Re: [PATCH v2 2/6] configure: avoid new clang 11+ warnings

[Daniele Buono <dbuono@linux.vnet.ibm.com>]

So what I would do for the warning in IplParameterBlock is the following
(if I got the comments right):

- Remove IplBlockFcp from IplParameterBlock
- Keep IPLBlockPV and, in its declaration, use
struct IPLBlockPVComp components[0];

Now for the IplBlockFcp struct declaration, it does not seem to be used
anywhere now.
I could either keep it as it was before (with the variable-size array)
or remove it entirely.
I guess this is more a question for the maintainers, what is your
preference here?

Daniele

On 10/27/2020 7:38 AM, Cornelia Huck wrote:
> On Tue, 27 Oct 2020 12:26:21 +0100
> Thomas Huth <thuth@redhat.com> wrote:
> 
>> On 26/10/2020 16.12, Paolo Bonzini wrote:
>>> On 26/10/20 16:03, Daniele Buono wrote:
>>>> Hi Paolo,
>>>> I reorganized UASStatus to put uas_iu at the end and it works fine.
>>>> Unfortunately, this uncovered another part of the code with a similar
>>>> issue (variable sized type not at the end of the struct), here:
>>>>
>>>> In file included from ../qemu-cfi-v3/target/s390x/diag.c:21:
>>>> ../qemu-cfi-v3/hw/s390x/ipl.h:161:23: error: field 'iplb' with variable
>>>> sized type 'IplParameterBlock' (aka 'union IplParameterBlock') not at
>>>> the end of a struct or class is a GNU extension
>>>> [-Werror,-Wgnu-variable-sized-type-not-at-end]
>>>>      IplParameterBlock iplb;
>>>>                        ^
>>>> ../qemu-cfi-v3/hw/s390x/ipl.h:162:23: error: field 'iplb_pv' with
>>>> variable sized type 'IplParameterBlock' (aka 'union IplParameterBlock')
>>>> not at the end of a struct or class is a GNU extension
>>>> [-Werror,-Wgnu-variable-sized-type-not-at-end]
>>>>      IplParameterBlock iplb_pv;
>>>>
>>>> My understanding is that each of these IplParameterBlock may contain
>>>> either a IPLBlockPV or a IplBlockFcp, which both end with a variable
>>>> sized field (an array).
>>>>
>>>> Adding maintainers of s390x to see if they have a suggested solution to
>>>> avoid disabling the warning.
>>>
>>> This one seems okay because the union constrains the size to 4K. If
>>> "[0]" is enough to shut up the compiler, I'd say do that.
>>
>> The "IplBlockFcp fcp" part seems to be completely unused, so I think you
>> could even remove that IplBlockFcp struct. For IPLBlockPV I agree with
>> Paolo, it's likely easiest to use [0] for that struct.
> 
> The fcp block had probably been added for completeness' sake, but we do
> not support list-directed IPL anyway. Not sure if we actually want it,
> as we use a different mechanism for IPLing from SCSI devices. So yes,
> maybe we should just drop it.
> 
>