do not read buffers out of bounds -- load only the 4 bytes we know exist here. Could also have done a vld1_lane_f32 but doing so here, without the overhead of initializing the unused lane, would have triggered used-of-uninitialized-value errors in tools such as ASan. Note that this code is sub-optimal before or after this change: we should be reading either 2 or 4 float32 values per load-instruction (2 for ARM in-order cores with an affinity for 8-byte loads; 4 for ARM out-of-order cores able to dual-issue 16-byte load instructions with arithmetic instructions). Before or after this patch, we are only loading 4 bytes of useful data here (even if before this patch, we were technically loading 8, only to use only the 4 first).

1 files changed, 3 insertions, 3 deletions

diff --git a/Eigen/src/Core/products/GeneralBlockPanelKernel.h b/Eigen/src/Core/products/GeneralBlockPanelKernel.h
index 5619a4588..9ca865bd1 100644
--- a/Eigen/src/Core/products/GeneralBlockPanelKernel.h
+++ b/Eigen/src/Core/products/GeneralBlockPanelKernel.h
@@ -859,7 +859,7 @@ template<>
 struct gebp_traits <float, float, false, false,Architecture::NEON>
  : gebp_traits<float,float,false,false,Architecture::Generic>
 {
-  typedef float32x2_t RhsPacket;
+  typedef float RhsPacket;
 
   EIGEN_STRONG_INLINE void broadcastRhs(const RhsScalar* b, RhsPacket& b0, RhsPacket& b1, RhsPacket& b2, RhsPacket& b3)
   {
@@ -871,7 +871,7 @@ struct gebp_traits <float, float, false, false,Architecture::NEON>
 
   EIGEN_STRONG_INLINE void loadRhs(const RhsScalar* b, RhsPacket& dest) const
   {
-     dest = vld1_f32(b);
+     dest = *b;
   }
 
   EIGEN_STRONG_INLINE void loadRhsQuad(const RhsScalar* b, RhsPacket& dest) const
@@ -881,7 +881,7 @@ struct gebp_traits <float, float, false, false,Architecture::NEON>
 
   EIGEN_STRONG_INLINE void madd(const LhsPacket& a, const RhsPacket& b, AccPacket& c, RhsPacket& /*tmp*/) const
   {
-    c = vfmaq_lane_f32(c, a, b, 0);
+    c = vfmaq_n_f32(c, a, b);
   }
 };