Root cause (verified against our exact IDF tree, not the community guess): the "258" in "sdio_write_task: Failed to send data: 258" is NOT a timeout (that is 263). 258 = 0x102 = ESP_ERR_INVALID_ARG. On the ESP32-P4, block- mode CMD53 writes require the SOURCE buffer to be 64-byte (cache-line) aligned; the IDF sdmmc driver rejects a misaligned source with INVALID_ARG BEFORE any bus activity. esp_hosts write loop then declares "Unrecoverable host sdio state" and reboots the whole P4. The audio TX payload is not 64-aligned, so streaming mic audio wedged on the very FIRST frame (which is exactly what we saw: listening -> instant Failed to send -> reboot). This also explains why buffer/queue/clock/retry tuning all did nothing: the write never reached the bus. And why our symptom was instant, not after ~100 writes (the community block-mode-desync theory) — it is the first misaligned buffer, every time. Fix: vendored esp_hosted 2.12.11 as an editable local component (overrides the registry copy) and bounce a misaligned TX payload through one aligned DMA scratch buffer in hosted_sdio_write_block (port_esp_hosted_host_sdio.c). TX is serialized by the bus lock so a single static bounce buffer is safe; freed in hosted_sdio_deinit. Host-only change — no C6 reflash. VERIFIED ON HARDWARE (autonomous self-test): 40s of continuous mic-audio upstream streaming — the traffic that previously wedged on the first frame — ran clean, zero timeouts, zero reboots. A guarded SDIO_TX_SELFTEST harness is kept (compiled out) for future SDIO stress testing. Credit: root cause + patch designed via multi-agent investigation; the precise 258=INVALID_ARG decode (correcting the upstream community timeout assumption) came from checking our actual esp_err.h. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
140 lines
4.2 KiB
C
140 lines
4.2 KiB
C
/* WebSocket link to the DeskLock gateway.
|
|
* Protocol: docs/architecture.md "WebSocket protocol" — keep in sync.
|
|
*/
|
|
|
|
#include <string.h>
|
|
|
|
#include "cJSON.h"
|
|
#include "esp_app_desc.h"
|
|
#include "esp_log.h"
|
|
#include "esp_websocket_client.h"
|
|
|
|
#include "desklock.h"
|
|
#include "secrets.h"
|
|
|
|
static const char *TAG = "gw";
|
|
|
|
static esp_websocket_client_handle_t s_ws;
|
|
static volatile bool s_connected;
|
|
|
|
static void handle_text(const char *data, int len)
|
|
{
|
|
cJSON *root = cJSON_ParseWithLength(data, len);
|
|
if (root == NULL) {
|
|
return;
|
|
}
|
|
const cJSON *type = cJSON_GetObjectItem(root, "type");
|
|
if (!cJSON_IsString(type)) {
|
|
cJSON_Delete(root);
|
|
return;
|
|
}
|
|
face_activity();
|
|
|
|
if (strcmp(type->valuestring, "state") == 0) {
|
|
const cJSON *value = cJSON_GetObjectItem(root, "value");
|
|
if (cJSON_IsString(value)) {
|
|
if (strcmp(value->valuestring, "thinking") == 0) {
|
|
face_set(FACE_EFFORT);
|
|
} else if (strcmp(value->valuestring, "idle") == 0 && !audio_is_playing()) {
|
|
face_set(FACE_IDLE);
|
|
}
|
|
}
|
|
} else if (strcmp(type->valuestring, "transcript") == 0) {
