Merge branch 'bugfix/fix_mrm_crash_and_rtc_default_code' into 'master'

jason-mao · jason-mao · commit d48369fe4940 · 2025-09-19T10:23:57.000+08:00
bugfix(examples): Fix mrm crash and rtc flash tone offset and sync framesize enum

See merge request adf/esp-adf-internal!1452
diff --git a/examples/advanced_examples/multi-room/main/multi_room.c b/examples/advanced_examples/multi-room/main/multi_room.c
@@ -92,7 +92,10 @@ static void _multi_room_play_task(void *para)
 static esp_err_t multi_room_play_start(const char *url)
 {
     audio_element_set_uri(http_stream_reader, url);
-    audio_element_process_init(http_stream_reader);
+    if (ESP_OK != audio_element_process_init(http_stream_reader)) {
+        ESP_LOGE(TAG, "Audio element process init failed, can not start multi_room_play service");
+        return ESP_FAIL;
+    }
     audio_element_run(http_stream_reader);
 
     play_task_run = true;
@@ -103,7 +106,7 @@ static esp_err_t multi_room_play_start(const char *url)
                             DEFAULT_MRM_TASK_PRIO,
                             true,
                             0) != ESP_OK) {
-        ESP_LOGE(TAG, "Can not start multi_room_play service");
+        ESP_LOGE(TAG, "Play task create failed, can not start multi_room_play service");
         return ESP_FAIL;
     }
 
diff --git a/examples/protocols/components/av_stream/av_stream.c b/examples/protocols/components/av_stream/av_stream.c
@@ -713,8 +713,10 @@ int av_audio_get_vol(av_stream_handle_t av_stream, int *vol)
 static void init_jpeg_encoder(av_stream_handle_t av_stream)
 {
     jpeg_enc_config_t config = DEFAULT_JPEG_ENC_CONFIG();
-    config.width = av_resolution[av_stream->config.hal.video_framesize].width;
-    config.height = av_resolution[av_stream->config.hal.video_framesize].height;
+    uint16_t width = 0, height = 0;
+    av_stream_get_video_framesize(av_stream->config.hal.video_framesize, &width, &height);
+    config.width = width;
+    config.height = height;
     config.src_type = JPEG_PIXEL_FORMAT_YCbYCr;
     config.subsampling = JPEG_SUBSAMPLE_420;
     // config.rotate = JPEG_ROTATE_0D;
@@ -726,9 +728,11 @@ static int init_h264_encoder(av_stream_handle_t av_stream)
 {
 #if CONFIG_IDF_TARGET_ESP32S3
     esp_h264_enc_cfg_t cfg = DEFAULT_H264_ENCODER_CONFIG();
+    uint16_t width = 0, height = 0;
+    av_stream_get_video_framesize(av_stream->config.hal.video_framesize, &width, &height);
     cfg.pic_type = ESP_H264_RAW_FMT_YUV422;
-    cfg.width = av_resolution[av_stream->config.hal.video_framesize].width;
-    cfg.height = av_resolution[av_stream->config.hal.video_framesize].height;
+    cfg.width = width;
+    cfg.height = height;
     cfg.fps = VIDEO_FPS;
     cfg.gop_size = VIDEO_FPS * 3;
     cfg.target_bitrate = 400000;
diff --git a/examples/protocols/components/av_stream/av_stream_hal/av_stream_hal.h b/examples/protocols/components/av_stream/av_stream_hal/av_stream_hal.h
@@ -29,6 +29,7 @@
 #include "i2s_stream.h"
 #include "board.h"
 #include "esp_lcd_panel_ops.h"
+#include "esp_camera.h"
 
 #ifdef __cplusplus
 extern "C" {
@@ -51,41 +52,6 @@ extern "C" {
 
 #define USB_CAMERA_FRAME_INTERVAL(fps) ((10000000) / fps)
 
