2018-11-09 02:28:47

by Darryl T. Agostinelli

Subject: [PATCH] Suppress the sparse warning ./include/linux/slab.h:332:43: warning: dubious: x & !y

Signed-off-by: Darryl T. Agostinelli <[email protected]>
---
include/linux/slab.h | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 918f374e7156..883b7f56bf35 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -317,6 +317,7 @@ static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags)
int is_dma = 0;
int type_dma = 0;
int is_reclaimable;
+ int y;

#ifdef CONFIG_ZONE_DMA
is_dma = !!(flags & __GFP_DMA);
@@ -329,7 +330,10 @@ static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags)
* If an allocation is both __GFP_DMA and __GFP_RECLAIMABLE, return
* KMALLOC_DMA and effectively ignore __GFP_RECLAIMABLE
*/
- return type_dma + (is_reclaimable & !is_dma) * KMALLOC_RECLAIM;
+
+ y = (is_reclaimable & (is_dma == 0 ? 1 : 0));
+
+ return type_dma + y * KMALLOC_RECLAIM;
}

/*
--
2.17.1

2018-11-14 22:26:22

by David Rientjes

Subject: Re: [PATCH] Suppress the sparse warning ./include/linux/slab.h:332:43: warning: dubious: x & !y

On Thu, 8 Nov 2018, Darryl T. Agostinelli wrote:

> Signed-off-by: Darryl T. Agostinelli <[email protected]>
> ---
> include/linux/slab.h | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/slab.h b/include/linux/slab.h
> index 918f374e7156..883b7f56bf35 100644
> --- a/include/linux/slab.h
> +++ b/include/linux/slab.h
> @@ -317,6 +317,7 @@ static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags)
> int is_dma = 0;
> int type_dma = 0;
> int is_reclaimable;
> + int y;
>
> #ifdef CONFIG_ZONE_DMA
> is_dma = !!(flags & __GFP_DMA);
> @@ -329,7 +330,10 @@ static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags)
> * If an allocation is both __GFP_DMA and __GFP_RECLAIMABLE, return
> * KMALLOC_DMA and effectively ignore __GFP_RECLAIMABLE
> */
> - return type_dma + (is_reclaimable & !is_dma) * KMALLOC_RECLAIM;
> +
> + y = (is_reclaimable & (is_dma == 0 ? 1 : 0));
> +
> + return type_dma + y * KMALLOC_RECLAIM;
> }
>
> /*

I agree with you that the function as written is less than pretty :) How
does the assembly change as a result of this code change, however? This
will be in the kmalloc() path so impacting the assembly to fix a sparse
warning may not be warranted.