[llvm-commits] [llvm] r77407 - in /llvm/trunk: include/llvm/IntrinsicsX86.td lib/Target/X86/X86ISelLowering.cpp lib/Target/X86/X86ISelLowering.h lib/Target/X86/X86InstrSSE.td test/CodeGen/X86/sse41.ll

[Chris Lattner]

On Jul 28, 2009, at 10:10 PM, Eli Friedman wrote:

> On Tue, Jul 28, 2009 at 9:44 PM, Chris Lattner<clattner at apple.com>  
> wrote:
>>  From my understanding, I think that ptest is commutative (but maybe
>> only for "z", but not the others?).  If unconditionally true, you can
>> declare this with [SDNPCommutative] as a third argument to SDNode and
>> the instruction should be figured out to be commutative.  If only  
>> true
>> in some cases, more trickery will be required.
>
> For PTEST, the ZF flag is commutative, but the CF flag is not.

Ok, that's what I feared.  This means that we need conditional flag  
liveness stuff?  Is there a way to know (e.g. in regalloc, when doing  
a spill) that only one flag is used?  Do we have flag subregs?

Perhaps this should just go in the readme-sse.txt file to worry about  
later.

-Chris

>
>
>> It should allow the instruction to fold the load instead of  
>> producing:
>>
>> _test:
>>        movl    4(%esp), %eax
>>        movaps  (%eax), %xmm1
>>        ptest   %xmm0, %xmm1
>>        sete    %al
>>        movzbl  %al, %eax
>>        ret
>>
>> Does "ptest" need to be added to the load folding table in case it is
>> isel'd as reg/reg but an operand gets spilled (thus the regalloc
>> should fold the load by forming reg/mem)?
>
> I think the issue here is that it doesn't know it's safe to commute  
> the ptest...
>
> -Eli
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits