[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <alpine.DEB.2.00.1004081209380.21040@chino.kir.corp.google.com>
Date: Thu, 8 Apr 2010 12:12:09 -0700 (PDT)
From: David Rientjes <rientjes@...gle.com>
To: Xiaotian Feng <dfeng@...hat.com>
cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Christoph Lameter <cl@...ux-foundation.org>,
Pekka Enberg <penberg@...helsinki.fi>,
Matt Mackall <mpm@...enic.com>,
Vegard Nossum <vegard.nossum@...il.com>,
Dmitry Monakhov <dmonakhov@...nvz.org>,
Catalin Marinas <catalin.marinas@....com>
Subject: Re: [PATCH] slab: fix caller tracking on !CONFIG_DEBUG_SLAB &&
CONFIG_TRACING
On Thu, 8 Apr 2010, Xiaotian Feng wrote:
> diff --git a/include/linux/slab.h b/include/linux/slab.h
> index 4884462..1a0625c 100644
> --- a/include/linux/slab.h
> +++ b/include/linux/slab.h
> @@ -267,7 +267,7 @@ static inline void *kmem_cache_alloc_node(struct kmem_cache *cachep,
> * allocator where we care about the real place the memory allocation
> * request comes from.
> */
> -#if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_SLUB)
> +#if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_SLUB) || defined(CONFIG_TRACING)
> extern void *__kmalloc_track_caller(size_t, gfp_t, unsigned long);
> #define kmalloc_track_caller(size, flags) \
> __kmalloc_track_caller(size, flags, _RET_IP_)
> @@ -285,7 +285,7 @@ extern void *__kmalloc_track_caller(size_t, gfp_t, unsigned long);
> * standard allocator where we care about the real place the memory
> * allocation request comes from.
> */
> -#if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_SLUB)
> +#if defined(CONFIG_DEBUG_SLAB) || defined(CONFIG_SLUB) || defined(CONFIG_TRACING)
> extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, unsigned long);
> #define kmalloc_node_track_caller(size, flags, node) \
> __kmalloc_node_track_caller(size, flags, node, \
This doesn't work if the underlying slab allocator doesn't define
__kmalloc_node_track_caller() regardless of whether CONFIG_TRACING is
enabled or not. SLOB, for example, never defines it, and that's why the
conditional exists in the way it currently does.
This is your patch with CONFIG_EMBEDDED && CONFIG_SLOB:
mm/built-in.o: In function `__krealloc':
(.text+0x1283c): undefined reference to `__kmalloc_track_caller'
mm/built-in.o: In function `kmemdup':
(.text+0x128b4): undefined reference to `__kmalloc_track_caller'
mm/built-in.o: In function `kstrndup':
(.text+0x128fc): undefined reference to `__kmalloc_track_caller'
mm/built-in.o: In function `kstrdup':
(.text+0x12943): undefined reference to `__kmalloc_track_caller'
mm/built-in.o: In function `memdup_user':
(.text+0x129f7): undefined reference to `__kmalloc_track_caller'
drivers/built-in.o:(.text+0xc48a4): more undefined references to `__kmalloc_track_caller' follow
net/built-in.o: In function `__alloc_skb':
(.text+0x8dc6): undefined reference to `__kmalloc_node_track_caller'
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists