Hi, Ferruh,
Thanks for your comment.
All has been fixed in v2, please check it out.
Thanks.
在 2021/4/13 7:08, Ferruh Yigit 写道:
On 4/10/2021 10:18 AM, Min Hu (Connor) wrote:
This patch adds more sanity checks in control path APIs.
Fixes: 214ed1acd125 ("ethdev: add iterator to match devargs input")
Fixes: 3d98f921fbe9 ("ethdev: unify prefix for static functions and
variables")
Fixes: 0366137722a0 ("ethdev: check for invalid device name")
Fixes: d948f596fee2 ("ethdev: fix port data mismatched in multiple
process model")
Fixes: 5b7ba31148a8 ("ethdev: add port ownership")
Fixes: f8244c6399d9 ("ethdev: increase port id range")
Cc: sta...@dpdk.org
Signed-off-by: Min Hu (Connor) <humi...@huawei.com>
Thanks Connor for the patch, please find a few minor comments below.
<...>
@@ -3822,6 +3903,9 @@ rte_eth_dev_rss_reta_query(uint16_t port_id,
struct rte_eth_dev *dev;
int ret;
+ if (reta_conf == NULL)
+ return -EINVAL;
+
The 'eth_check_reta_mask()' & 'eth_check_reta_entry()' static fucntions
are also checking 'reta_conf' against NULL, they may be removed after
these checks.
<...>
@@ -4447,6 +4552,9 @@ rte_eth_dev_callback_process(struct rte_eth_dev
*dev,
struct rte_eth_dev_callback dev_cb;
int rc = 0;
+ if (dev == NULL)
+ return -EINVAL;
+
This is internal API, not sure if needed to verify 'dev' here, it won't
hurt of course, but just to prevent unnecessary checks. No strong option.
Same comment for all internal functions that changed.
<...>
@@ -5370,6 +5502,9 @@ rte_eth_dev_get_dcb_info(uint16_t port_id,
{
struct rte_eth_dev *dev;
+ if (dcb_info == NULL)
+ return -EINVAL;
+
Function documentation in the header file, doesn't list '-EINVAL' as a
return value, whenever a new return value is returned, it needs to be
documented.
.
