On 09/03/2012 07:33 PM, Glauber Costa wrote:
> On 08/24/2012 08:17 PM, Christoph Lameter wrote:
>> -__kmem_cache_create (struct kmem_cache *cachep, const char *name, size_t size, size_t align,
>> - unsigned long flags, void (*ctor)(void *))
>> +__kmem_cache_create (struct kmem_cache *cachep, unsigned long flags)
>> {
>> size_t left_over, slab_size, ralign;
>> gfp_t gfp;
>> @@ -2385,9 +2383,9 @@ __kmem_cache_create (struct kmem_cache *
>> * unaligned accesses for some archs when redzoning is used, and makes
>> * sure any on-slab bufctl's are also correctly aligned.
>> */
>> - if (size & (BYTES_PER_WORD - 1)) {
>> - size += (BYTES_PER_WORD - 1);
>> - size &= ~(BYTES_PER_WORD - 1);
>> + if (cachep->size & (BYTES_PER_WORD - 1)) {
>> + cachep->size += (BYTES_PER_WORD - 1);
>> + cachep->size &= ~(BYTES_PER_WORD - 1);
>> }
>
> There are still one reference to "size" inside this function that will
> break the build. This reference is enclosed inside CONFIG_DEBUG.
>
Actually, Christoph, it would be a lot cleaner if you would just do
size_t size = cachep->size;
in the beginning of this function. The resulting patch size would be a
lot smaller since you don't need to patch the references, and would
avoid mistakes like that altogether.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>