On 11/11/2020 1:05 PM, Conor Walsh wrote:
When DPDK is compiled with gcc 7.5 with the optimization level set to 1
gcc sees the offset variable in txgbe_ethdev.c as possibly being
uninitialised. To correct this error offset has been initialised to 0 in
txgbe_ethdev.c
Signed-off-by: Conor Walsh <conor.wa...@intel.com>
---
drivers/net/txgbe/txgbe_ethdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/txgbe/txgbe_ethdev.c b/drivers/net/txgbe/txgbe_ethdev.c
index 9f533603a1..7fd56c2698 100644
--- a/drivers/net/txgbe/txgbe_ethdev.c
+++ b/drivers/net/txgbe/txgbe_ethdev.c
@@ -2247,7 +2247,7 @@ txgbe_dev_xstats_get_(struct rte_eth_dev *dev, uint64_t
*values,
/* Extended stats from txgbe_hw_stats */
for (i = 0; i < limit; i++) {
- uint32_t offset;
+ uint32_t offset = 0;
This is false positive, right?
Is there an actual case that 'offset' is used without initialization?
'txgbe_get_offset_by_id()' return value seems used only to detect fail/success,
can you please try with 'txgbe_get_offset_by_id()' returning a simple '-1' on
failure, instead of current "-(int)(id + 1)", if it helps with the compiler warning?
