Masahiro Yamada <yamada.masahiro@xxxxxxxxxxxxx> writes:
> 2018-07-03 16:32 GMT+09:00 Dirk Gouders <dirk@xxxxxxxxxxx>:
>> Masahiro Yamada <yamada.masahiro@xxxxxxxxxxxxx> writes:
>>
>>> 2018-06-29 18:12 GMT+09:00 Dirk Gouders <dirk@xxxxxxxxxxx>:
>>>> I could not verify the comment for that symbol flag.
>>>
>>> Good catch.
>>> I forgot to fix up the comment
>>> in commit 104daea149c4.
>>>
>>>
>>>> I could only find that flag set for choices and the defconfig_list
>>>> symbol in a dump of all symbols, which corresponds to the only two
>>>> locations in the code where that flag is being set explicitely.
>>>>
>>>> Signed-off-by: Dirk Gouders <dirk@xxxxxxxxxxx>
>>>> Cc: Sam Ravnborg <sam@xxxxxxxxxxxx>
>>>> ---
>>>> scripts/kconfig/expr.h | 4 +++-
>>>> 1 file changed, 3 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h
>>>> index 94a383b21df6..0f53e44f14d6 100644
>>>> --- a/scripts/kconfig/expr.h
>>>> +++ b/scripts/kconfig/expr.h
>>>> @@ -141,7 +141,9 @@ struct symbol {
>>>> #define SYMBOL_OPTIONAL 0x0100 /* choice is optional - values can be 'n' */
>>>> #define SYMBOL_WRITE 0x0200 /* write symbol to file (KCONFIG_CONFIG) */
>>>> #define SYMBOL_CHANGED 0x0400 /* ? */
>>>> -#define SYMBOL_AUTO 0x1000 /* value from environment variable */
>>>> +#define SYMBOL_AUTO 0x1000 /* Symbols of type choice and the
>>>> + * symbol with option defconfig_list
>>>> + * have this flag set */
>>>
>>> Hmm. This explanation is not very helpful in my opinion.
>>> Could you reword that?
>>>
>>> In my understanding, symbols with SYMBOL_AUTO
>>> are never written out to file.
>>
>> Yes, that's right, sym_calc_value() clears SYMBOL_WRITE for those
>> symbols and the comment should describe the effect of the flag, not it's
>> users. I will read a bit more to see if the flag has more effects and
>> reword the comment.
>>
>> Sidenote: probably, AUTO is a misnomer for this flag.
>
> I agree.
>
>
>> I would expect
>> something like automatically generated symbols but the symbols that have
>> this flag set are rather -- hmm, perhaps auxiliary symbols?
>
> Good.
> Another idea might be SYMBOL_NO_WRITE ??
> (it seems almost self-commenting)
Oh sorry, there was an overlap with me sending out v2 without
recognizing and reading your reply.
I like SYMBOL_NO_WRITE, and would prepare v3 if nobody has objections.
Dirk
--
To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
