This patch remove useless if(ptr) check to kfree.

Signed-off-by: Bernard Zhao <zhaojunkui2008@126.com>
---
 drivers/gpu/host1x/fence.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/host1x/fence.c b/drivers/gpu/host1x/fence.c
index ecab72882192..05b36bfc8b74 100644
--- a/drivers/gpu/host1x/fence.c
+++ b/drivers/gpu/host1x/fence.c
@@ -93,8 +93,7 @@ static void host1x_syncpt_fence_release(struct dma_fence *f)
 {
 	struct host1x_syncpt_fence *sf = to_host1x_fence(f);
 
-	if (sf->waiter)
-		kfree(sf->waiter);
+	kfree(sf->waiter);
 
 	dma_fence_free(f);
 }
-- 
2.33.1

On 4/13/22 09:39, Bernard Zhao wrote:
> This patch remove useless if(ptr) check to kfree.
> 
> Signed-off-by: Bernard Zhao <zhaojunkui2008@126.com>
> ---
>   drivers/gpu/host1x/fence.c | 3 +--
>   1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/host1x/fence.c b/drivers/gpu/host1x/fence.c
> index ecab72882192..05b36bfc8b74 100644
> --- a/drivers/gpu/host1x/fence.c
> +++ b/drivers/gpu/host1x/fence.c
> @@ -93,8 +93,7 @@ static void host1x_syncpt_fence_release(struct dma_fence *f)
>   {
>   	struct host1x_syncpt_fence *sf = to_host1x_fence(f);
>   
> -	if (sf->waiter)
> -		kfree(sf->waiter);
> +	kfree(sf->waiter);
>   
>   	dma_fence_free(f);
>   }

I personally dislike leaving the NULL checks off with calls to kfree. 
With the NULL check there, while reading the code it is obvious that the 
intention is that the value can be either a valid pointer or NULL. IMHO 
with C's type system/conventions we need this kind of contextual 
information to understand the code easily and avoid bugs.

Mikko