On Mon, Jun 29, 2009 at 04:12:34PM +0200, Heiko Carstens wrote:
> slqb returns ZERO_SIZE_PTR instead of NULL for large size requests it cannot
> handle.
Ah, thank you for debugging this. Great.
> The patch below would fix it. But I think its too ugly. So I leave it up to
> Nick to come up with a real and nice patch ;)
The patch isn't so bad :) It is for the constant case anyway so it
should optimize away. But I think I have a similar problem for
the non constant case too, so I will look at doing another patch.
Thanks,
Nick
>
> diff --git a/include/linux/slqb_def.h b/include/linux/slqb_def.h
> index 7b4a601..9d03485 100644
> --- a/include/linux/slqb_def.h
> +++ b/include/linux/slqb_def.h
> @@ -187,7 +187,7 @@ static __always_inline int kmalloc_index(size_t size)
> if (unlikely(!size))
> return 0;
> if (unlikely(size > 1UL << KMALLOC_SHIFT_SLQB_HIGH))
> - return 0;
> + return -1;
>
> if (unlikely(size <= KMALLOC_MIN_SIZE))
> return KMALLOC_SHIFT_LOW;
> @@ -219,7 +219,7 @@ static __always_inline int kmalloc_index(size_t size)
> if (size <= 512 * 1024) return 19;
> if (size <= 1024 * 1024) return 20;
> if (size <= 2 * 1024 * 1024) return 21;
> - return -1;
> + return -2;
> }
>
> #ifdef CONFIG_ZONE_DMA
> @@ -239,8 +239,12 @@ static __always_inline struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags)
> int index;
>
> index = kmalloc_index(size);
> - if (unlikely(index == 0))
> - return ZERO_SIZE_PTR;
> + if (unlikely(index <= 0)) {
> + if (index == 0)
> + return ZERO_SIZE_PTR;
> + if (index == -1)
> + return NULL;
> + }
>
> if (likely(!(flags & SLQB_DMA)))
> return &kmalloc_caches[index];
--
To unsubscribe from this list: send the line "unsubscribe linux-next" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