-typedef enum {
-    AV_FRAMESIZE_96X96,    // 96x96
-    AV_FRAMESIZE_QQVGA,    // 160x120
-    AV_FRAMESIZE_QCIF,     // 176x144
-    AV_FRAMESIZE_HQVGA,    // 240x176
-    AV_FRAMESIZE_240X240,  // 240x240
-    AV_FRAMESIZE_QVGA,     // 320x240
-    AV_FRAMESIZE_CIF,      // 400x296
-    AV_FRAMESIZE_HVGA,     // 480x320
-    AV_FRAMESIZE_VGA,      // 640x480
-    AV_FRAMESIZE_SVGA,     // 800x600
-    AV_FRAMESIZE_XGA,      // 1024x768
-    AV_FRAMESIZE_HD,       // 1280x720
-    AV_FRAMESIZE_SXGA,     // 1280x1024
-    AV_FRAMESIZE_UXGA,     // 1600x1200
-    // 3MP Sensors
-    AV_FRAMESIZE_FHD,      // 1920x1080
-    AV_FRAMESIZE_P_HD,     //  720x1280
-    AV_FRAMESIZE_P_3MP,    //  864x1536
-    AV_FRAMESIZE_QXGA,     // 2048x1536
-    // 5MP Sensors
-    AV_FRAMESIZE_QHD,      // 2560x1440
-    AV_FRAMESIZE_WQXGA,    // 2560x1600
-    AV_FRAMESIZE_P_FHD,    // 1080x1920
-    AV_FRAMESIZE_QSXGA,    // 2560x1920
-    AV_FRAMESIZE_INVALID
-} av_framesize_t;
-
-typedef struct {
-    const uint16_t width;
-    const uint16_t height;
-} av_resolution_info_t;
-
-extern const av_resolution_info_t av_resolution[];
-
 /**
  * @brief AV stream hal configurations
  */
@@ -97,7 +63,7 @@ typedef struct {
     bool                        video_soft_enc;     /*!< Use software JPEG / H264 encoder */
     uint32_t                    audio_samplerate;   /*!< Audio sample rate */
     uint32_t                    audio_framesize;    /*!< Audio frame size */
-    av_framesize_t              video_framesize;    /*!< Video frame size */
+    framesize_t                 video_framesize;    /*!< Video frame size */
 } av_stream_hal_config_t;
 
 /**
@@ -150,6 +116,15 @@ int av_stream_audio_read(char *buf, int len, TickType_t wait_time, bool uac_en);
  */
 int av_stream_audio_write(char *buf, int len, TickType_t wait_time, bool uac_en);
 
+/**
+ * @brief         Get actual width and height from enum value
+ * 
+ * @param         framesize: Video framesize setting
+ * @param[out]    width: Video width
+ * @param[out]    height: Video height
+ */
+void av_stream_get_video_framesize(framesize_t framesize, uint16_t *width, uint16_t *height);
+
 /**
  * @brief      Intialize camera hal
  *
diff --git a/examples/protocols/components/av_stream/av_stream_hal/av_stream_hal_camera.c b/examples/protocols/components/av_stream/av_stream_hal/av_stream_hal_camera.c
@@ -28,32 +28,112 @@
 
 static const char *TAG = "AV_STREAM_HAL";
 
