[polly] r271264 - [GSoC 2016] [Polly] [FIX] Determination of statements that contain matrix

[Roman Gareev via llvm-commits]

Author: romangareev
Date: Tue May 31 06:22:21 2016
New Revision: 271264

URL: http://llvm.org/viewvc/llvm-project?rev=271264&view=rev
Log:
[GSoC 2016] [Polly] [FIX] Determination of statements that contain matrix
multiplication

Fix small issues related to characters, operators  and descriptions of tests.

Differential Revision: http://reviews.llvm.org/D20806

Modified:
    polly/trunk/include/polly/ScheduleOptimizer.h
    polly/trunk/lib/Transform/ScheduleOptimizer.cpp
    polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts.ll
    polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts_2.ll

Modified: polly/trunk/include/polly/ScheduleOptimizer.h
URL: http://llvm.org/viewvc/llvm-project/polly/trunk/include/polly/ScheduleOptimizer.h?rev=271264&r1=271263&r2=271264&view=diff
==============================================================================
--- polly/trunk/include/polly/ScheduleOptimizer.h (original)
+++ polly/trunk/include/polly/ScheduleOptimizer.h Tue May 31 06:22:21 2016
@@ -179,7 +179,7 @@ private:
   ///    the outer loop).
   /// 4. all memory accesses of the statement except from the last one, are
   ///    read memory access and the last one is write memory access.
-  /// 5. all subscripts of the last memory access of the statement don’t
+  /// 5. all subscripts of the last memory access of the statement don't
   ///    contain the variable used in the inner loop.
   /// If this is the case, we could try to use an approach that is similar to
   /// the one used to get close-to-peak performance of matrix multiplications.

Modified: polly/trunk/lib/Transform/ScheduleOptimizer.cpp
URL: http://llvm.org/viewvc/llvm-project/polly/trunk/lib/Transform/ScheduleOptimizer.cpp?rev=271264&r1=271263&r2=271264&view=diff
==============================================================================
--- polly/trunk/lib/Transform/ScheduleOptimizer.cpp (original)
+++ polly/trunk/lib/Transform/ScheduleOptimizer.cpp Tue May 31 06:22:21 2016
@@ -233,7 +233,7 @@ static __isl_give isl_set *addExtentCons
 /// 2. Extend it to a set, which has exactly VectorWidth iterations for
 ///    any prefix from the set that was built on the previous step.
 /// 3. Subtract loop domain from it, project out the vector loop dimension and
-///    get a set of prefixes, which don’t have exactly VectorWidth iterations.
+///    get a set of prefixes, which don't have exactly VectorWidth iterations.
 /// 4. Subtract it from all prefixes of the vector loop and get the desired
 ///    set.
 ///
@@ -431,7 +431,7 @@ static bool isInputDimUsed(__isl_take is
 ///    loop to the outer loop).
 /// 2. all memory accesses of the statement except from the last one, are
 ///    read memory access and the last one is write memory access.
-/// 3. all subscripts of the last memory access of the statement don’t contain
+/// 3. all subscripts of the last memory access of the statement don't contain
 ///    the variable used in the inner loop.
 ///
 /// @param PartialSchedule The PartialSchedule that contains a SCoP statement
@@ -445,14 +445,14 @@ static bool containsMatrMult(__isl_keep
   auto MemA = ScpStmt->begin();
   for (unsigned i = 0; i < ScpStmt->size() - 2 && MemA != ScpStmt->end();
        i++, MemA++)
-    if (!(*MemA)->isRead() or
-        ((*MemA)->isArrayKind() and
-         !((*MemA)->isStrideOne(isl_map_copy(PartialSchedule)) or
+    if (!(*MemA)->isRead() ||
+        ((*MemA)->isArrayKind() &&
+         !((*MemA)->isStrideOne(isl_map_copy(PartialSchedule)) ||
            (*MemA)->isStrideZero(isl_map_copy(PartialSchedule)))))
       return false;
   MemA++;
-  if (!(*MemA)->isWrite() or !(*MemA)->isArrayKind() or
-      !((*MemA)->isStrideOne(isl_map_copy(PartialSchedule)) or
+  if (!(*MemA)->isWrite() || !(*MemA)->isArrayKind() ||
+      !((*MemA)->isStrideOne(isl_map_copy(PartialSchedule)) ||
         (*MemA)->isStrideZero(isl_map_copy(PartialSchedule))))
     return false;
   auto DimNum = isl_map_dim(PartialSchedule, isl_dim_in);

Modified: polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts.ll
URL: http://llvm.org/viewvc/llvm-project/polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts.ll?rev=271264&r1=271263&r2=271264&view=diff
==============================================================================
--- polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts.ll (original)
+++ polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts.ll Tue May 31 06:22:21 2016
@@ -1,8 +1,20 @@
 ; RUN: opt %loadPolly -polly-opt-isl -debug < %s 2>&1| FileCheck %s
 ; RUN: opt %loadPolly -polly-opt-isl -polly-pattern-matching-based-opts=true -debug < %s 2>&1| FileCheck %s --check-prefix=PATTERN-MATCHING-OPTS
 ; REQUIRES: asserts
+;
+;    /* C := alpha*A*B + beta*C */
+;    for (i = 0; i < _PB_NI; i++)
+;      for (j = 0; j < _PB_NJ; j++)
+;        {
+;	   C[i][j] *= beta;
+;	   for (k = 0; k < _PB_NK; ++k)
+;	     C[i][j] += alpha * A[i][k] * B[k][j];
+;        }
+;
 ; CHECK-NOT: The matrix multiplication pattern was detected
 ; PATTERN-MATCHING-OPTS: The matrix multiplication pattern was detected
+;
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 
 define internal void @kernel_gemm(i32 %arg, i32 %arg1, i32 %arg2, double %arg3, double %arg4, [1056 x double]* %arg5, [1024 x double]* %arg6, [1056 x double]* %arg7) {
 bb:

Modified: polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts_2.ll
URL: http://llvm.org/viewvc/llvm-project/polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts_2.ll?rev=271264&r1=271263&r2=271264&view=diff
==============================================================================
--- polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts_2.ll (original)
+++ polly/trunk/test/ScheduleOptimizer/pattern-matching-based-opts_2.ll Tue May 31 06:22:21 2016
@@ -1,6 +1,22 @@
 ; RUN: opt %loadPolly -polly-opt-isl -polly-pattern-matching-based-opts=true -debug < %s 2>&1 | FileCheck %s
 ; REQUIRES: asserts
+;
+;    /* C := alpha*A*B + beta*C */
+;    for (i = 0; i < _PB_NI; i++)
+;      for (j = 0; j < _PB_NJ; j += 2)
+;        {
+;	   C[i][j] *= beta;
+;	   for (k = 0; k < _PB_NK; ++k)
+;	     C[i][j] += alpha * A[i][k] * B[k][j];
+;        }
+;
+; Check that we won’t detect the matrix multiplication pattern,
+; if, for example, there are memory accesses that have stride 2
+; after the interchanging of loops.
+;
 ; CHECK-NOT: The matrix multiplication pattern was detected
+;
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 
 define internal void @kernel_gemm(i32 %arg, i32 %arg1, i32 %arg2, double %arg3, double %arg4, [1056 x double]* %arg5, [1024 x double]* %arg6, [1056 x double]* %arg7) {
 bb: