Closed dqdqdq31 closed 3 years ago
jeromecoutant
commented
4 years ago Hi
crashed with mbed-os-6.0.0-beta-1 But it works with mbed-os-6.0.0-alpha-3
Ideally, you should find the guilty commit between these 2 tags :-)
"trace-level": { "value": "TRACE_LEVEL_DEBUG"
Could you remove it to check ?
"cellular.debug-at" : true,
Same check to do
0xc0170
commented
4 years ago Please do definitely git bisect or at least what has changed for you setup (target code/cellular or something else) ?
dqdqdq31
commented
4 years ago I found the difference. I add component SPIF without pins of spif as below, even though I don't implement or call any SPIF driver, it still crashes on mbed-os-6.0.0-beta-1 (but not crash on 6.0.0-alpha-3). "target.components_add" : ["SPIF"], "spif-driver.SPI_FREQ" : 40000000
Besides, remove SPIF configs above, it crashed as below:
[INFO][CELL]: AT TX (8): AT^SICS? Mbed Bootloader No Update image [DBG ] Active firmware up-to-date booting...
[DBG ][CELL]: GEMALTO_CINTERION flow control: RTS 52 CTS 51 [INFO][CELL]: New CellularContext (0x2000CA50)
++ MbedOS Error Info ++ Error Status: 0x80FF0144 Code: 324 Module: 255 Error Message: Assertion failed: _device.get_property(AT_CellularDevice::PROPERTY_SOCKET_COUNT) > 0 Location: 0x804E275 File: .\mbed-os\features\cellular\framework\AT\AT_CellularStack.cpp+33 Error Value: 0x0 Current Thread: main Id: 0x20009280 Entry: 0x804FADD StackSize: 0x1400 StackMem: 0x2000A568 SP: 0x2000B788 For more info, visit: https://mbed.com/s/error?error=0x80FF0144&tgt=NUCLEO_F429ZI -- MbedOS Error Info --
= System will be rebooted due to a fatal error = = Reboot count(=1) reached maximum, system will halt after rebooting = Mbed Bootloader No Update image [DBG ] Active firmware up-to-date booting...
I only modify mbed_app.json and replace mbed_cloud_dev_credentials.c and update_default_resources.c. Here's my mbed_app.json
{ "macros": [ "ARM_UC_USE_PAL_BLOCKDEVICE=1" ], "target_overrides": { "*": { "target.features_add" : ["BOOTLOADER", "STORAGE"], "platform.stdio-baud-rate" : 115200, "platform.stdio-convert-newlines" : true, "platform.stdio-buffered-serial" : true, "platform.stdio-flush-at-exit" : true, "rtos.main-thread-stack-size" : 5120, "events.shared-stacksize" : 2048, "events.shared-eventsize" : 2048, "update-client.storage-locations" : 1, "mbed-trace.enable" : null,
"target.OUTPUT_EXT" : "hex"
},
"NUCLEO_F429ZI": {
"target.network-default-interface-type" : "CELLULAR",
"target.bootloader_img" : "tools/mbed-bootloader-nucleo_f429zi-internal_flash-no_rot-v4.1.0.bin",
"target.header_offset" : "0x8000",
"target.app_offset" : "0x8400",
"target.restrict_size" : "0xF7C00",
"update-client.bootloader-details" : "0x080078CC",
"update-client.application-details" : "(MBED_ROM_START + MBED_BOOTLOADER_SIZE)",
"update-client.storage-address" : "(MBED_CONF_STORAGE_TDB_INTERNAL_INTERNAL_BASE_ADDRESS+MBED_CONF_STORAGE_TDB_INTERNAL_INTERNAL_SIZE)",
"update-client.storage-size" : "(1024*1024-MBED_CONF_STORAGE_TDB_INTERNAL_INTERNAL_SIZE)",
"update-client.storage-page" : 1,
"mbed-cloud-client.update-storage" : "ARM_UCP_FLASHIAP",
"storage_tdb_internal.internal_base_address": "(MBED_ROM_START+1024*1024)",
"storage_tdb_internal.internal_size" : "(128*1024)",
"storage.storage_type" : "TDB_INTERNAL",
"mbed-trace.enable" : true,
"cellular.debug-at" : true,
"cellular.use-apn-lookup" : false,
"lwip.ppp-enabled" : false,
"target.macros_add": [
"CELLULAR_DEVICE=GEMALTO_CINTERION"
],
"target.components_add" : ["GEMALTO_CINTERION"],
"GEMALTO_CINTERION.tx" : "PD_5",
"GEMALTO_CINTERION.rx" : "PD_6",
"GEMALTO_CINTERION.rts" : "PD_4",
"GEMALTO_CINTERION.cts" : "PD_3",
"GEMALTO_CINTERION.provide-default" : "true",
"mbed-client-pal.pal-max-frag-len" : 1,
"mbed-client.sn-coap-max-blockwise-payload-size": 256
}
},
"config": {
"developer-mode": {
"help" : "Enable Developer mode to skip Factory enrollment",
"options" : [null, 1],
"value" : 1
