2017-03-08 13:10:29

by SF Markus Elfring

Subject: [PATCH 08/26] IB/ocrdma: Improve size determinations in ocrdma_mbx_rdma_stats()

From: Markus Elfring <[email protected]>
Date: Tue, 7 Mar 2017 20:16:16 +0100

Replace the specification of two data structures by pointer dereferences
as the parameter for the operator "sizeof" to make the corresponding size
determinations a bit safer according to the Linux coding style convention.

Signed-off-by: Markus Elfring <[email protected]>
---
drivers/infiniband/hw/ocrdma/ocrdma_hw.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
index 1f7d88d7569e..0628600bee56 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
@@ -1302,7 +1302,7 @@ int ocrdma_mbx_rdma_stats(struct ocrdma_dev *dev, bool reset)
mqe->u.nonemb_req.sge[0].len = dev->stats_mem.size;

/* Cache the old stats */
- memcpy(old_stats, req, sizeof(struct ocrdma_rdma_stats_resp));
+ memcpy(old_stats, req, sizeof(*old_stats));
memset(req, 0, dev->stats_mem.size);

ocrdma_init_mch((struct ocrdma_mbx_hdr *)req,
@@ -1315,7 +1315,7 @@ int ocrdma_mbx_rdma_stats(struct ocrdma_dev *dev, bool reset)
status = ocrdma_nonemb_mbx_cmd(dev, mqe, dev->stats_mem.va);
if (status)
/* Copy from cache, if mbox fails */
- memcpy(req, old_stats, sizeof(struct ocrdma_rdma_stats_resp));
+ memcpy(req, old_stats, sizeof(*old_stats));
else
ocrdma_le32_to_cpu(req, dev->stats_mem.size);

--
2.12.0

2017-03-08 14:22:57

by Yuval Shaia

Subject: Re: [PATCH 08/26] IB/ocrdma: Improve size determinations in ocrdma_mbx_rdma_stats()

On Wed, Mar 08, 2017 at 01:58:49PM +0100, SF Markus Elfring wrote:
> From: Markus Elfring <[email protected]>
> Date: Tue, 7 Mar 2017 20:16:16 +0100
>
> Replace the specification of two data structures by pointer dereferences
> as the parameter for the operator "sizeof" to make the corresponding size
> determinations a bit safer according to the Linux coding style convention.
>
> Signed-off-by: Markus Elfring <[email protected]>
> ---
> drivers/infiniband/hw/ocrdma/ocrdma_hw.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
> index 1f7d88d7569e..0628600bee56 100644
> --- a/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
> +++ b/drivers/infiniband/hw/ocrdma/ocrdma_hw.c
> @@ -1302,7 +1302,7 @@ int ocrdma_mbx_rdma_stats(struct ocrdma_dev *dev, bool reset)
> mqe->u.nonemb_req.sge[0].len = dev->stats_mem.size;
>
> /* Cache the old stats */
> - memcpy(old_stats, req, sizeof(struct ocrdma_rdma_stats_resp));
> + memcpy(old_stats, req, sizeof(*old_stats));

Isn't sizeof(req) smaller than sizeof(*old_stats)?

> memset(req, 0, dev->stats_mem.size);
>
> ocrdma_init_mch((struct ocrdma_mbx_hdr *)req,
> @@ -1315,7 +1315,7 @@ int ocrdma_mbx_rdma_stats(struct ocrdma_dev *dev, bool reset)
> status = ocrdma_nonemb_mbx_cmd(dev, mqe, dev->stats_mem.va);
> if (status)
> /* Copy from cache, if mbox fails */
> - memcpy(req, old_stats, sizeof(struct ocrdma_rdma_stats_resp));
> + memcpy(req, old_stats, sizeof(*old_stats));

Fix itself looks fine but original code seems strange and makes me anxious
since resp is (much) bigger than req.

> else
> ocrdma_le32_to_cpu(req, dev->stats_mem.size);
>
> --
> 2.12.0
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html