-const av_resolution_info_t av_resolution[AV_FRAMESIZE_INVALID] = {
-    {   96,   96 }, /* 96x96 */
-    {  160,  120 }, /* QQVGA */
-    {  176,  144 }, /* QCIF  */
-    {  240,  176 }, /* HQVGA */
-    {  240,  240 }, /* 240x240 */
-    {  320,  240 }, /* QVGA  */
-    {  400,  296 }, /* CIF   */
-    {  480,  320 }, /* HVGA  */
-    {  640,  480 }, /* VGA   */
-    {  800,  600 }, /* SVGA  */
-    { 1024,  768 }, /* XGA   */
-    { 1280,  720 }, /* HD    */
-    { 1280, 1024 }, /* SXGA  */
-    { 1600, 1200 }, /* UXGA  */
-    // 3MP Sensors
-    { 1920, 1080 }, /* FHD   */
-    {  720, 1280 }, /* Portrait HD   */
-    {  864, 1536 }, /* Portrait 3MP   */
-    { 2048, 1536 }, /* QXGA  */
-    // 5MP Sensors
-    { 2560, 1440 }, /* QHD    */
-    { 2560, 1600 }, /* WQXGA  */
-    { 1088, 1920 }, /* Portrait FHD   */
-    { 2560, 1920 }, /* QSXGA  */
-};
+void av_stream_get_video_framesize(framesize_t framesize, uint16_t *width, uint16_t *height)
+{
+    switch (framesize) {
+        case FRAMESIZE_96X96:
+            *width = 96;
+            *height = 96;
+            break;
+        case FRAMESIZE_QQVGA:
+            *width = 160;
+            *height = 120;
+            break;
+        case FRAMESIZE_128X128:
+            *width = 128;
+            *height = 128;
+            break;
+        case FRAMESIZE_QCIF:
+            *width = 176;
+            *height = 144;
+            break;
+        case FRAMESIZE_HQVGA:
+            *width = 240;
+            *height = 176;
+            break;
+        case FRAMESIZE_240X240:
+            *width = 240;
+            *height = 240;
+            break;
+        case FRAMESIZE_QVGA:
+        default:
+            *width = 320;
+            *height = 240;
+            break;
+        case FRAMESIZE_320X320:
+            *width = 320;
+            *height = 320;
+            break;
+        case FRAMESIZE_CIF:
+            *width = 400;
+            *height = 296;
+            break;
+        case FRAMESIZE_HVGA:
+            *width = 480;
+            *height = 320;
+            break;
+        case FRAMESIZE_VGA:
+            *width = 640;
+            *height = 480;
+            break;
+        case FRAMESIZE_SVGA:    
+            *width = 800;
+            *height = 600;
+            break;
+        case FRAMESIZE_XGA:
+            *width = 1024;
+            *height = 768;
+            break;
+        case FRAMESIZE_HD:
+            *width = 1280;
+            *height = 720;
+            break;
+        case FRAMESIZE_SXGA:
+            *width = 1280;
+            *height = 1024;
+            break;
+        case FRAMESIZE_UXGA:
+            *width = 1600;
+            *height = 1200;
+            break;
+        case FRAMESIZE_FHD:
+            *width = 1920;
+            *height = 1080;
+            break;
+        case FRAMESIZE_P_HD:
+            *width = 720;
+            *height = 1280;
+            break;
+        case FRAMESIZE_P_3MP:
+            *width = 864;
+            *height = 1536;
+            break;
+        case FRAMESIZE_QXGA:
+            *width = 2048;
+            *height = 1536;
+            break;
+        case FRAMESIZE_QHD:
+            *width = 2560;
+            *height = 1440;
+            break;
+        case FRAMESIZE_WQXGA:
+            *width = 2560;
+            *height = 1600;
+            break;
+        case FRAMESIZE_P_FHD:
+            *width = 1080;
+            *height = 1920;
+            break;
+        case FRAMESIZE_QSXGA:
+            *width = 2560;
+            *height = 1920;
+            break;
+        case FRAMESIZE_5MP:
+            *width = 2592;
+            *height = 1944;
+            break;
+    }
+}
 
 #if CONFIG_IDF_TARGET_ESP32
 int av_stream_camera_init(av_stream_hal_config_t *config, void *cb, void *arg)
@@ -92,9 +172,12 @@ static esp_err_t usb_camera_init(av_stream_hal_config_t *config, void *camera_cb
 
     /* the quick demo skip the standred get descriptors process,
     users need to get params from camera descriptors from demo print */
+    uint16_t width = 0, height = 0;
+    av_stream_get_video_framesize(config->video_framesize, &width, &height);
+    ESP_LOGI(TAG, "camera frame size: %dx%d", width, height);
     uvc_config_t uvc_config = {
-        .frame_width = av_resolution[config->video_framesize].width,
-        .frame_height = av_resolution[config->video_framesize].height,
+        .frame_width = width,
+        .frame_height = height,
         .frame_interval = USB_CAMERA_FRAME_INTERVAL(VIDEO_FPS),
         .xfer_buffer_size = VIDEO_MAX_SIZE,
         .xfer_buffer_a = xfer_buffer_a,
diff --git a/examples/protocols/esp-rtc/README_CN.md b/examples/protocols/esp-rtc/README_CN.md
@@ -71,10 +71,10 @@ idf.py -p PORT flash monitor
 
 ### 下载 flash 提示音
 
-- 此应用额外需要下载一个提示音的 bin 到 flash 中：
+- 此应用额外需要下载一个提示音的 bin 到 flash 中，注意 partition size 0x230000 需要与 partitions.csv 中的信息匹配：
 
 ```
-python $ADF_PATH/esp-idf/components/esptool_py/esptool/esptool.py --chip esp32s3 --port /dev/ttyUSB0 --baud 921600 --before default_reset --after hard_reset write_flash -z --flash_mode dio --flash_freq 40m --flash_size detect 0x210000 ../components/audio_flash_tone/bin/audio-esp.bin
+python $ADF_PATH/esp-idf/components/esptool_py/esptool/esptool.py --chip esp32s3 --port /dev/ttyUSB0 --baud 921600 --before default_reset --after hard_reset write_flash -z --flash_mode dio --flash_freq 40m --flash_size detect 0x230000 ../components/audio_flash_tone/bin/audio-esp.bin
 ```
 
