Merge branch 'feature/esp32c3_apptrace' into 'master'

apptrace: refactoring & esp32c3 support

See merge request espressif/esp-idf!11702

components/app_trace/CMakeLists.txt

@@ -6,7 +6,23 @@ set(srcs
 
 set(include_dirs "include")
 
-if(CONFIG_SYSVIEW_ENABLE)
+set(priv_include_dirs "private_include" "port/include")
+
+if(CONFIG_APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE)
+    list(APPEND srcs
+        "app_trace_membufs_proto.c")
+
+    if(CONFIG_IDF_TARGET_ARCH_XTENSA)
+        list(APPEND srcs
+            "port/xtensa/port.c")
+    endif()
+    if(CONFIG_IDF_TARGET_ARCH_RISCV)
+        list(APPEND srcs
+            "port/riscv/port.c")
+    endif()
+endif()
+
+if(CONFIG_APPTRACE_SV_ENABLE)
     list(APPEND include_dirs
         sys_view/Config
         sys_view/SEGGER
@@ -16,7 +32,7 @@ if(CONFIG_SYSVIEW_ENABLE)
         "sys_view/SEGGER/SEGGER_SYSVIEW.c"
         "sys_view/Sample/Config/SEGGER_SYSVIEW_Config_FreeRTOS.c"
         "sys_view/Sample/OS/SEGGER_SYSVIEW_FreeRTOS.c"
-        "sys_view/esp32/SEGGER_RTT_esp32.c"
+        "sys_view/esp/SEGGER_RTT_esp.c"
         "sys_view/ext/heap_trace_module.c"
         "sys_view/ext/logging.c")
 endif()
@@ -30,6 +46,7 @@ endif()
 
 idf_component_register(SRCS "${srcs}"
                        INCLUDE_DIRS "${include_dirs}"
+                       PRIV_INCLUDE_DIRS "${priv_include_dirs}"
                        PRIV_REQUIRES soc
                        LDFRAGMENTS linker.lf)
 

components/app_trace/Kconfig

@@ -4,29 +4,44 @@ menu "Application Level Tracing"
         prompt "Data Destination"
         default APPTRACE_DEST_NONE
         help
-            Select destination for application trace: trace memory or none (to disable).
+            Select destination for application trace: JTAG or none (to disable).
 
-        config APPTRACE_DEST_TRAX
-            bool "Trace memory"
+        config APPTRACE_DEST_JTAG
+            bool "JTAG"
+            select APPTRACE_DEST_TRAX if IDF_TARGET_ARCH_XTENSA
+            select APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE
             select APPTRACE_ENABLE
+
         config APPTRACE_DEST_NONE
             bool "None"
     endchoice
 
-    config APPTRACE_ENABLE
+    config APPTRACE_DEST_TRAX
         bool
-        depends on !ESP32_TRAX && !ESP32S2_TRAX
+        depends on IDF_TARGET_ARCH_XTENSA && !ESP32_TRAX && !ESP32S2_TRAX
         select ESP32_MEMMAP_TRACEMEM
         select ESP32S2_MEMMAP_TRACEMEM
         select ESP32_MEMMAP_TRACEMEM_TWOBANKS
         select ESP32S2_MEMMAP_TRACEMEM_TWOBANKS
         default n
+        help
+            Enables/disable TRAX tracing HW.
+
+    config APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE
+        bool
+        default n
+        help
+            Enables/disable swapping memory buffers tracing protocol.
+
+    config APPTRACE_ENABLE
+        bool
+        default n
         help
             Enables/disable application tracing module.
 
     config APPTRACE_LOCK_ENABLE
         bool
-        default !SYSVIEW_ENABLE
+        default !APPTRACE_SV_ENABLE
         help
             Enables/disable application tracing module internal sync lock.
 
@@ -41,16 +56,23 @@ menu "Application Level Tracing"
 
     config APPTRACE_POSTMORTEM_FLUSH_THRESH
         int "Threshold for flushing last trace data to host on panic"
-        depends on APPTRACE_DEST_TRAX
+        depends on APPTRACE_ENABLE
         range 0 16384
         default 0
         help
             Threshold for flushing last trace data to host on panic in post-mortem mode.
             This is minimal amount of data needed to perform flush. In bytes.
 
+    config APPTRACE_BUF_SIZE
+        int "Size of the apptrace buffer"
+        depends on APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE && !APPTRACE_DEST_TRAX
+        default 16384
+        help
+            Size of the memory buffer for trace data in bytes.
+
     config APPTRACE_PENDING_DATA_SIZE_MAX
         int "Size of the pending data buffer"
-        depends on APPTRACE_DEST_TRAX
+        depends on APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE
         default 0
         help
             Size of the buffer for events in bytes. It is useful for buffering events from
@@ -59,151 +81,151 @@ menu "Application Level Tracing"
 
     menu "FreeRTOS SystemView Tracing"
         depends on APPTRACE_ENABLE
-        config SYSVIEW_ENABLE
+        config APPTRACE_SV_ENABLE
             bool "SystemView Tracing Enable"
             depends on APPTRACE_ENABLE
             default n
             help
                 Enables supporrt for SEGGER SystemView tracing functionality.
 
-        choice SYSVIEW_TS_SOURCE
+        choice APPTRACE_SV_TS_SOURCE
             prompt "Timer to use as timestamp source"
-            depends on SYSVIEW_ENABLE
-            default SYSVIEW_TS_SOURCE_CCOUNT if FREERTOS_UNICORE && !PM_ENABLE
-            default SYSVIEW_TS_SOURCE_TIMER_00 if !FREERTOS_UNICORE && !PM_ENABLE
-            default SYSVIEW_TS_SOURCE_ESP_TIMER if PM_ENABLE
+            depends on APPTRACE_SV_ENABLE
+            default APPTRACE_SV_TS_SOURCE_CCOUNT if FREERTOS_UNICORE && !PM_ENABLE && !IDF_TARGET_ESP32C3
+            default APPTRACE_SV_TS_SOURCE_TIMER_00 if !FREERTOS_UNICORE && !PM_ENABLE && !IDF_TARGET_ESP32C3
+            default APPTRACE_SV_TS_SOURCE_ESP_TIMER if PM_ENABLE || IDF_TARGET_ESP32C3
             help
                 SystemView needs to use a hardware timer as the source of timestamps
                 when tracing. This option selects the timer for it.
 
-            config SYSVIEW_TS_SOURCE_CCOUNT
+            config APPTRACE_SV_TS_SOURCE_CCOUNT
                 bool "CPU cycle counter (CCOUNT)"
-                depends on FREERTOS_UNICORE && !PM_ENABLE
+                depends on FREERTOS_UNICORE && !PM_ENABLE && !IDF_TARGET_ESP32C3
 
-            config SYSVIEW_TS_SOURCE_TIMER_00
+            config APPTRACE_SV_TS_SOURCE_TIMER_00
                 bool "Timer 0, Group 0"
-                depends on !PM_ENABLE
+                depends on !PM_ENABLE && !IDF_TARGET_ESP32C3
 
-            config SYSVIEW_TS_SOURCE_TIMER_01
+            config APPTRACE_SV_TS_SOURCE_TIMER_01
                 bool "Timer 1, Group 0"
-                depends on !PM_ENABLE
+                depends on !PM_ENABLE && !IDF_TARGET_ESP32C3
 
-            config SYSVIEW_TS_SOURCE_TIMER_10
+            config APPTRACE_SV_TS_SOURCE_TIMER_10
                 bool "Timer 0, Group 1"
-                depends on !PM_ENABLE
+                depends on !PM_ENABLE && !IDF_TARGET_ESP32C3
 
-            config SYSVIEW_TS_SOURCE_TIMER_11
+            config APPTRACE_SV_TS_SOURCE_TIMER_11
                 bool "Timer 1, Group 1"
-                depends on !PM_ENABLE
+                depends on !PM_ENABLE && !IDF_TARGET_ESP32C3
 
-            config SYSVIEW_TS_SOURCE_ESP_TIMER
+            config APPTRACE_SV_TS_SOURCE_ESP_TIMER
                 bool "esp_timer high resolution timer"
 
         endchoice
 
-        config SYSVIEW_MAX_TASKS
+        config APPTRACE_SV_MAX_TASKS
             int "Maximum supported tasks"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             range 1 64
             default 16
             help
                 Configures maximum supported tasks in sysview debug
 
-        config SYSVIEW_BUF_WAIT_TMO
+        config APPTRACE_SV_BUF_WAIT_TMO
             int "Trace buffer wait timeout"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default 500
             help
                 Configures timeout (in us) to wait for free space in trace buffer.
                 Set to -1 to wait forever and avoid lost events.
 
-        config SYSVIEW_EVT_OVERFLOW_ENABLE
+        config APPTRACE_SV_EVT_OVERFLOW_ENABLE
             bool "Trace Buffer Overflow Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Trace Buffer Overflow" event.
 
-        config SYSVIEW_EVT_ISR_ENTER_ENABLE
+        config APPTRACE_SV_EVT_ISR_ENTER_ENABLE
             bool "ISR Enter Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "ISR Enter" event.
 
-        config SYSVIEW_EVT_ISR_EXIT_ENABLE
+        config APPTRACE_SV_EVT_ISR_EXIT_ENABLE
             bool "ISR Exit Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "ISR Exit" event.
 
-        config SYSVIEW_EVT_ISR_TO_SCHEDULER_ENABLE
+        config APPTRACE_SV_EVT_ISR_TO_SCHED_ENABLE
             bool "ISR Exit to Scheduler Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "ISR to Scheduler" event.
 
-        config SYSVIEW_EVT_TASK_START_EXEC_ENABLE
+        config APPTRACE_SV_EVT_TASK_START_EXEC_ENABLE
             bool "Task Start Execution Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Task Start Execution" event.
 
-        config SYSVIEW_EVT_TASK_STOP_EXEC_ENABLE
+        config APPTRACE_SV_EVT_TASK_STOP_EXEC_ENABLE
             bool "Task Stop Execution Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Task Stop Execution" event.
 
-        config SYSVIEW_EVT_TASK_START_READY_ENABLE
+        config APPTRACE_SV_EVT_TASK_START_READY_ENABLE
             bool "Task Start Ready State Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Task Start Ready State" event.
 
-        config SYSVIEW_EVT_TASK_STOP_READY_ENABLE
+        config APPTRACE_SV_EVT_TASK_STOP_READY_ENABLE
             bool "Task Stop Ready State Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Task Stop Ready State" event.
 
-        config SYSVIEW_EVT_TASK_CREATE_ENABLE
+        config APPTRACE_SV_EVT_TASK_CREATE_ENABLE
             bool "Task Create Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Task Create" event.
 
-        config SYSVIEW_EVT_TASK_TERMINATE_ENABLE
+        config APPTRACE_SV_EVT_TASK_TERMINATE_ENABLE
             bool "Task Terminate Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Task Terminate" event.
 
-        config SYSVIEW_EVT_IDLE_ENABLE
+        config APPTRACE_SV_EVT_IDLE_ENABLE
             bool "System Idle Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "System Idle" event.
 
-        config SYSVIEW_EVT_TIMER_ENTER_ENABLE
+        config APPTRACE_SV_EVT_TIMER_ENTER_ENABLE
             bool "Timer Enter Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Timer Enter" event.
 
-        config SYSVIEW_EVT_TIMER_EXIT_ENABLE
+        config APPTRACE_SV_EVT_TIMER_EXIT_ENABLE
             bool "Timer Exit Event"
-            depends on SYSVIEW_ENABLE
+            depends on APPTRACE_SV_ENABLE
             default y
             help
                 Enables "Timer Exit" event.
@@ -212,7 +234,7 @@ menu "Application Level Tracing"
 
     config APPTRACE_GCOV_ENABLE
         bool "GCOV to Host Enable"
-        depends on APPTRACE_ENABLE && !SYSVIEW_ENABLE
+        depends on APPTRACE_ENABLE && !APPTRACE_SV_ENABLE
         select ESP_DEBUG_STUBS_ENABLE
         default n
         help

