xen/gnttab: add deferred freeing logic

Rather than just leaking pages that can't be freed at the point where
access permission for the backend domain gets revoked, put them on a
list and run a timer to (infrequently) retry freeing them. (This can
particularly happen when unloading a frontend driver when devices are
still present, and the backend still has them in non-closed state or
hasn't finished closing them yet.)

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

1 files changed, 96 insertions, 10 deletions

diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index b4d4eac761d..9f514bb561a 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -426,10 +426,8 @@ static int gnttab_end_foreign_access_ref_v1(grant_ref_t ref, int readonly)
         nflags = *pflags;
         do {
                 flags = nflags;
-                if (flags & (GTF_reading|GTF_writing)) {
-                        printk(KERN_ALERT "WARNING: g.e. still in use!\n");
+                if (flags & (GTF_reading|GTF_writing))
                         return 0;
-                }
         } while ((nflags = sync_cmpxchg(pflags, flags, 0)) != flags);
 
         return 1;
@@ -458,12 +456,103 @@ static int gnttab_end_foreign_access_ref_v2(grant_ref_t ref, int readonly)
         return 1;
 }
 
-int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly)
+static inline int _gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly)
 {
         return gnttab_interface->end_foreign_access_ref(ref, readonly);
 }
+
+int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly)
+{
+        if (_gnttab_end_foreign_access_ref(ref, readonly))
+                return 1;
+        pr_warn("WARNING: g.e. %#x still in use!\n", ref);
+        return 0;
+}
 EXPORT_SYMBOL_GPL(gnttab_end_foreign_access_ref);
 
+struct deferred_entry {
+        struct list_head list;
+        grant_ref_t ref;
+        bool ro;
+        uint16_t warn_delay;
+        struct page *page;
+};
+static LIST_HEAD(deferred_list);
+static void gnttab_handle_deferred(unsigned long);
+static DEFINE_TIMER(deferred_timer, gnttab_handle_deferred, 0, 0);
+
+static void gnttab_handle_deferred(unsigned long unused)
+{
+        unsigned int nr = 10;
+        struct deferred_entry *first = NULL;
+        unsigned long flags;
+
+        spin_lock_irqsave(&gnttab_list_lock, flags);
+        while (nr--) {
+                struct deferred_entry *entry
+                        = list_first_entry(&deferred_list,
+                                           struct deferred_entry, list);
+
+                if (entry == first)
+                        break;
+                list_del(&entry->list);
+                spin_unlock_irqrestore(&gnttab_list_lock, flags);
+                if (_gnttab_end_foreign_access_ref(entry->ref, entry->ro)) {
+                        put_free_entry(entry->ref);
+                        if (entry->page) {
+                                pr_debug("freeing g.e. %#x (pfn %#lx)\n",
+                                         entry->ref, page_to_pfn(entry->page));
+                                __free_page(entry->page);
+                        } else
+                                pr_info("freeing g.e. %#x\n", entry->ref);
+                        kfree(entry);
+                        entry = NULL;
+                } else {
+                        if (!--entry->warn_delay)
+                                pr_info("g.e. %#x still pending\n",
+                                        entry->ref);
+                        if (!first)
+                                first = entry;
+                }
+                spin_lock_irqsave(&gnttab_list_lock, flags);
+                if (entry)
+                        list_add_tail(&entry->list, &deferred_list);
+                else if (list_empty(&deferred_list))
+                        break;
+        }
+        if (!list_empty(&deferred_list) && !timer_pending(&deferred_timer)) {
+                deferred_timer.expires = jiffies + HZ;
+                add_timer(&deferred_timer);
+        }
+        spin_unlock_irqrestore(&gnttab_list_lock, flags);
+}
+
+static void gnttab_add_deferred(grant_ref_t ref, bool readonly,
+                                struct page *page)
+{
+        struct deferred_entry *entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
+        const char *what = KERN_WARNING "leaking";
+
+        if (entry) {
+                unsigned long flags;
+
+                entry->ref = ref;
+                entry->ro = readonly;
+                entry->page = page;
+                entry->warn_delay = 60;
+                spin_lock_irqsave(&gnttab_list_lock, flags);
+                list_add_tail(&entry->list, &deferred_list);
+                if (!timer_pending(&deferred_timer)) {
+                        deferred_timer.expires = jiffies + HZ;
+                        add_timer(&deferred_timer);
+                }
+                spin_unlock_irqrestore(&gnttab_list_lock, flags);
+                what = KERN_DEBUG "deferring";
+        }
+        printk("%s g.e. %#x (pfn %#lx)\n",
+               what, ref, page ? page_to_pfn(page) : -1);
+}
+
 void gnttab_end_foreign_access(grant_ref_t ref, int readonly,
                                unsigned long page)
 {
@@ -471,12 +560,9 @@ void gnttab_end_foreign_access(grant_ref_t ref, int readonly,
                 put_free_entry(ref);
                 if (page != 0)
                         free_page(page);
-        } else {
-                /* XXX This needs to be fixed so that the ref and page are
-                   placed on a list to be freed up later. */
-                printk(KERN_WARNING
-                       "WARNING: leaking g.e. and page still in use!\n");
-        }
+        } else
+                gnttab_add_deferred(ref, readonly,
+                                    page ? virt_to_page(page) : NULL);
 }
 EXPORT_SYMBOL_GPL(gnttab_end_foreign_access);