mbox series

[0/5] kvargs: promote or remove experimental api

Message ID 20210924155740.20510-1-olivier.matz@6wind.com (mailing list archive)
Headers
Series kvargs: promote or remove experimental api |

Message

Olivier Matz Sept. 24, 2021, 3:57 p.m. UTC
  This patchset promotes 2 functions rte_kvargs_parse_delim() and
rte_kvargs_get() as stable.

It also replaces rte_kvargs_strcmp() by a new one
rte_kvargs_get_with_value(), which is easier to use.

Olivier Matz (5):
  kvargs: promote delimited parsing as stable
  kvargs: promote get from key as stable
  kvargs: new function to get from key and value
  kvargs: remove experimental function to compare string
  kvargs: fix comments style

 drivers/bus/auxiliary/auxiliary_params.c |  9 ++--
 drivers/bus/vdev/vdev_params.c           | 13 +-----
 lib/kvargs/rte_kvargs.c                  | 32 +++++++------
 lib/kvargs/rte_kvargs.h                  | 59 ++++++++++++------------
 lib/kvargs/version.map                   |  7 +--
 5 files changed, 57 insertions(+), 63 deletions(-)
  

Comments

Xueming Li Sept. 25, 2021, 7:02 a.m. UTC | #1
On Fri, 2021-09-24 at 17:57 +0200, Olivier Matz wrote:
> This patchset promotes 2 functions rte_kvargs_parse_delim() and
> rte_kvargs_get() as stable.
> 
> It also replaces rte_kvargs_strcmp() by a new one
> rte_kvargs_get_with_value(), which is easier to use.
> 
> Olivier Matz (5):
>   kvargs: promote delimited parsing as stable
>   kvargs: promote get from key as stable
>   kvargs: new function to get from key and value
>   kvargs: remove experimental function to compare string
>   kvargs: fix comments style
> 
>  drivers/bus/auxiliary/auxiliary_params.c |  9 ++--
>  drivers/bus/vdev/vdev_params.c           | 13 +-----
>  lib/kvargs/rte_kvargs.c                  | 32 +++++++------
>  lib/kvargs/rte_kvargs.h                  | 59 ++++++++++++------------
>  lib/kvargs/version.map                   |  7 +--
>  5 files changed, 57 insertions(+), 63 deletions(-)
> 

The whole series looks good to me, the new function that match key and
value directly looks more efficient, thanks!

Reviewed-by: Xueming Li <xuemingl@nvidia.com>