Re: [RFC PATCH 01/19] lockdep: annotate reclaim context(__GFP_NOFS)
From: Peter Zijlstra
Date: Thu Jan 22 2009 - 14:40:51 EST
On Thu, 2009-01-22 at 18:37 +0100, Peter Zijlstra wrote:
>
> +void lockdep_trace_alloc(gfp_t gfp_mask)
> +{
> + struct task_struct *curr = current;
> +
> + /* this guy won't enter reclaim */
> + if (curr->flags & PF_MEMALLOC)
> + return;
> +
> + /* We're only interested __GFP_FS allocations for now */
> + if (!(gfp_mask & __GFP_FS))
> + return;
> +
> + if (unlikely(!debug_locks))
> + return;
> + if (DEBUG_LOCKS_WARN_ON(irqs_disabled()))
> + return;
> +
> + mark_held_locks(curr, RECLAIM_FS);
> +}
Not quite seeing how this happens though...
[ 6.998258] WARNING: at /usr/src/linux-2.6/kernel/lockdep.c:2266 lockdep_trace_alloc+0x58/0x69()
[ 7.003081] Hardware name: To Be Filled By O.E.M.
[ 7.003081] Modules linked in: shpchp
[ 7.003081] Pid: 726, comm: grep Not tainted 2.6.29-rc2-tip #53
[ 7.003081] Call Trace:
[ 7.003081] [<ffffffff8023fcc0>] warn_slowpath+0xd8/0x114
[ 7.003081] [<ffffffff80235dc6>] ? enqueue_task_fair+0x235/0x286
[ 7.003081] [<ffffffff80261d7f>] ? trace_hardirqs_off+0xd/0xf
[ 7.003081] [<ffffffff8054fc8a>] ? _spin_unlock_irqrestore+0x3d/0x4c
[ 7.003081] [<ffffffff8023c1c6>] ? try_to_wake_up+0x2ba/0x2cc
[ 7.003081] [<ffffffff80262c05>] ? mark_held_locks+0x8a/0xa9
[ 7.003081] [<ffffffff8054fc91>] ? _spin_unlock_irqrestore+0x44/0x4c
[ 7.003081] [<ffffffff80262e67>] ? trace_hardirqs_on+0xd/0xf
[ 7.003081] [<ffffffff802333cd>] ? __wake_up+0x48/0x55
[ 7.003081] [<ffffffff80262c7c>] lockdep_trace_alloc+0x58/0x69
[ 7.003081] [<ffffffff8029e34f>] __alloc_pages_internal+0x72/0x439
[ 7.003081] [<ffffffff802e6f7e>] ? unlock_buffer+0x17/0x19
[ 7.003081] [<ffffffff802e9117>] ? __bread+0x45/0x92
[ 7.003081] [<ffffffff803201cf>] ? ext3_get_branch+0x7b/0xee
[ 7.003081] [<ffffffff8029aae8>] ? mempool_alloc_slab+0x16/0x18
[ 7.003081] [<ffffffff802c1aee>] __slab_alloc+0x1a2/0x569
[ 7.003081] [<ffffffff8029aae8>] ? mempool_alloc_slab+0x16/0x18
[ 7.003081] [<ffffffff80262c05>] ? mark_held_locks+0x8a/0xa9
[ 7.003081] [<ffffffff8029aae8>] ? mempool_alloc_slab+0x16/0x18
[ 7.003081] [<ffffffff802c1f30>] kmem_cache_alloc+0x7b/0xbd
[ 7.003081] [<ffffffff8029aae8>] mempool_alloc_slab+0x16/0x18
[ 7.003081] [<ffffffff8029abce>] mempool_alloc+0x4f/0x107
[ 7.003081] [<ffffffff802ea151>] bio_alloc_bioset+0x30/0x106
[ 7.003081] [<ffffffff802ea2a0>] bio_alloc+0x15/0x24
[ 7.003081] [<ffffffff802ed6cf>] mpage_alloc+0x2b/0x86
[ 7.003081] [<ffffffff802ee15c>] do_mpage_readpage+0x3f8/0x4eb
[ 7.003081] [<ffffffff80262e67>] ? trace_hardirqs_on+0xd/0xf
[ 7.003081] [<ffffffff80298ed9>] ? add_to_page_cache_locked+0x9f/0xae
[ 7.003081] [<ffffffff802ee37a>] mpage_readpages+0xd4/0x117
[ 7.003081] [<ffffffff803217ab>] ? ext3_get_block+0x0/0x101
[ 7.003081] [<ffffffff803217ab>] ? ext3_get_block+0x0/0x101
[ 7.003081] [<ffffffff8029e3d7>] ? __alloc_pages_internal+0xfa/0x439
[ 7.003081] [<ffffffff80320e75>] ext3_readpages+0x1f/0x21
[ 7.003081] [<ffffffff802a0573>] __do_page_cache_readahead+0x165/0x1fe
[ 7.003081] [<ffffffff802a0481>] ? __do_page_cache_readahead+0x73/0x1fe
[ 7.003081] [<ffffffff80298a37>] ? find_get_page+0x0/0xcf
[ 7.003081] [<ffffffff802a0667>] do_page_cache_readahead+0x5b/0x68
[ 7.003081] [<ffffffff802992f3>] filemap_fault+0x16f/0x341
[ 7.003081] [<ffffffff802aac3a>] __do_fault+0x59/0x3c5
[ 7.003081] [<ffffffff802acdf4>] handle_mm_fault+0x318/0x6a8
[ 7.003081] [<ffffffff80552337>] do_page_fault+0x1fc/0x213
[ 7.003081] [<ffffffff8055031f>] page_fault+0x1f/0x30
[ 7.003081] ---[ end trace f2b79ed5a3eafb2f ]---
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/