On Sat, Jan 20, 2018 at 9:50 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Sat, Jan 20, 2018 at 8:31 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>> On Fri, Jan 19, 2018 at 7:57 PM, Martin Sebor <msebor@gmail.com> wrote:
>>> On 01/19/2018 10:14 AM, Martin Sebor wrote:
>>>>
>>
>>> After reading the Clang code review for the warning
>>> (https://reviews.llvm.org/D20561) and experimenting with a few
>>> more test cases I noticed some additional false negatives that
>>> I think would be worthwhile diagnosing:
>>>
>>>   struct A {
>>>     int i;
>>>   } __attribute__ ((packed));
>>>
>>>   int f (struct A *p)
>>>   {
>>>     return *&p->i;
>>>   }
>>
>> I now got
>>
>> [hjl@gnu-tools-1 pr51628]$ cat a1.i
>> struct A {
>>   int i;
>> } __attribute__ ((packed));
>>
>> int f (struct A *p)
>> {
>>   return *&p->i;
>> }
>> [hjl@gnu-tools-1 pr51628]$ make a1.s
>> /export/build/gnu/gcc/build-x86_64-linux/gcc/xgcc
>> -B/export/build/gnu/gcc/build-x86_64-linux/gcc/ -O2 -S a1.i
>> a1.i: In function ‘f’:
>> a1.i:7:10: warning: taking address of packed member of ‘struct A’ may
>> result in an unaligned pointer value [-Waddress-of-packed-member]
>>    return *&p->i;
>>           ^~~~~~
>> [hjl@gnu-tools-1 pr51628]$
>
>
> It is wrong to warn "*&p->i;" since GCC always converts to "p->i;".  Unless
> we want to warn
>
> [hjl@gnu-tools-1 pr51628]$ cat g5.i
> struct A {
>   int i;
> } __attribute__ ((packed));
>
> int f (struct A *p)
> {
>   return p->i;
> }
>
> we shouldn't warn:
>
> [hjl@gnu-tools-1 pr51628]$ cat a1.i
> struct A {
>   int i;
> } __attribute__ ((packed));
>
> int f (struct A *p)
> {
>   return *&p->i;
> }
> [hjl@gnu-tools-1 pr51628]$
>
> There is no unaligned load here.  I am testing a new patch.
>

Here is the updated patch.  Tested on x86-64 and i686.


-- 
H.J.