PGO: Static profile affects loop alignment where it's not supposed to do

[EgorBo]

Extracted from dotnet/performance:

    private static uint[] input_uint = Enumerable.Range(0, 1000).Select(i => (uint)i).ToArray();

    [MethodImpl(MethodImplOptions.NoInlining)]
    public int TrailingZeroCount_uint()
    {
        int sum = 0;
        uint[] input = input_uint;
        for (int i = 0; i < input.Length; i++)
            sum += BitOperations.TrailingZeroCount(input[i]);
        return sum;
    }

Codegen diff: https://www.diffchecker.com/dIVV1Iwy
Left: ReadyToRun=1, Right: ReadyToRun=0

My theory that when we collect the profile for TrailingZeroCount we don't have BMI on that machine so it messes weights up which then inmported into this loop and mark it as non-profitable to align.

mibc dump:

"Method": "[S.P.CoreLib]System.Numerics.BitOperations.TrailingZeroCount(uint32)",
"InstrumentationData": [
  {
    "ILOffset": 0,
    "InstrumentationKind": "NumRuns",
    "Other": 7
  },
  {
    "ILOffset": 7,
    "InstrumentationKind": "EdgeIntCount",
    "Other": 0,
    "Data": 162837
  },
  {
    "ILOffset": 20,
    "InstrumentationKind": "EdgeIntCount",
    "Other": 0,
    "Data": 0
  },
  {
    "ILOffset": 30,
    "InstrumentationKind": "EdgeIntCount",
    "Other": 0,
    "Data": 0
  },
  {
    "ILOffset": 37,
    "InstrumentationKind": "EdgeIntCount",
    "Other": 0,
    "Data": 0
  }
]

Hm.. however, looks like MIBC is correct for BitOperations.TrailingZeroCount, "ILOffset": 7 stands for

call         unsigned int32 System.Runtime.Intrinsics.X86.Bmi1::TrailingZeroCount(unsigned int32)

[ghost]

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Extracted from dotnet/performance:

    private static uint[] input_uint = Enumerable.Range(0, 1000).Select(i => (uint)i).ToArray();

    [MethodImpl(MethodImplOptions.NoInlining)]
    public int TrailingZeroCount_uint()
    {
        int sum = 0;
        uint[] input = input_uint;
        for (int i = 0; i < input.Length; i++)
            sum += BitOperations.TrailingZeroCount(input[i]);
        return sum;
    }

Codegen diff: https://www.diffchecker.com/dIVV1Iwy
Left: ReadyToRun=1, Right: ReadyToRun=0

My theory that when we collect the profile for TrailingZeroCount we don't have BMI on that machine so it messes weights up which then inmported into this loop and mark it as non-profitable to align.

Author:	EgorBo
Assignees:	-
Labels:	tenet-performance, area-CodeGen-coreclr
Milestone:	-

[EgorBo]

cc @AndyAyersMS I think you might find this issue curious too, somehow static pgo marks loop body "cold" or rather not hot enough.

Weights with ReadyToRun=1 (using StandardOptimization.mibc):

-----------------------------------------------------------------------------------------------------------------------------------------
BBnum BBid ref try hnd preds           weight   IBC  lp [IL range]     [jump]      [EH region]         [flags]
-----------------------------------------------------------------------------------------------------------------------------------------
BB01 [0014]  1                             1           [???..???)-> BB09 ( cond )                     internal idxlen 
BB02 [0015]  1       BB01                  1           [???..???)-> BB07 ( cond )                     internal 
BB10 [0021]  1       BB02                  1           [00C..???)                                     internal idxlen LoopPH 
BB05 [0001]  2       BB05,BB10             0.49  50  0 [00C..021)-> BB05 ( cond )                     i Loop Loop0 idxlen bwd bwd-target IBC 
BB06 [0016]  1       BB05                  1           [???..???)-> BB09 (always)                     internal 
BB07 [0018]  2       BB02,BB07             0.01   1    [00C..021)-> BB07 ( cond )                     i idxlen bwd IBC 
BB09 [0003]  3       BB01,BB06,BB07        1           [021..023)        (return)                     i 
-----------------------------------------------------------------------------------------------------------------------------------------

Weights for ReadyToRun=0:

-----------------------------------------------------------------------------------------------------------------------------------------
BBnum BBid ref try hnd preds           weight    lp [IL range]     [jump]      [EH region]         [flags]
-----------------------------------------------------------------------------------------------------------------------------------------
BB01 [0014]  1                             1       [???..???)-> BB09 ( cond )                     internal idxlen 
BB02 [0015]  1       BB01                  1       [???..???)-> BB07 ( cond )                     internal 
BB10 [0021]  1       BB02                  1       [00C..???)                                     internal idxlen LoopPH 
BB05 [0001]  2       BB05,BB10             3.96  0 [00C..021)-> BB05 ( cond )                     i Loop Loop0 idxlen bwd bwd-target align 
BB06 [0016]  1       BB05                  1       [???..???)-> BB09 (always)                     internal 
BB07 [0018]  2       BB02,BB07             0.04    [00C..021)-> BB07 ( cond )                     i idxlen bwd 
BB09 [0003]  3       BB01,BB06,BB07        1       [021..023)        (return)                     i 
-----------------------------------------------------------------------------------------------------------------------------------------

[EgorBo]

I assume it's that general problem of propagating weights from profiled inlinees to non-profiled callee.