[llvm] [LoopInterchange] Motivating example for interchange enablement. NFC. (PR #171631)

[Ryotaro Kasuga via llvm-commits]

================
@@ -0,0 +1,543 @@
+; RUN: opt < %s -passes=loop-interchange -cache-line-size=64 -pass-remarks-missed='loop-interchange' -pass-remarks-output=%t -S
+; RUN: FileCheck --input-file=%t %s
+
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128-Fn32"
+
+; The IR test case below is a full and representative example of a more complex
+; loop nest structure where we would like to see loops 'j' and 'JL' get
+; interchanged. This is the pseudo-code of the loop nest to get and idea of
+; the stucture:
+;
+;      for L=1 to NX
+;       for M=1 to NY
+;        for i=1 to NX
+;         for j=1 to NY
+;          for IL=1 to NX
+;           for JL=1 to NY
+;             GlobC(j,JL,M) + GlobG(j,JL,M) + GlobE(j,JL,M) + GlobI(j,JL,M)
+;           End
+;          End
----------------
kasuga-fj wrote:

```suggestion
;          for IL=1 to NX
;           load GlobC(i,IL,L)
;           load GlobG(i,IL,L)
;           load GlobE(i,IL,L)
;           load GlobI(i,IL,L)
;           for JL=1 to NY
;             load GlobD(j,JL,M)
;             load GlobH(j,JL,M)
;             load GlobF(j,JL,M)
;             load GlobJ(j,JL,M)
;             store GlobL(NY*i+j,NY*IL+JL)
;           End
;          End
```

If I understand correctly, this pseudo code is not consistent with the IR. If you find this format awkward, please feel free to correct it.

https://github.com/llvm/llvm-project/pull/171631