Btrfs: Don't try to compress pages past i_size

The compression code had some checks to make sure we were only
compressing bytes inside of i_size, but it wasn't catching every
case.  To make things worse, some incorrect math about the number
of bytes remaining would make it try to compress more pages than the
file really had.

The fix used here is to fall back to the non-compression code in this
case, which does all the proper cleanup of delalloc and other accounting.

Signed-off-by: Chris Mason <chris.mason@oracle.com>

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5792816d4fca8f40c5be6dad4945b8f6dfc29b38..9b43a6f303b8ee8e1fbfd2f15e5230316a3b300e 100644 (file)
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -360,6 +360,19 @@ again:
        nr_pages = (end >> PAGE_CACHE_SHIFT) - (start >> PAGE_CACHE_SHIFT) + 1;
        nr_pages = min(nr_pages, (128 * 1024UL) / PAGE_CACHE_SIZE);
 
+       /*
+        * we don't want to send crud past the end of i_size through
+        * compression, that's just a waste of CPU time.  So, if the
+        * end of the file is before the start of our current
+        * requested range of bytes, we bail out to the uncompressed
+        * cleanup code that can deal with all of this.
+        *
+        * It isn't really the fastest way to fix things, but this is a
+        * very uncommon corner.
+        */
+       if (actual_end <= start)
+               goto cleanup_and_bail_uncompressed;
+
        total_compressed = actual_end - start;
 
        /* we want to make sure that amount of ram required to uncompress
@@ -504,6 +517,7 @@ again:
                        goto again;
                }
        } else {
+cleanup_and_bail_uncompressed:
                /*
                 * No compression, but we still need to write the pages in
                 * the file we've been given so far.  redirty the locked