[PATCH v2 078/169] Correct SPL use of EFI_UNICODE_COLLATION_PROTOCOL2

[Heinrich Schuchardt]

On 2/7/23 00:38, Tom Rini wrote:
> On Tue, Feb 07, 2023 at 12:36:51AM +0100, Heinrich Schuchardt wrote:
>> On 2/5/23 23:39, Simon Glass wrote:
>>> This converts 1 usage of this option to the non-SPL form, since there is
>>> no SPL_EFI_UNICODE_COLLATION_PROTOCOL2 defined in Kconfig
>>>
>>> Signed-off-by: Simon Glass <sjg at chromium.org>
>>> ---
>>>
>>> (no changes since v1)
>>>
>>>    lib/efi_loader/efi_root_node.c | 2 +-
>>>    1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/lib/efi_loader/efi_root_node.c b/lib/efi_loader/efi_root_node.c
>>> index 21a014d7c21..108c14b95bd 100644
>>> --- a/lib/efi_loader/efi_root_node.c
>>> +++ b/lib/efi_loader/efi_root_node.c
>>> @@ -68,7 +68,7 @@ efi_status_t efi_root_node_register(void)
>>>    		 &efi_guid_dt_fixup_protocol,
>>>    		 &efi_dt_fixup_prot,
>>>    #endif
>>> -#if CONFIG_IS_ENABLED(EFI_UNICODE_COLLATION_PROTOCOL2)
>>> +#if IS_ENABLED(CONFIG_EFI_UNICODE_COLLATION_PROTOCOL2)
>>
>> I never received this patch in my inbox. Expect series with more than 50
>> mails not even to be copied to the spam folder. They are outright rejected
>> by my mail provider.
>>
>> I cannot see any problem solved by this patch. Why did you send it?
> 
> You should look in to setting up lei to fetch the list then, as this is
> well explained in the cover letter.

lei = circle of flowers is what I find in a dictionary. Do I miss an 
idiomatic expression?

The cover letter describes an observation but does not point out any 
problem relating to this observation.

We don't have a CONFIG_SPL_EFI_LOADER. In vsnprintf_internal() we use 
CONFIG_IS_ENABLED(EFI_LOADER). The code inside the #if condition is not 
compiled in SPL. This is the desired behavior.

How is moveconfig meant to accept this?

Is anything wrong about this use of CONFIG_IS_ENABLED(EFI_LOADER)?

What is the reasoning behind wanting to use CONFIG_IS_ENABLED() only if 
an SPL config option exists?

Where is the documentation change describing that CONFIG_IS_ENABLED() 
should only be used if an SPL config option exists?

Best regards

Heinrich