[PowerPC] Correctly specify the cache line size for Power 7, 8 and 9.

Fixes PPCTTIImpl::getCacheLineSize() returning the wrong cache line size for
newer ppc processors.

Commiting on behalf of Stefan Pintilie.
Differential Revision: https://reviews.llvm.org/D33656

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304317 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 61 insertions, 3 deletions

diff --git a/lib/Target/PowerPC/PPCTargetTransformInfo.cpp b/lib/Target/PowerPC/PPCTargetTransformInfo.cpp
index 83768da3a9a..5559cdc5fe4 100644
--- a/lib/Target/PowerPC/PPCTargetTransformInfo.cpp
+++ b/lib/Target/PowerPC/PPCTargetTransformInfo.cpp
@@ -244,9 +244,18 @@ unsigned PPCTTIImpl::getRegisterBitWidth(bool Vector) {
 }
 
 unsigned PPCTTIImpl::getCacheLineSize() {
-  // This is currently only used for the data prefetch pass which is only
-  // enabled for BG/Q by default.
-  return CacheLineSize;
+  // Check first if the user specified a custom line size.
+  if (CacheLineSize.getNumOccurrences() > 0)
+    return CacheLineSize;
+
+  // On P7, P8 or P9 we have a cache line size of 128.
+  unsigned Directive = ST->getDarwinDirective();
+  if (Directive == PPC::DIR_PWR7 || Directive == PPC::DIR_PWR8 ||
+      Directive == PPC::DIR_PWR9)
+    return 128;
+
+  // On other processors return a default of 64 bytes.
+  return 64;
 }
 
 unsigned PPCTTIImpl::getPrefetchDistance() {
diff --git a/test/CodeGen/PowerPC/ppc64-get-cache-line-size.ll b/test/CodeGen/PowerPC/ppc64-get-cache-line-size.ll
new file mode 100644
index 00000000000..7ca5332865c
--- /dev/null
+++ b/test/CodeGen/PowerPC/ppc64-get-cache-line-size.ll
@@ -0,0 +1,49 @@
+; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -enable-ppc-prefetching=true | FileCheck %s
+; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -enable-ppc-prefetching=true -ppc-loop-prefetch-cache-line=64 | FileCheck %s -check-prefix=CHECK-DCBT
+; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 -enable-ppc-prefetching=true | FileCheck %s
+; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 -enable-ppc-prefetching=true -ppc-loop-prefetch-cache-line=64 | FileCheck %s -check-prefix=CHECK-DCBT
+; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 -enable-ppc-prefetching=true | FileCheck %s
+; RUN: llc < %s -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 -enable-ppc-prefetching=true -ppc-loop-prefetch-cache-line=64 | FileCheck %s -check-prefix=CHECK-DCBT
+; RUN: llc < %s -march=ppc64 -mcpu=a2 -enable-ppc-prefetching=true | FileCheck %s -check-prefix=CHECK-DCBT
+
+; Function Attrs: nounwind
+define signext i32 @check_cache_line() local_unnamed_addr {
+entry:
+  %call = tail call i32* bitcast (i32* (...)* @magici to i32* ()*)()
+  %call115 = tail call signext i32 bitcast (i32 (...)* @iter to i32 ()*)()
+  %cmp16 = icmp sgt i32 %call115, 0
+  br i1 %cmp16, label %for.body, label %for.cond.cleanup
+
+for.cond.cleanup:                                 ; preds = %for.body, %entry
+  %res.0.lcssa = phi i32 [ 0, %entry ], [ %add5, %for.body ]
+  ret i32 %res.0.lcssa
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %res.017 = phi i32 [ %add5, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i32, i32* %call, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %res.017
+  %1 = add nuw nsw i64 %indvars.iv, 16
+  %arrayidx4 = getelementptr inbounds i32, i32* %call, i64 %1
+  %2 = load i32, i32* %arrayidx4, align 4
+  %add5 = add nsw i32 %add, %2
+  %indvars.iv.next = add nuw i64 %indvars.iv, 1
+  %call1 = tail call signext i32 bitcast (i32 (...)* @iter to i32 ()*)()
+  %3 = sext i32 %call1 to i64
+  %cmp = icmp slt i64 %indvars.iv.next, %3
+  br i1 %cmp, label %for.body, label %for.cond.cleanup
+; CHECK-LABEL: check_cache_line
+; CHECK: dcbt
+; CHECK-NOT: dcbt
+; CHECK: blr
+; CHECK-DCBT-LABEL: check_cache_line
+; CHECK-DCBT: dcbt
+; CHECK-DCBT: dcbt
+; CHECK-DCBT: blr
+}
+
+declare i32* @magici(...) local_unnamed_addr
+
+declare signext i32 @iter(...) local_unnamed_addr
+