The patch titled
mm: non syncing lock_page()
has been added to the -mm tree. Its filename is
mm-non-syncing-lock_page.patch
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: mm: non syncing lock_page()
From: Nick Piggin <npiggin@xxxxxxx>
lock_page needs the caller to have a reference on the page->mapping inode
due to sync_page, ergo set_page_dirty_lock is obviously buggy according to
its comments.
Solve it by introducing a new lock_page_nosync which does not do a sync_page.
akpm: unpleasant solution to an unpleasant problem. If it goes wrong it could
cause great slowdowns while the lock_page() caller waits for kblockd to
perform the unplug. And if a filesystem has special sync_page() requirements
(none presently do), permanent hangs are possible.
otoh, set_page_dirty_lock() is usually (always?) called against userspace
pages. They are always up-to-date, so there shouldn't be any pending read I/O
against these pages.
Signed-off-by: Nick Piggin <npiggin@xxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxx>
---
include/linux/pagemap.h | 15 +++++++++++++++
mm/filemap.c | 17 +++++++++++++++++
mm/page-writeback.c | 2 +-
3 files changed, 33 insertions(+), 1 deletion(-)
diff -puN include/linux/pagemap.h~mm-non-syncing-lock_page include/linux/pagemap.h
--- a/include/linux/pagemap.h~mm-non-syncing-lock_page
+++ a/include/linux/pagemap.h
@@ -130,14 +130,29 @@ static inline pgoff_t linear_page_index(
}
extern void FASTCALL(__lock_page(struct page *page));
+extern void FASTCALL(__lock_page_nosync(struct page *page));
extern void FASTCALL(unlock_page(struct page *page));
+/*
+ * lock_page may only be called if we have the page's inode pinned.
+ */
static inline void lock_page(struct page *page)
{
might_sleep();
if (TestSetPageLocked(page))
__lock_page(page);
}
+
+/*
+ * lock_page_nosync should only be used if we can't pin the page's inode.
+ * Doesn't play quite so well with block device plugging.
+ */
+static inline void lock_page_nosync(struct page *page)
+{
+ might_sleep();
+ if (TestSetPageLocked(page))
+ __lock_page_nosync(page);
+}
/*
* This is exported only for wait_on_page_locked/wait_on_page_writeback.
diff -puN mm/filemap.c~mm-non-syncing-lock_page mm/filemap.c
--- a/mm/filemap.c~mm-non-syncing-lock_page
+++ a/mm/filemap.c
@@ -488,6 +488,12 @@ struct page *page_cache_alloc_cold(struc
EXPORT_SYMBOL(page_cache_alloc_cold);
#endif
+static int __sleep_on_page_lock(void *word)
+{
+ io_schedule();
+ return 0;
+}
+
/*
* In order to wait for pages to become available there must be
* waitqueues associated with pages. By using a hash table of
@@ -577,6 +583,17 @@ void fastcall __lock_page(struct page *p
}
EXPORT_SYMBOL(__lock_page);
+/*
+ * Variant of lock_page that does not require the caller to hold a reference
+ * on the page's mapping.
+ */
+void fastcall __lock_page_nosync(struct page *page)
+{
+ DEFINE_WAIT_BIT(wait, &page->flags, PG_locked);
+ __wait_on_bit_lock(page_waitqueue(page), &wait, __sleep_on_page_lock,
+ TASK_UNINTERRUPTIBLE);
+}
+
/**
* find_get_page - find and get a page reference
* @mapping: the address_space to search
diff -puN mm/page-writeback.c~mm-non-syncing-lock_page mm/page-writeback.c
--- a/mm/page-writeback.c~mm-non-syncing-lock_page
+++ a/mm/page-writeback.c
@@ -701,7 +701,7 @@ int set_page_dirty_lock(struct page *pag
{
int ret;
- lock_page(page);
+ lock_page_nosync(page);
ret = set_page_dirty(page);
unlock_page(page);
return ret;
_
Patches currently in -mm which might be from npiggin@xxxxxxx are
mm-vm_bug_on.patch
radix-tree-rcu-lockless-readside.patch
redo-radix-tree-fixes.patch
adix-tree-rcu-lockless-readside-update.patch
adix-tree-rcu-lockless-readside-fix-2.patch
update-some-mm-comments.patch
mm-remove_mapping-safeness.patch
mm-non-syncing-lock_page.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
