You could use:
u16 uninitialized_var(vlan);
instead.
Although this in the special QP data flow, I still prefer to avoid adding extra code (even setting
initial values at procedure entry). The line above will also do the job. "uninitialized_var"
is used elsewhere in the driver. See, for example, mlx4_ib_post_send() in the same file (qp.c).
-Jack
On Friday 28 September 2012 14:48, Paul Bolle wrote:
> Building qp.o (part of the "Mellanox ConnectX HCA support" driver)
> triggers this GCC warning:
> drivers/infiniband/hw/mlx4/qp.c: In function ‘mlx4_ib_post_send’:
> drivers/infiniband/hw/mlx4/qp.c:1828:30: warning: ‘vlan’ may be used uninitialized in this function [-Wmaybe-uninitialized]
> drivers/infiniband/hw/mlx4/qp.c:1718:6: note: ‘vlan’ was declared here
>
> Looking at the code it is clear 'vlan' is only set and used if 'is_eth'
> is non-zero. But there's no harm in initializing 'vlan' to 0 (which
> matches ib_get_cached_gid()'s default return) to silence GCC.
>
> Signed-off-by: Paul Bolle <[email protected]>
> ---
> 0) I noticed this warning while building v3.6-rc7 on current Fedora 17,
> using Fedora's default config.
>
> 1) Compile tested only. I tested against v3.6-rc7, with commit
> a41262bb5721f2b708ee8b23f67be2f2e16a2fab ("IB/mlx4: SR-IOV IB context
> objects and proxy/tunnel SQP") from linux-next cherry-picked, to take
> into account a trivial context change in linux-next.
>
> drivers/infiniband/hw/mlx4/qp.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
> index a862251..71fdda6 100644
> --- a/drivers/infiniband/hw/mlx4/qp.c
> +++ b/drivers/infiniband/hw/mlx4/qp.c
> @@ -1715,7 +1715,7 @@ static int build_mlx_header(struct mlx4_ib_sqp *sqp, struct ib_send_wr *wr,
> int is_eth;
> int is_vlan = 0;
> int is_grh;
> - u16 vlan;
> + u16 vlan = 0;
> int err = 0;
>
> send_size = 0;
On Wed, 2012-10-10 at 09:23 +0200, Jack Morgenstein wrote:
> You could use:
>
> u16 uninitialized_var(vlan);
>
> instead.
I guess we'd better just wait and see whether uninitialized_var()
survives before discussing your suggestion (see the thread starting at
https://lkml.org/lkml/2012/10/26/508 ).
> Although this in the special QP data flow, I still prefer to avoid adding extra code (even setting
> initial values at procedure entry). The line above will also do the job. "uninitialized_var"
> is used elsewhere in the driver. See, for example, mlx4_ib_post_send() in the same file (qp.c).
Paul Bolle