[llvm-commits] [llvm] r66366 - in /llvm/trunk: lib/Transforms/Scalar/ScalarReplAggregates.cpp test/Transforms/ScalarRepl/2008-06-22-LargeArray.ll test/Transforms/ScalarRepl/vector_memcpy.ll

[Chris Lattner]

On Mar 8, 2009, at 3:36 AM, Duncan Sands wrote:
> Hi Chris,
>
>> +    // If this is a memcpy or memmove into or out of the whole  
>> allocation, we
>> +    // can handle it like a load or store of the scalar type.
>
> a load+store of a floating point type might trap, while a memcpy/ 
> memmove would
> not.

How so?  Load of double works on any bit pattern, so long as the  
address is correct and the alignment is right etc.

>  Also, a memcpy/memmove of i33 may not the same as a load+store, due  
> to
> the extra bits off the end (load of an i33 is only defined if the  
> original
> value was stored using an i33 store - this way we can assume any  
> extra bits
> are zero, because the store puts zero there; if someone wanted to  
> avoid this
> undefinedness they might have used memcpy/memmove).

This is converting a memcpy to a load/store of the same size, so it  
must be a multiple of a byte in size.

-Chris