[dpdk-dev] config: remove useless lines for DPAA2

  Some comments are not relevant in a config which only overrides
the default config.

The option CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER is already disabled
by default so it can be removed from this file.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
---
 config/defconfig_arm64-dpaa2-linuxapp-gcc | 7 -------
 1 file changed, 7 deletions(-)
  

On Tuesday 03 April 2018 03:36 AM, Thomas Monjalon wrote:
> Some comments are not relevant in a config which only overrides
> the default config.
> 
> The option CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER is already disabled
> by default so it can be removed from this file.
> 
> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> ---
>   config/defconfig_arm64-dpaa2-linuxapp-gcc | 7 -------
>   1 file changed, 7 deletions(-)
> 
> diff --git a/config/defconfig_arm64-dpaa2-linuxapp-gcc b/config/defconfig_arm64-dpaa2-linuxapp-gcc
> index ecac994bf..96f478a06 100644
> --- a/config/defconfig_arm64-dpaa2-linuxapp-gcc
> +++ b/config/defconfig_arm64-dpaa2-linuxapp-gcc
> @@ -9,9 +9,6 @@
>   CONFIG_RTE_MACHINE="dpaa2"
>   CONFIG_RTE_ARCH_ARM_TUNE="cortex-a72"
>   
> -#
> -# Compile Environment Abstraction Layer
> -#
>   CONFIG_RTE_MAX_LCORE=16
>   CONFIG_RTE_MAX_NUMA_NODES=1
>   CONFIG_RTE_CACHE_LINE_SIZE=64
> @@ -22,8 +19,4 @@ CONFIG_RTE_PKTMBUF_HEADROOM=128
>   CONFIG_RTE_EAL_NUMA_AWARE_HUGEPAGES=n
>   CONFIG_RTE_LIBRTE_VHOST_NUMA=n
>   
> -#
> -# Compile Support Libraries for DPAA2
> -#
>   CONFIG_RTE_LIBRTE_DPAA2_USE_PHYS_IOVA=n
> -CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER=n

I think DEBUG_DRIVER should exist in local config so that a developer 
can easily toggle it without modifying the common_base, which should 
serves as repository rather than toggle.

I do see the problem that having an option in common_base and then 
having it again in local config without overriding (=n) - is non-intuitive.

But, I still feel it is easier to control changes in local config - 
(make T=<local config>) is easily visible.
  

03/04/2018 08:55, Shreyansh Jain:
> On Tuesday 03 April 2018 03:36 AM, Thomas Monjalon wrote:
> > Some comments are not relevant in a config which only overrides
> > the default config.
> > 
> > The option CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER is already disabled
> > by default so it can be removed from this file.
> > 
> > Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> > ---
> >   config/defconfig_arm64-dpaa2-linuxapp-gcc | 7 -------
> >   1 file changed, 7 deletions(-)
> 
> I think DEBUG_DRIVER should exist in local config so that a developer 
> can easily toggle it without modifying the common_base, which should 
> serves as repository rather than toggle.

No, the file to be modified is .config in the build directory,
not the default config.

> I do see the problem that having an option in common_base and then 
> having it again in local config without overriding (=n) - is non-intuitive.
> 
> But, I still feel it is easier to control changes in local config - 
> (make T=<local config>) is easily visible.

It is not a local config file.
The local config is .config.
  

On Tuesday 03 April 2018 12:33 PM, Thomas Monjalon wrote:
> 03/04/2018 08:55, Shreyansh Jain:
>> On Tuesday 03 April 2018 03:36 AM, Thomas Monjalon wrote:
>>> Some comments are not relevant in a config which only overrides
>>> the default config.
>>>
>>> The option CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER is already disabled
>>> by default so it can be removed from this file.
>>>
>>> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
>>> ---
>>>    config/defconfig_arm64-dpaa2-linuxapp-gcc | 7 -------
>>>    1 file changed, 7 deletions(-)
>>
>> I think DEBUG_DRIVER should exist in local config so that a developer
>> can easily toggle it without modifying the common_base, which should
>> serves as repository rather than toggle.
> 
> No, the file to be modified is .config in the build directory,
> not the default config.

In my environment I prefer changing the platform config rather than 
local .config as that way I can track changes through git.

> 
>> I do see the problem that having an option in common_base and then
>> having it again in local config without overriding (=n) - is non-intuitive.
>>
>> But, I still feel it is easier to control changes in local config -
>> (make T=<local config>) is easily visible.
> 
> It is not a local config file.
> The local config is .config.

Sorry, allow me to rephrase my previous wordings:

s/local config/platform config/

Anyways, this is just my personal preference - not a big inconvenience.
Thanks for the patch.

Acked-By: Shreyansh Jain <shreyansh.jain@nxp.com>
  

03/04/2018 11:43, Shreyansh Jain:
> On Tuesday 03 April 2018 12:33 PM, Thomas Monjalon wrote:
> > 03/04/2018 08:55, Shreyansh Jain:
> >> On Tuesday 03 April 2018 03:36 AM, Thomas Monjalon wrote:
> >>> Some comments are not relevant in a config which only overrides
> >>> the default config.
> >>>
> >>> The option CONFIG_RTE_LIBRTE_DPAA2_DEBUG_DRIVER is already disabled
> >>> by default so it can be removed from this file.
> >>>
> >>> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> >>> ---
> >>>    config/defconfig_arm64-dpaa2-linuxapp-gcc | 7 -------
> >>>    1 file changed, 7 deletions(-)
> >>
> >> I think DEBUG_DRIVER should exist in local config so that a developer
> >> can easily toggle it without modifying the common_base, which should
> >> serves as repository rather than toggle.
> > 
> > No, the file to be modified is .config in the build directory,
> > not the default config.
> 
> In my environment I prefer changing the platform config rather than 
> local .config as that way I can track changes through git.
> 
> > 
> >> I do see the problem that having an option in common_base and then
> >> having it again in local config without overriding (=n) - is non-intuitive.
> >>
> >> But, I still feel it is easier to control changes in local config -
> >> (make T=<local config>) is easily visible.
> > 
> > It is not a local config file.
> > The local config is .config.
> 
> Sorry, allow me to rephrase my previous wordings:
> 
> s/local config/platform config/
> 
> Anyways, this is just my personal preference - not a big inconvenience.
> Thanks for the patch.
> 
> Acked-By: Shreyansh Jain <shreyansh.jain@nxp.com>

Applied