components/app_trace/app_trace_membufs_proto.c

@@ -0,0 +1,364 @@
+#include <sys/param.h>
+#include <string.h>
+#include "sdkconfig.h"
+#include "esp_log.h"
+#include "esp_app_trace_membufs_proto.h"
+
+/** Trace data header. Every user data chunk is prepended with this header.
+ * User allocates block with esp_apptrace_buffer_get and then fills it with data,
+ * in multithreading environment it can happen that tasks gets buffer and then gets interrupted,
+ * so it is possible that user data are incomplete when  memory block is exposed to the host.
+ * In this case host SW will see that wr_sz < block_sz and will report error.
+ */
+typedef struct {
+#if CONFIG_APPTRACE_SV_ENABLE
+    uint8_t   block_sz; // size of allocated block for user data
+    uint8_t   wr_sz;    // size of actually written data
+#else
+    uint16_t   block_sz; // size of allocated block for user data
+    uint16_t   wr_sz;    // size of actually written data
+#endif
+} esp_tracedata_hdr_t;
+
+/** TODO: docs
+ */
+typedef struct {
+    uint16_t   block_sz; // size of allocated block for user data
+} esp_hostdata_hdr_t;
+
+#if CONFIG_APPTRACE_SV_ENABLE
+#define ESP_APPTRACE_USR_BLOCK_CORE(_cid_)          (0)
+#define ESP_APPTRACE_USR_BLOCK_LEN(_v_)             (_v_)
+#define ESP_APPTRACE_USR_DATA_LEN_MAX(_hw_data_)    255UL
+#else
+#define ESP_APPTRACE_USR_BLOCK_CORE(_cid_)      ((_cid_) << 15)
+#define ESP_APPTRACE_USR_BLOCK_LEN(_v_)         (~(1 << 15) & (_v_))
+#define ESP_APPTRACE_USR_DATA_LEN_MAX(_hw_data_)       (ESP_APPTRACE_INBLOCK(_hw_data_)->sz - sizeof(esp_tracedata_hdr_t))
+#endif
+#define ESP_APPTRACE_USR_BLOCK_RAW_SZ(_s_)     ((_s_) + sizeof(esp_tracedata_hdr_t))
+
+#define ESP_APPTRACE_INBLOCK_MARKER(_hw_data_)          ((_hw_data_)->state.markers[(_hw_data_)->state.in_block % 2])
+#define ESP_APPTRACE_INBLOCK_MARKER_UPD(_hw_data_, _v_)   do {(_hw_data_)->state.markers[(_hw_data_)->state.in_block % 2] += (_v_);}while(0)
+#define ESP_APPTRACE_INBLOCK(_hw_data_)             (&(_hw_data_)->blocks[(_hw_data_)->state.in_block % 2])
+
+const static char *TAG = "esp_apptrace";
+
+static uint32_t esp_apptrace_membufs_down_buffer_write_nolock(esp_apptrace_membufs_proto_data_t *proto, uint8_t *data, uint32_t size);
+
+
+esp_err_t esp_apptrace_membufs_init(esp_apptrace_membufs_proto_data_t *proto, const esp_apptrace_mem_block_t blocks_cfg[2])
+{
+    // disabled by default
+    esp_apptrace_rb_init(&proto->rb_down, NULL, 0);
+    // membufs proto init
+    for (unsigned i = 0; i < 2; i++) {
+        proto->blocks[i].start = blocks_cfg[i].start;
+        proto->blocks[i].sz = blocks_cfg[i].sz;
+        proto->state.markers[i] = 0;
+    }
+    proto->state.in_block = 0;
+#if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+    esp_apptrace_rb_init(&proto->rb_pend, proto->pending_data,
+                        sizeof(proto->pending_data));
+#endif
+    return ESP_OK;
+}
+
+void esp_apptrace_membufs_down_buffer_config(esp_apptrace_membufs_proto_data_t *data, uint8_t *buf, uint32_t size)
+{
+    esp_apptrace_rb_init(&data->rb_down, buf, size);
+}
+
+// assumed to be protected by caller from multi-core/thread access
+static esp_err_t esp_apptrace_membufs_swap(esp_apptrace_membufs_proto_data_t *proto)
+{
+    int prev_block_num = proto->state.in_block % 2;
+    int new_block_num = prev_block_num ? (0) : (1);
+    esp_err_t res = ESP_OK;
+
+    res = proto->hw->swap_start(proto->state.in_block);
+    if (res != ESP_OK) {
+        return res;
+    }
+
+    proto->state.markers[new_block_num] = 0;
+    // switch to new block
+    proto->state.in_block++;
+
+    proto->hw->swap(new_block_num);
+
+    // handle data from host
+    esp_hostdata_hdr_t *hdr = (esp_hostdata_hdr_t *)proto->blocks[new_block_num].start;
+    // ESP_APPTRACE_LOGV("Host data %d, sz %d @ %p", proto->hw->host_data_pending(), hdr->block_sz, hdr);
+    if (proto->hw->host_data_pending() && hdr->block_sz > 0) {
+        // TODO: add support for multiple blocks from host, currently there is no need for that
+        uint8_t *p = proto->blocks[new_block_num].start + proto->blocks[new_block_num].sz;
+        ESP_APPTRACE_LOGD("Recvd %d bytes from host [%x %x %x %x %x %x %x %x .. %x %x %x %x %x %x %x %x]", hdr->block_sz,
+            *(proto->blocks[new_block_num].start+0), *(proto->blocks[new_block_num].start+1),
+            *(proto->blocks[new_block_num].start+2), *(proto->blocks[new_block_num].start+3),
+            *(proto->blocks[new_block_num].start+4), *(proto->blocks[new_block_num].start+5),
+            *(proto->blocks[new_block_num].start+6), *(proto->blocks[new_block_num].start+7),
+            *(p-8), *(p-7), *(p-6), *(p-5), *(p-4), *(p-3), *(p-2), *(p-1));
+        uint32_t sz = esp_apptrace_membufs_down_buffer_write_nolock(proto, (uint8_t *)(hdr+1), hdr->block_sz);
+        if (sz != hdr->block_sz) {
+            ESP_APPTRACE_LOGE("Failed to write %d bytes to down buffer (%d %d)!", hdr->block_sz - sz, hdr->block_sz, sz);
+        }
+        hdr->block_sz = 0;
+    }
+#if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+    // copy pending data to  block if any
+    while (proto->state.markers[new_block_num] < proto->blocks[new_block_num].sz) {
+        uint32_t read_sz = esp_apptrace_rb_read_size_get(&proto->rb_pend);
+        if (read_sz == 0) {
+            break; // no more data in pending buffer
+        }
+        if (read_sz > proto->blocks[new_block_num].sz - proto->state.markers[new_block_num]) {
+            read_sz = proto->blocks[new_block_num].sz - proto->state.markers[new_block_num];
+        }
+        uint8_t *ptr = esp_apptrace_rb_consume(&proto->rb_pend, read_sz);
+        if (!ptr) {
+            assert(false && "Failed to consume pended bytes!!");
+            break;
+        }
+        ESP_APPTRACE_LOGD("Pump %d pend bytes [%x %x %x %x : %x %x %x %x : %x %x %x %x : %x %x...%x %x]",
+            read_sz, *(ptr+0), *(ptr+1), *(ptr+2), *(ptr+3), *(ptr+4),
+            *(ptr+5), *(ptr+6), *(ptr+7), *(ptr+8), *(ptr+9), *(ptr+10), *(ptr+11), *(ptr+12), *(ptr+13), *(ptr+read_sz-2), *(ptr+read_sz-1));
+        memcpy(proto->blocks[new_block_num].start + proto->state.markers[new_block_num], ptr, read_sz);
+        proto->state.markers[new_block_num] += read_sz;
+    }
+#endif
+    proto->hw->swap_end(proto->state.in_block, proto->state.markers[prev_block_num]);
+    return res;
+}
+
+static esp_err_t esp_apptrace_membufs_swap_waitus(esp_apptrace_membufs_proto_data_t *proto, esp_apptrace_tmo_t *tmo)
+{
+    int res;
+
+    while ((res = esp_apptrace_membufs_swap(proto)) != ESP_OK) {
+        res = esp_apptrace_tmo_check(tmo);
+        if (res != ESP_OK) {
+            break;
+        }
+    }
+    return res;
+}
+
+uint8_t *esp_apptrace_membufs_down_buffer_get(esp_apptrace_membufs_proto_data_t *proto, uint32_t *size, esp_apptrace_tmo_t *tmo)
+{
+    uint8_t *ptr = NULL;
+
+    while (1) {
+        uint32_t sz = esp_apptrace_rb_read_size_get(&proto->rb_down);
+        if (sz != 0) {
+            *size = MIN(*size, sz);
+            ptr = esp_apptrace_rb_consume(&proto->rb_down, *size);
+            if (!ptr) {
+                assert(false && "Failed to consume bytes from down buffer!");
+            }
+            break;
+        }
+        // may need to flush
+        if (proto->hw->host_data_pending()) {
+            ESP_APPTRACE_LOGD("force flush");
+            int res = esp_apptrace_membufs_swap_waitus(proto, tmo);
+            if (res != ESP_OK) {
+                ESP_APPTRACE_LOGE("Failed to switch to another block to recv data from host!");
+                /*do not return error because data can be in down buffer already*/
+            }
+        } else {
+            // check tmo only if there is no data from host
+            int res = esp_apptrace_tmo_check(tmo);
+            if (res != ESP_OK) {
+                return NULL;
+            }
+        }
+    }
+    return ptr;
+}
+
+esp_err_t esp_apptrace_membufs_down_buffer_put(esp_apptrace_membufs_proto_data_t *proto, uint8_t *ptr, esp_apptrace_tmo_t *tmo)
+{
+    /* nothing todo */
+    return ESP_OK;
+}
+
+static uint32_t esp_apptrace_membufs_down_buffer_write_nolock(esp_apptrace_membufs_proto_data_t *proto, uint8_t *data, uint32_t size)
+{
+    uint32_t total_sz = 0;
+
+    while (total_sz < size) {
+        ESP_APPTRACE_LOGD("esp_apptrace_trax_down_buffer_write_nolock WRS %d-%d-%d %d", proto->rb_down.wr, proto->rb_down.rd,
+            proto->rb_down.cur_size, size);
+        uint32_t wr_sz = esp_apptrace_rb_write_size_get(&proto->rb_down);
+        if (wr_sz == 0) {
+            break;
+        }
+
+        if (wr_sz > size - total_sz) {
+            wr_sz = size - total_sz;
+        }
+        ESP_APPTRACE_LOGD("esp_apptrace_trax_down_buffer_write_nolock wr %d", wr_sz);
+        uint8_t *ptr = esp_apptrace_rb_produce(&proto->rb_down, wr_sz);
+        if (!ptr) {
+            assert(false && "Failed to produce bytes to down buffer!");
+        }
+        ESP_APPTRACE_LOGD("esp_apptrace_trax_down_buffer_write_nolock wr %d to 0x%x from 0x%x", wr_sz, ptr, data + total_sz + wr_sz);
+        memcpy(ptr, data + total_sz, wr_sz);
+        total_sz += wr_sz;
+        ESP_APPTRACE_LOGD("esp_apptrace_trax_down_buffer_write_nolock wr %d/%d", wr_sz, total_sz);
+    }
+    return total_sz;
+}
+
+static inline uint8_t *esp_apptrace_membufs_wait4buf(esp_apptrace_membufs_proto_data_t *proto, uint16_t size, esp_apptrace_tmo_t *tmo, int *pended)
+{
+    uint8_t *ptr = NULL;
+
+    int res = esp_apptrace_membufs_swap_waitus(proto, tmo);
+    if (res != ESP_OK) {
+        return NULL;
+    }
+#if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+    // check if we still have pending data
+    if (esp_apptrace_rb_read_size_get(&proto->rb_pend) > 0) {
+        // if after block switch we still have pending data (not all pending data have been pumped to block)
+        // alloc new pending buffer
+        *pended = 1;
+        ptr = esp_apptrace_rb_produce(&proto->rb_pend, size);
+        if (!ptr) {
+            ESP_APPTRACE_LOGE("Failed to alloc pend buf 1: w-r-s %d-%d-%d!", proto->rb_pend.wr, proto->rb_pend.rd, proto->rb_pend.cur_size);
+        }
+    } else
+#endif
+    {
+        // update block pointers
+        if (ESP_APPTRACE_INBLOCK_MARKER(proto) + size > ESP_APPTRACE_INBLOCK(proto)->sz) {
+#if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+            *pended = 1;
+            ptr = esp_apptrace_rb_produce(&proto->rb_pend, size);
+            if (ptr == NULL) {
+                ESP_APPTRACE_LOGE("Failed to alloc pend buf 2: w-r-s %d-%d-%d!", proto->rb_pend.wr, proto->rb_pend.rd, proto->rb_pend.cur_size);
+            }
+#endif
+        } else {
+            *pended = 0;
+            ptr = ESP_APPTRACE_INBLOCK(proto)->start + ESP_APPTRACE_INBLOCK_MARKER(proto);
+        }
+    }
+
+    return ptr;
+}
+
+static inline uint8_t *esp_apptrace_membufs_pkt_start(uint8_t *ptr, uint16_t size)
+{
+    // it is safe to use cpu_hal_get_core_id() in macro call because arg is used only once inside it
+    ((esp_tracedata_hdr_t *)ptr)->block_sz = ESP_APPTRACE_USR_BLOCK_CORE(cpu_hal_get_core_id()) | size;
+    ((esp_tracedata_hdr_t *)ptr)->wr_sz = 0;
+    return ptr + sizeof(esp_tracedata_hdr_t);
+}
+
+static inline void esp_apptrace_membufs_pkt_end(uint8_t *ptr)
+{
+    esp_tracedata_hdr_t *hdr = (esp_tracedata_hdr_t *)(ptr - sizeof(esp_tracedata_hdr_t));
+    // update written size
+    hdr->wr_sz = hdr->block_sz;
+}
+
+uint8_t *esp_apptrace_membufs_up_buffer_get(esp_apptrace_membufs_proto_data_t *proto, uint32_t size, esp_apptrace_tmo_t *tmo)
+{
+    uint8_t *buf_ptr = NULL;
+
+    if (size > ESP_APPTRACE_USR_DATA_LEN_MAX(proto)) {
+        ESP_APPTRACE_LOGE("Too large user data size %d!", size);
+        return NULL;
+    }
+
+    // check for data in the pending buffer
+#if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+    if (esp_apptrace_rb_read_size_get(&proto->rb_pend) > 0) {
+        // if we have buffered data try to switch  block
+        esp_apptrace_membufs_swap(proto);
+        // if switch was successful, part or all pended data have been copied to  block
+    }
+    if (esp_apptrace_rb_read_size_get(&proto->rb_pend) > 0) {
+        // if we have buffered data alloc new pending buffer
+        ESP_APPTRACE_LOGD("Get %d bytes from PEND buffer", size);
+        buf_ptr = esp_apptrace_rb_produce(&proto->rb_pend, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size));
+        if (buf_ptr == NULL) {
+            int pended_buf;
+            buf_ptr = esp_apptrace_membufs_wait4buf(proto, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size), tmo, &pended_buf);
+            if (buf_ptr && !pended_buf) {
+                ESP_APPTRACE_LOGD("Get %d bytes from block", size);
+                // update cur block marker
+                ESP_APPTRACE_INBLOCK_MARKER_UPD(proto, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size));
+            }
+        }
+    } else {
+#else
+    if (1) {
+#endif
+        if (ESP_APPTRACE_INBLOCK_MARKER(proto) + ESP_APPTRACE_USR_BLOCK_RAW_SZ(size) > ESP_APPTRACE_INBLOCK(proto)->sz) {
+            #if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+            ESP_APPTRACE_LOGD("Block full. Get %d bytes from PEND buffer", size);
+            buf_ptr = esp_apptrace_rb_produce(&proto->rb_pend, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size));
+            #endif
+            if (buf_ptr == NULL) {
+                int pended_buf;
+                ESP_APPTRACE_LOGD(" full. Get %d bytes from pend buffer", size);
+                buf_ptr = esp_apptrace_membufs_wait4buf(proto, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size), tmo, &pended_buf);
+                if (buf_ptr && !pended_buf) {
+                    ESP_APPTRACE_LOGD("Got %d bytes from block", size);
+                    // update cur block marker
+                    ESP_APPTRACE_INBLOCK_MARKER_UPD(proto, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size));
+                }
+            }
+        } else {
+            ESP_APPTRACE_LOGD("Get %d bytes from  buffer", size);
+            // fit to curr  nlock
+            buf_ptr = ESP_APPTRACE_INBLOCK(proto)->start + ESP_APPTRACE_INBLOCK_MARKER(proto);
+            // update cur block marker
+            ESP_APPTRACE_INBLOCK_MARKER_UPD(proto, ESP_APPTRACE_USR_BLOCK_RAW_SZ(size));
+        }
+    }
+    if (buf_ptr) {
+        buf_ptr = esp_apptrace_membufs_pkt_start(buf_ptr, size);
+    }
+
+    return buf_ptr;
+}
+
+esp_err_t esp_apptrace_membufs_up_buffer_put(esp_apptrace_membufs_proto_data_t *proto, uint8_t *ptr, esp_apptrace_tmo_t *tmo)
+{
+    esp_apptrace_membufs_pkt_end(ptr);
+    // TODO: mark block as busy in order not to re-use it for other tracing calls until it is completely written
+    // TODO: avoid potential situation when all memory is consumed by low prio tasks which can not complete writing due to
+    // higher prio tasks and the latter can not allocate buffers at all
+    // this is abnormal situation can be detected on host which will receive only uncompleted buffers
+    // workaround: use own memcpy which will kick-off dead tracing calls
+    return ESP_OK;
+}
+
+esp_err_t esp_apptrace_membufs_flush_nolock(esp_apptrace_membufs_proto_data_t *proto, uint32_t min_sz, esp_apptrace_tmo_t *tmo)
+{
+    int res = ESP_OK;
+
+    if (ESP_APPTRACE_INBLOCK_MARKER(proto) < min_sz) {
+        ESP_APPTRACE_LOGI("Ignore flush request for min %d bytes. Bytes in  block: %d.", min_sz, ESP_APPTRACE_INBLOCK_MARKER(proto));
+        return ESP_OK;
+    }
+    // switch  block while size of data (including that in pending buffer) is more than min size
+    while (ESP_APPTRACE_INBLOCK_MARKER(proto) > min_sz) {
+        ESP_APPTRACE_LOGD("Try to flush %d bytes. Wait until block switch for %lld us", ESP_APPTRACE_INBLOCK_MARKER(proto), tmo->tmo);
+        res = esp_apptrace_membufs_swap_waitus(proto, tmo);
+        if (res != ESP_OK) {
+            if (tmo->tmo != ESP_APPTRACE_TMO_INFINITE)
+                ESP_APPTRACE_LOGW("Failed to switch to another block in %lld us!", tmo->tmo);
+            else
+                ESP_APPTRACE_LOGE("Failed to switch to another block in %lld us!", tmo->tmo);
+            return res;
+        }
+    }
+
+    return res;
+}

components/app_trace/app_trace_util.c

@@ -8,33 +8,42 @@
 #include "freertos/task.h"
 #include "esp_app_trace_util.h"
 #include "sdkconfig.h"
-#if CONFIG_IDF_TARGET_ESP32
-#include "esp32/clk.h"
-#elif CONFIG_IDF_TARGET_ESP32S2
-#include "esp32s2/clk.h"
-#elif CONFIG_IDF_TARGET_ESP32S3
-#include "esp32s3/clk.h"
-#elif CONFIG_IDF_TARGET_ESP32C3
-#include "esp32c3/clk.h"
+
+///////////////////////////////////////////////////////////////////////////////
+///////////////////////////////// Locks /////////////////////////////////////
+///////////////////////////////////////////////////////////////////////////////
+
+#if ESP_APPTRACE_PRINT_LOCK
+static esp_apptrace_lock_t s_log_lock = {.irq_stat = 0, .portmux = portMUX_INITIALIZER_UNLOCKED};
+#endif
+
+int esp_apptrace_log_lock(void)
+{
+#if ESP_APPTRACE_PRINT_LOCK
+    esp_apptrace_tmo_t tmo;
+    esp_apptrace_tmo_init(&tmo, ESP_APPTRACE_TMO_INFINITE);
+    int ret = esp_apptrace_lock_take(&s_log_lock, &tmo);
+    return ret;
+#else
+    return 0;
 #endif
+}
+
+void esp_apptrace_log_unlock(void)
+{
+ #if ESP_APPTRACE_PRINT_LOCK
+    esp_apptrace_lock_give(&s_log_lock);
+#endif
+}
 
 ///////////////////////////////////////////////////////////////////////////////
 ///////////////////////////////// TIMEOUT /////////////////////////////////////
 ///////////////////////////////////////////////////////////////////////////////
 
-#define ESP_APPTRACE_CPUTICKS2US(_t_, _cpu_freq_)       ((_t_)/(_cpu_freq_/1000000))
-#define ESP_APPTRACE_US2CPUTICKS(_t_, _cpu_freq_)       ((_t_)*(_cpu_freq_/1000000))
-
 esp_err_t esp_apptrace_tmo_check(esp_apptrace_tmo_t *tmo)
 {
-    int cpu_freq = esp_clk_cpu_freq();
-    if (tmo->tmo != ESP_APPTRACE_TMO_INFINITE) {
-        unsigned cur = portGET_RUN_TIME_COUNTER_VALUE();
-        if (tmo->start <= cur) {
-            tmo->elapsed = ESP_APPTRACE_CPUTICKS2US(cur - tmo->start, cpu_freq);
-        } else {
-            tmo->elapsed = ESP_APPTRACE_CPUTICKS2US(0xFFFFFFFF - tmo->start + cur, cpu_freq);
-        }
+    if (tmo->tmo != (int64_t)-1) {
+        tmo->elapsed = esp_timer_get_time() - tmo->start;
         if (tmo->elapsed >= tmo->tmo) {
             return ESP_ERR_TIMEOUT;
         }

components/app_trace/component.mk

@@ -4,14 +4,21 @@
 
 COMPONENT_SRCDIRS := .
 
+ifdef CONFIG_APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE
+COMPONENT_SRCDIRS += port/xtensa
+endif
+
 COMPONENT_ADD_INCLUDEDIRS = include
 
+COMPONENT_PRIV_INCLUDEDIRS = private_include \
+	port/include
+
 COMPONENT_ADD_LDFLAGS = -lapp_trace
 
 # do not produce gcov info for this module, it is used as transport for gcov
 CFLAGS := $(subst --coverage,,$(CFLAGS))
 
-ifdef CONFIG_SYSVIEW_ENABLE
+ifdef CONFIG_APPTRACE_SV_ENABLE
 
 COMPONENT_ADD_INCLUDEDIRS += \
 	sys_view/Config \
@@ -23,7 +30,7 @@ COMPONENT_SRCDIRS += \
 	sys_view/SEGGER \
 	sys_view/Sample/OS \
 	sys_view/Sample/Config \
-	sys_view/esp32 \
+	sys_view/esp \
 	sys_view/ext
 else
 COMPONENT_SRCDIRS += gcov

components/app_trace/gcov/gcov_rtio.c

@@ -129,7 +129,7 @@ void esp_gcov_dump(void)
     // disable IRQs on this CPU, other CPU is halted by OpenOCD
     unsigned irq_state = portENTER_CRITICAL_NESTED();
 #if !CONFIG_FREERTOS_UNICORE
-    int other_core = xPortGetCoreID() ? 0 : 1;
+    int other_core = cpu_hal_get_core_id() ? 0 : 1;
     esp_cpu_stall(other_core);
 #endif
     while (!esp_apptrace_host_is_connected(ESP_APPTRACE_DEST_TRAX)) {

components/app_trace/heap_trace_tohost.c

@@ -9,7 +9,7 @@
 #include "esp_heap_trace.h"
 #undef HEAP_TRACE_SRCFILE
 
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
 #include "esp_app_trace.h"
 #include "esp_sysview_trace.h"
 #endif
@@ -18,7 +18,7 @@
 
 #ifdef CONFIG_HEAP_TRACING_TOHOST
 
-#if !CONFIG_SYSVIEW_ENABLE
+#if !CONFIG_APPTRACE_SV_ENABLE
 #error None of the heap tracing backends is enabled! You must enable SystemView compatible tracing to use this feature.
 #endif
 
@@ -34,7 +34,7 @@ esp_err_t heap_trace_init_tohost(void)
 
 esp_err_t heap_trace_start(heap_trace_mode_t mode_param)
 {
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
     esp_err_t ret = esp_sysview_heap_trace_start((uint32_t)-1);
     if (ret != ESP_OK) {
         return ret;
@@ -47,7 +47,7 @@ esp_err_t heap_trace_start(heap_trace_mode_t mode_param)
 esp_err_t heap_trace_stop(void)
 {
     esp_err_t ret = ESP_ERR_NOT_SUPPORTED;
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
     ret = esp_sysview_heap_trace_stop();
 #endif
     s_tracing = false;
@@ -80,7 +80,7 @@ static IRAM_ATTR void record_allocation(const heap_trace_record_t *record)
     if (!s_tracing) {
         return;
     }
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
     esp_sysview_heap_trace_alloc(record->address, record->size, record->alloced_by);
 #endif
 }
@@ -95,7 +95,7 @@ static IRAM_ATTR void record_free(void *p, void **callers)
     if (!s_tracing) {
         return;
     }
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
     esp_sysview_heap_trace_free(p, callers);
 #endif
 }

components/app_trace/include/esp_app_trace.h

@@ -18,8 +18,11 @@ extern "C" {
  * Application trace data destinations bits.
  */
 typedef enum {
-    ESP_APPTRACE_DEST_TRAX = 0x1,	///< JTAG destination
-    ESP_APPTRACE_DEST_UART0 = 0x2,	///< UART destination
+    ESP_APPTRACE_DEST_JTAG = 1,                         ///< JTAG destination
+    ESP_APPTRACE_DEST_TRAX = ESP_APPTRACE_DEST_JTAG, 	///< xxx_TRAX name is obsolete, use more common xxx_JTAG
+    ESP_APPTRACE_DEST_UART0,	                        ///< UART0 destination
+    ESP_APPTRACE_DEST_MAX = ESP_APPTRACE_DEST_UART0,
+    ESP_APPTRACE_DEST_NUM
 } esp_apptrace_dest_t;
 
