[patch 3/10] Allow truncate to discard unmapped buffers

Andrew Morton (akpm@zip.com.au)
Sun, 05 May 2002 13:54:28 -0700


The buffer state "uptodate, dirty, unmapped" is legal. It occurs when
a mapped page with attached buffers which is over a hole is dirtied.

So discard_buffer() needs to be able to discard those buffers as well;
otherwise the page ends up clean, with dirty buffers and is unfreeable.

=====================================

--- 2.5.13/fs/buffer.c~truncate-leak Sun May 5 13:32:00 2002
+++ 2.5.13-akpm/fs/buffer.c Sun May 5 13:32:37 2002
@@ -1091,17 +1091,15 @@ EXPORT_SYMBOL(set_bh_page);
/*
* Called when truncating a buffer on a page completely.
*/
-static void discard_buffer(struct buffer_head * bh)
+static /* inline */ void discard_buffer(struct buffer_head * bh)
{
- if (buffer_mapped(bh)) {
- clear_buffer_dirty(bh);
- lock_buffer(bh);
- bh->b_bdev = NULL;
- clear_buffer_mapped(bh);
- clear_buffer_req(bh);
- clear_buffer_new(bh);
- unlock_buffer(bh);
- }
+ lock_buffer(bh);
+ clear_buffer_dirty(bh);
+ bh->b_bdev = NULL;
+ clear_buffer_mapped(bh);
+ clear_buffer_req(bh);
+ clear_buffer_new(bh);
+ unlock_buffer(bh);
}

/**
--- 2.5.13/fs/jbd/transaction.c~truncate-leak Sun May 5 13:32:00 2002
+++ 2.5.13-akpm/fs/jbd/transaction.c Sun May 5 13:32:37 2002
@@ -1587,7 +1587,7 @@ static int __journal_try_to_free_buffer(
goto out;
}

- if (!buffer_uptodate(bh))
+ if (!buffer_uptodate(bh)) /* AKPM: why? */
goto out;

if (jh->b_next_transaction != 0)
@@ -1775,9 +1775,6 @@ static int journal_unmap_buffer(journal_

BUFFER_TRACE(bh, "entry");

- if (!buffer_mapped(bh))
- return 1;
-
/* It is safe to proceed here without the
* journal_datalist_spinlock because the buffers cannot be
* stolen by try_to_free_buffers as long as we are holding the
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/