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

[Sebastian Pop via llvm-commits]

================
@@ -0,0 +1,155 @@
+; RUN: opt < %s -passes=loop-interchange -cache-line-size=64 -pass-remarks='loop-interchange' -pass-remarks-missed='loop-interchange' -pass-remarks-output=%t -disable-output -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"
+
+; This is a reduced test case for the example in  "large-nested-6d.ll". For a
+; full description of the purpose this test and its complexities, see that file.
+;
+; This reproducer contains the perfectly nested sub part of that bigger loop
+; nest:
+;
+;        for i=1 to NX
+;         for j=1 to NY
+;          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
+;         End
+;        End
+;
+; This reproducer is useful to focus on only on the 2nd challenge: the data
+; dependence analysis problem, and not worry about the rest of loop nest
+; structure.
----------------
sebpop wrote:

Thanks for reducing the testcase even further and focusing on only one aspect of the interchange at a time.

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