abort handler: Fix abort stack trace when abort() called in ISR

Previously, this resulted in task stack frames turning up incorrectly in the backtrace, ie

Backtrace: 0x400d22a0:0x3ffb0fa0 0x40085a3c:0x3ffb0fc0 0x400f32c4:0x3ffb0fe0 0x40081965:0x3ffb1010
0x400d22a0: esp_vApplicationIdleHook at /home/esp/esp-idf/components/esp32/./freertos_hooks.c:
52
0x40085a3c: prvIdleTask at /home/esp/esp-idf/components/freertos/./tasks.c:4431
0x400f32c4: i2c_isr_handler_default at /home/esp/esp-idf/components/driver/./i2c.c:598
0x40081965: _xt_lowint1 at xtensa_vectors.o:?

Fix is to implement abort() via an unhandled exception rather than a breakpoint, I think
because of relative priority of exception types.

Another approach would be to assign a software-only INUM to abort()ing and defined a
PANIC_RSN_ABORTED, but this is more complex and interrupt numbers are more scarce than RAM!

components/esp32/panic.c

@@ -119,7 +119,9 @@ static __attribute__((noreturn)) inline void invoke_abort()
     esp_apptrace_flush_nolock(ESP_APPTRACE_DEST_TRAX, ESP_APPTRACE_TRAX_BLOCK_SIZE*CONFIG_ESP32_APPTRACE_ONPANIC_HOST_FLUSH_TRAX_THRESH/100, CONFIG_ESP32_APPTRACE_ONPANIC_HOST_FLUSH_TMO);
 #endif
     while(1) {
+        if (esp_cpu_in_ocd_debug_mode()) {
             __asm__ ("break 0,0");
+        }
         *((int*) 0) = 0;
     }
 }
@@ -127,7 +129,7 @@ static __attribute__((noreturn)) inline void invoke_abort()
 void abort()
 {
 #if !CONFIG_ESP32_PANIC_SILENT_REBOOT
-    ets_printf("abort() was called at PC 0x%08x on core %d\n", (intptr_t)__builtin_return_address(0) - 3, xPortGetCoreID());
+    ets_printf("abort() was called at PC 0x%08x on core %d\r\n", (intptr_t)__builtin_return_address(0) - 3, xPortGetCoreID());
 #endif
     invoke_abort();
 }
@@ -146,6 +148,7 @@ static const char *edesc[] = {
     "Cp4Dis", "Cp5Dis", "Cp6Dis", "Cp7Dis"
 };
 
+#define NUM_EDESCS (sizeof(edesc) / sizeof(char *))
 
 static void commonErrorHandler(XtExcFrame *frame);
 
@@ -197,7 +200,6 @@ void panicHandler(XtExcFrame *frame)
     panicPutStr("Guru Meditation Error: Core ");
     panicPutDec(core_id);
     panicPutStr(" panic'ed (");
-    if (!abort_called) {
     panicPutStr(reason);
     panicPutStr(")\r\n");
     if (frame->exccause == PANIC_RSN_DEBUGEXCEPTION) {
@@ -228,9 +230,6 @@ void panicHandler(XtExcFrame *frame)
         if (debugRsn&XCHAL_DEBUGCAUSE_DEBUGINT_MASK) panicPutStr("DebugIntr ");
         panicPutStr("\r\n");
     }
-        } else {
-            panicPutStr("abort)\r\n");
-        }
 
     if (esp_cpu_in_ocd_debug_mode()) {
 #if CONFIG_ESP32_APPTRACE_ENABLE
@@ -245,9 +244,10 @@ void panicHandler(XtExcFrame *frame)
 void xt_unhandled_exception(XtExcFrame *frame)
 {
     haltOtherCore();
+    if (!abort_called) {
         panicPutStr("Guru Meditation Error of type ");
         int exccause = frame->exccause;
-    if (exccause < 40) {
+        if (exccause < NUM_EDESCS) {
             panicPutStr(edesc[exccause]);
         } else {
             panicPutStr("Unknown");
@@ -267,6 +267,7 @@ void xt_unhandled_exception(XtExcFrame *frame)
             return;
         }
         panicPutStr(". Exception was unhandled.\r\n");
+    }
     commonErrorHandler(frame);
 }
 
@@ -389,7 +390,7 @@ void esp_restart_noos() __attribute__ ((noreturn));
   We arrive here after a panic or unhandled exception, when no OCD is detected. Dump the registers to the
   serial port and either jump to the gdb stub, halt the CPU or reboot.
 */
-static void commonErrorHandler(XtExcFrame *frame)
+static __attribute__((noreturn)) void commonErrorHandler(XtExcFrame *frame)
 {
     int *regs = (int *)frame;
     int x, y;