[FFmpeg-devel] [PATCH 1/2] dnn_backend_native_layer_conv2d.c: fix memory allocation bug in multithread function.

[xujunzz at sjtu.edu.cn]

From: Xu Jun <xujunzz at sjtu.edu.cn>

Before patch, memory was allocated in each thread functions,
which may cause more than one time of memory allocation and
cause crash.

After patch, memory is allocated in the main thread once,
an index was parsed into thread functions. Bug fixed.

Signed-off-by: Xu Jun <xujunzz at sjtu.edu.cn>
---
 .../dnn/dnn_backend_native_layer_conv2d.c     | 55 +++++++++----------
 1 file changed, 27 insertions(+), 28 deletions(-)

diff --git a/libavfilter/dnn/dnn_backend_native_layer_conv2d.c b/libavfilter/dnn/dnn_backend_native_layer_conv2d.c
index c52725aa2b..5ed1851512 100644
--- a/libavfilter/dnn/dnn_backend_native_layer_conv2d.c
+++ b/libavfilter/dnn/dnn_backend_native_layer_conv2d.c
@@ -32,6 +32,7 @@ typedef struct thread_common_param{
     int32_t output_operand_index;
     const void *parameters;
     NativeContext *ctx;
+    float *output_data;
     int thread_num;
 } thread_common_param;
 
@@ -111,7 +112,6 @@ static void * dnn_execute_layer_conv2d_thread(void *threadarg)
     thread_param *thread_param = (struct thread_param *)threadarg;
     thread_common_param *thread_common_param = thread_param->thread_common_param;
     DnnOperand *operands = thread_common_param->operands;
-    float *output;
     int32_t input_operand_index = thread_common_param->input_operand_indexes[0];
     int number = operands[input_operand_index].dims[0];
     int height = operands[input_operand_index].dims[1];
@@ -130,24 +130,7 @@ static void * dnn_execute_layer_conv2d_thread(void *threadarg)
     int thread_start = thread_stride * thread_param->thread_index + pad_size;
     int thread_end = (thread_param->thread_index == thread_common_param->thread_num - 1) ? (height - pad_size) : (thread_start + thread_stride);
 
-    DnnOperand *output_operand = &operands[thread_common_param->output_operand_index];
-    output_operand->dims[0] = number;
-    output_operand->dims[1] = height - pad_size * 2;
-    output_operand->dims[2] = width - pad_size * 2;
-    output_operand->dims[3] = conv_params->output_num;
-    output_operand->data_type = operands[input_operand_index].data_type;
-    output_operand->length = calculate_operand_data_length(output_operand);
-    if (output_operand->length <= 0) {
-        av_log(thread_common_param->ctx, AV_LOG_ERROR, "The output data length overflow\n");
-        return (void *)DNN_ERROR;
-    }
-    output_operand->data = av_realloc(output_operand->data, output_operand->length);
-    if (!output_operand->data) {
-        av_log(thread_common_param->ctx, AV_LOG_ERROR, "Failed to reallocate memory for output\n");
-        return (void *)DNN_ERROR;
-    }
-
-    output = output_operand->data;
+    float *output = thread_common_param->output_data;
     output += (conv_params->output_num) * (width - 2 * pad_size) * (thread_start - pad_size);
 
     av_assert0(channel == conv_params->input_num);
@@ -213,8 +196,6 @@ int dnn_execute_layer_conv2d(DnnOperand *operands, const int32_t *input_operand_
     pthread_t *thread_id = av_malloc(thread_num * sizeof(pthread_t));
 #endif
     thread_param **thread_param = av_malloc(thread_num * sizeof(*thread_param));
-    void *res;
-    int error_flag = DNN_SUCCESS;
 
     //struct used to pass parameters
     thread_common_param thread_common_param;
@@ -223,6 +204,28 @@ int dnn_execute_layer_conv2d(DnnOperand *operands, const int32_t *input_operand_
     thread_common_param.output_operand_index = output_operand_index;
     thread_common_param.parameters = parameters;
     thread_common_param.ctx = ctx;
+
+    //alloc memory
+    const ConvolutionalParams *conv_params = (const ConvolutionalParams *)(parameters);
+    int pad_size = (conv_params->padding_method == VALID) ? (conv_params->kernel_size - 1) / 2 * conv_params->dilation : 0;
+    DnnOperand *output_operand = &operands[output_operand_index];
+    output_operand->dims[0] = operands[input_operand_indexes[0]].dims[0];
+    output_operand->dims[1] = operands[input_operand_indexes[0]].dims[1] - pad_size * 2;
+    output_operand->dims[2] = operands[input_operand_indexes[0]].dims[2] - pad_size * 2;
+    output_operand->dims[3] = conv_params->output_num;
+    output_operand->data_type = operands[input_operand_indexes[0]].data_type;
+    output_operand->length = calculate_operand_data_length(output_operand);
+    if (output_operand->length <= 0) {
+        av_log(ctx, AV_LOG_ERROR, "The output data length overflow\n");
+        return DNN_ERROR;
+    }
+    output_operand->data = av_realloc(output_operand->data, output_operand->length);
+    if (!output_operand->data) {
+        av_log(ctx, AV_LOG_ERROR, "Failed to reallocate memory for output\n");
+        return DNN_ERROR;
+    }
+    thread_common_param.output_data = output_operand->data;
+
 #if HAVE_PTHREAD_CANCEL
     thread_common_param.thread_num = thread_num;
 
@@ -236,9 +239,7 @@ int dnn_execute_layer_conv2d(DnnOperand *operands, const int32_t *input_operand_
 
     //join threads, res gets function return
     for (int i = 0; i < thread_num; i++){
-        pthread_join(thread_id[i], &res);
-        if ((int)res != DNN_SUCCESS)
-            error_flag = (int)res;
+        pthread_join(thread_id[i], NULL);
     }
 
     //release memory
@@ -252,12 +253,10 @@ int dnn_execute_layer_conv2d(DnnOperand *operands, const int32_t *input_operand_
     thread_param[0] = av_malloc(sizeof(thread_param));
     thread_param[0]->thread_common_param = &thread_common_param;
     thread_param[0]->thread_index = 0;
-    res = dnn_execute_layer_conv2d_thread((void *)thread_param[0]);
-    if ((int)res != DNN_SUCCESS)
-        error_flag = (int)res;
+    dnn_execute_layer_conv2d_thread((void *)thread_param[0]);
     av_free(thread_param[0]);
 #endif
 
     av_free(thread_param);
-    return error_flag;
+    return DNN_SUCCESS;
 }
-- 
2.28.0