 /**

components/app_trace/include/esp_app_trace_util.h

@@ -12,6 +12,7 @@ extern "C" {
 
 #include "freertos/FreeRTOS.h"
 #include "esp_err.h"
+#include "esp_timer.h"
 
 /** Infinite waiting timeout */
 #define ESP_APPTRACE_TMO_INFINITE               ((uint32_t)-1)
@@ -22,9 +23,9 @@ extern "C" {
  *  periodically to check timeout for expiration.
  */
 typedef struct {
-    uint32_t   start;   ///< time interval start (in CPU ticks)
-    uint32_t   tmo;     ///< timeout value (in us)
-    uint32_t   elapsed; ///< elapsed time (in us)
+    int64_t   start;   ///< time interval start (in us)
+    int64_t   tmo;     ///< timeout value (in us)
+    int64_t   elapsed; ///< elapsed time (in us)
 } esp_apptrace_tmo_t;
 
 /**
@@ -35,23 +36,23 @@ typedef struct {
 */
 static inline void esp_apptrace_tmo_init(esp_apptrace_tmo_t *tmo, uint32_t user_tmo)
 {
-    tmo->start = portGET_RUN_TIME_COUNTER_VALUE();
-    tmo->tmo = user_tmo;
+    tmo->start = esp_timer_get_time();
+    tmo->tmo = user_tmo == ESP_APPTRACE_TMO_INFINITE ? (int64_t)-1 : (int64_t)user_tmo;
     tmo->elapsed = 0;
 }
 
 /**
  * @brief Checks timeout for expiration.
  *
- * @param tmo Pointer to timeout structure to be initialized.
+ * @param tmo Pointer to timeout structure.
  *
- * @return ESP_OK on success, otherwise \see esp_err_t
+ * @return number of remaining us till tmo.
  */
 esp_err_t esp_apptrace_tmo_check(esp_apptrace_tmo_t *tmo);
 
 static inline uint32_t esp_apptrace_tmo_remaining_us(esp_apptrace_tmo_t *tmo)
 {
-    return tmo->tmo != ESP_APPTRACE_TMO_INFINITE ? (tmo->elapsed - tmo->tmo) : ESP_APPTRACE_TMO_INFINITE;
+    return tmo->tmo != (int64_t)-1 ? (tmo->elapsed - tmo->tmo) : ESP_APPTRACE_TMO_INFINITE;
 }
 
 /** Tracing module synchronization lock */
@@ -160,6 +161,30 @@ uint32_t esp_apptrace_rb_read_size_get(esp_apptrace_rb_t *rb);
  */
 uint32_t esp_apptrace_rb_write_size_get(esp_apptrace_rb_t *rb);
 
+int esp_apptrace_log_lock(void);
+void esp_apptrace_log_unlock(void);
+
+#define ESP_APPTRACE_LOG( format, ... )   \
+    do { \
+        esp_apptrace_log_lock(); \
+        esp_rom_printf(format, ##__VA_ARGS__); \
+        esp_apptrace_log_unlock(); \
+    } while(0)
+
+#define ESP_APPTRACE_LOG_LEV( _L_, level, format, ... )   \
+    do { \
+        if (LOG_LOCAL_LEVEL >= level) { \
+            ESP_APPTRACE_LOG(LOG_FORMAT(_L_, format), esp_log_early_timestamp(), TAG, ##__VA_ARGS__); \
+        } \
+    } while(0)
+
+#define ESP_APPTRACE_LOGE( format, ... )  ESP_APPTRACE_LOG_LEV(E, ESP_LOG_ERROR, format, ##__VA_ARGS__)
+#define ESP_APPTRACE_LOGW( format, ... )  ESP_APPTRACE_LOG_LEV(W, ESP_LOG_WARN, format, ##__VA_ARGS__)
+#define ESP_APPTRACE_LOGI( format, ... )  ESP_APPTRACE_LOG_LEV(I, ESP_LOG_INFO, format, ##__VA_ARGS__)
+#define ESP_APPTRACE_LOGD( format, ... )  ESP_APPTRACE_LOG_LEV(D, ESP_LOG_DEBUG, format, ##__VA_ARGS__)
+#define ESP_APPTRACE_LOGV( format, ... )  ESP_APPTRACE_LOG_LEV(V, ESP_LOG_VERBOSE, format, ##__VA_ARGS__)
+#define ESP_APPTRACE_LOGO( format, ... )  ESP_APPTRACE_LOG_LEV(E, ESP_LOG_NONE, format, ##__VA_ARGS__)
+
 #ifdef __cplusplus
 }
 #endif

components/app_trace/include/esp_sysview_trace.h

@@ -12,7 +12,7 @@ extern "C" {
 
 #include <stdarg.h>
 #include "esp_err.h"
-#include "SEGGER_RTT.h" // SEGGER_RTT_ESP32_Flush
+#include "SEGGER_RTT.h" // SEGGER_RTT_ESP_Flush
 #include "esp_app_trace_util.h" // ESP_APPTRACE_TMO_INFINITE
 
 /**
@@ -24,7 +24,7 @@ extern "C" {
  */
 static inline esp_err_t esp_sysview_flush(uint32_t tmo)
 {
-    SEGGER_RTT_ESP32_Flush(0, tmo);
+    SEGGER_RTT_ESP_Flush(0, tmo);
     return ESP_OK;
 }
 

components/app_trace/linker.lf

@@ -3,17 +3,17 @@ archive: libapp_trace.a
 entries:
     app_trace (noflash)
     app_trace_util (noflash)
-    if SYSVIEW_ENABLE = y:
+    if APPTRACE_SV_ENABLE = y:
       SEGGER_SYSVIEW (noflash)
-      SEGGER_RTT_esp32 (noflash)
+      SEGGER_RTT_esp (noflash)
       SEGGER_SYSVIEW_Config_FreeRTOS (noflash)
       SEGGER_SYSVIEW_FreeRTOS (noflash)
 
 [mapping:app_trace_driver]
 archive: libdriver.a
 entries:
-    if SYSVIEW_TS_SOURCE_TIMER_00 = y || SYSVIEW_TS_SOURCE_TIMER_01 = y
-       || SYSVIEW_TS_SOURCE_TIMER_10 = y || SYSVIEW_TS_SOURCE_TIMER_11 = y:
+    if APPTRACE_SV_TS_SOURCE_TIMER_00 = y || APPTRACE_SV_TS_SOURCE_TIMER_01 = y
+       || APPTRACE_SV_TS_SOURCE_TIMER_10 = y || APPTRACE_SV_TS_SOURCE_TIMER_11 = y:
       timer (noflash)
     else:
       * (default)

components/app_trace/port/include/esp_app_trace_port.h

@@ -0,0 +1,43 @@
+// Copyright 2020 Espressif Systems (Shanghai) PTE LTD
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+#ifndef ESP_APP_TRACE_PORT_H_
+#define ESP_APP_TRACE_PORT_H_
+
+#include "esp_app_trace_util.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** Apptrace HW interface. */
+typedef struct {
+    esp_err_t (*init)(void *hw_data);
+    uint8_t *(*get_up_buffer)(void *hw_data, uint32_t, esp_apptrace_tmo_t *);
+    esp_err_t (*put_up_buffer)(void *hw_data, uint8_t *, esp_apptrace_tmo_t *);
+    esp_err_t (*flush_up_buffer_nolock)(void *hw_data, uint32_t, esp_apptrace_tmo_t *);
+    esp_err_t (*flush_up_buffer)(void *hw_data, esp_apptrace_tmo_t *);
+    void (*down_buffer_config)(void *hw_data, uint8_t *buf, uint32_t size);
+    uint8_t *(*get_down_buffer)(void *hw_data, uint32_t *, esp_apptrace_tmo_t *);
+    esp_err_t (*put_down_buffer)(void *hw_data, uint8_t *, esp_apptrace_tmo_t *);
+    bool (*host_is_connected)(void *hw_data);
+} esp_apptrace_hw_t;
+
+esp_apptrace_hw_t *esp_apptrace_jtag_hw_get(void **data);
+esp_apptrace_hw_t *esp_apptrace_uart_hw_get(int num, void **data);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif

components/app_trace/port/riscv/port.c

@@ -0,0 +1,374 @@
+#include "esp_log.h"
+#include "esp_app_trace_membufs_proto.h"
+#include "esp_app_trace_port.h"
+
+/** RISCV HW transport data */
+typedef struct {
+    uint8_t                             inited; // initialization state flags for every core
+#if CONFIG_APPTRACE_LOCK_ENABLE
+    esp_apptrace_lock_t                 lock;   // sync lock
+#endif
+    esp_apptrace_membufs_proto_data_t   membufs;
+} esp_apptrace_riscv_data_t;
+
+/** RISCV memory host iface control block */
+typedef struct {
+    uint32_t                    ctrl;
+    // - Guard field. If this register is not zero then CPU is changing this struct and
+    //   this guard field holds address of the instruction which application will execute when CPU finishes with those modifications.
+    uint32_t                    stat;
+    esp_apptrace_mem_block_t *  mem_blocks;
+} esp_apptrace_riscv_ctrl_block_t;
+
+#define RISCV_APPTRACE_SYSNR    0x64
+
+#define ESP_APPTRACE_RISCV_BLOCK_LEN_MSK         0x7FFFUL
+#define ESP_APPTRACE_RISCV_BLOCK_LEN(_l_)        ((_l_) & ESP_APPTRACE_RISCV_BLOCK_LEN_MSK)
+#define ESP_APPTRACE_RISCV_BLOCK_LEN_GET(_v_)    ((_v_) & ESP_APPTRACE_RISCV_BLOCK_LEN_MSK)
+#define ESP_APPTRACE_RISCV_BLOCK_ID_MSK          0x7FUL
+#define ESP_APPTRACE_RISCV_BLOCK_ID(_id_)        (((_id_) & ESP_APPTRACE_RISCV_BLOCK_ID_MSK) << 15)
+#define ESP_APPTRACE_RISCV_BLOCK_ID_GET(_v_)     (((_v_) >> 15) & ESP_APPTRACE_RISCV_BLOCK_ID_MSK)
+#define ESP_APPTRACE_RISCV_HOST_DATA             (1 << 22)
+#define ESP_APPTRACE_RISCV_HOST_CONNECT          (1 << 23)
+
+#define ESP_APPTRACE_RISCV_INITED(_hw_)          ((_hw_)->inited & (1 << 0/*cpu_hal_get_core_id()*/))
+
+static esp_err_t esp_apptrace_riscv_init(esp_apptrace_riscv_data_t *hw_data);
+static esp_err_t esp_apptrace_riscv_flush(esp_apptrace_riscv_data_t *hw_data, esp_apptrace_tmo_t *tmo);
+static esp_err_t esp_apptrace_riscv_flush_nolock(esp_apptrace_riscv_data_t *hw_data, uint32_t min_sz, esp_apptrace_tmo_t *tmo);
+static uint8_t *esp_apptrace_riscv_up_buffer_get(esp_apptrace_riscv_data_t *hw_data, uint32_t size, esp_apptrace_tmo_t *tmo);
+static esp_err_t esp_apptrace_riscv_up_buffer_put(esp_apptrace_riscv_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo);
+static void esp_apptrace_riscv_down_buffer_config(esp_apptrace_riscv_data_t *hw_data, uint8_t *buf, uint32_t size);
+static uint8_t *esp_apptrace_riscv_down_buffer_get(esp_apptrace_riscv_data_t *hw_data, uint32_t *size, esp_apptrace_tmo_t *tmo);
+static esp_err_t esp_apptrace_riscv_down_buffer_put(esp_apptrace_riscv_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo);
+static bool esp_apptrace_riscv_host_is_connected(esp_apptrace_riscv_data_t *hw_data);
+static esp_err_t esp_apptrace_riscv_buffer_swap_start(uint32_t curr_block_id);
+static esp_err_t esp_apptrace_riscv_buffer_swap(uint32_t new_block_id);
+static esp_err_t esp_apptrace_riscv_buffer_swap_end(uint32_t new_block_id, uint32_t prev_block_len);
+static bool esp_apptrace_riscv_host_data_pending(void);
+
+
+const static char *TAG = "esp_apptrace";
+
+static esp_apptrace_riscv_ctrl_block_t s_tracing_ctrl[portNUM_PROCESSORS];
+
+esp_apptrace_hw_t *esp_apptrace_uart_hw_get(int num, void **data)
+{
+    return NULL;
+}
+
+esp_apptrace_hw_t *esp_apptrace_jtag_hw_get(void **data)
+{
+#if CONFIG_APPTRACE_DEST_JTAG
+    static esp_apptrace_membufs_proto_hw_t s_trace_proto_hw = {
+        .swap_start = esp_apptrace_riscv_buffer_swap_start,
+        .swap = esp_apptrace_riscv_buffer_swap,
+        .swap_end = esp_apptrace_riscv_buffer_swap_end,
+        .host_data_pending = esp_apptrace_riscv_host_data_pending,
+    };
+    static esp_apptrace_riscv_data_t s_trace_hw_data = {
+        .membufs = {
+            .hw = &s_trace_proto_hw,
+        },
+    };
+    static esp_apptrace_hw_t s_trace_hw = {
+        .init = (esp_err_t (*)(void *))esp_apptrace_riscv_init,
+        .get_up_buffer = (uint8_t *(*)(void *, uint32_t, esp_apptrace_tmo_t *))esp_apptrace_riscv_up_buffer_get,
+        .put_up_buffer = (esp_err_t (*)(void *, uint8_t *, esp_apptrace_tmo_t *))esp_apptrace_riscv_up_buffer_put,
+        .flush_up_buffer_nolock = (esp_err_t (*)(void *, uint32_t, esp_apptrace_tmo_t *))esp_apptrace_riscv_flush_nolock,
+        .flush_up_buffer = (esp_err_t (*)(void *, esp_apptrace_tmo_t *))esp_apptrace_riscv_flush,
+        .down_buffer_config = (void (*)(void *, uint8_t *, uint32_t ))esp_apptrace_riscv_down_buffer_config,
+        .get_down_buffer = (uint8_t *(*)(void *, uint32_t *, esp_apptrace_tmo_t *))esp_apptrace_riscv_down_buffer_get,
+        .put_down_buffer = (esp_err_t (*)(void *, uint8_t *, esp_apptrace_tmo_t *))esp_apptrace_riscv_down_buffer_put,
+        .host_is_connected = (bool (*)(void *))esp_apptrace_riscv_host_is_connected,
+    };
+    *data = &s_trace_hw_data;
+    return &s_trace_hw;
+#else
+    return NULL;
+#endif
+}
+
+/* Advertises apptrace control block address to host.
+   This function can be overriden with custom implementation,
+   e.g. OpenOCD flasher stub use own implementation of it. */
+__attribute__((weak)) int esp_apptrace_advertise_ctrl_block(void *ctrl_block_addr)
+{
+    register int sys_nr = RISCV_APPTRACE_SYSNR;
+    register int host_ret = 0;
+
+    if (!esp_cpu_in_ocd_debug_mode()) {
+        return 0;
+    }
+    __asm__ volatile ( \
+        ".option push\n" \
+        ".option norvc\n" \
+        "mv a0, %[sys_nr]\n" \
+        "mv a1, %[arg1]\n" \
+        "slli    zero,zero,0x1f\n" \
+        "ebreak\n" \
+        "srai    zero,zero,0x7\n" \
+        "mv %[host_ret], a0\n" \
+    	".option pop\n" \
+        :[host_ret]"=r"(host_ret)
+        :[sys_nr]"r"(sys_nr),[arg1]"r"(ctrl_block_addr):"a0","a1");
+    return host_ret;
+}
+
+/* Returns up buffers config.
+   This function can be overriden with custom implementation,
+   e.g. OpenOCD flasher stub use own implementation of it. */
+__attribute__((weak)) void esp_apptrace_get_up_buffers(esp_apptrace_mem_block_t mem_blocks_cfg[2])
+{
+    static uint8_t s_mem_blocks[2][CONFIG_APPTRACE_BUF_SIZE];
+
+    mem_blocks_cfg[0].start = s_mem_blocks[0];
+    mem_blocks_cfg[0].sz = CONFIG_APPTRACE_BUF_SIZE;
+    mem_blocks_cfg[1].start = s_mem_blocks[1];
+    mem_blocks_cfg[1].sz = CONFIG_APPTRACE_BUF_SIZE;
+}
+
+static esp_err_t esp_apptrace_riscv_lock(esp_apptrace_riscv_data_t *hw_data, esp_apptrace_tmo_t *tmo)
+{
+#if CONFIG_APPTRACE_LOCK_ENABLE
+    esp_err_t ret = esp_apptrace_lock_take(&hw_data->lock, tmo);
+    if (ret != ESP_OK) {
+        return ESP_FAIL;
+    }
+#endif
+    return ESP_OK;
+}
+
+static esp_err_t esp_apptrace_riscv_unlock(esp_apptrace_riscv_data_t *hw_data)
+{
+    esp_err_t ret = ESP_OK;
+#if CONFIG_APPTRACE_LOCK_ENABLE
+    ret = esp_apptrace_lock_give(&hw_data->lock);
+#endif
+    return ret;
+}
+
+/*****************************************************************************************/
+/***************************** Apptrace HW iface *****************************************/
+/*****************************************************************************************/
+
+static esp_err_t esp_apptrace_riscv_init(esp_apptrace_riscv_data_t *hw_data)
+{
+    int core_id = cpu_hal_get_core_id();
+
+    if (hw_data->inited == 0) {
+        esp_apptrace_mem_block_t mem_blocks_cfg[2];
+        esp_apptrace_get_up_buffers(mem_blocks_cfg);
+        esp_err_t res = esp_apptrace_membufs_init(&hw_data->membufs, mem_blocks_cfg);
+        if (res != ESP_OK) {
+            ESP_APPTRACE_LOGE("Failed to init membufs proto (%d)!", res);
+            return res;
+        }
+#if CONFIG_APPTRACE_LOCK_ENABLE
+        esp_apptrace_lock_init(&hw_data->lock);
+#endif
+    }
+    hw_data->inited |= 1 << core_id;
+    ESP_APPTRACE_LOGI("Apptrace initialized on CPU%d. Tracing control block @ %p.", core_id, &s_tracing_ctrl[core_id]);
+    s_tracing_ctrl[core_id].mem_blocks = hw_data->membufs.blocks;
+    for (int i = 0; i < 2; i++) {
+        ESP_APPTRACE_LOGD("Mem buf[%d] %d bytes @ %p (%p/%p)", i,
+            s_tracing_ctrl[core_id].mem_blocks[i].sz, s_tracing_ctrl[core_id].mem_blocks[i].start,
+            &(s_tracing_ctrl[core_id].mem_blocks[i].start), &(s_tracing_ctrl[core_id].mem_blocks[i].sz));
+    }
+    // notify host about control block address
+    int res = esp_apptrace_advertise_ctrl_block(&s_tracing_ctrl[core_id]);
+    assert(res == 0 && "Falied to send config to host!");
+
+    return ESP_OK;
+}
+
+static uint8_t *esp_apptrace_riscv_up_buffer_get(esp_apptrace_riscv_data_t *hw_data, uint32_t size, esp_apptrace_tmo_t *tmo)
+{
+    uint8_t *ptr;
+
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return NULL;
+    }
+    esp_err_t res = esp_apptrace_riscv_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return NULL;
+    }
+
+    ptr = esp_apptrace_membufs_up_buffer_get(&hw_data->membufs, size, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    if (esp_apptrace_riscv_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }
+    return ptr;
+}
+
+static esp_err_t esp_apptrace_riscv_up_buffer_put(esp_apptrace_riscv_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    // Can avoid locking because esp_apptrace_membufs_up_buffer_put() just modifies buffer's header
+    esp_err_t res = esp_apptrace_membufs_up_buffer_put(&hw_data->membufs, ptr, tmo);
+    return res;
+}
+
+static void esp_apptrace_riscv_down_buffer_config(esp_apptrace_riscv_data_t *hw_data, uint8_t *buf, uint32_t size)
+{
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return;
+    }
+    esp_apptrace_membufs_down_buffer_config(&hw_data->membufs, buf, size);
+}
+
+static uint8_t *esp_apptrace_riscv_down_buffer_get(esp_apptrace_riscv_data_t *hw_data, uint32_t *size, esp_apptrace_tmo_t *tmo)
+{
+    uint8_t *ptr;
+
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return NULL;
+    }
+    esp_err_t res = esp_apptrace_riscv_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return NULL;
+    }
+
+    ptr = esp_apptrace_membufs_down_buffer_get(&hw_data->membufs, size, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    if (esp_apptrace_riscv_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }
+    return ptr;
+}
+
+static esp_err_t esp_apptrace_riscv_down_buffer_put(esp_apptrace_riscv_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    // Can avoid locking because esp_apptrace_membufs_down_buffer_put() does nothing
+    /*esp_err_t res = esp_apptrace_riscv_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return res;
+    }*/
+
+    esp_err_t res = esp_apptrace_membufs_down_buffer_put(&hw_data->membufs, ptr, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    /*if (esp_apptrace_riscv_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }*/
+    return res;
+}
+
+static bool esp_apptrace_riscv_host_is_connected(esp_apptrace_riscv_data_t *hw_data)
+{
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return false;
+    }
+    return s_tracing_ctrl[cpu_hal_get_core_id()].ctrl & ESP_APPTRACE_RISCV_HOST_CONNECT ? true : false;
+}
+
+static esp_err_t esp_apptrace_riscv_flush_nolock(esp_apptrace_riscv_data_t *hw_data, uint32_t min_sz, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    return esp_apptrace_membufs_flush_nolock(&hw_data->membufs, min_sz, tmo);
+}
+
+static esp_err_t esp_apptrace_riscv_flush(esp_apptrace_riscv_data_t *hw_data, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_RISCV_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    esp_err_t res = esp_apptrace_riscv_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return res;
+    }
+
+    res = esp_apptrace_membufs_flush_nolock(&hw_data->membufs, 0, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    if (esp_apptrace_riscv_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }
+    return res;
+}
+
+/*****************************************************************************************/
+/************************** Membufs proto HW iface ***************************************/
+/*****************************************************************************************/
+
+static inline void esp_apptrace_riscv_buffer_swap_lock(void)
+{
+    extern uint32_t __esp_apptrace_riscv_updated;
+
+    // indicate to host that we are about to update.
+    // this is used only to place CPU into streaming mode at tracing startup
+    // before starting streaming host can halt us after we read  ESP_APPTRACE_RISCV_CTRL_REG and before we updated it
+    // HACK: in this case host will set breakpoint just after ESP_APPTRACE_RISCV_CTRL_REG update,
+    // here we set address to set bp at
+    // enter ERI update critical section
+    s_tracing_ctrl[cpu_hal_get_core_id()].stat = (uint32_t)&__esp_apptrace_riscv_updated;
+}
+
+static __attribute__((noinline)) void esp_apptrace_riscv_buffer_swap_unlock(void)
+{
+    // exit ERI update critical section
+    s_tracing_ctrl[cpu_hal_get_core_id()].stat = 0;
+    // TODO: currently host sets breakpoint, use break instruction to stop;
+    // it will allow to use ESP_APPTRACE_RISCV_STAT_REG for other purposes
+    asm volatile (
+        "    .global     __esp_apptrace_riscv_updated\n"
+        "__esp_apptrace_riscv_updated:\n"); // host will set bp here to resolve collision at streaming start
+}
+
+static esp_err_t esp_apptrace_riscv_buffer_swap_start(uint32_t curr_block_id)
+{
+    esp_err_t res = ESP_OK;
+
+    esp_apptrace_riscv_buffer_swap_lock();
+
+    uint32_t ctrl_reg = s_tracing_ctrl[cpu_hal_get_core_id()].ctrl;
+    uint32_t host_connected = ESP_APPTRACE_RISCV_HOST_CONNECT & ctrl_reg;
+    if (host_connected) {
+        uint32_t acked_block = ESP_APPTRACE_RISCV_BLOCK_ID_GET(ctrl_reg);
+        uint32_t host_to_read = ESP_APPTRACE_RISCV_BLOCK_LEN_GET(ctrl_reg);
+        if (host_to_read != 0 || acked_block != (curr_block_id & ESP_APPTRACE_RISCV_BLOCK_ID_MSK)) {
+            ESP_APPTRACE_LOGD("[%d]: Can not switch %x %d %x %x/%lx", cpu_hal_get_core_id(), ctrl_reg, host_to_read, acked_block,
+                curr_block_id & ESP_APPTRACE_RISCV_BLOCK_ID_MSK, curr_block_id);
+            res = ESP_ERR_NO_MEM;
+            goto _on_err;
+        }
+    }
+    return ESP_OK;
+_on_err:
+    esp_apptrace_riscv_buffer_swap_unlock();
+    return res;
+}
+
+static esp_err_t esp_apptrace_riscv_buffer_swap_end(uint32_t new_block_id, uint32_t prev_block_len)
+{
+    uint32_t ctrl_reg = s_tracing_ctrl[cpu_hal_get_core_id()].ctrl;
+    uint32_t host_connected = ESP_APPTRACE_RISCV_HOST_CONNECT & ctrl_reg;
+    s_tracing_ctrl[cpu_hal_get_core_id()].ctrl = ESP_APPTRACE_RISCV_BLOCK_ID(new_block_id) |
+              host_connected | ESP_APPTRACE_RISCV_BLOCK_LEN(prev_block_len);
+    esp_apptrace_riscv_buffer_swap_unlock();
+    return ESP_OK;
+}
+
+static esp_err_t esp_apptrace_riscv_buffer_swap(uint32_t new_block_id)
+{
+    /* do nothing */
+    return ESP_OK;
+}
+
+static bool esp_apptrace_riscv_host_data_pending(void)
+{
+    uint32_t ctrl_reg = s_tracing_ctrl[cpu_hal_get_core_id()].ctrl;
+    // ESP_APPTRACE_LOGV("%s() 0x%x", __func__, ctrl_reg);
+    return (ctrl_reg & ESP_APPTRACE_RISCV_HOST_DATA) ? true : false;
+}

components/app_trace/port/xtensa/port.c

@@ -0,0 +1,543 @@
+//
+// How It Works
+// ************
+
+// 1. Components Overview
+// ======================
+
+// Xtensa has useful feature: TRAX debug module. It allows recording program execution flow at run-time without disturbing CPU.
+// Exectution flow data are written to configurable Trace RAM block. Besides accessing Trace RAM itself TRAX module also allows to read/write
+// trace memory via its registers by means of JTAG, APB or ERI transactions.
+// ESP32 has two Xtensa cores with separate TRAX modules on them and provides two special memory regions to be used as trace memory.
+// Chip allows muxing access to those trace memory blocks in such a way that while one block is accessed by CPUs another one can be accessed by host
+// by means of reading/writing TRAX registers via JTAG. Blocks muxing is configurable at run-time and allows switching trace memory blocks between
+// accessors in round-robin fashion so they can read/write separate memory blocks without disturbing each other.
+// This module implements application tracing feature based on above mechanisms. It allows to transfer arbitrary user data to/from
+// host via JTAG with minimal impact on system performance. This module is implied to be used in the following tracing scheme.
+
+//                                                        ------>------                                         ----- (host components) -----
+//                                                        |           |                                         |                           |
+// -------------------   -----------------------     -----------------------     ----------------    ------     ---------   -----------------
+// |trace data source|-->|target tracing module|<--->|TRAX_MEM0 | TRAX_MEM1|---->|TRAX_DATA_REGS|<-->|JTAG|<--->|OpenOCD|-->|trace data sink|
+// -------------------   -----------------------     -----------------------     ----------------    ------     ---------   -----------------
+//                                 |                      |           |                                |
+//                                 |                      ------<------          ----------------      |
+//                                 |<------------------------------------------->|TRAX_CTRL_REGS|<---->|
+//                                                                               ----------------
+
+// In general tracing goes in the following way. User application requests tracing module to send some data by calling esp_apptrace_buffer_get(),
+// module allocates necessary buffer in current input trace block. Then user fills received buffer with data and calls esp_apptrace_buffer_put().
+// When current input trace block is filled with app data it is exposed to host and the second block becomes input one and buffer filling restarts.
+// While target application fills one TRAX block host reads another one via JTAG.
+// This module also allows communication in the opposite direction: from host to target. As it was said ESP32 and host can access different TRAX blocks
+// simultaneously, so while target writes trace data to one block host can write its own data (e.g. tracing commands) to another one then when
+// blocks are switched host receives trace data and target receives data written by host application. Target user application can read host data
+// by calling esp_apptrace_read() API.
+// To control buffer switching and for other communication purposes this implementation uses some TRAX registers. It is safe since HW TRAX tracing
+// can not be used along with application tracing feature so these registers are freely readable/writeable via JTAG from host and via ERI from ESP32 cores.
+// Overhead of this implementation on target CPU is produced only by allocating/managing buffers and copying of data.
+// On the host side special OpenOCD command must be used to read trace data.
+
+// 2. TRAX Registers layout
+// ========================
+
+// This module uses two TRAX HW registers to communicate with host SW (OpenOCD).
+//  - Control register uses TRAX_DELAYCNT as storage. Only lower 24 bits of TRAX_DELAYCNT are writable. Control register has the following bitfields:
+//   | 31..XXXXXX..24 | 23 .(host_connect). 23| 22..(block_id)..15 | 14..(block_len)..0 |
+//    14..0  bits - actual length of user data in trace memory block. Target updates it every time it fills memory block and exposes it to host.
+//                  Host writes zero to this field when it finishes reading exposed block;
+//    21..15 bits - trace memory block transfer ID. Block counter. It can overflow. Updated by target, host should not modify it. Actually can be 2 bits;
+//    22     bit  - 'host data present' flag. If set to one there is data from host, otherwise - no host data;
+//    23     bit  - 'host connected' flag. If zero then host is not connected and tracing module works in post-mortem mode, otherwise in streaming mode;
+// - Status register uses TRAX_TRIGGERPC as storage. If this register is not zero then current CPU is changing TRAX registers and
+//   this register holds address of the instruction which application will execute when it finishes with those registers modifications.
+//   See 'Targets Connection' setion for details.
+
+// 3. Modes of operation
+// =====================
+
+// This module supports two modes of operation:
+//  - Post-mortem mode. This is the default mode. In this mode application tracing module does not check whether host has read all the data from block
+//    exposed to it and switches block in any case. The mode does not need host interaction for operation and so can be useful when only the latest
+//    trace data are necessary, e.g. for analyzing crashes. On panic the latest data from current input block are exposed to host and host can read them.
+//    It can happen that system panic occurs when there are very small amount of data which are not exposed to host yet (e.g. crash just after the
+//    TRAX block switch). In this case the previous 16KB of collected data will be dropped and host will see the latest, but very small piece of trace.
+//    It can be insufficient to diagnose the problem. To avoid such situations there is menuconfig option
+//    CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH
+//    which controls the threshold for flushing data in case of panic.
+//  - Streaming mode. Tracing module enters this mode when host connects to target and sets respective bits in control registers (per core).
+//    In this mode before switching the block tracing module waits for the host to read all the data from the previously exposed block.
+//    On panic tracing module also waits (timeout is configured via menuconfig via CONFIG_APPTRACE_ONPANIC_HOST_FLUSH_TMO) for the host to read all data.
+
+// 4. Communication Protocol
+// =========================
+
+// 4.1 Trace Memory Blocks
+// -----------------------
+
+// Communication is controlled via special register. Host periodically polls control register on each core to find out if there are any data available.
+// When current input memory block is filled it is exposed to host and 'block_len' and 'block_id' fields are updated in the control register.
+// Host reads new register value and according to it's value starts reading data from exposed block. Meanwhile target starts filling another trace block.
+// When host finishes reading the block it clears 'block_len' field in control register indicating to the target that it is ready to accept the next one.
+// If the host has some data to transfer to the target it writes them to trace memory block before clearing 'block_len' field. Then it sets
+// 'host_data_present' bit and clears 'block_len' field in control register. Upon every block switch target checks 'host_data_present' bit and if it is set
+// reads them to down buffer before writing any trace data to switched TRAX block.
+
+// 4.2 User Data Chunks Level
+// --------------------------
+
+// Since trace memory block is shared between user data chunks and data copying is performed on behalf of the API user (in its normal context) in
+// multithreading environment it can happen that task/ISR which copies data is preempted by another high prio task/ISR. So it is possible situation
+// that task/ISR will fail to complete filling its data chunk before the whole trace block is exposed to the host. To handle such conditions tracing
+// module prepends all user data chunks with header which contains allocated buffer size and actual data length within it. OpenOCD command
+// which reads application traces reports error when it reads incomplete user data block.
+// Data which are transffered from host to target are also prepended with a header. Down channel data header is simple and consists of one two bytes field
+// containing length of host data following the header.
+
+// 4.3 Data Buffering
+// ------------------
+
+// It takes some time for the host to read TRAX memory block via JTAG. In streaming mode it can happen that target has filled its TRAX block, but host
+// has not completed reading of the previous one yet. So in this case time critical tracing calls (which can not be delayed for too long time due to
+// the lack of free memory in TRAX block) can be dropped. To avoid such scenarios tracing module implements data buffering. Buffered data will be sent
+// to the host later when TRAX block switch occurs. The maximum size of the buffered data is controlled by menuconfig option
+// CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX.
+
+// 4.4 Target Connection/Disconnection
+// -----------------------------------
+
+// When host is going to start tracing in streaming mode it needs to put both ESP32 cores into initial state when 'host connected' bit is set
+// on both cores. To accomplish this host halts both cores and sets this bit in TRAX registers. But target code can be halted in state when it has read control
+// register but has not updated its value. To handle such situations target code indicates to the host that it is updating control register by writing
+// non-zero value to status register. Actually it writes address of the instruction which it will execute when it finishes with
+// the registers update. When target is halted during control register update host sets breakpoint at the address from status register and resumes CPU.
+// After target code finishes with register update it is halted on breakpoint, host detects it and safely sets 'host connected' bit. When both cores
+// are set up they are resumed. Tracing starts without further intrusion into CPUs work.
+// When host is going to stop tracing in streaming mode it needs to disconnect targets. Disconnection process is done using the same algorithm
+// as for connecting, but 'host connected' bits are cleared on ESP32 cores.
+
+// 5. Module Access Synchronization
+// ================================
+
+// Access to internal module's data is synchronized with custom mutex. Mutex is a wrapper for portMUX_TYPE and uses almost the same sync mechanism as in
+// vPortCPUAcquireMutex/vPortCPUReleaseMutex. The mechanism uses S32C1I Xtensa instruction to implement exclusive access to module's data from tasks and
+// ISRs running on both cores. Also custom mutex allows specifying timeout for locking operation. Locking routine checks underlaying mutex in cycle until
+// it gets its ownership or timeout expires. The differences of application tracing module's mutex implementation from vPortCPUAcquireMutex/vPortCPUReleaseMutex are:
+// - Support for timeouts.
+// - Local IRQs for CPU which owns the mutex are disabled till the call to unlocking routine. This is made to avoid possible task's prio inversion.
+//   When low prio task takes mutex and enables local IRQs gets preempted by high prio task which in its turn can try to acquire mutex using infinite timeout.
+//   So no local task switch occurs when mutex is locked. But this does not apply to tasks on another CPU.
+//   WARNING: Priority inversion can happen when low prio task works on one CPU and medium and high prio tasks work on another.
+// WARNING: Care must be taken when selecting timeout values for trace calls from ISRs. Tracing module does not care about watchdogs when waiting
+// on internal locks and for host to complete previous block reading, so if timeout value exceeds watchdog's one it can lead to the system reboot.
+
+// 6. Timeouts
+// ===========
+
+// Timeout mechanism is based on xthal_get_ccount() routine and supports timeout values in microseconds.
+// There are two situations when task/ISR can be delayed by tracing API call. Timeout mechanism takes into account both conditions:
+// - Trace data are locked by another task/ISR. When wating on trace data lock.
+// - Current TRAX memory input block is full when working in streaming mode (host is connected). When waiting for host to complete previous block reading.
+// When wating for any of above conditions xthal_get_ccount() is called periodically to calculate time elapsed from trace API routine entry. When elapsed
+// time exceeds specified timeout value operation is canceled and ESP_ERR_TIMEOUT code is returned.
+#include "sdkconfig.h"
+#include "soc/soc.h"
+#include "soc/dport_access.h"
+#if CONFIG_IDF_TARGET_ESP32
+#include "soc/dport_reg.h"
+#elif CONFIG_IDF_TARGET_ESP32S2
+#include "soc/sensitive_reg.h"
+#endif
+#include "eri.h"
+#include "trax.h"
+#include "esp_log.h"
+#include "esp_app_trace_membufs_proto.h"
+#include "esp_app_trace_port.h"
+
+// TODO: move these (and same definitions in trax.c to dport_reg.h)
+#if CONFIG_IDF_TARGET_ESP32
+#define TRACEMEM_MUX_PROBLK0_APPBLK1            0
+#define TRACEMEM_MUX_BLK0_ONLY                  1
+#define TRACEMEM_MUX_BLK1_ONLY                  2
+#define TRACEMEM_MUX_PROBLK1_APPBLK0            3
+#elif CONFIG_IDF_TARGET_ESP32S2
+#define TRACEMEM_MUX_BLK0_NUM                   19
+#define TRACEMEM_MUX_BLK1_NUM                   20
+#define TRACEMEM_BLK_NUM2ADDR(_n_)              (0x3FFB8000UL + 0x4000UL*((_n_)-4))
+#endif
+
+// TRAX is disabled, so we use its registers for our own purposes
+// | 31..XXXXXX..24 | 23 .(host_connect). 23 | 22 .(host_data). 22| 21..(block_id)..15 | 14..(block_len)..0 |
+#define ESP_APPTRACE_TRAX_CTRL_REG              ERI_TRAX_DELAYCNT
+#define ESP_APPTRACE_TRAX_STAT_REG              ERI_TRAX_TRIGGERPC
+
+#define ESP_APPTRACE_TRAX_BLOCK_LEN_MSK         0x7FFFUL