 ## 如何使用例程
diff --git a/examples/protocols/esp-rtc/main/rtc_service.c b/examples/protocols/esp-rtc/main/rtc_service.c
@@ -143,10 +143,12 @@ esp_rtc_handle_t rtc_service_start(av_stream_handle_t av_stream, const char *uri
     AUDIO_NULL_CHECK(TAG, av_stream, return NULL);
     media_lib_add_default_adapter();
 
+    uint16_t width = 0, height = 0;
+    av_stream_get_video_framesize(VIDEO_FRAME_SIZE, &width, &height);
     esp_rtc_video_info_t vcodec_info = {
         .vcodec = RTC_VCODEC_MJPEG,
-        .width = av_resolution[VIDEO_FRAME_SIZE].width,
-        .height = av_resolution[VIDEO_FRAME_SIZE].height,
+        .width = width,
+        .height = height,
         .fps = VIDEO_FPS,
         .len = VIDEO_MAX_SIZE,
     };
diff --git a/examples/protocols/esp-rtc/main/rtc_service.h b/examples/protocols/esp-rtc/main/rtc_service.h
@@ -30,7 +30,7 @@
 #include "av_stream.h"
 #include "esp_rtc.h"
 
-#define VIDEO_FRAME_SIZE AV_FRAMESIZE_QVGA
+#define VIDEO_FRAME_SIZE FRAMESIZE_QVGA
 
 #ifdef __cplusplus
 extern "C" {
diff --git a/examples/protocols/esp-rtsp/main/rtsp_service.c b/examples/protocols/esp-rtsp/main/rtsp_service.c
@@ -110,10 +110,12 @@ esp_rtsp_handle_t rtsp_service_start(av_stream_handle_t av_stream, esp_rtsp_mode
     AUDIO_NULL_CHECK(TAG, av_stream, return NULL);
     media_lib_add_default_adapter();
 
+    uint16_t width = 0, height = 0;
+    av_stream_get_video_framesize(RTSP_FRAME_SIZE, &width, &height);
     esp_rtsp_video_info_t vcodec_info = {
         .vcodec = RTSP_VCODEC_MJPEG,
-        .width = av_resolution[RTSP_FRAME_SIZE].width,
-        .height = av_resolution[RTSP_FRAME_SIZE].height,
+        .width = width,
+        .height = height,
         .fps = VIDEO_FPS,
         .len = VIDEO_MAX_SIZE,
     };
diff --git a/examples/protocols/esp-rtsp/main/rtsp_service.h b/examples/protocols/esp-rtsp/main/rtsp_service.h
@@ -31,7 +31,7 @@
 #define RTSP_SERVER_PORT            8554
 #define RTSP_STACK_SZIE             4*1024
 #define RTSP_TASK_PRIO              5
-#define RTSP_FRAME_SIZE             AV_FRAMESIZE_HVGA
+#define RTSP_FRAME_SIZE             FRAMESIZE_HVGA
 
 #ifdef __cplusplus
 extern "C" {
diff --git a/examples/protocols/voip/README.md b/examples/protocols/voip/README.md
@@ -91,7 +91,7 @@ See the Getting Started Guide for full steps to configure and use [ESP-IDF Progr
 
 ### Download flash tone
 
-- To use this example, you need to download a tone onto flash:
+- To use this example, you need to download a tone onto flash, note that partition size 0x210000 needs to match the information in partitions.csv:
 
 ```
   python $ADF_PATH/esp-idf/components/esptool_py/esptool/esptool.py --chip esp32 --port /dev/ttyUSB0 --baud 921600 --before default_reset --after hard_reset write_flash -z --flash_mode dio --flash_freq 40m --flash_size detect 0x210000 ../components/audio_flash_tone/bin/audio-esp.bin
diff --git a/examples/protocols/voip/README_CN.md b/examples/protocols/voip/README_CN.md
@@ -93,7 +93,7 @@ idf.py -p PORT flash monitor
 
 ### 下载 flash 提示音
 
-- 此应用额外需要下载一个提示音的 bin 到 flash 中：
+- 此应用额外需要下载一个提示音的 bin 到 flash 中，注意partition size 0x210000 需要与 partitions.csv 中的信息匹配：
 
 ```
   python $ADF_PATH/esp-idf/components/esptool_py/esptool/esptool.py --chip esp32 --port /dev/ttyUSB0 --baud 921600 --before default_reset --after hard_reset write_flash -z --flash_mode dio --flash_freq 40m --flash_size detect 0x210000 ../components/audio_flash_tone/bin/audio-esp.bin
