[llvm] r176408 - recommit r172363 & r171325 (reverted in r172756)

Shuxin Yang shuxin.llvm at gmail.com
Sun Mar 3 16:43:37 PST 2013


On 03/03/2013 04:24 PM, Arnold Schwaighofer wrote:
> Hi Shuxin,
>
> On Mar 3, 2013, at 5:45 PM, Shuxin Yang <shuxin.llvm at gmail.com> wrote:
>
>> Hi, Arnold:
>>
>>   You are right. As I put in the report to the pr14988, my concerns have two parts:
>>    1) in alias analysis context, while Nuno's change dose not seem correctness problem for what we have today.
>>        It likely has problem in the future, as it deviate the original meaning of class ObjectSizeOffsetVisitor.
>>        I gave a contrived problem in the previous mail.
>>
> I think we are mostly in wild agreement here :).
>
> I want APIs with clearly defined semantics as do you.
>
> I am saying if we want Nuno's functionality we (he ;) will have to clean up code:
>
> Introduce several functions and use them appropriately, possibly rename
> ObjectSizeOffsetVisitor to ObjectsSizeOffsetVisitor (or something similar), etc . This will prevent any future confusion.
>
> Future code will have to respect exising code or use different functions: getUnderlyingObjectSize vs getUnderlyingObjectsSize, etc.
>
> By introducing a clearly defined API, future code will be aware that there is a difference and bugs that could arise because of unawareness can be prevented.
>
>
Yes, that is on right track!






More information about the llvm-commits mailing list