+#define ESP_APPTRACE_TRAX_BLOCK_LEN(_l_)        ((_l_) & ESP_APPTRACE_TRAX_BLOCK_LEN_MSK)
+#define ESP_APPTRACE_TRAX_BLOCK_LEN_GET(_v_)    ((_v_) & ESP_APPTRACE_TRAX_BLOCK_LEN_MSK)
+#define ESP_APPTRACE_TRAX_BLOCK_ID_MSK          0x7FUL
+#define ESP_APPTRACE_TRAX_BLOCK_ID(_id_)        (((_id_) & ESP_APPTRACE_TRAX_BLOCK_ID_MSK) << 15)
+#define ESP_APPTRACE_TRAX_BLOCK_ID_GET(_v_)     (((_v_) >> 15) & ESP_APPTRACE_TRAX_BLOCK_ID_MSK)
+#define ESP_APPTRACE_TRAX_HOST_DATA             (1 << 22)
+#define ESP_APPTRACE_TRAX_HOST_CONNECT          (1 << 23)
+
+#define ESP_APPTRACE_TRAX_INITED(_hw_)          ((_hw_)->inited & (1 << cpu_hal_get_core_id()))
+
+#if CONFIG_IDF_TARGET_ESP32
+static uint8_t * const s_trax_blocks[] = {
+    (uint8_t *) 0x3FFFC000,
+    (uint8_t *) 0x3FFF8000
+};
+#elif CONFIG_IDF_TARGET_ESP32S2
+static uint8_t * const s_trax_blocks[] = {
+    (uint8_t *)TRACEMEM_BLK_NUM2ADDR(TRACEMEM_MUX_BLK0_NUM),
+    (uint8_t *)TRACEMEM_BLK_NUM2ADDR(TRACEMEM_MUX_BLK1_NUM)
+};
+#endif
+
+#define ESP_APPTRACE_TRAX_BLOCK_SIZE            (0x4000UL)
+
+/** TRAX HW transport data */
+typedef struct {
+    uint8_t                             inited;
+#if CONFIG_APPTRACE_LOCK_ENABLE
+    esp_apptrace_lock_t                 lock;   // sync lock
+#endif
+    esp_apptrace_membufs_proto_data_t   membufs;
+} esp_apptrace_trax_data_t;
+
+
+static esp_err_t esp_apptrace_trax_init(esp_apptrace_trax_data_t *hw_data);
+static esp_err_t esp_apptrace_trax_flush(esp_apptrace_trax_data_t *hw_data, esp_apptrace_tmo_t *tmo);
+static esp_err_t esp_apptrace_trax_flush_nolock(esp_apptrace_trax_data_t *hw_data, uint32_t min_sz, esp_apptrace_tmo_t *tmo);
+static uint8_t *esp_apptrace_trax_up_buffer_get(esp_apptrace_trax_data_t *hw_data, uint32_t size, esp_apptrace_tmo_t *tmo);
+static esp_err_t esp_apptrace_trax_up_buffer_put(esp_apptrace_trax_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo);
+static void esp_apptrace_trax_down_buffer_config(esp_apptrace_trax_data_t *hw_data, uint8_t *buf, uint32_t size);
+static uint8_t *esp_apptrace_trax_down_buffer_get(esp_apptrace_trax_data_t *hw_data, uint32_t *size, esp_apptrace_tmo_t *tmo);
+static esp_err_t esp_apptrace_trax_down_buffer_put(esp_apptrace_trax_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo);
+static bool esp_apptrace_trax_host_is_connected(esp_apptrace_trax_data_t *hw_data);
+static esp_err_t esp_apptrace_trax_buffer_swap_start(uint32_t curr_block_id);
+static esp_err_t esp_apptrace_trax_buffer_swap(uint32_t new_block_id);
+static esp_err_t esp_apptrace_trax_buffer_swap_end(uint32_t new_block_id, uint32_t prev_block_len);
+static bool esp_apptrace_trax_host_data_pending(void);
+
+
+const static char *TAG = "esp_apptrace";
+
+esp_apptrace_hw_t *esp_apptrace_uart_hw_get(int num, void **data)
+{
+    return NULL;
+}
+
+esp_apptrace_hw_t *esp_apptrace_jtag_hw_get(void **data)
+{
+#if CONFIG_APPTRACE_DEST_JTAG
+    static esp_apptrace_membufs_proto_hw_t s_trax_proto_hw = {
+        .swap_start = esp_apptrace_trax_buffer_swap_start,
+        .swap = esp_apptrace_trax_buffer_swap,
+        .swap_end = esp_apptrace_trax_buffer_swap_end,
+        .host_data_pending = esp_apptrace_trax_host_data_pending,
+    };
+    static esp_apptrace_trax_data_t s_trax_hw_data = {
+        .membufs = {
+            .hw = &s_trax_proto_hw,
+        },
+    };
+    static esp_apptrace_hw_t s_trax_hw = {
+        .init = (esp_err_t (*)(void *))esp_apptrace_trax_init,
+        .get_up_buffer = (uint8_t *(*)(void *, uint32_t, esp_apptrace_tmo_t *))esp_apptrace_trax_up_buffer_get,
+        .put_up_buffer = (esp_err_t (*)(void *, uint8_t *, esp_apptrace_tmo_t *))esp_apptrace_trax_up_buffer_put,
+        .flush_up_buffer_nolock = (esp_err_t (*)(void *, uint32_t, esp_apptrace_tmo_t *))esp_apptrace_trax_flush_nolock,
+        .flush_up_buffer = (esp_err_t (*)(void *, esp_apptrace_tmo_t *))esp_apptrace_trax_flush,
+        .down_buffer_config = (void (*)(void *, uint8_t *, uint32_t ))esp_apptrace_trax_down_buffer_config,
+        .get_down_buffer = (uint8_t *(*)(void *, uint32_t *, esp_apptrace_tmo_t *))esp_apptrace_trax_down_buffer_get,
+        .put_down_buffer = (esp_err_t (*)(void *, uint8_t *, esp_apptrace_tmo_t *))esp_apptrace_trax_down_buffer_put,
+        .host_is_connected = (bool (*)(void *))esp_apptrace_trax_host_is_connected,
+    };
+    *data = &s_trax_hw_data;
+    return &s_trax_hw;
+#else
+    return NULL;
+#endif
+}
+
+static esp_err_t esp_apptrace_trax_lock(esp_apptrace_trax_data_t *hw_data, esp_apptrace_tmo_t *tmo)
+{
+#if CONFIG_APPTRACE_LOCK_ENABLE
+    esp_err_t ret = esp_apptrace_lock_take(&hw_data->lock, tmo);
+    if (ret != ESP_OK) {
+        return ESP_FAIL;
+    }
+#endif
+    return ESP_OK;
+}
+
+static esp_err_t esp_apptrace_trax_unlock(esp_apptrace_trax_data_t *hw_data)
+{
+    esp_err_t ret = ESP_OK;
+#if CONFIG_APPTRACE_LOCK_ENABLE
+    ret = esp_apptrace_lock_give(&hw_data->lock);
+#endif
+    return ret;
+}
+
+static inline void esp_apptrace_trax_hw_init(void)
+{
+    // Stop trace, if any (on the current CPU)
+    eri_write(ERI_TRAX_TRAXCTRL, TRAXCTRL_TRSTP);
+    eri_write(ERI_TRAX_TRAXCTRL, TRAXCTRL_TMEN);
+    eri_write(ESP_APPTRACE_TRAX_CTRL_REG, ESP_APPTRACE_TRAX_BLOCK_ID(0));
+    // this is for OpenOCD to let him know where stub entries vector is resided
+    // must be read by host before any transfer using TRAX
+    eri_write(ESP_APPTRACE_TRAX_STAT_REG, 0);
+
+    ESP_APPTRACE_LOGI("Initialized TRAX on CPU%d", cpu_hal_get_core_id());
+}
+
+static inline void esp_apptrace_trax_select_memory_block(int block_num)
+{
+    // select memory block to be exposed to the TRAX module (accessed by host)
+#if CONFIG_IDF_TARGET_ESP32
+    DPORT_WRITE_PERI_REG(DPORT_TRACEMEM_MUX_MODE_REG, block_num ? TRACEMEM_MUX_BLK0_ONLY : TRACEMEM_MUX_BLK1_ONLY);
+#elif CONFIG_IDF_TARGET_ESP32S2
+    WRITE_PERI_REG(DPORT_PMS_OCCUPY_3_REG, block_num ? BIT(TRACEMEM_MUX_BLK0_NUM-4) : BIT(TRACEMEM_MUX_BLK1_NUM-4));
+#endif
+}
+
+static inline void esp_apptrace_trax_memory_enable(void)
+{
+#if CONFIG_IDF_TARGET_ESP32
+        /* Enable trace memory on PRO CPU */
+        DPORT_WRITE_PERI_REG(DPORT_PRO_TRACEMEM_ENA_REG, DPORT_PRO_TRACEMEM_ENA_M);
+#if CONFIG_FREERTOS_UNICORE == 0
+        /* Enable trace memory on APP CPU */
+        DPORT_WRITE_PERI_REG(DPORT_APP_TRACEMEM_ENA_REG, DPORT_APP_TRACEMEM_ENA_M);
+#endif
+#endif
+}
+
+/*****************************************************************************************/
+/***************************** Apptrace HW iface *****************************************/
+/*****************************************************************************************/
+
+static esp_err_t esp_apptrace_trax_init(esp_apptrace_trax_data_t *hw_data)
+{
+    int core_id = cpu_hal_get_core_id();
+
+    // 'esp_apptrace_trax_init()' is called on every core, so ensure to do main initialization only once
+    if (core_id == 0) {
+        esp_apptrace_mem_block_t mem_blocks_cfg[2] = {
+            {
+                .start = s_trax_blocks[0],
+                .sz = ESP_APPTRACE_TRAX_BLOCK_SIZE
+            },
+            {
+                .start = s_trax_blocks[1],
+                .sz = ESP_APPTRACE_TRAX_BLOCK_SIZE
+            },
+        };
+        esp_err_t res = esp_apptrace_membufs_init(&hw_data->membufs, mem_blocks_cfg);
+        if (res != ESP_OK) {
+            ESP_APPTRACE_LOGE("Failed to init membufs proto (%d)!", res);
+            return res;
+        }
+#if CONFIG_APPTRACE_LOCK_ENABLE
+        esp_apptrace_lock_init(&hw_data->lock);
+#endif
+        esp_apptrace_trax_memory_enable();
+        esp_apptrace_trax_select_memory_block(0);
+    }
+    // init TRAX on this CPU
+    esp_apptrace_trax_hw_init();
+    hw_data->inited |= 1 << core_id;
+
+    return ESP_OK;
+}
+
+static uint8_t *esp_apptrace_trax_up_buffer_get(esp_apptrace_trax_data_t *hw_data, uint32_t size, esp_apptrace_tmo_t *tmo)
+{
+    uint8_t *ptr;
+
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return NULL;
+    }
+    esp_err_t res = esp_apptrace_trax_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return NULL;
+    }
+
+    ptr = esp_apptrace_membufs_up_buffer_get(&hw_data->membufs, size, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    if (esp_apptrace_trax_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }
+    return ptr;
+}
+
+static esp_err_t esp_apptrace_trax_up_buffer_put(esp_apptrace_trax_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    // Can avoid locking because esp_apptrace_membufs_up_buffer_put() just modifies buffer's header
+    esp_err_t res = esp_apptrace_membufs_up_buffer_put(&hw_data->membufs, ptr, tmo);
+    return res;
+}
+
+static void esp_apptrace_trax_down_buffer_config(esp_apptrace_trax_data_t *hw_data, uint8_t *buf, uint32_t size)
+{
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return;
+    }
+    esp_apptrace_membufs_down_buffer_config(&hw_data->membufs, buf, size);
+}
+
+static uint8_t *esp_apptrace_trax_down_buffer_get(esp_apptrace_trax_data_t *hw_data, uint32_t *size, esp_apptrace_tmo_t *tmo)
+{
+    uint8_t *ptr;
+
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return NULL;
+    }
+    esp_err_t res = esp_apptrace_trax_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return NULL;
+    }
+
+    ptr = esp_apptrace_membufs_down_buffer_get(&hw_data->membufs, size, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    if (esp_apptrace_trax_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }
+    return ptr;
+}
+
+static esp_err_t esp_apptrace_trax_down_buffer_put(esp_apptrace_trax_data_t *hw_data, uint8_t *ptr, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    // Can avoid locking because esp_apptrace_membufs_down_buffer_put() does nothing
+    /*esp_err_t res = esp_apptrace_trax_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return res;
+    }*/
+
+    esp_err_t res = esp_apptrace_membufs_down_buffer_put(&hw_data->membufs, ptr, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    /*if (esp_apptrace_trax_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }*/
+    return res;
+}
+
+static bool esp_apptrace_trax_host_is_connected(esp_apptrace_trax_data_t *hw_data)
+{
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return false;
+    }
+    return eri_read(ESP_APPTRACE_TRAX_CTRL_REG) & ESP_APPTRACE_TRAX_HOST_CONNECT ? true : false;
+}
+
+static esp_err_t esp_apptrace_trax_flush_nolock(esp_apptrace_trax_data_t *hw_data, uint32_t min_sz, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    return esp_apptrace_membufs_flush_nolock(&hw_data->membufs, min_sz, tmo);
+}
+
+static esp_err_t esp_apptrace_trax_flush(esp_apptrace_trax_data_t *hw_data, esp_apptrace_tmo_t *tmo)
+{
+    if (!ESP_APPTRACE_TRAX_INITED(hw_data)) {
+        return ESP_ERR_INVALID_STATE;
+    }
+    esp_err_t res = esp_apptrace_trax_lock(hw_data, tmo);
+    if (res != ESP_OK) {
+        return res;
+    }
+
+    res = esp_apptrace_membufs_flush_nolock(&hw_data->membufs, 0, tmo);
+
+    // now we can safely unlock apptrace to allow other tasks/ISRs to get other buffers and write their data
+    if (esp_apptrace_trax_unlock(hw_data) != ESP_OK) {
+        assert(false && "Failed to unlock apptrace data!");
+    }
+    return res;
+}
+
+/*****************************************************************************************/
+/************************** Membufs proto HW iface ***************************************/
+/*****************************************************************************************/
+
+static inline void esp_apptrace_trax_buffer_swap_lock(void)
+{
+    extern uint32_t __esp_apptrace_trax_eri_updated;
+
+    // indicate to host that we are about to update.
+    // this is used only to place CPU into streaming mode at tracing startup
+    // before starting streaming host can halt us after we read  ESP_APPTRACE_TRAX_CTRL_REG and before we updated it
+    // HACK: in this case host will set breakpoint just after ESP_APPTRACE_TRAX_CTRL_REG update,
+    // here we set address to set bp at
+    // enter ERI update critical section
+    eri_write(ESP_APPTRACE_TRAX_STAT_REG, (uint32_t)&__esp_apptrace_trax_eri_updated);
+}
+
+static __attribute__((noinline)) void esp_apptrace_trax_buffer_swap_unlock(void)
+{
+    // exit ERI update critical section
+    eri_write(ESP_APPTRACE_TRAX_STAT_REG, 0x0);
+    // TODO: currently host sets breakpoint, use break instruction to stop;
+    // it will allow to use ESP_APPTRACE_TRAX_STAT_REG for other purposes
+    asm volatile (
+        "    .global     __esp_apptrace_trax_eri_updated\n"
+        "__esp_apptrace_trax_eri_updated:\n"); // host will set bp here to resolve collision at streaming start
+}
+
+static esp_err_t esp_apptrace_trax_buffer_swap_start(uint32_t curr_block_id)
+{
+    esp_err_t res = ESP_OK;
+
+    esp_apptrace_trax_buffer_swap_lock();
+
+    uint32_t ctrl_reg = eri_read(ESP_APPTRACE_TRAX_CTRL_REG);
+    uint32_t host_connected = ESP_APPTRACE_TRAX_HOST_CONNECT & ctrl_reg;
+    if (host_connected) {
+        uint32_t acked_block = ESP_APPTRACE_TRAX_BLOCK_ID_GET(ctrl_reg);
+        uint32_t host_to_read = ESP_APPTRACE_TRAX_BLOCK_LEN_GET(ctrl_reg);
+        if (host_to_read != 0 || acked_block != (curr_block_id & ESP_APPTRACE_TRAX_BLOCK_ID_MSK)) {
+            ESP_APPTRACE_LOGD("HC[%d]: Can not switch %x %d %x %x/%lx", cpu_hal_get_core_id(), ctrl_reg, host_to_read, acked_block,
+                curr_block_id & ESP_APPTRACE_TRAX_BLOCK_ID_MSK, curr_block_id);
+            res = ESP_ERR_NO_MEM;
+            goto _on_err;
+        }
+    }
+    return ESP_OK;
+_on_err:
+    esp_apptrace_trax_buffer_swap_unlock();
+    return res;
+}
+
+static esp_err_t esp_apptrace_trax_buffer_swap_end(uint32_t new_block_id, uint32_t prev_block_len)
+{
+    uint32_t ctrl_reg = eri_read(ESP_APPTRACE_TRAX_CTRL_REG);
+    uint32_t host_connected = ESP_APPTRACE_TRAX_HOST_CONNECT & ctrl_reg;
+    eri_write(ESP_APPTRACE_TRAX_CTRL_REG, ESP_APPTRACE_TRAX_BLOCK_ID(new_block_id) |
+              host_connected | ESP_APPTRACE_TRAX_BLOCK_LEN(prev_block_len));
+    esp_apptrace_trax_buffer_swap_unlock();
+    return ESP_OK;
+}
+
+static esp_err_t esp_apptrace_trax_buffer_swap(uint32_t new_block_id)
+{
+    esp_apptrace_trax_select_memory_block(new_block_id);
+    return ESP_OK;
+}
+
+static bool esp_apptrace_trax_host_data_pending(void)
+{
+    uint32_t ctrl_reg = eri_read(ESP_APPTRACE_TRAX_CTRL_REG);
+    return (ctrl_reg & ESP_APPTRACE_TRAX_HOST_DATA) ? true : false;
+}

