ceph: don't truncate dirty pages in invalidate work thread
Instead of truncating the whole range of pages, we skip those
pages that are dirty or in the middle of writeback. Those pages
will be cleared later when the writeback completes.
Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index d7d5d49..7abe1ae 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -10,6 +10,7 @@
#include <linux/namei.h>
#include <linux/writeback.h>
#include <linux/vmalloc.h>
+#include <linux/pagevec.h>
#include "super.h"
#include "decode.h"
@@ -1280,6 +1281,49 @@
}
/*
+ * invalidate any pages that are not dirty or under writeback. this
+ * includes pages that are clean and mapped.
+ */
+static void ceph_invalidate_nondirty_pages(struct address_space *mapping)
+{
+ struct pagevec pvec;
+ pgoff_t next = 0;
+ int i;
+
+ pagevec_init(&pvec, 0);
+ while (pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
+ for (i = 0; i < pagevec_count(&pvec); i++) {
+ struct page *page = pvec.pages[i];
+ pgoff_t index;
+ int skip_page =
+ (PageDirty(page) || PageWriteback(page));
+
+ if (!skip_page)
+ skip_page = !trylock_page(page);
+
+ /*
+ * We really shouldn't be looking at the ->index of an
+ * unlocked page. But we're not allowed to lock these
+ * pages. So we rely upon nobody altering the ->index
+ * of this (pinned-by-us) page.
+ */
+ index = page->index;
+ if (index > next)
+ next = index;
+ next++;
+
+ if (skip_page)
+ continue;
+
+ generic_error_remove_page(mapping, page);
+ unlock_page(page);
+ }
+ pagevec_release(&pvec);
+ cond_resched();
+ }
+}
+
+/*
* Invalidate inode pages in a worker thread. (This can't be done
* in the message handler context.)
*/
@@ -1305,7 +1349,7 @@
orig_gen = ci->i_rdcache_gen;
spin_unlock(&inode->i_lock);
- truncate_inode_pages(&inode->i_data, 0);
+ ceph_invalidate_nondirty_pages(inode->i_mapping);
spin_lock(&inode->i_lock);
if (orig_gen == ci->i_rdcache_gen) {