Re: [PATCH v2 3/3] io: Add FORTIFY_SOURCE check for fcntl arguments

[Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>]

On 30/05/23 06:50, Florian Weimer wrote:
> * Sergey Bugaev:
> 
>> Hello,
>>
>> On Tue, May 30, 2023 at 10:41 AM Florian Weimer <fweimer@redhat.com> wrote:
>>> The noinline attribute does exactly what it says: it prevents inlining,
>>> but it does not prevent inter-procedural analysis.  So if a function
>>> returns a constrant, the compiler will still use that constant
>>> elsewhere.  Newer GCC versions support the noipa attribute.  If you drop
>>> the static, you can use the weak attribute for compatibility with pretty
>>> much all GCC versions.
>>>
>>> But I think using volatile (without noinline for clarity) is fine here.
>>
>> So do I read this right that you're saying I should just do
>>
>> static int
>> hide_constant (int value)
>> {
>>   volatile int v = value;
>>   return v;
>> }
>>
>> ? After the yesterday's discussion with Adhemerval, I was thinking of doing
>>
>> static inline int
>> hide_constant (int value)
>> {
>>   asm ("" : "+rm" (value));
>>   return value;
>> }
>>
>> This does *not* prevent the whole computation from being optimized out
>> if the result is unused (i.e. there's no 'volatile'), but it does
>> prevent the compiler from assuming anything about the return value.
> 
> I think both variants are fine in this case.  If Adhemerval prefers the
> second, please use that.

Using the second allows us to provide an internal macro that can consolidate
the DO_NOT_OPTIMIZE on benchtests as well.