[llvm-bugs] [Bug 37697] New: [llvm-exegesis] Idealized Resource Pressure misinterprets ProcResGroups

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=37697

            Bug ID: 37697
           Summary: [llvm-exegesis] Idealized Resource Pressure
                    misinterprets ProcResGroups
           Product: new-bugs
           Version: unspecified
          Hardware: PC
                OS: Windows NT
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: new bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: llvm-dev at redking.me.uk
                CC: andrea.dibiagio at gmail.com, clement.courbet at gmail.com,
                    gchatelet at google.com, llvm-bugs at lists.llvm.org,
                    spatel+llvm at rotateright.com

I've seeing cases where the Idealized Resource Pressure calculation is
idealising for use on resources that the model doesn't refer to.

These seem to be caused by having common root groups. 

For instance, on btver2:

def JFPU0 : ProcResource<1>; // Vector/FPU Pipe0: VALU0/VIMUL/FPA
def JFPU1 : ProcResource<1>; // Vector/FPU Pipe1: VALU1/STC/FPM

// Fpu Pipe Scheduler
def JFPU01 : ProcResGroup<[JFPU0, JFPU1]> {
  let BufferSize=18;
}

For instructions that only use JFPU0 or JFPU1, we see idealized predictions
using JFPU01 but then get applied to the other resources:

WriteProcRes:
JFPU01: 1
JFPU0: 1

Idealized Resource Pressure
JFPU0: 0.50 <--- HUH?
JFPU1: 1.50

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20180606/510318d4/attachment.html>