Messages in this thread | | | Date | Wed, 21 Dec 2011 12:28:43 -0800 | From | Andrew Morton <> | Subject | Re: [PATCH v2] vfs: __read_cache_page should use gfp argument rather than GFP_KERNEL |
| |
On Wed, 21 Dec 2011 11:05:48 -0600 Dave Kleikamp <dave.kleikamp@oracle.com> wrote:
> [ updated to remove now-obsolete comment in read_cache_page_gfp()] > > lockdep reports a deadlock in jfs because a special inode's rw semaphore > is taken recursively. The mapping's gfp mask is GFP_NOFS, but is not used > when __read_cache_page() calls add_to_page_cache_lru().
Well hang on, it's not just a lockdep splat. The kernel actually will deadlock if we reenter JFS via this GFP_KERNEL allocation attempt, yes?
Was that GFP_NOFS allocation recently added to JFS? If not then we should backport this deadlock fix into -stable, no?
| |