[cfe-dev] problems with vector code

[Nate Begeman]

On Mar 23, 2010, at 11:00 AM, Jochen Wilhelmy wrote:

> 
>> That would probably be because you're compiling for x86-64 and the x86-64 ABI handling code canonicalizes vector types to<2 x double>  for argument passing.
>>   
> but what happens to the rest of the elements?
> 
> -Jochen

In the example given, a <4 x float> was being passed as a <2 x double> .  They are both 128 bits, and if the example had continued on, you'd have seen a bit convert from <2 x double> to <4 x float> in the entry block of the function, converting it back to its appropriate type without any loss of data.

Cheers,
Nate



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20100323/810b5bd7/attachment.html>