Re: Question on FIELD_PREP() for static array

[John Garry <john.garry@huawei.com>]

On 11/10/2018 16:23, Johannes Berg wrote:

Hi

> On Thu, 2018-10-11 at 15:24 +0100, John Garry wrote:
>>
>>> +#define BUILD_BUG_ON_RET_ZERO(cond) (sizeof(char[1 - 2*!!(cond)]) - 1)
>>> +#define BUILD_BUG_ON_NOT_POW2_RET_ZERO(n)	BUILD_BUG_ON_RET_ZERO(((n) & ((n) - 1)) != 0)
>
>> Seems reasonable. However I did try this and was getting compiler
>> warnings about VLA, from a non-constant being fed into
>> BUILD_BUG_ON_RET_ZERO(), related to sizeof char[]:
>> drivers/iio/adc/meson_saradc.c:375:2: warning: ISO C90 forbids variable
>> length array [-Wvla]
>>    regval = FIELD_PREP(MESON_SAR_ADC_CHAN_LIST_ENTRY_MASK(0),
>
> Hmm, what's the code there?

Nothing special, it was just a sample. Here'e the code:
     regval = FIELD_PREP(MESON_SAR_ADC_CHAN_LIST_ENTRY_MASK(0),
                 chan->address);

So val is a variable, and I find if remove both of the 
BUILD_BUG_ON_RET_ZERO()' which use __bf_shf() then it goes away.

>
> I don't see why the compiler should think it's a variable length?
>
>> Surely __NLA_ENSURE is getting a similar issue as it uses a similar
>> principle, no? I see that this is in -next now, but could not this macro
>> or derivatives being referenced.
>
> Yeah, I have a patch now to reference it, but I don't see anything from
> -Wvla with gcc 8.1?

I'm using a 7.3.1-based toolchain

>
> See
> https://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next.git/commit/?id=3d7af878357acd9e37fc156928106f1a969c8942
> and its parent.
>
> Do you see -Wvla warnings there? Any idea how I could reproduce them?

I'll try it, thanks

John

>
> johannes
>
> .
>