From 79e0a9f32aa067850ed21e9744490ccec3cc4dd7 Mon Sep 17 00:00:00 2001
From: Matt Pharr <matt.pharr@intel.com>
Date: Fri, 8 Jun 2012 14:56:18 -0700
Subject: [PATCH] Fix codegen bug with foreach_tiled.

When the outermost dimension(s) were partially active, but the innermost
dimension was all on, we'd inadvertently use an incorrect "all on"
execution mask.

Fixes issues #177 and #200.
---
 stmt.cpp | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/stmt.cpp b/stmt.cpp
index 974af871..25d1d3d5 100644
--- a/stmt.cpp
+++ b/stmt.cpp
@@ -1649,11 +1649,12 @@ ForeachStmt::EmitCode(FunctionEmitContext *ctx) const {
     // width.  Set the mask and jump to the masked loop body.
     ctx->SetCurrentBasicBlock(bbAllInnerPartialOuter); {
         llvm::Value *mask;
-        if (extrasMaskPtrs.size() == 0)
+        if (nDims == 1)
             // 1D loop; we shouldn't ever get here anyway
             mask = LLVMMaskAllOff;
         else
-            mask = ctx->LoadInst(extrasMaskPtrs.back());
+            mask = ctx->LoadInst(extrasMaskPtrs[nDims-2]);
+
         ctx->SetInternalMask(mask);
 
         ctx->StoreInst(LLVMTrue, stepIndexAfterMaskedBodyPtr);