[dpdk-dev] [PATCH] app/testpmd: fix start index for showing FEC array

Rahul Lakkireddy rahul.lakkireddy at chelsio.com
Fri Dec 25 14:49:33 CET 2020


+dev at dpdk.org which seems to have been dropped by accident.

On Friday, December 12/25/20, 2020 at 09:03:43 +0800, Min Hu (Connor) wrote:
> 
> 
> 在 2020/12/24 19:25, Rahul Lakkireddy 写道:
> >On Thursday, December 12/24/20, 2020 at 17:36:27 +0800, Min Hu (Connor) wrote:
> >>
> >>
> >>在 2020/12/23 20:31, Rahul Lakkireddy 写道:
> >>>On Monday, December 12/21/20, 2020 at 17:07:21 +0800, Min Hu (Connor) wrote:
> >>>>
> >>>>
> >>>>在 2020/12/21 6:47, Rahul Lakkireddy 写道:
> >>>>>From: Karra Satwik <kaara.satwik at chelsio.com>
> >>>>>
> >>>>>Start from index 0 when going through the FEC array. This will allow
> >>>>>"off" to get printed for RTE_ETH_FEC_NOFEC mode.
> >>>>>
> >>>>>Fixes: b19da32e3151 ("app/testpmd: add FEC command")
> >>>>>Cc: stable at dpdk.org
> >>>>>
> >>>>>Signed-off-by: Karra Satwik <kaara.satwik at chelsio.com>
> >>>>>Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy at chelsio.com>
> >>>>>---
> >>>>>  app/test-pmd/config.c | 2 +-
> >>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>>>>
> >>>>>diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
> >>>>>index 3f6c8642b..a6a5baa4e 100644
> >>>>>--- a/app/test-pmd/config.c
> >>>>>+++ b/app/test-pmd/config.c
> >>>>>@@ -3701,7 +3701,7 @@ show_fec_capability(unsigned int num, struct rte_eth_fec_capa *speed_fec_capa)
> >>>>>  		printf("%s : ",
> >>>>>  			rte_eth_link_speed_to_str(speed_fec_capa[i].speed));
> >>>>>-		for (j = RTE_ETH_FEC_AUTO; j < RTE_DIM(fec_mode_name); j++) {
> >>>>>+		for (j = 0; j < RTE_DIM(fec_mode_name); j++) {
> >>>>
> >>>>As RTE_ETH_FEC_NOFEC is mode which every device  has, so we think it
> >>>>should not be regarged as "capabilities".
> >>>>Thanks.
> >>>>
> >>>
> >>>We had gotten several requests asking if device supported turning
> >>>FEC "off" because it was not listed in capabilities. Hence, the
> >>>motiviation for this patch to explicitly show that "off" is
> >>>supported.
> >>
> >>HI, we have referred to other netcard in kernel driver mode, it shows like
> >>this:
> >>
> >>[root at centos197-test_dpdk]$ethtool  --show-fec  eth9
> >>FEC parameters for eth9:
> >>Configured FEC encodings: Auto BaseR
> >>Active FEC encoding: Off
> >>
> >>Here, "Configured FEC encodings" means the capability it supports, it
> >>does not include "off", although it can be configured using "off".
> >>thanks.
> >>
> >
> >It is the same with our own card too using our kernel driver and
> >have gotten the same questions in the past with our kernel driver
> >too.
> >
> ># ethtool --show-fec enp2s0f4
> >FEC parameters for enp2s0f4:
> >Configured FEC encodings: Auto BaseR RS
> >Active FEC encoding: RS
> >
> >We don't have any strong opinion on this. We just wanted to let
> >testpmd show the 'off' caps since it's also an available option.
> >If the intention is confusing, then sure we will drop the patch.
> >Let us know your feedback.
> Hi, I've got your opinion, but I think the display about FEC in dpdk
> testpmd had better be in accordance with that in kernel ethtool.
> IF the two is different, it may confuse users, thanks.
> So, what about everybody? any opinion will be welcome.
> 
> >
> >>
> >>
> >>>
> >>>
> >>>>>  			if (RTE_ETH_FEC_MODE_TO_CAPA(j) &
> >>>>>  						speed_fec_capa[i].capa)
> >>>>>  				printf("%s ", fec_mode_name[j].name);
> >>>>>
> >>>.
> >>>
> >.
> >


More information about the dev mailing list