[TRIVIAL] silence invalidate_bdev() a bit

Rusty Trivial Russell (rusty@rustcorp.com.au)
Tue, 20 Aug 2002 18:27:52 +1000


[ Experts, does this make sense? The Debian kernel tree makes it a
KERN_DEBUG printk instead. ]

From: Christoph Hellwig <hch@lst.de>

The 2.4.1x invalidate_bdev() is a little to verbose and warns about
conditions that can easily happen in practice. Silence up those
printks a little like most vendor trees already do.

--- trivial-2.4.20-pre4/fs/buffer.c.orig 2002-08-20 18:00:33.000000000 +1000
+++ trivial-2.4.20-pre4/fs/buffer.c 2002-08-20 18:00:33.000000000 +1000
@@ -695,13 +695,13 @@
/* All buffers in the lru lists are mapped */
if (!buffer_mapped(bh))
BUG();
- if (buffer_dirty(bh))
+ if (buffer_dirty(bh) && destroy_dirty_buffers)
printk("invalidate: dirty buffer\n");
if (!atomic_read(&bh->b_count)) {
if (destroy_dirty_buffers || !buffer_dirty(bh)) {
remove_inode_queue(bh);
}
- } else
+ } else if (!bdev->bd_openers)
printk("invalidate: busy buffer\n");

write_unlock(&hash_table_lock);

-- 
  Don't blame me: the Monkey is driving
  File: Christoph Hellwig <hch@lst.de>: [PATCH] silence invalidate_bdev() a bit
-
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/