Messages in this thread Patch in this message | | | From | Steven Whitehouse <> | Subject | [PATCH 17/32] GFS2: Improve bug trap code in ->releasepage() | Date | Thu, 19 May 2011 09:47:12 +0100 |
| |
If the buffer is dirty or pinned, then as well as printing a warning, we should also refuse to release the page in question.
Currently this can occur if there is a race between mmap()ed writers and O_DIRECT on the same file. With the addition of ->launder_page() in the future, we should be able to close this gap.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 0f5c4f9..802ac5e 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -1076,8 +1076,8 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) bd = bh->b_private; if (bd && bd->bd_ail) goto cannot_release; - gfs2_assert_warn(sdp, !buffer_pinned(bh)); - gfs2_assert_warn(sdp, !buffer_dirty(bh)); + if (buffer_pinned(bh) || buffer_dirty(bh)) + goto not_possible; bh = bh->b_this_page; } while(bh != head); gfs2_log_unlock(sdp); @@ -1107,6 +1107,10 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) } while (bh != head); return try_to_free_buffers(page); + +not_possible: /* Should never happen */ + WARN_ON(buffer_dirty(bh)); + WARN_ON(buffer_pinned(bh)); cannot_release: gfs2_log_unlock(sdp); return 0; -- 1.7.4
| |