
-
- balloon_set_new_target(target_bytes >> PAGE_SHIFT);
-
- return count;
-}
-
-static SYSDEV_ATTR(target_kb, S_IRUGO | S_IWUSR,
- show_target_kb, store_target_kb);
-
-
-static ssize_t show_target(struct sys_device *dev, struct sysdev_attribute *attr,
- char *buf)
-{
- return sprintf(buf, "%llu\n",
- (unsigned long long)balloon_stats.target_pages
- << PAGE_SHIFT);
-}
-
-static ssize_t store_target(struct sys_device *dev,
- struct sysdev_attribute *attr,
- const char *buf,
- size_t count)
-{
- char *endchar;
- unsigned long long target_bytes;
-
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
-
- target_bytes = memparse(buf, &endchar);
-
- balloon_set_new_target(target_bytes >> PAGE_SHIFT);
-
- return count;
-}
-
-static SYSDEV_ATTR(target, S_IRUGO | S_IWUSR,
- show_target, store_target);
-
-
-static struct sysdev_attribute *balloon_attrs[] = {
- &attr_target_kb,
- &attr_target,
-};
-
-static struct attribute *balloon_info_attrs[] = {
- &attr_current_kb.attr,
- &attr_low_kb.attr,
- &attr_high_kb.attr,
- &attr_driver_kb.attr,
- NULL
-};
-
-static struct attribute_group balloon_info_group = {
- .name = "info",
- .attrs = balloon_info_attrs,
-};
-
-static struct sysdev_class balloon_sysdev_class = {
- .name = BALLOON_CLASS_NAME,
-};
-
-static int register_balloon(struct sys_device *sysdev)
-{
- int i, error;
-
- error = sysdev_class_register(&balloon_sysdev_class);
- if (error)
- return error;
-
- sysdev->id = 0;
- sysdev->cls = &balloon_sysdev_class;
-
- error = sysdev_register(sysdev);
- if (error) {
- sysdev_class_unregister(&balloon_sysdev_class);
- return error;
- }
-
- for (i = 0; i < ARRAY_SIZE(balloon_attrs); i++) {
- error = sysdev_create_file(sysdev, balloon_attrs[i]);
- if (error)
- goto fail;
- }
-
- error = sysfs_create_group(&sysdev->kobj, &balloon_info_group);
- if (error)
- goto fail;
-
- return 0;
-
- fail:
- while (--i >= 0)
- sysdev_remove_file(sysdev, balloon_attrs[i]);
- sysdev_unregister(sysdev);
- sysdev_class_unregister(&balloon_sysdev_class);
- return error;
-}
-
MODULE_LICENSE("GPL");
git.openpandora.org / pandora-kernel.git / blobdiff