Re: [PATCH 1/24] make atomic_read() behave consistently on alpha

[Segher Boessenkool <segher@kernel.crashing.org>]

>>>> That means GCC cannot compile Linux; it already optimises
>>>> some accesses to scalars to smaller accesses when it knows
>>>> it is allowed to.  Not often though, since it hardly ever
>>>> helps in the cost model it employs.
>>>
>>> Please give an example code snippet + gcc version + arch
>>> to back this up.
>>
>> 	unsigned char f(unsigned long *p)
>> 	{
>> 	        return *p & 1;
>> 	}
>
> This doesn't really matter since we only care about the LSB.

It is exactly what I claimed, and what you asked proof of.

> Do you have an example where gcc reads it non-atmoically and
> we care about all parts?

Like I explained in the original mail; no, I suspect such
a testcase will be really hard to construct, esp. as a small
testcase.  I have no reason to believe it is impossible to
do so though -- maybe someone else can write trickier code
than I can, in which case, please do so.


Segher