components/app_trace/private_include/esp_app_trace_membufs_proto.h

@@ -0,0 +1,70 @@
+// Copyright 2020 Espressif Systems (Shanghai) PTE LTD
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+#ifndef ESP_APP_TRACE_MEMBUFS_PROTO_H_
+#define ESP_APP_TRACE_MEMBUFS_PROTO_H_
+
+#include "esp_app_trace_util.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** TRAX HW transport state */
+typedef struct {
+    uint32_t                   in_block;     // input block ID
+    // TODO: change to uint16_t
+    uint32_t                   markers[2];   // block filling level markers
+} esp_apptrace_membufs_state_t;
+
+/** memory block parameters,
+ * should be packed, because it is read from the host */
+typedef struct {
+    uint8_t   *start;   // start address
+    uint32_t   sz;      // size
+} esp_apptrace_mem_block_t;
+
+typedef struct {
+    esp_err_t (*swap_start)(uint32_t curr_block_id);
+    esp_err_t (*swap)(uint32_t new_block_id);
+    esp_err_t (*swap_end)(uint32_t new_block_id, uint32_t prev_block_len);
+    bool (*host_data_pending)(void);
+} esp_apptrace_membufs_proto_hw_t;
+
+typedef struct {
+    esp_apptrace_membufs_proto_hw_t *       hw;
+    volatile esp_apptrace_membufs_state_t   state;       // state
+    esp_apptrace_mem_block_t                blocks[2];   // memory blocks
+#if CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX > 0
+    // ring buffer control struct for pending user blocks
+    esp_apptrace_rb_t                   rb_pend;
+    // storage for pending user blocks
+    uint8_t                             pending_data[CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX + 1];
+#endif
+    // ring buffer control struct for data from host (down buffer)
+    esp_apptrace_rb_t                       rb_down;
+} esp_apptrace_membufs_proto_data_t;
+
+esp_err_t esp_apptrace_membufs_init(esp_apptrace_membufs_proto_data_t *proto, const esp_apptrace_mem_block_t blocks_cfg[2]);
+void esp_apptrace_membufs_down_buffer_config(esp_apptrace_membufs_proto_data_t *data, uint8_t *buf, uint32_t size);
+uint8_t *esp_apptrace_membufs_down_buffer_get(esp_apptrace_membufs_proto_data_t *proto, uint32_t *size, esp_apptrace_tmo_t *tmo);
+esp_err_t esp_apptrace_membufs_down_buffer_put(esp_apptrace_membufs_proto_data_t *proto, uint8_t *ptr, esp_apptrace_tmo_t *tmo);
+uint8_t *esp_apptrace_membufs_up_buffer_get(esp_apptrace_membufs_proto_data_t *proto, uint32_t size, esp_apptrace_tmo_t *tmo);
+esp_err_t esp_apptrace_membufs_up_buffer_put(esp_apptrace_membufs_proto_data_t *proto, uint8_t *ptr, esp_apptrace_tmo_t *tmo);
+esp_err_t esp_apptrace_membufs_flush_nolock(esp_apptrace_membufs_proto_data_t *proto, uint32_t min_sz, esp_apptrace_tmo_t *tmo);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif

components/app_trace/sdkconfig.rename

@@ -3,10 +3,34 @@
 
 CONFIG_ESP32_APPTRACE_DESTINATION      					CONFIG_APPTRACE_DESTINATION
 CONFIG_ESP32_APPTRACE_DEST_NONE							CONFIG_APPTRACE_DEST_NONE
-CONFIG_ESP32_APPTRACE_DEST_TRAX							CONFIG_APPTRACE_DEST_TRAX
+CONFIG_ESP32_APPTRACE_DEST_TRAX							CONFIG_APPTRACE_DEST_JTAG
 CONFIG_ESP32_APPTRACE_ENABLE							CONFIG_APPTRACE_ENABLE
 CONFIG_ESP32_APPTRACE_LOCK_ENABLE						CONFIG_APPTRACE_LOCK_ENABLE
 CONFIG_ESP32_APPTRACE_ONPANIC_HOST_FLUSH_TMO			CONFIG_APPTRACE_ONPANIC_HOST_FLUSH_TMO
 CONFIG_ESP32_APPTRACE_POSTMORTEM_FLUSH_TRAX_THRESH      CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH
 CONFIG_ESP32_APPTRACE_PENDING_DATA_SIZE_MAX				CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX
 CONFIG_ESP32_GCOV_ENABLE								CONFIG_APPTRACE_GCOV_ENABLE
+
+CONFIG_SYSVIEW_ENABLE							        CONFIG_APPTRACE_SV_ENABLE
+CONFIG_SYSVIEW_TS_SOURCE							    CONFIG_APPTRACE_SV_TS_SOURCE
+CONFIG_SYSVIEW_TS_SOURCE_CCOUNT							CONFIG_APPTRACE_SV_TS_SOURCE_CCOUNT
+CONFIG_SYSVIEW_TS_SOURCE_TIMER_00						CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_00
+CONFIG_SYSVIEW_TS_SOURCE_TIMER_01						CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_01
+CONFIG_SYSVIEW_TS_SOURCE_TIMER_10						CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_10
+CONFIG_SYSVIEW_TS_SOURCE_TIMER_11						CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_11
+CONFIG_SYSVIEW_TS_SOURCE_ESP_TIMER						CONFIG_APPTRACE_SV_TS_SOURCE_ESP_TIMER
+CONFIG_SYSVIEW_MAX_TASKS							    CONFIG_APPTRACE_SV_MAX_TASKS
+CONFIG_SYSVIEW_BUF_WAIT_TMO							    CONFIG_APPTRACE_SV_BUF_WAIT_TMO
+CONFIG_SYSVIEW_EVT_OVERFLOW_ENABLE						CONFIG_APPTRACE_SV_EVT_OVERFLOW_ENABLE
+CONFIG_SYSVIEW_EVT_ISR_ENTER_ENABLE						CONFIG_APPTRACE_SV_EVT_ISR_ENTER_ENABLE
+CONFIG_SYSVIEW_EVT_ISR_EXIT_ENABLE						CONFIG_APPTRACE_SV_EVT_ISR_EXIT_ENABLE
+CONFIG_SYSVIEW_EVT_ISR_TO_SCHEDULER_ENABLE				CONFIG_APPTRACE_SV_EVT_ISR_TO_SCHED_ENABLE
+CONFIG_SYSVIEW_EVT_TASK_START_EXEC_ENABLE				CONFIG_APPTRACE_SV_EVT_TASK_START_EXEC_ENABLE
+CONFIG_SYSVIEW_EVT_TASK_STOP_EXEC_ENABLE				CONFIG_APPTRACE_SV_EVT_TASK_STOP_EXEC_ENABLE
+CONFIG_SYSVIEW_EVT_TASK_START_READY_ENABLE				CONFIG_APPTRACE_SV_EVT_TASK_START_READY_ENABLE
+CONFIG_SYSVIEW_EVT_TASK_STOP_READY_ENABLE				CONFIG_APPTRACE_SV_EVT_TASK_STOP_READY_ENABLE
+CONFIG_SYSVIEW_EVT_TASK_CREATE_ENABLE					CONFIG_APPTRACE_SV_EVT_TASK_CREATE_ENABLE
+CONFIG_SYSVIEW_EVT_TASK_TERMINATE_ENABLE				CONFIG_APPTRACE_SV_EVT_TASK_TERMINATE_ENABLE
+CONFIG_SYSVIEW_EVT_IDLE_ENABLE							CONFIG_APPTRACE_SV_EVT_IDLE_ENABLE
+CONFIG_SYSVIEW_EVT_TIMER_ENTER_ENABLE					CONFIG_APPTRACE_SV_EVT_TIMER_ENTER_ENABLE
+CONFIG_SYSVIEW_EVT_TIMER_EXIT_ENABLE					CONFIG_APPTRACE_SV_EVT_TIMER_EXIT_ENABLE

components/app_trace/sys_view/SEGGER/SEGGER_RTT.h

@@ -159,8 +159,8 @@ unsigned     SEGGER_RTT_WriteNoLock             (unsigned BufferIndex, const voi
 unsigned     SEGGER_RTT_WriteSkipNoLock         (unsigned BufferIndex, const void* pBuffer, unsigned NumBytes);
 unsigned     SEGGER_RTT_WriteString             (unsigned BufferIndex, const char* s);
 void         SEGGER_RTT_WriteWithOverwriteNoLock(unsigned BufferIndex, const void* pBuffer, unsigned NumBytes);
-void         SEGGER_RTT_ESP32_FlushNoLock       (unsigned long min_sz, unsigned long tmo);
-void         SEGGER_RTT_ESP32_Flush             (unsigned long min_sz, unsigned long tmo);
+void         SEGGER_RTT_ESP_FlushNoLock       (unsigned long min_sz, unsigned long tmo);
+void         SEGGER_RTT_ESP_Flush             (unsigned long min_sz, unsigned long tmo);
 //
 // Function macro for performance optimization
 //

components/app_trace/sys_view/Sample/Config/SEGGER_SYSVIEW_Config_FreeRTOS.c

@@ -73,6 +73,8 @@ Revision: $Rev: 3734 $
 #include "esp32/clk.h"
 #elif CONFIG_IDF_TARGET_ESP32S2
 #include "esp32s2/clk.h"
+#elif CONFIG_IDF_TARGET_ESP32C3
+#include "esp32c3/clk.h"
 #endif
 
 
@@ -89,11 +91,17 @@ extern const SEGGER_SYSVIEW_OS_API SYSVIEW_X_OS_TraceAPI;
 
 // The target device name
 #define SYSVIEW_DEVICE_NAME     CONFIG_IDF_TARGET
+// The target core name
+#if CONFIG_IDF_TARGET_ARCH_XTENSA
+#define SYSVIEW_CORE_NAME       "xtensa"
+#elif CONFIG_IDF_TARGET_ARCH_RISCV
+#define SYSVIEW_CORE_NAME       "riscv"
+#endif
 
 // Determine which timer to use as timestamp source
-#if CONFIG_SYSVIEW_TS_SOURCE_CCOUNT
+#if CONFIG_APPTRACE_SV_TS_SOURCE_CCOUNT
 #define TS_USE_CCOUNT 1
-#elif CONFIG_SYSVIEW_TS_SOURCE_ESP_TIMER
+#elif CONFIG_APPTRACE_SV_TS_SOURCE_ESP_TIMER
 #define TS_USE_ESP_TIMER 1
 #else
 #define TS_USE_TIMERGROUP 1
@@ -109,13 +117,13 @@ extern const SEGGER_SYSVIEW_OS_API SYSVIEW_X_OS_TraceAPI;
 #define SYSVIEW_TIMESTAMP_FREQ  (esp_clk_apb_freq() / SYSVIEW_TIMER_DIV)
 
 // Timer ID and group ID
-#if defined(CONFIG_SYSVIEW_TS_SOURCE_TIMER_00) || defined(CONFIG_SYSVIEW_TS_SOURCE_TIMER_01)
+#if defined(CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_00) || defined(CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_01)
 #define TS_TIMER_ID 0
 #else
 #define TS_TIMER_ID 1
 #endif // TIMER_00 || TIMER_01
 
-#if defined(CONFIG_SYSVIEW_TS_SOURCE_TIMER_00) || defined(CONFIG_SYSVIEW_TS_SOURCE_TIMER_10)
+#if defined(CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_00) || defined(CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_10)
 #define TS_TIMER_GROUP 0
 #else
 #define TS_TIMER_GROUP 1
@@ -143,12 +151,16 @@ extern const SEGGER_SYSVIEW_OS_API SYSVIEW_X_OS_TraceAPI;
 // The lowest RAM address used for IDs (pointers)
 #define SYSVIEW_RAM_BASE        (SOC_DROM_LOW)
 
+#if CONFIG_IDF_TARGET_ESP32 || CONFIG_IDF_TARGET_ESP32S2
 #if CONFIG_FREERTOS_CORETIMER_0
     #define SYSTICK_INTR_ID (ETS_INTERNAL_TIMER0_INTR_SOURCE+ETS_INTERNAL_INTR_SOURCE_OFF)
 #endif
 #if CONFIG_FREERTOS_CORETIMER_1
     #define SYSTICK_INTR_ID (ETS_INTERNAL_TIMER1_INTR_SOURCE+ETS_INTERNAL_INTR_SOURCE_OFF)
 #endif
+#elif CONFIG_IDF_TARGET_ESP32C3
+    #define SYSTICK_INTR_ID (ETS_SYSTIMER_TARGET0_EDGE_INTR_SOURCE+ETS_INTERNAL_INTR_SOURCE_OFF)
+#endif
 
 // SystemView is single core specific: it implies that SEGGER_SYSVIEW_LOCK()
 // disables IRQs (disables rescheduling globally). So we can not use finite timeouts for locks and return error
@@ -167,11 +179,13 @@ static esp_apptrace_lock_t s_sys_view_lock = {.mux = portMUX_INITIALIZER_UNLOCKE
 */
 static void _cbSendSystemDesc(void) {
     char irq_str[32];
-    SEGGER_SYSVIEW_SendSysDesc("N="SYSVIEW_APP_NAME",D="SYSVIEW_DEVICE_NAME",C=Xtensa,O=FreeRTOS");
+    SEGGER_SYSVIEW_SendSysDesc("N="SYSVIEW_APP_NAME",D="SYSVIEW_DEVICE_NAME",C="SYSVIEW_CORE_NAME",O=FreeRTOS");
     snprintf(irq_str, sizeof(irq_str), "I#%d=SysTick", SYSTICK_INTR_ID);
     SEGGER_SYSVIEW_SendSysDesc(irq_str);
     size_t isr_count = sizeof(esp_isr_names)/sizeof(esp_isr_names[0]);
     for (size_t i = 0; i < isr_count; ++i) {
+        if (esp_isr_names[i] == NULL || (ETS_INTERNAL_INTR_SOURCE_OFF + i) == SYSTICK_INTR_ID)
+            continue;
         snprintf(irq_str, sizeof(irq_str), "I#%d=%s", ETS_INTERNAL_INTR_SOURCE_OFF + i, esp_isr_names[i]);
         SEGGER_SYSVIEW_SendSysDesc(irq_str);
     }
@@ -213,43 +227,43 @@ void SEGGER_SYSVIEW_Conf(void) {
                         &SYSVIEW_X_OS_TraceAPI, _cbSendSystemDesc);
     SEGGER_SYSVIEW_SetRAMBase(SYSVIEW_RAM_BASE);
 
-#if !CONFIG_SYSVIEW_EVT_OVERFLOW_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_OVERFLOW_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_OVERFLOW;
 #endif
-#if !CONFIG_SYSVIEW_EVT_ISR_ENTER_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_ISR_ENTER_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_ISR_ENTER;
 #endif
-#if !CONFIG_SYSVIEW_EVT_ISR_EXIT_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_ISR_EXIT_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_ISR_EXIT;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TASK_START_EXEC_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TASK_START_EXEC_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TASK_START_EXEC;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TASK_STOP_EXEC_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TASK_STOP_EXEC_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TASK_STOP_EXEC;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TASK_START_READY_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TASK_START_READY_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TASK_START_READY;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TASK_STOP_READY_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TASK_STOP_READY_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TASK_STOP_READY;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TASK_CREATE_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TASK_CREATE_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TASK_CREATE;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TASK_TERMINATE_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TASK_TERMINATE_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TASK_TERMINATE;
 #endif
-#if !CONFIG_SYSVIEW_EVT_IDLE_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_IDLE_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_IDLE;
 #endif
-#if !CONFIG_SYSVIEW_EVT_ISR_TO_SCHEDULER_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_ISR_TO_SCHED_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_ISR_TO_SCHEDULER;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TIMER_ENTER_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TIMER_ENTER_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TIMER_ENTER;
 #endif
-#if !CONFIG_SYSVIEW_EVT_TIMER_EXIT_ENABLE
+#if !CONFIG_APPTRACE_SV_EVT_TIMER_EXIT_ENABLE
     disable_evts |= SYSVIEW_EVTMASK_TIMER_EXIT;
 #endif
   SEGGER_SYSVIEW_DisableEvents(disable_evts);

components/app_trace/sys_view/Sample/OS/SEGGER_SYSVIEW_FreeRTOS.h

@@ -80,7 +80,7 @@ Notes:
   #define portSTACK_GROWTH              ( -1 )
 #endif
 
-#define SYSVIEW_FREERTOS_MAX_NOF_TASKS  CONFIG_SYSVIEW_MAX_TASKS
+#define SYSVIEW_FREERTOS_MAX_NOF_TASKS  CONFIG_APPTRACE_SV_MAX_TASKS
 
 /*********************************************************************
 *
@@ -290,12 +290,12 @@ Notes:
   #define traceTASK_SWITCHED_IN()                   if(prvGetTCBFromHandle(NULL) == xTaskGetIdleTaskHandle()) {           \
                                                       SEGGER_SYSVIEW_OnIdle();                                          \
                                                     } else {                                                            \
-                                                      SEGGER_SYSVIEW_OnTaskStartExec((U32)pxCurrentTCB[xPortGetCoreID()]); \
+                                                      SEGGER_SYSVIEW_OnTaskStartExec((U32)pxCurrentTCB[cpu_hal_get_core_id()]); \
                                                     }
 #else
   #define traceTASK_SWITCHED_IN()                   {                                                                   \
-                                                      if (memcmp(pxCurrentTCB[xPortGetCoreID()]->pcTaskName, "IDLE", 5) != 0) { \
-                                                        SEGGER_SYSVIEW_OnTaskStartExec((U32)pxCurrentTCB[xPortGetCoreID()]);    \
+                                                      if (memcmp(pxCurrentTCB[cpu_hal_get_core_id()]->pcTaskName, "IDLE", 5) != 0) { \
+                                                        SEGGER_SYSVIEW_OnTaskStartExec((U32)pxCurrentTCB[cpu_hal_get_core_id()]);    \
                                                       } else {                                                          \
                                                         SEGGER_SYSVIEW_OnIdle();                                        \
                                                       }                                                                 \
@@ -305,14 +305,15 @@ Notes:
 #define traceMOVED_TASK_TO_READY_STATE(pxTCB)       SEGGER_SYSVIEW_OnTaskStartReady((U32)pxTCB)
 #define traceREADDED_TASK_TO_READY_STATE(pxTCB)
 
