diff --git a/paddle/phi/api/yaml/legacy_backward.yaml b/paddle/phi/api/yaml/legacy_backward.yaml
index cfe7930c8c32cd..a41538806923c5 100755
--- a/paddle/phi/api/yaml/legacy_backward.yaml
+++ b/paddle/phi/api/yaml/legacy_backward.yaml
@@ -209,7 +209,7 @@
     func : GeneralTernaryGradInferMeta
     param: [input, filter, grad_out]
   kernel :
-    func : conv2d_grad_grad
+    func : conv2d_double_grad
   optional : grad_input_grad, grad_filter_grad
 
 - backward_op : conv2d_transpose_double_grad
diff --git a/paddle/phi/kernels/cpu/conv_grad_kernel.cc b/paddle/phi/kernels/cpu/conv_grad_kernel.cc
index 273b04df887fdc..0b4480b7bd4496 100644
--- a/paddle/phi/kernels/cpu/conv_grad_kernel.cc
+++ b/paddle/phi/kernels/cpu/conv_grad_kernel.cc
@@ -123,9 +123,12 @@ PD_REGISTER_KERNEL(depthwise_conv2d_grad,
 PD_REGISTER_KERNEL(
     conv3d_grad, CPU, ALL_LAYOUT, phi::Conv3DGradKernel, float, double) {}
 
-PD_REGISTER_KERNEL(
-    conv2d_grad_grad, CPU, ALL_LAYOUT, phi::ConvGradGradKernel, float, double) {
-}
+PD_REGISTER_KERNEL(conv2d_double_grad,
+                   CPU,
+                   ALL_LAYOUT,
+                   phi::ConvGradGradKernel,
+                   float,
+                   double) {}
 
 PD_REGISTER_KERNEL(conv3d_double_grad,
                    CPU,
diff --git a/paddle/phi/kernels/gpu/conv_grad_kernel.cu b/paddle/phi/kernels/gpu/conv_grad_kernel.cu
index 037a619b7a29b9..af54f82025ac6c 100644
--- a/paddle/phi/kernels/gpu/conv_grad_kernel.cu
+++ b/paddle/phi/kernels/gpu/conv_grad_kernel.cu
@@ -55,6 +55,9 @@ PD_REGISTER_KERNEL(
 PD_REGISTER_KERNEL(
     conv3d_grad, GPU, ALL_LAYOUT, phi::Conv3DGradKernel, float, double) {}
 
-PD_REGISTER_KERNEL(
-    conv2d_grad_grad, GPU, ALL_LAYOUT, phi::ConvGradGradKernel, float, double) {
-}
+PD_REGISTER_KERNEL(conv2d_double_grad,
+                   GPU,
+                   ALL_LAYOUT,
+                   phi::ConvGradGradKernel,
+                   float,
+                   double) {}
diff --git a/paddle/phi/kernels/gpudnn/conv_grad_kernel.cu b/paddle/phi/kernels/gpudnn/conv_grad_kernel.cu
index d81c03ceabb5c6..a41144c5ee7361 100644
--- a/paddle/phi/kernels/gpudnn/conv_grad_kernel.cu
+++ b/paddle/phi/kernels/gpudnn/conv_grad_kernel.cu
@@ -1473,7 +1473,7 @@ PD_REGISTER_KERNEL(depthwise_conv2d_grad,
                    phi::DepthwiseConvCudnnGradKernel,
                    float,
                    phi::dtype::float16) {}
-PD_REGISTER_KERNEL(conv2d_grad_grad,
+PD_REGISTER_KERNEL(conv2d_double_grad,
                    GPUDNN,
                    ALL_LAYOUT,
                    phi::ConvCudnnGradGradKernel,
@@ -1512,7 +1512,7 @@ PD_REGISTER_KERNEL(conv3d_grad,
                    double,
                    phi::dtype::float16,
                    phi::dtype::bfloat16) {}
-PD_REGISTER_KERNEL(conv2d_grad_grad,
+PD_REGISTER_KERNEL(conv2d_double_grad,
                    GPUDNN,
                    ALL_LAYOUT,
                    phi::ConvCudnnGradGradKernel,
@@ -1555,7 +1555,7 @@ PD_REGISTER_KERNEL(conv3d_grad,
                    double,
                    phi::dtype::float16) {}
 
-PD_REGISTER_KERNEL(conv2d_grad_grad,
+PD_REGISTER_KERNEL(conv2d_double_grad,
                    GPUDNN,
                    ALL_LAYOUT,
                    phi::ConvCudnnGradGradKernel,
diff --git a/paddle/phi/ops/compat/conv2d_sig.cc b/paddle/phi/ops/compat/conv2d_sig.cc
index 6963d6a06d8203..4e63ee19580ed4 100644
--- a/paddle/phi/ops/compat/conv2d_sig.cc
+++ b/paddle/phi/ops/compat/conv2d_sig.cc
@@ -42,7 +42,7 @@ KernelSignature Conv2dGradOpArgumentMapping(const ArgumentMappingContext& ctx) {
 
 KernelSignature Conv2dDoubleGradOpArgumentMapping(
     const ArgumentMappingContext& ctx) {
-  return KernelSignature("conv2d_grad_grad",
+  return KernelSignature("conv2d_double_grad",
                          {"Input", "Filter", "DOutput", "DDInput", "DDFilter"},
                          {"strides",
                           "paddings",