From 1a97e58ffa40962dd49d1baf67d66418a6450089 Mon Sep 17 00:00:00 2001
From: Fan <fanhaozh@38f9d3e5ca30.ant.amazon.com>
Date: Thu, 15 Aug 2019 16:47:47 +0800
Subject: [PATCH] add comments for the tricky block matrix multiplication

---
 src/operator/numpy/linalg/np_gesvd-inl.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/operator/numpy/linalg/np_gesvd-inl.h b/src/operator/numpy/linalg/np_gesvd-inl.h
index efcc65ae3f47..7ce3078dae66 100644
--- a/src/operator/numpy/linalg/np_gesvd-inl.h
+++ b/src/operator/numpy/linalg/np_gesvd-inl.h
@@ -236,6 +236,7 @@ struct gesvd_backward {
 
     // G3:
     gemm::op(tempM, V, dA, DType(1.0), DType(1.0), false, false, s);
+    // dA <- dot(UT, dA). Loop over (k, m, m) blocks to avoid large temporary memory
     for (int i = 0; i < n; i += m) {
       int ncols = n - i < m ? n - i : m;
       Tensor<xpu, 3, DType> t = Tensor<xpu, 3, DType>(dA.dptr_ + i,