|
|
const cJSON *text = cJSON_GetObjectItem(root, "text");
|
|
ESP_LOGI(TAG, "transcript: %s", cJSON_IsString(text) ? text->valuestring : "?");
|
|
} else if (strcmp(type->valuestring, "reply_text") == 0) {
|
|
const cJSON *text = cJSON_GetObjectItem(root, "text");
|
|
ESP_LOGI(TAG, "tatlock: %s", cJSON_IsString(text) ? text->valuestring : "?");
|
|
} else if (strcmp(type->valuestring, "audio_start") == 0) {
|
|
audio_playback_begin();
|
|
face_set(FACE_SPEAKING);
|
|
} else if (strcmp(type->valuestring, "audio_end") == 0) {
|
|
audio_playback_end();
|
|
} else if (strcmp(type->valuestring, "error") == 0) {
|
|
face_set(FACE_RAGE);
|
|
}
|
|
cJSON_Delete(root);
|
|
}
|
|
|
|
static void ws_event(void *arg, esp_event_base_t base, int32_t id, void *data)
|
|
{
|
|
(void)arg;
|
|
(void)base;
|
|
esp_websocket_event_data_t *ev = (esp_websocket_event_data_t *)data;
|
|
switch (id) {
|
|
case WEBSOCKET_EVENT_CONNECTED: {
|
|
ESP_LOGI(TAG, "gateway connected");
|
|
s_connected = true;
|
|
char hello[128];
|
|
snprintf(hello, sizeof(hello),
|
|
"{\"type\":\"hello\",\"device\":\"desklock-p4\",\"fw\":\"%s\"}",
|
|
esp_app_get_description()->version);
|
|
gw_send_text(hello);
|
|
face_set(FACE_IDLE);
|
|
#if SDIO_TX_SELFTEST
|
|
extern void sdio_tx_selftest_kick(void);
|
|
sdio_tx_selftest_kick();
|
|
#endif
|
|
break;
|
|
}
|
|
case WEBSOCKET_EVENT_DISCONNECTED:
|
|
case WEBSOCKET_EVENT_ERROR:
|
|
if (s_connected) {
|
|
ESP_LOGW(TAG, "gateway lost");
|
|
}
|
|
s_connected = false;
|
|
audio_capture_stop();
|
|
app_on_disconnect();
|
|
/* Stay calm: the rain draining away signals "unreachable". A sustained
|
|
* outage escalates to the x_x error face in face.c (clock_cb). */
|
|
face_set(FACE_IDLE);
|
|
break;
|
|
case WEBSOCKET_EVENT_DATA:
|
|
if (ev->op_code == 0x01) { /* text */
|
|
handle_text(ev->data_ptr, ev->data_len);
|
|
} else if (ev->op_code == 0x02 && ev->data_len > 0) { /* binary PCM */
|
|
audio_playback_feed((const uint8_t *)ev->data_ptr, ev->data_len);
|
|
}
|
|
break;
|
|
default:
|
|
break;
|
|
}
|
|
}
|
|
|
|
void gw_start(void)
|
|
{
|
|
if (s_ws != NULL) {
|
|
return;
|
|
}
|
|
esp_websocket_client_config_t cfg = {
|
|
.uri = GATEWAY_WS_URI,
|
|
.buffer_size = 8192,
|
|
.reconnect_timeout_ms = 5000,
|
|
.network_timeout_ms = 10000,
|
|
.ping_interval_sec = 8, /* keep the link visibly alive through long Tatlock turns */
|
|
};
|
|
s_ws = esp_websocket_client_init(&cfg);
|
|
esp_websocket_register_events(s_ws, WEBSOCKET_EVENT_ANY, ws_event, NULL);
|
|
esp_websocket_client_start(s_ws);
|
|
ESP_LOGI(TAG, "connecting to %s", GATEWAY_WS_URI);
|
|
}
|
|
|
|
bool gw_connected(void)
|
|
{
|
|
return s_connected;
|
|
}
|
|
|
|
void gw_send_text(const char *json)
|
|
{
|
|
if (s_connected) {
|
|
esp_websocket_client_send_text(s_ws, json, strlen(json), pdMS_TO_TICKS(1500));
|
|
}
|
|
}
|
|
|
|
void gw_send_bin(const uint8_t *data, size_t len)
|
|
{
|
|
if (s_connected) {
|
|
esp_websocket_client_send_bin(s_ws, (const char *)data, len, pdMS_TO_TICKS(1500));
|
|
}
|
|
}
|