[LLVMdev] HazardRecognizer and RegisterAllocation

Scott Michel scottm at aero.org
Mon Jan 19 18:39:45 PST 2009


Dan:

CellSPU could clearly benefit from the post-RA scheduler. In fact, we  
were thinking about writing a machine pass of our own.

One thing that does "disturb" me is that both HazardRecognizer and  
post-RA sched assume there's only one kind of NOP. For Cell, there  
are two, depending upon the pipeline being filled. Pipe 0 takes  
"ENOP" whereas Pipe 1 take "LNOP" (depends on the low order PC bits.)


-scooter

On Jan 19, 2009, at 2:42 PM, Dan Gohman wrote:

>
> On Jan 19, 2009, at 11:01 AM, Patrick Boettcher wrote:
>>
>>
>>>> And more generally: Is the hazardRecognizer the right and only way
>>>> to
>>>> solve our NOOP-minimizing problem?
>>>
>>> Perhaps you want to do this after register allocation is done.  
>>> Dan is
>>> developing the post-allocation scheduler. You can try it out.
>>
>> Interesting. Can it already be found SVN? I will search the mail
>> archive
>> later, if not.
>
> Yes, it is in SVN. It's new, and so far it's only being used in  
> limited
> ways, and not anything like your specific problem. We don't currently
> have any targets in-tree that require no-ops, so it may not address  
> all
> your needs out of the box, but patches are welcome :-).
>
> Dan
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev




More information about the llvm-dev mailing list