[PATCH] D11605: [LoopUnswitch] Add block frequency analysis to recognize hot/cold regions

[Chen Li via llvm-commits]

> On Sep 10, 2015, at 12:04 PM, Xinliang David Li <davidxl at google.com> wrote:
> 
> Right. I won't trust static profile data to tell you what is cold or hot.
> 
> In fact I think the right way to determine coldness is to use the
> actual profile count of the loop header to determine loop is cold or
> not -- the actual exec count of a loop header can be computed by
> multiplying the relative frequency with the the F.getEntryCount().   A
> common threshold can be 0 or 1, which means the loop rarely executes.

That makes sense. I will introduce profile count in an improvement patch later.

thanks,
chen


> 
> David
> 
> 
> 
> On Thu, Sep 10, 2015 at 11:48 AM, Chen Li <meloli87 at gmail.com> wrote:
>> chenli added inline comments.
>> 
>> ================
>> Comment at: lib/Transforms/Scalar/LoopUnswitch.cpp:441
>> @@ +440,3 @@
>> +
>> +  if (LoopUnswitchWithBlockFrequency) {
>> +    BranchProbabilityInfo BPI(*F, *LI);
>> ----------------
>> davidxl wrote:
>>> Note that this also triggers the check when PGO is not enabled (e.g, with static prediction).
>>> 
>>> To check if PGO is on, do:
>>> 
>>>  HasProfileData = F.getEntryCount().hasValue();
>>>  if (HasProfileData) {
>>> 
>>> ..
>> Do you mean that without PGO enabled, this should never be triggered?
>> 
>> 
>> http://reviews.llvm.org/D11605
>> 
>> 
>>