On Mon, Jun 28, 2021 at 07:29:11AM -0700, Tyler Retzlaff wrote:

On Mon, Jun 28, 2021 at 01:55:02PM +0300, Andrew Rybchenko wrote:

quoted

On 6/24/21 1:34 AM, Jie Zhou wrote:

quoted

Replace parse_fec_mode misleading return type name mode with fec_capa

Fixes: b19da32e3151 ("app/testpmd: add FEC command")
Cc: stable@dpdk.org

Signed-off-by: Jie Zhou <redacted>
Signed-off-by: Jie Zhou <redacted>

[snip]

quoted

diff --git a/app/test-pmd/testpmd.h b/app/test-pmd/testpmd.h
index 283b5e3680..9ae4d90dd1 100644
--- a/app/test-pmd/testpmd.h
+++ b/app/test-pmd/testpmd.h

@@ -885,7 +885,7 @@ void show_tx_pkt_segments(void);
 void set_tx_pkt_times(unsigned int *tx_times);
 void show_tx_pkt_times(void);
 void set_tx_pkt_split(const char *name);
-int parse_fec_mode(const char *name, enum rte_eth_fec_mode *mode);
+int parse_fec_mode(const char *name, uint32_t *fec_capa);

I guess that the real reason behind is to fix implicit
conversion of enum pointer to/from uint32_t pointer.
I guess the problem is different signness of enum on
Windows compiler.

yes, compilers targeting targets will select `int' once all constants of
the enumeration list are defined.

quoted

If so, please, put real motivation of the changeset in summary.
It should be human-readable (and do not contain function name).
Explain details in the description.

Yes, I agree that mode is misleading here and should be mentioned
in the description, but I guess it is not the root cause.
May be I'm wrong.

Your understanding on the root cause is corret. Will fix the description in V15. Thanks.