2021-01-14 07:13:36

by Jason Wang

[permalink] [raw]
Subject: [PATCH V2] mlx5: vdpa: fix possible uninitialized var

When compiling with -Werror=maybe-uninitialized, gcc may complains the
possible uninitialized umem. Since the callers won't pass value other
than 1 to 3, making 3 as default to fix the compiler warning.

Signed-off-by: Jason Wang <[email protected]>
---
drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index f1d54814db97..07ccc61cd6f6 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -703,7 +703,7 @@ static void umem_destroy(struct mlx5_vdpa_net *ndev, struct mlx5_vdpa_virtqueue
case 2:
umem = &mvq->umem2;
break;
- case 3:
+ default:
umem = &mvq->umem3;
break;
}
--
2.25.1


2021-01-14 07:26:06

by Eli Cohen

[permalink] [raw]
Subject: Re: [PATCH V2] mlx5: vdpa: fix possible uninitialized var

On Thu, Jan 14, 2021 at 03:09:04PM +0800, Jason Wang wrote:
> When compiling with -Werror=maybe-uninitialized, gcc may complains the
Maybe you want to fix to: gcc may complain about possible...

Other than that:
Acked-by: Eli Cohen <[email protected]>

> possible uninitialized umem. Since the callers won't pass value other
> than 1 to 3, making 3 as default to fix the compiler warning.
>
> Signed-off-by: Jason Wang <[email protected]>
> ---
> drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> index f1d54814db97..07ccc61cd6f6 100644
> --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
> +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> @@ -703,7 +703,7 @@ static void umem_destroy(struct mlx5_vdpa_net *ndev, struct mlx5_vdpa_virtqueue
> case 2:
> umem = &mvq->umem2;
> break;
> - case 3:
> + default:
> umem = &mvq->umem3;
> break;
> }
> --
> 2.25.1
>

2021-01-17 10:47:53

by Michael S. Tsirkin

[permalink] [raw]
Subject: Re: [PATCH V2] mlx5: vdpa: fix possible uninitialized var

On Thu, Jan 14, 2021 at 03:09:04PM +0800, Jason Wang wrote:
> When compiling with -Werror=maybe-uninitialized, gcc may complains the
> possible uninitialized umem. Since the callers won't pass value other
> than 1 to 3, making 3 as default to fix the compiler warning.
>
> Signed-off-by: Jason Wang <[email protected]>

Of course if it's actually something else this is not DTRT ...
As long as we worry about this warning, let's have defence in depth:
case 3:
umem = &mvq->umem3;
break;
+ default:
+ umem = NULL;
+ BUG();
+ break;


> ---
> drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> index f1d54814db97..07ccc61cd6f6 100644
> --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
> +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> @@ -703,7 +703,7 @@ static void umem_destroy(struct mlx5_vdpa_net *ndev, struct mlx5_vdpa_virtqueue
> case 2:
> umem = &mvq->umem2;
> break;
> - case 3:
> + default:
> umem = &mvq->umem3;
> break;
> }
> --
> 2.25.1

2021-01-18 05:56:44

by Jason Wang

[permalink] [raw]
Subject: Re: [PATCH V2] mlx5: vdpa: fix possible uninitialized var


On 2021/1/17 下午6:30, Michael S. Tsirkin wrote:
> On Thu, Jan 14, 2021 at 03:09:04PM +0800, Jason Wang wrote:
>> When compiling with -Werror=maybe-uninitialized, gcc may complains the
>> possible uninitialized umem. Since the callers won't pass value other
>> than 1 to 3, making 3 as default to fix the compiler warning.
>>
>> Signed-off-by: Jason Wang <[email protected]>
> Of course if it's actually something else this is not DTRT ...


It breaks some donw stream build (which mandates
-Werror=maybe-uninitialized)


> As long as we worry about this warning, let's have defence in depth:
> case 3:
> umem = &mvq->umem3;
> break;
> + default:
> + umem = NULL;
> + BUG();
> + break;


Checkpatch doesn't like this:

WARNING: Avoid crashing the kernel - try using WARN_ON & recovery code
rather than BUG() or BUG_ON()
#20: FILE: drivers/vdpa/mlx5/net/mlx5_vnet.c:695:
+        BUG();

But if you stick, I can go this way.

Thanks


>
>> ---
>> drivers/vdpa/mlx5/net/mlx5_vnet.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
>> index f1d54814db97..07ccc61cd6f6 100644
>> --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
>> +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
>> @@ -703,7 +703,7 @@ static void umem_destroy(struct mlx5_vdpa_net *ndev, struct mlx5_vdpa_virtqueue
>> case 2:
>> umem = &mvq->umem2;
>> break;
>> - case 3:
>> + default:
>> umem = &mvq->umem3;
>> break;
>> }
>> --
>> 2.25.1