-#define traceMOVED_TASK_TO_DELAYED_LIST()           SEGGER_SYSVIEW_OnTaskStopReady((U32)pxCurrentTCB[xPortGetCoreID()],  (1u << 2))
-#define traceMOVED_TASK_TO_OVERFLOW_DELAYED_LIST()  SEGGER_SYSVIEW_OnTaskStopReady((U32)pxCurrentTCB[xPortGetCoreID()],  (1u << 2))
+#define traceMOVED_TASK_TO_DELAYED_LIST()           SEGGER_SYSVIEW_OnTaskStopReady((U32)pxCurrentTCB[cpu_hal_get_core_id()],  (1u << 2))
+#define traceMOVED_TASK_TO_OVERFLOW_DELAYED_LIST()  SEGGER_SYSVIEW_OnTaskStopReady((U32)pxCurrentTCB[cpu_hal_get_core_id()],  (1u << 2))
 #define traceMOVED_TASK_TO_SUSPENDED_LIST(pxTCB)    SEGGER_SYSVIEW_OnTaskStopReady((U32)pxTCB,         ((3u << 3) | 3))
 
 #define traceISR_EXIT_TO_SCHEDULER()                SEGGER_SYSVIEW_RecordExitISRToScheduler()
 #define traceISR_EXIT()                             SEGGER_SYSVIEW_RecordExitISR()
 #define traceISR_ENTER(_n_)                         SEGGER_SYSVIEW_RecordEnterISR(_n_)
 
+
 /*********************************************************************
 *
 *       API functions

components/app_trace/sys_view/esp32/SEGGER_RTT_esp32.c → components/app_trace/sys_view/esp/SEGGER_RTT_esp.c

@@ -20,10 +20,10 @@ const static char *TAG = "segger_rtt";
 // size of down channel data buf
 #define SYSVIEW_DOWN_BUF_SIZE   32
 #define SEGGER_STOP_WAIT_TMO    1000000 //us
-#if CONFIG_SYSVIEW_BUF_WAIT_TMO == -1
+#if CONFIG_APPTRACE_SV_BUF_WAIT_TMO == -1
 #define SEGGER_HOST_WAIT_TMO    ESP_APPTRACE_TMO_INFINITE
 #else
-#define SEGGER_HOST_WAIT_TMO    CONFIG_SYSVIEW_BUF_WAIT_TMO
+#define SEGGER_HOST_WAIT_TMO    CONFIG_APPTRACE_SV_BUF_WAIT_TMO
 #endif
 
 static uint8_t s_events_buf[SYSVIEW_EVENTS_BUF_SZ];
@@ -39,7 +39,7 @@ static uint8_t s_down_buf[SYSVIEW_DOWN_BUF_SIZE];
 
 /*********************************************************************
 *
-*       SEGGER_RTT_ESP32_FlushNoLock()
+*       SEGGER_RTT_ESP_FlushNoLock()
 *
 *  Function description
 *    Flushes buffered events.
@@ -51,7 +51,7 @@ static uint8_t s_down_buf[SYSVIEW_DOWN_BUF_SIZE];
 *  Return value
 *    None.
 */
-void SEGGER_RTT_ESP32_FlushNoLock(unsigned long min_sz, unsigned long tmo)
+void SEGGER_RTT_ESP_FlushNoLock(unsigned long min_sz, unsigned long tmo)
 {
     esp_err_t res;
     if (s_events_buf_filled > 0) {
@@ -70,7 +70,7 @@ void SEGGER_RTT_ESP32_FlushNoLock(unsigned long min_sz, unsigned long tmo)
 
 /*********************************************************************
 *
-*       SEGGER_RTT_ESP32_Flush()
+*       SEGGER_RTT_ESP_Flush()
 *
 *  Function description
 *    Flushes buffered events.
@@ -82,10 +82,10 @@ void SEGGER_RTT_ESP32_FlushNoLock(unsigned long min_sz, unsigned long tmo)
 *  Return value
 *    None.
 */
-void SEGGER_RTT_ESP32_Flush(unsigned long min_sz, unsigned long tmo)
+void SEGGER_RTT_ESP_Flush(unsigned long min_sz, unsigned long tmo)
 {
     SEGGER_SYSVIEW_LOCK();
-    SEGGER_RTT_ESP32_FlushNoLock(min_sz, tmo);
+    SEGGER_RTT_ESP_FlushNoLock(min_sz, tmo);
     SEGGER_SYSVIEW_UNLOCK();
 }
 
@@ -147,7 +147,7 @@ unsigned SEGGER_RTT_WriteSkipNoLock(unsigned BufferIndex, const void* pBuffer, u
       ESP_LOGE(TAG, "Too large event %u bytes!", NumBytes);
       return 0;
   }
-  if (xPortGetCoreID()) { // dual core specific code
+  if (cpu_hal_get_core_id()) { // dual core specific code
     // use the highest - 1 bit of event ID to indicate core ID
     // the highest bit can not be used due to event ID encoding method
     // this reduces supported ID range to [0..63] (for 1 byte IDs) plus [128..16383] (for 2 bytes IDs)
@@ -167,7 +167,7 @@ unsigned SEGGER_RTT_WriteSkipNoLock(unsigned BufferIndex, const void* pBuffer, u
   memcpy(&s_events_buf[s_events_buf_filled], pBuffer, NumBytes);
   s_events_buf_filled += NumBytes;
   if (event_id == SYSVIEW_EVTID_TRACE_STOP) {
-    SEGGER_RTT_ESP32_FlushNoLock(0, SEGGER_STOP_WAIT_TMO);
+    SEGGER_RTT_ESP_FlushNoLock(0, SEGGER_STOP_WAIT_TMO);
   }
   return NumBytes;
 }

components/app_trace/sys_view/ext/heap_trace_module.c

@@ -52,7 +52,7 @@ esp_err_t esp_sysview_heap_trace_start(uint32_t tmo)
 esp_err_t esp_sysview_heap_trace_stop(void)
 {
     ESP_EARLY_LOGV(TAG, "%s", __func__);
-    SEGGER_RTT_ESP32_Flush(0, ESP_APPTRACE_TMO_INFINITE);
+    SEGGER_RTT_ESP_Flush(0, ESP_APPTRACE_TMO_INFINITE);
     return ESP_OK;
 }
 

components/app_trace/test/test_trace.c

@@ -88,7 +88,7 @@ static void esp_apptrace_test_timer_init(int timer_group, int timer_idx, uint32_
     timer_enable_intr(timer_group, timer_idx);
 }
 
-#if CONFIG_SYSVIEW_ENABLE == 0
+#if CONFIG_APPTRACE_SV_ENABLE == 0
 #define ESP_APPTRACE_TEST_WRITE(_b_, _s_)            esp_apptrace_write(ESP_APPTRACE_DEST_TRAX, _b_, _s_, ESP_APPTRACE_TMO_INFINITE)
 #define ESP_APPTRACE_TEST_WRITE_FROM_ISR(_b_, _s_)   esp_apptrace_write(ESP_APPTRACE_DEST_TRAX, _b_, _s_, 0UL)
 #define ESP_APPTRACE_TEST_WRITE_NOWAIT(_b_, _s_)     esp_apptrace_write(ESP_APPTRACE_DEST_TRAX, _b_, _s_, 0)
@@ -214,7 +214,7 @@ static void esp_apptrace_dummy_task(void *p)
 
     i = 0;
     while (!arg->stop) {
-        ESP_APPTRACE_TEST_LOGD("%x: dummy task work %d.%d", xTaskGetCurrentTaskHandle(), xPortGetCoreID(), i++);
+        ESP_APPTRACE_TEST_LOGD("%x: dummy task work %d.%d", xTaskGetCurrentTaskHandle(), cpu_hal_get_core_id(), i++);
         if (tmo_ticks) {
             vTaskDelay(tmo_ticks);
         }
@@ -259,7 +259,7 @@ static void esp_apptrace_test_task(void *p)
                 ESP_APPTRACE_TEST_LOGE("Failed to timer_isr_register (%d)!", res);
                 goto on_fail;
             }
-            *(uint32_t *)arg->timers[i].data.buf = ((uint32_t)inth[i]) | (1 << 31) | (xPortGetCoreID() ? 0x1 : 0);
+            *(uint32_t *)arg->timers[i].data.buf = ((uint32_t)inth[i]) | (1 << 31) | (cpu_hal_get_core_id() ? 0x1 : 0);
             ESP_APPTRACE_TEST_LOGI("%x: start timer %x period %u us", xTaskGetCurrentTaskHandle(), inth[i], arg->timers[i].data.period);
             res = timer_start(arg->timers[i].group, arg->timers[i].id);
             if (res != ESP_OK) {
@@ -269,7 +269,7 @@ static void esp_apptrace_test_task(void *p)
         }
     }
 
-    *(uint32_t *)arg->data.buf = (uint32_t)xTaskGetCurrentTaskHandle() | (xPortGetCoreID() ? 0x1 : 0);
+    *(uint32_t *)arg->data.buf = (uint32_t)xTaskGetCurrentTaskHandle() | (cpu_hal_get_core_id() ? 0x1 : 0);
     arg->data.wr_cnt = 0;
     arg->data.wr_err = 0;
     while (!arg->stop) {
@@ -749,7 +749,7 @@ static void esp_logtrace_task(void *p)
         ESP_LOGI(TAG, "%p: sample print 4 %c", xTaskGetCurrentTaskHandle(), ((i & 0xFF) % 95) + 32);
         ESP_LOGI(TAG, "%p: sample print 5 %f", xTaskGetCurrentTaskHandle(), 1.0);
         ESP_LOGI(TAG, "%p: sample print 6 %f", xTaskGetCurrentTaskHandle(), 3.45);
-        ESP_LOGI(TAG, "%p: logtrace task work %d.%d", xTaskGetCurrentTaskHandle(), xPortGetCoreID(), i);
+        ESP_LOGI(TAG, "%p: logtrace task work %d.%d", xTaskGetCurrentTaskHandle(), cpu_hal_get_core_id(), i);
         if (++i == 10000) {
             break;
         }

components/esp32c3/memprot.c

@@ -28,6 +28,8 @@
 #include "esp32c3/rom/ets_sys.h"
 #include "esp_log.h"
 
+#include "soc/cpu.h"
+
 extern int _iram_text_end;
 static const char *TAG = "memprot";
 
@@ -532,6 +534,11 @@ void esp_memprot_set_prot_int(bool invoke_panic_handler, bool lock_feature, void
         esp_memprot_set_monitor_en(MEMPROT_DRAM0_SRAM, false);
     }
 
+    // do not enable if being debugged
+    if (esp_cpu_in_ocd_debug_mode()) {
+        return;
+    }
+
     //panic handling
     if (invoke_panic_handler) {
         if (use_iram0) {

components/esp_hw_support/intr_alloc.c

@@ -107,10 +107,6 @@ static uint32_t non_iram_int_mask[SOC_CPU_CORES_NUM];
 static uint32_t non_iram_int_disabled[SOC_CPU_CORES_NUM];
 static bool non_iram_int_disabled_flag[SOC_CPU_CORES_NUM];
 
-#if CONFIG_SYSVIEW_ENABLE
-extern uint32_t port_switch_flag[];
-#endif
-
 static portMUX_TYPE spinlock = portMUX_INITIALIZER_UNLOCKED;
 
 //Inserts an item into vector_desc list so that the list is sorted
@@ -415,16 +411,12 @@ static void IRAM_ATTR shared_intr_isr(void *arg)
     while(sh_vec) {
         if (!sh_vec->disabled) {
             if ((sh_vec->statusreg == NULL) || (*sh_vec->statusreg & sh_vec->statusmask)) {
-#if CONFIG_SYSVIEW_ENABLE
                 traceISR_ENTER(sh_vec->source+ETS_INTERNAL_INTR_SOURCE_OFF);
-#endif
                 sh_vec->isr(sh_vec->arg);
-#if CONFIG_SYSVIEW_ENABLE
                 // check if we will return to scheduler or to interrupted task after ISR
-                if (!port_switch_flag[cpu_hal_get_core_id()]) {
+                if (!os_task_switch_is_pended(cpu_hal_get_core_id())) {
                     traceISR_EXIT();
                 }
-#endif
             }
         }
         sh_vec=sh_vec->next;
@@ -432,18 +424,18 @@ static void IRAM_ATTR shared_intr_isr(void *arg)
     portEXIT_CRITICAL_ISR(&spinlock);
 }
 
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
 //Common non-shared isr handler wrapper.
 static void IRAM_ATTR non_shared_intr_isr(void *arg)
 {
     non_shared_isr_arg_t *ns_isr_arg=(non_shared_isr_arg_t*)arg;
     portENTER_CRITICAL_ISR(&spinlock);
     traceISR_ENTER(ns_isr_arg->source+ETS_INTERNAL_INTR_SOURCE_OFF);
-    // FIXME: can we call ISR and check port_switch_flag after releasing spinlock?
-    // when CONFIG_SYSVIEW_ENABLE = 0 ISRs for non-shared IRQs are called without spinlock
+    // FIXME: can we call ISR and check os_task_switch_is_pended() after releasing spinlock?
+    // when CONFIG_APPTRACE_SV_ENABLE = 0 ISRs for non-shared IRQs are called without spinlock
     ns_isr_arg->isr(ns_isr_arg->isr_arg);
     // check if we will return to scheduler or to interrupted task after ISR
-    if (!port_switch_flag[cpu_hal_get_core_id()]) {
+    if (!os_task_switch_is_pended(cpu_hal_get_core_id())) {
         traceISR_EXIT();
     }
     portEXIT_CRITICAL_ISR(&spinlock);
@@ -539,7 +531,7 @@ esp_err_t esp_intr_alloc_intrstatus(int source, int flags, uint32_t intrstatusre
         //Mark as unusable for other interrupt sources. This is ours now!
         vd->flags=VECDESC_FL_NONSHARED;
         if (handler) {
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
             non_shared_isr_arg_t *ns_isr_arg=malloc(sizeof(non_shared_isr_arg_t));
             if (!ns_isr_arg) {
                 portEXIT_CRITICAL(&spinlock);
@@ -688,7 +680,7 @@ esp_err_t esp_intr_free(intr_handle_t handle)
 
     if ((handle->vector_desc->flags&VECDESC_FL_NONSHARED) || free_shared_vector) {
         ESP_EARLY_LOGV(TAG, "esp_intr_free: Disabling int, killing handler");
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
         if (!free_shared_vector) {
             void *isr_arg = interrupt_controller_hal_get_int_handler_arg(handle->vector_desc->intno);
             if (isr_arg) {

components/esp_system/panic.c

@@ -40,7 +40,7 @@
 
 #if CONFIG_APPTRACE_ENABLE
 #include "esp_app_trace.h"
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
 #include "SEGGER_RTT.h"
 #endif
 
@@ -266,8 +266,8 @@ void esp_panic_handler(panic_info_t *info)
         panic_print_str(" and returning...\r\n");
         disable_all_wdts();
 #if CONFIG_APPTRACE_ENABLE
-#if CONFIG_SYSVIEW_ENABLE
-        SEGGER_RTT_ESP32_FlushNoLock(CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH, APPTRACE_ONPANIC_HOST_FLUSH_TMO);
+#if CONFIG_APPTRACE_SV_ENABLE
+        SEGGER_RTT_ESP_FlushNoLock(CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH, APPTRACE_ONPANIC_HOST_FLUSH_TMO);
 #else
         esp_apptrace_flush_nolock(ESP_APPTRACE_DEST_TRAX, CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH,
                                   APPTRACE_ONPANIC_HOST_FLUSH_TMO);
@@ -307,8 +307,8 @@ void esp_panic_handler(panic_info_t *info)
 
 #if CONFIG_APPTRACE_ENABLE
     disable_all_wdts();
-#if CONFIG_SYSVIEW_ENABLE
-    SEGGER_RTT_ESP32_FlushNoLock(CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH, APPTRACE_ONPANIC_HOST_FLUSH_TMO);
+#if CONFIG_APPTRACE_SV_ENABLE
+    SEGGER_RTT_ESP_FlushNoLock(CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH, APPTRACE_ONPANIC_HOST_FLUSH_TMO);
 #else
     esp_apptrace_flush_nolock(ESP_APPTRACE_DEST_TRAX, CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH,
                               APPTRACE_ONPANIC_HOST_FLUSH_TMO);
@@ -379,8 +379,8 @@ void __attribute__((noreturn, no_sanitize_undefined)) panic_abort(const char *de
     s_panic_abort_details = (char *) details;
 
 #if CONFIG_APPTRACE_ENABLE
-#if CONFIG_SYSVIEW_ENABLE
-    SEGGER_RTT_ESP32_FlushNoLock(CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH, APPTRACE_ONPANIC_HOST_FLUSH_TMO);
+#if CONFIG_APPTRACE_SV_ENABLE
+    SEGGER_RTT_ESP_FlushNoLock(CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH, APPTRACE_ONPANIC_HOST_FLUSH_TMO);
 #else
     esp_apptrace_flush_nolock(ESP_APPTRACE_DEST_TRAX, CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH,
                               APPTRACE_ONPANIC_HOST_FLUSH_TMO);

components/esp_system/port/cpu_start.c

@@ -272,6 +272,13 @@ void IRAM_ATTR call_start_cpu0(void)
 #endif
 
 #ifdef __riscv
+    if (cpu_hal_is_debugger_attached()) {
+        /* Let debugger some time to detect that target started, halt it, enable ebreaks and resume.
+           500ms should be enough. */
+        for (uint32_t ms_num = 0; ms_num < 2; ms_num++) {
+            esp_rom_delay_us(100000);
+        }
+    }
     // Configure the global pointer register
     // (This should be the first thing IDF app does, as any other piece of code could be
     // relaxed by the linker to access something relative to __global_pointer$)

components/esp_system/startup.c

@@ -297,19 +297,6 @@ static void do_core_init(void)
 
     esp_err_t err __attribute__((unused));
 
-    // [refactor-todo] move this to secondary init
-#if CONFIG_APPTRACE_ENABLE
-    err = esp_apptrace_init();
-    assert(err == ESP_OK && "Failed to init apptrace module on PRO CPU!");
-#endif
-#if CONFIG_SYSVIEW_ENABLE
-    SEGGER_SYSVIEW_Conf();
-#endif
-
-#if CONFIG_ESP_DEBUG_STUBS_ENABLE
-    esp_dbg_stubs_init();
-#endif
-
     err = esp_pthread_init();
     assert(err == ESP_OK && "Failed to init pthread module!");
 
@@ -450,6 +437,18 @@ IRAM_ATTR ESP_SYSTEM_INIT_FN(init_components0, BIT(0))
     esp_sleep_enable_gpio_switch(true);
 #endif
 
+#if CONFIG_APPTRACE_ENABLE
+    esp_err_t err = esp_apptrace_init();
+    assert(err == ESP_OK && "Failed to init apptrace module on PRO CPU!");
+#endif
+#if CONFIG_APPTRACE_SV_ENABLE
+    SEGGER_SYSVIEW_Conf();
+#endif
+
+#if CONFIG_ESP_DEBUG_STUBS_ENABLE
+    esp_dbg_stubs_init();
+#endif
+
 #if defined(CONFIG_PM_ENABLE)
     esp_pm_impl_init();
 #endif

components/freertos/CMakeLists.txt

@@ -42,7 +42,7 @@ else()  # RISC-V
         port/riscv
         .)
 
