From a98271b0817adbae934054bd48320704049f0239 Mon Sep 17 00:00:00 2001
From: Wang Huan <wanghuan29@baidu.com>
Date: Fri, 18 Apr 2025 07:37:01 +0000
Subject: [PATCH 1/2] refine forrange

---
 paddle/phi/kernels/funcs/for_range.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/paddle/phi/kernels/funcs/for_range.h b/paddle/phi/kernels/funcs/for_range.h
index 7ff8ac7177a3de..fab56a238d075f 100644
--- a/paddle/phi/kernels/funcs/for_range.h
+++ b/paddle/phi/kernels/funcs/for_range.h
@@ -60,7 +60,7 @@ __global__ static void ForRangeElemwiseOp(Function func, unsigned int limit) {
 template <typename Function>
 __global__ static void ForRangeElemwiseOpLargeSize(Function func,
                                                    size_t limit) {
-  size_t idx =
+  ` size_t idx =
       static_cast<size_t>(blockIdx.x) * static_cast<size_t>(blockDim.x) +
       static_cast<size_t>(threadIdx.x);
   if (idx < limit) {
@@ -89,7 +89,8 @@ struct ForRange<phi::GPUContext> {
     if (grid_size == 1) {
       ForRangeElemwiseOpGridIsOne<<<1, block_size, 0, dev_ctx_.stream()>>>(
           func);
-    } else if (limit_ > std::numeric_limits<unsigned int>::max()) {
+    } else if (block_size * grid_size >
+               std::numeric_limits<unsigned int>::max()) {
       ForRangeElemwiseOpLargeSize<<<grid_size,
                                     block_size,
                                     0,

From 2e5fa4dcecac9cda3cfd155b5c2cb6f7b76f492e Mon Sep 17 00:00:00 2001
From: Wang Huan <wanghuan29@baidu.com>
Date: Fri, 18 Apr 2025 07:38:22 +0000
Subject: [PATCH 2/2] refine forrange

---
 paddle/phi/kernels/funcs/for_range.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/paddle/phi/kernels/funcs/for_range.h b/paddle/phi/kernels/funcs/for_range.h
index fab56a238d075f..227b1cdf504903 100644
--- a/paddle/phi/kernels/funcs/for_range.h
+++ b/paddle/phi/kernels/funcs/for_range.h
@@ -60,7 +60,7 @@ __global__ static void ForRangeElemwiseOp(Function func, unsigned int limit) {
 template <typename Function>
 __global__ static void ForRangeElemwiseOpLargeSize(Function func,
                                                    size_t limit) {
-  ` size_t idx =
+  size_t idx =
       static_cast<size_t>(blockIdx.x) * static_cast<size_t>(blockDim.x) +
       static_cast<size_t>(threadIdx.x);
   if (idx < limit) {