[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
clattner at apple.com
Sun Mar 8 12:54:43 PDT 2009
On Mar 8, 2009, at 12:52 PM, Duncan Sands wrote:
>>>> This is converting a memcpy to a load/store of the same size, so it
>>>> must be a multiple of a byte in size.
>>> How is the size measured? By getTypeSizeInBits,
>>> or getTypePaddedSizeInBits? The first one is ok, the others not.
>> Padded Size. Why do you think it isn't safe? Have an example?
> Sure, consider an i24. The padded size is 32 bits. A memcpy will
> copy 32 bits, a load/store will copy 24 bits.
This xform would turn it into an i32, not an i24.
More information about the llvm-commits