-    set(required_components esp_timer)
+    set(required_components app_trace esp_timer)
 endif()
 
 list(APPEND srcs
@@ -61,7 +61,7 @@ if(CONFIG_ESP32_IRAM_AS_8BIT_ACCESSIBLE_MEMORY)
     list(APPEND srcs "port/xtensa/xtensa_loadstore_handler.S")
 endif()
 
-# app_trace is required by FreeRTOS headers only when CONFIG_SYSVIEW_ENABLE=y,
+# app_trace is required by FreeRTOS headers only when CONFIG_APPTRACE_SV_ENABLE=y,
 # but requirements can't depend on config options, so always require it.
 idf_component_register(SRCS "${srcs}"
                     INCLUDE_DIRS ${include_dirs}

components/freertos/port/riscv/include/freertos/FreeRTOSConfig.h

@@ -91,4 +91,13 @@
 #define configISR_STACK_SIZE                            (CONFIG_FREERTOS_ISR_STACKSIZE)
 #endif
 
+#ifndef __ASSEMBLER__
+#if CONFIG_APPTRACE_SV_ENABLE
+extern int xPortSwitchFlag;
+#define os_task_switch_is_pended(_cpu_) (xPortSwitchFlag)
+#else
+#define os_task_switch_is_pended(_cpu_) (false)
+#endif
+#endif
+
 #endif // FREERTOS_CONFIG_RISCV_H

components/freertos/port/riscv/port.c

@@ -346,6 +346,7 @@ void vPortYieldOtherCore(BaseType_t coreid)
 
 void vPortYieldFromISR( void )
 {
+    traceISR_EXIT_TO_SCHEDULER();
     uxSchedulerRunning = 1;
     xPortSwitchFlag = 1;
 }

components/freertos/port/xtensa/include/freertos/FreeRTOSConfig.h

@@ -154,4 +154,13 @@ int xt_clock_freq(void) __attribute__((deprecated));
 #define configISR_STACK_SIZE                            ((CONFIG_FREERTOS_ISR_STACKSIZE + configSTACK_ALIGNMENT - 1) & (~(configSTACK_ALIGNMENT - 1)))
 #endif
 
+#ifndef __ASSEMBLER__
+#if CONFIG_APPTRACE_SV_ENABLE
+extern uint32_t port_switch_flag[];
+#define os_task_switch_is_pended(_cpu_) (port_switch_flag[_cpu_])
+#else
+#define os_task_switch_is_pended(_cpu_) (false)
+#endif
+#endif
+
 #endif // FREERTOS_CONFIG_XTENSA_H

components/hal/esp32c3/include/hal/cpu_ll.h

@@ -16,6 +16,8 @@
 #include <stdint.h>
 
 #include "soc/soc_caps.h"
+#include "soc/dport_access.h"
+#include "soc/system_reg.h"
 #include "esp_bit_defs.h"
 #include "soc/assist_debug_reg.h"
 #include "esp_attr.h"
@@ -153,6 +155,11 @@ static inline void cpu_ll_set_vecbase(const void* vecbase)
 
 static inline void cpu_ll_waiti(void)
 {
+    if (cpu_ll_is_debugger_attached() && DPORT_REG_GET_BIT(SYSTEM_CPU_PER_CONF_REG, SYSTEM_CPU_WAIT_MODE_FORCE_ON) == 0) {
+        /* when SYSTEM_CPU_WAIT_MODE_FORCE_ON is disabled in WFI mode SBA access to memory does not work for debugger,
+           so do not enter that mode when debugger is connected */
+        return;
+    }
     asm volatile ("wfi\n");
 }
 

components/soc/esp32c3/interrupts.c

@@ -14,9 +14,9 @@
 
 #include "soc/interrupts.h"
 
-const char *const esp_isr_names[ETS_MAX_INTR_SOURCE] = {
+const char *const esp_isr_names[] = {
     [0] = "WIFI_MAC",
-    [1] = "WIFI_NMI",
+    [1] = "WIFI_MAC_NMI",
     [2] = "WIFI_PWR",
     [3] = "WIFI_BB",
     [4] = "BT_MAC",
@@ -26,7 +26,7 @@ const char *const esp_isr_names[ETS_MAX_INTR_SOURCE] = {
     [8] = "RWBLE",
     [9] = "RWBT_NMI",
     [10] = "RWBLE_NMI",
-    [11] = "I2C",
+    [11] = "I2C_MASTER",
     [12] = "SLC0",
     [13] = "SLC1",
     [14] = "APB_CTRL",
@@ -40,7 +40,7 @@ const char *const esp_isr_names[ETS_MAX_INTR_SOURCE] = {
     [22] = "UART1",
     [23] = "LEDC",
     [24] = "EFUSE",
-    [25] = "CAN",
+    [25] = "TWAI",
     [26] = "USB",
     [27] = "RTC_CORE",
     [28] = "RMT",
@@ -65,16 +65,16 @@ const char *const esp_isr_names[ETS_MAX_INTR_SOURCE] = {
     [47] = "RSA",
     [48] = "AES",
     [49] = "SHA",
-    [50] = "ETS_FROM_CPU_INTR0",
-    [51] = "ETS_FROM_CPU_INTR1",
-    [52] = "ETS_FROM_CPU_INTR2",
-    [53] = "ETS_FROM_CPU_INTR3",
-    [54] = "ETS_ASSIST_DEBUG",
-    [55] = "ETS_DMA_APBPERI_PMS",
-    [56] = "ETS_CORE0_IRAM0_PMS",
-    [57] = "ETS_CORE0_DRAM0_PMS",
-    [58] = "ETS_CORE0_PIF_PMS",
-    [59] = "ETS_CORE0_PIF_PMS_SIZE",
-    [60] = "ETS_BAK_PMS_VIOLATE",
-    [61] = "ETS_CACHE_CORE0_ACS",
+    [50] = "FROM_CPU_INTR0",
+    [51] = "FROM_CPU_INTR1",
+    [52] = "FROM_CPU_INTR2",
+    [53] = "FROM_CPU_INTR3",
+    [54] = "ASSIST_DEBUG",
+    [55] = "DMA_APBPERI_PMS",
+    [56] = "CORE0_IRAM0_PMS",
+    [57] = "CORE0_DRAM0_PMS",
+    [58] = "CORE0_PIF_PMS",
+    [59] = "CORE0_PIF_PMS_SIZE",
+    [60] = "BAK_PMS_VIOLATE",
+    [61] = "CACHE_CORE0_ACS",
 };

components/xtensa/xtensa_intr.c

@@ -143,7 +143,7 @@ xt_handler xt_set_interrupt_handler(int n, xt_handler f, void * arg)
     return ((old == &xt_unhandled_interrupt) ? 0 : old);
 }
 
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
 void * xt_get_interrupt_handler_arg(int n)
 {
     xt_handler_table_entry * entry;

docs/en/api-guides/app_trace.rst

@@ -326,10 +326,10 @@ Another useful IDF feature built on top of application tracing library is the sy
 How To Use It
 """""""""""""
 
-Support for this feature is enabled by *Component config > Application Level Tracing > FreeRTOS SystemView Tracing* (:ref:`CONFIG_SYSVIEW_ENABLE`) menuconfig option. There are several other options enabled under the same menu:
+Support for this feature is enabled by *Component config > Application Level Tracing > FreeRTOS SystemView Tracing* (:ref:`CONFIG_APPTRACE_SV_ENABLE`) menuconfig option. There are several other options enabled under the same menu:
 
-1. {IDF_TARGET_NAME} timer to use as SystemView timestamp source: (:ref:`CONFIG_SYSVIEW_TS_SOURCE`) selects the source of timestamps for SystemView events. In single core mode timestamps are generated using {IDF_TARGET_NAME} internal cycle counter running at maximum 240 Mhz (~4 ns granularity). In dual-core mode external timer working at 40 Mhz is used, so timestamp granularity is 25 ns.
-2. Individually enabled or disabled collection of SystemView events (``CONFIG_SYSVIEW_EVT_XXX``):
+1. {IDF_TARGET_NAME} timer to use as SystemView timestamp source: (:ref:`CONFIG_APPTRACE_SV_TS_SOURCE`) selects the source of timestamps for SystemView events. In single core mode timestamps are generated using {IDF_TARGET_NAME} internal cycle counter running at maximum 240 Mhz (~4 ns granularity). In dual-core mode external timer working at 40 Mhz is used, so timestamp granularity is 25 ns.
+2. Individually enabled or disabled collection of SystemView events (``CONFIG_APPTRACE_SV_EVT_XXX``):
 
     - Trace Buffer Overflow Event
     - ISR Enter Event

docs/en/api-reference/system/heap_debug.rst

@@ -273,7 +273,7 @@ Once you've identified the code which you think is leaking:
 
 - In the project configuration menu, navigate to ``Component settings`` -> ``Heap Memory Debugging`` -> :ref:`CONFIG_HEAP_TRACING_DEST` and select ``Host-Based``.
 - In the project configuration menu, navigate to ``Component settings`` -> ``Application Level Tracing`` -> :ref:`CONFIG_APPTRACE_DESTINATION` and select ``Trace memory``.
-- In the project configuration menu, navigate to ``Component settings`` -> ``Application Level Tracing`` -> ``FreeRTOS SystemView Tracing`` and enable :ref:`CONFIG_SYSVIEW_ENABLE`.
+- In the project configuration menu, navigate to ``Component settings`` -> ``Application Level Tracing`` -> ``FreeRTOS SystemView Tracing`` and enable :ref:`CONFIG_APPTRACE_SV_ENABLE`.
 - Call the function :cpp:func:`heap_trace_init_tohost` early in the program, to initialize JTAG heap tracing module.
 - Call the function :cpp:func:`heap_trace_start` to begin recording all mallocs/frees in the system. Call this immediately before the piece of code which you suspect is leaking memory.
   In host-based mode argument to this function is ignored and heap tracing module behaves like ``HEAP_TRACE_ALL`` was passed: all allocations and deallocations are sent to the host.

docs/zh_CN/api-guides/app_trace.rst

@@ -326,10 +326,10 @@ IDF 中另一个基于应用层跟踪库的实用功能是系统级跟踪，它
 如何使用
 """"""""
 
-若需使用这个功能，需要在 menuconfig 中开启 :ref:`CONFIG_SYSVIEW_ENABLE` 选项，具体路径为： *Component config > Application Level Tracing > FreeRTOS SystemView Tracing* 。在同一个菜单栏下还开启了其他几个选项：
+若需使用这个功能，需要在 menuconfig 中开启 :ref:`CONFIG_APPTRACE_SV_ENABLE` 选项，具体路径为： *Component config > Application Level Tracing > FreeRTOS SystemView Tracing* 。在同一个菜单栏下还开启了其他几个选项：
 
-1. *{IDF_TARGET_NAME} timer to use as SystemView timestamp source* （:ref:`CONFIG_SYSVIEW_TS_SOURCE`）选择 SystemView 事件使用的时间戳来源。在单核模式下，使用 {IDF_TARGET_NAME} 内部的循环计数器生成时间戳，其最大的工作频率是 240 MHz（时间戳粒度大约为 4 ns）。在双核模式下，使用工作在 40 MHz 的外部定时器，因此时间戳粒度为 25 ns。
-2. 可以单独启用或禁用的 SystemView 事件集合（``CONFIG_SYSVIEW_EVT_XXX``）：
+1. *{IDF_TARGET_NAME} timer to use as SystemView timestamp source* （:ref:`CONFIG_APPTRACE_SV_TS_SOURCE`）选择 SystemView 事件使用的时间戳来源。在单核模式下，使用 {IDF_TARGET_NAME} 内部的循环计数器生成时间戳，其最大的工作频率是 240 MHz（时间戳粒度大约为 4 ns）。在双核模式下，使用工作在 40 MHz 的外部定时器，因此时间戳粒度为 25 ns。
+2. 可以单独启用或禁用的 SystemView 事件集合（``CONFIG_APPTRACE_SV_EVT_XXX``）：
 
     - Trace Buffer Overflow Event
     - ISR Enter Event

examples/system/app_trace_to_host/sdkconfig.defaults

@@ -1,5 +1,5 @@
 # Enable application tracing by default
-CONFIG_APPTRACE_DEST_TRAX=y
+CONFIG_APPTRACE_DEST_JTAG=y
 CONFIG_APPTRACE_ENABLE=y
 # Disable WiFi stack by default
 CONFIG_WIFI_ENABLED=n

examples/system/gcov/sdkconfig.defaults

@@ -1,4 +1,4 @@
-CONFIG_APPTRACE_DEST_TRAX=y
+CONFIG_APPTRACE_DEST_JTAG=y
 # CONFIG_APPTRACE_DEST_NONE is not set
 CONFIG_APPTRACE_ENABLE=y
 CONFIG_APPTRACE_LOCK_ENABLE=y

examples/system/sysview_tracing/main/sysview_tracing.c

@@ -26,7 +26,7 @@ typedef struct {
 } example_event_data_t;
 
 
-#if CONFIG_SYSVIEW_ENABLE
+#if CONFIG_APPTRACE_SV_ENABLE
 #if !CONFIG_USE_CUSTOM_EVENT_ID
 
 #define SYSVIEW_EXAMPLE_SEND_EVENT_ID     0
@@ -165,13 +165,13 @@ void app_main(void)
         },
 #if CONFIG_FREERTOS_UNICORE == 0
         {
-            .group = TIMER_GROUP_1,
-            .timer = TIMER_1,
+            .group = TIMER_GROUP_0,
+            .timer = TIMER_0,
         },
 #endif
     };
 
-#if CONFIG_SYSVIEW_ENABLE && CONFIG_USE_CUSTOM_EVENT_ID
+#if CONFIG_APPTRACE_SV_ENABLE && CONFIG_USE_CUSTOM_EVENT_ID
     // Currently OpenOCD does not support requesting module info from target. So do the following...
     // Wait untill SystemView module receives START command from host,
     // after that data can be sent to the host using onboard API,
@@ -183,7 +183,7 @@ void app_main(void)
 #endif
 
     example_timer_init(TIMER_GROUP_1, TIMER_0, 2000);
-    example_timer_init(TIMER_GROUP_1, TIMER_1, 4000);
+    example_timer_init(TIMER_GROUP_0, TIMER_0, 4000);
 
     xTaskCreatePinnedToCore(example_task, "svtrace0", 2048, &event_data[0], 3, &event_data[0].thnd, 0);
     ESP_LOGI(TAG, "Created task %p", event_data[0].thnd);

examples/system/sysview_tracing/sdkconfig.defaults

@@ -4,21 +4,21 @@ CONFIG_FREERTOS_UNICORE=y
 # 1ms tick period
 CONFIG_FREERTOS_HZ=1000
 # Enable application tracing by default
-CONFIG_APPTRACE_DEST_TRAX=y
+CONFIG_APPTRACE_DEST_JTAG=y
 CONFIG_APPTRACE_ENABLE=y
 # Enable FreeRTOS SystemView Tracing by default
-CONFIG_SYSVIEW_ENABLE=y
-CONFIG_SYSVIEW_TS_SOURCE_TIMER_00=y
-CONFIG_SYSVIEW_EVT_OVERFLOW_ENABLE=y
-CONFIG_SYSVIEW_EVT_ISR_ENTER_ENABLE=y
-CONFIG_SYSVIEW_EVT_ISR_EXIT_ENABLE=y
-CONFIG_SYSVIEW_EVT_ISR_TO_SCHEDULER_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_START_EXEC_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_STOP_EXEC_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_START_READY_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_STOP_READY_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_CREATE_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_TERMINATE_ENABLE=y
-CONFIG_SYSVIEW_EVT_IDLE_ENABLE=y
-CONFIG_SYSVIEW_EVT_TIMER_ENTER_ENABLE=y
-CONFIG_SYSVIEW_EVT_TIMER_EXIT_ENABLE=y
+CONFIG_APPTRACE_SV_ENABLE=y
+CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_00=y
+CONFIG_APPTRACE_SV_EVT_OVERFLOW_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_ISR_ENTER_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_ISR_EXIT_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_ISR_TO_SCHED_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_START_EXEC_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_STOP_EXEC_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_START_READY_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_STOP_READY_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_CREATE_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_TERMINATE_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_IDLE_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TIMER_ENTER_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TIMER_EXIT_ENABLE=y

examples/system/sysview_tracing_heap_log/sdkconfig.defaults

@@ -4,24 +4,25 @@ CONFIG_FREERTOS_UNICORE=y
 # 1ms tick period
 CONFIG_FREERTOS_HZ=1000
 # Enable application tracing by default
-CONFIG_APPTRACE_DEST_TRAX=y
+CONFIG_APPTRACE_DEST_JTAG=y
+CONFIG_APPTRACE_MEMBUFS_APPTRACE_PROTO_ENABLE=y
 CONFIG_APPTRACE_ENABLE=y
 # Enable FreeRTOS SystemView Tracing by default
-CONFIG_SYSVIEW_ENABLE=y
-CONFIG_SYSVIEW_TS_SOURCE_TIMER_00=y
-CONFIG_SYSVIEW_EVT_OVERFLOW_ENABLE=y
-CONFIG_SYSVIEW_EVT_ISR_ENTER_ENABLE=y
-CONFIG_SYSVIEW_EVT_ISR_EXIT_ENABLE=y
-CONFIG_SYSVIEW_EVT_ISR_TO_SCHEDULER_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_START_EXEC_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_STOP_EXEC_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_START_READY_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_STOP_READY_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_CREATE_ENABLE=y
-CONFIG_SYSVIEW_EVT_TASK_TERMINATE_ENABLE=y
-CONFIG_SYSVIEW_EVT_IDLE_ENABLE=y
-CONFIG_SYSVIEW_EVT_TIMER_ENTER_ENABLE=y
-CONFIG_SYSVIEW_EVT_TIMER_EXIT_ENABLE=y
+CONFIG_APPTRACE_SV_ENABLE=y
+CONFIG_APPTRACE_SV_TS_SOURCE_TIMER_00=y
+CONFIG_APPTRACE_SV_EVT_OVERFLOW_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_ISR_ENTER_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_ISR_EXIT_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_ISR_TO_SCHED_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_START_EXEC_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_STOP_EXEC_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_START_READY_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_STOP_READY_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_CREATE_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TASK_TERMINATE_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_IDLE_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TIMER_ENTER_ENABLE=y
+CONFIG_APPTRACE_SV_EVT_TIMER_EXIT_ENABLE=y
 # Disable color output in logs
 CONFIG_LOG_COLORS=n
 # Enable heap tracing to host