From: Arnd Bergmann <[email protected]>
The newly added opcode variable is not initialized anywhere:
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:322:18: error: variable 'opcode' is uninitialized when used here [-Werror,-Wuninitialized]
crsqe->opcode = opcode;
Not sure what the intention was here, but this is clearly pointless, so
instead use 0 in place of the variable.
Fixes: bcfee4ce3e013 ("RDMA/bnxt_re: remove redundant cmdq_bitmap")
Signed-off-by: Arnd Bergmann <[email protected]>
---
This is likely not the correct fix, so treat this as a bug report,
the patch is what I applied as a temporary workaround in my randconfig
tree.
---
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
index bb5aebafe1622..8d349d6a633fa 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
@@ -285,7 +285,6 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw,
struct pci_dev *pdev;
unsigned long flags;
u16 cookie;
- u8 opcode;
u8 *preq;
cmdq = &rcfw->cmdq;
@@ -319,7 +318,7 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw,
crsqe->is_internal_cmd = false;
crsqe->is_waiter_alive = true;
crsqe->is_in_used = true;
- crsqe->opcode = opcode;
+ crsqe->opcode = 0;
crsqe->req_size = __get_cmdq_base_cmd_size(msg->req, msg->req_sz);
if (__get_cmdq_base_resp_size(msg->req, msg->req_sz) && msg->sb) {
--
2.39.2
On Fri, Jun 16, 2023 at 2:35 PM Arnd Bergmann <[email protected]> wrote:
>
> From: Arnd Bergmann <[email protected]>
>
> The newly added opcode variable is not initialized anywhere:
>
> drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:322:18: error: variable 'opcode' is uninitialized when used here [-Werror,-Wuninitialized]
> crsqe->opcode = opcode;
This is already fixed and merged
https://lore.kernel.org/r/6ad1e44be2b560986da6fdc6b68da606413e9026.1686644105.git.leonro@nvidia.com
Thanks
>
> Not sure what the intention was here, but this is clearly pointless, so
> instead use 0 in place of the variable.
>
> Fixes: bcfee4ce3e013 ("RDMA/bnxt_re: remove redundant cmdq_bitmap")
> Signed-off-by: Arnd Bergmann <[email protected]>
> ---
> This is likely not the correct fix, so treat this as a bug report,
> the patch is what I applied as a temporary workaround in my randconfig
> tree.
> ---
> drivers/infiniband/hw/bnxt_re/qplib_rcfw.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
> index bb5aebafe1622..8d349d6a633fa 100644
> --- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
> +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
> @@ -285,7 +285,6 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw,
> struct pci_dev *pdev;
> unsigned long flags;
> u16 cookie;
> - u8 opcode;
> u8 *preq;
>
> cmdq = &rcfw->cmdq;
> @@ -319,7 +318,7 @@ static int __send_message(struct bnxt_qplib_rcfw *rcfw,
> crsqe->is_internal_cmd = false;
> crsqe->is_waiter_alive = true;
> crsqe->is_in_used = true;
> - crsqe->opcode = opcode;
> + crsqe->opcode = 0;
>
> crsqe->req_size = __get_cmdq_base_cmd_size(msg->req, msg->req_sz);
> if (__get_cmdq_base_resp_size(msg->req, msg->req_sz) && msg->sb) {
> --
> 2.39.2
>