[PATCH] spec/build/bsps: Fix blank variables

[Ryan Long]

On 3/10/2022 12:52 AM, Sebastian Huber wrote:
> On 09/03/2022 22:02, Ryan Long wrote:
>> Several variables in the installed Makefile.inc, target.cfg, and bsp.cfg
>> had values that were not being initialized to yes or no as they should
>> be. 
>
> What are "several variables"?

RTEMS_HAS_POSIX_API, HAS_POSIX_API, HAS_NETWORKING, 
RTEMS_HAS_MULTIPROCESSING, RTEMS_HAS_NETWORKING

Should I add that list to the commit message?

>
>> Reordering the files and adding an initialization using yesno() to
>> RTEMS_HAS_NETWORKING fixed this issue.
>> ---
>>   spec/build/bsps/grpmake.yml | 8 ++++----
>>   spec/build/bsps/optmake.yml | 1 +
>>   2 files changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/spec/build/bsps/grpmake.yml b/spec/build/bsps/grpmake.yml
>> index 0998efc..f4a21f2 100644
>> --- a/spec/build/bsps/grpmake.yml
>> +++ b/spec/build/bsps/grpmake.yml
>> @@ -30,15 +30,15 @@ install:
>>   ldflags: []
>>   links:
>>   - role: build-dependency
>> -  uid: makebsp
>> -- role: build-dependency
>>     uid: makecustom
>>   - role: build-dependency
>> +  uid: optmake
>> +- role: build-dependency
>> +  uid: makebsp
>> +- role: build-dependency
>>     uid: makeinc
>>   - role: build-dependency
>>     uid: maketarget
>> -- role: build-dependency
>> -  uid: optmake
>>   type: build
>>   use-after: []
>>   use-before: []
>> diff --git a/spec/build/bsps/optmake.yml b/spec/build/bsps/optmake.yml
>> index 28e9660..bea2c90 100644
>> --- a/spec/build/bsps/optmake.yml
>> +++ b/spec/build/bsps/optmake.yml
>> @@ -7,6 +7,7 @@ actions:
>>           return "no"
>>       conf.env["RTEMS_HAS_MULTIPROCESSING"] = yesno(conf, 
>> "RTEMS_MULTIPROCESSING")
>>       conf.env["RTEMS_HAS_POSIX_API"] = yesno(conf, "RTEMS_POSIX_API")
>> +    conf.env["RTEMS_HAS_NETWORKING"] = yesno(conf, "RTEMS_NETWORKING")
>>   build-type: option
>>   copyrights:
>>   - Copyright (C) 2020 embedded brains GmbH 
>> (http://www.embedded-brains.de)
>
> The RTEMS_NETWORKING option no longer exists, so this should be a 
> hared coded "no".
Alright, I'll make that change.