[PATCH 09/10] tools: zynqmp_psu_init_minimize: Add serdes_illcalib forward declaration

[Michal Simek]

On 6/20/22 15:38, Stefan Herbrechtsmeier wrote:
> Am 20.06.2022 um 09:18 schrieb Michal Simek:
>> On 6/20/22 09:07, Stefan Herbrechtsmeier wrote:
>>> Am 16.06.2022 um 17:13 schrieb Michal Simek:
>>>> On 6/8/22 18:20, Stefan Herbrechtsmeier wrote:
>>>>> [CAUTION: External Email]
>>>>>
>>>>> From: Stefan Herbrechtsmeier <stefan.herbrechtsmeier at weidmueller.com>
>>>>>
>>>>> A forward declaration for the serdes_illcalib function.
>>>>>
>>>>> Signed-off-by: Stefan Herbrechtsmeier <stefan.herbrechtsmeier at weidmueller.com>
>>>>> ---
>>>>>
>>>>>   tools/zynqmp_psu_init_minimize.sh | 5 +++++
>>>>>   1 file changed, 5 insertions(+)
>>>>>
>>>>> diff --git a/tools/zynqmp_psu_init_minimize.sh 
>>>>> b/tools/zynqmp_psu_init_minimize.sh
>>>>> index 31fbeac327..8411065e13 100755
>>>>> --- a/tools/zynqmp_psu_init_minimize.sh
>>>>> +++ b/tools/zynqmp_psu_init_minimize.sh
>>>>> @@ -108,6 +108,11 @@ cat << EOF >${TMP}
>>>>>   #include <asm/arch/psu_init_gpl.h>
>>>>>   #include <xil_io.h>
>>>>>
>>>>> +static int serdes_illcalib(u32 lane3_protocol, u32 lane3_rate,
>>>>> +                          u32 lane2_protocol, u32 lane2_rate,
>>>>> +                          u32 lane1_protocol, u32 lane1_rate,
>>>>> +                          u32 lane0_protocol, u32 lane0_rate);
>>>>> +
>>>>
>>>> Better to resort that functions to avoid these additional lines.
>>>
>>> Do you propose to move the serdes_illcalib and serdes_illcalib_pcie_gen1 
>>> functions via sed?
>>
>> swapping that functions should be enough. Whatever tool which does this job is 
>> fine.
> 
> Does the swap need to check for the serdes_illcalib function inside the 
> psu_serdes_init_data function or is a general reorder okay?

Everybody should call checkpatch and compile it. Tool is prepares that files but 
at the end it is up to user to fix the rest.
Taht's why general reorder is fine for me.

> 
>>> The psu_init_gpl.c for e-a2197-00-revA, zcu208-revA and zcu216-revA already 
>>> use a forward declaration. Would it be okay to add the forward declaration if 
>>> the function is used inside the source file?
>>
>> Better to just fix this. Can you please send the patch?
> 
> The e-a2197-00-revA/psu_init_gpl.c file has an additional forward declaration of 
> the dpll_prog function.

Please fix all of these.

Thanks,
Michal