},
"button-pinname": {
"help" : "PinName for button.",
"value" : "BUTTON1"
},
"led-pinname" : {
"help" : "PinName for led, which is attached to led blink resource.",
"value" : "LED_RED"
},
"bootloader-size": {
"help" : "Helper macro to enable calculation of rom regions. target.header_offset and target.app_offset still needs to be calculated manually, though.",
"value" : "(32*1024)",
"macro_name": "MBED_BOOTLOADER_SIZE"
},
"mbed-cloud-client-update-buffer-size": {
"value" : null
},
"trace-level": {
"help": "Options are TRACE_LEVEL_ERROR,TRACE_LEVEL_WARN,TRACE_LEVEL_INFO,TRACE_LEVEL_DEBUG, TRACE_LEVEL_DEBUG->TRACE_LEVEL_WARN",
"macro_name": "MBED_TRACE_MAX_LEVEL",
"value": "TRACE_LEVEL_DEBUG"
}
}}
My build commands: mbed target NUCLEO_F429ZI mbed config GCC_ARM_PATH C:\ARM-SDK\gcc-arm-none-eabi-7-2017-q4-major-win32\bin mbed toolchain GCC_ARM mbed compile
AriParkkila
commented
4 years ago Can you share AT log around the line "Cinterion model".
adbridge
commented
4 years ago @dqdqdq31 please put the issue template back in the header and fill in all the required fields. We reserve the right to close this issue if insufficient information is provided in the correct form. Thank you.
dqdqdq31
commented
4 years ago What I modified is
// Needed for PRIu64 on FreeRTOS
// Note: this macro is needed on armcc to get the the limit macros like UINT16_MAX
// Note: this macro is needed on armcc to get the the PRI*32 macros // from inttypes.h in a C++ code.
(MBED_CONF_NANOSTACK_HAL_EVENT_LOOP_USE_MBED_EVENTS == 1) && \ defined(MBED_CONF_EVENTS_SHARED_DISPATCH_FROM_APPLICATION) && \ (MBED_CONF_EVENTS_SHARED_DISPATCH_FROM_APPLICATION == 1)
// event based LED blinker, controlled via pattern_resource
static Blinky blinky;
static void main_application(void);
extern "C" int mbed_cloud_application_entrypoint(void)
int main(void)
{ return mcc_platform_run_program(main_application); }
// Pointers to the resources that will be created in main_application(). static M2MResource button_res; static M2MResource pattern_res; static M2MResource blink_res; static M2MResource unregister_res; static M2MResource* factory_reset_res;
void unregister(void);
// Pointer to mbedClient, used for calling close function. static SimpleM2MClient *client;
void counter_updated(const char*) { // Converts uint64_t to a string to remove the dependency for int64 printf implementation. char buffer[20+1]; (void) m2m::itoa_c(button_res->get_value_int(), buffer); printf("Counter resource set to %s\r\n", buffer); }
void pattern_updated(const char *) { printf("PUT received, new value: %s\r\n", pattern_res->get_value_string().c_str()); }
void blink_callback(void *) { String pattern_string = pattern_res->get_value_string(); printf("POST executed\r\n");
// The pattern is something like 500:200:500, so parse that.
// LED blinking is done while parsing.const bool restart_pattern = false;
if (blinky.start((char*)pattern_res->value(), pattern_res->value_length(), restart_pattern) == false) {
printf("out of memory error\r\n");
}blink_res->send_delayed_post_response();}
void notification_status_callback(const M2MBase& object, const M2MBase::MessageDeliveryStatus status, const M2MBase::MessageType /type/) { switch(status) { case M2MBase::MESSAGE_STATUS_BUILD_ERROR: printf("Message status callback: (%s) error when building CoAP message\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_RESEND_QUEUE_FULL: printf("Message status callback: (%s) CoAP resend queue full\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_SENT: printf("Message status callback: (%s) Message sent to server\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_DELIVERED: printf("Message status callback: (%s) Message delivered\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_SEND_FAILED: printf("Message status callback: (%s) Message sending failed\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_SUBSCRIBED: printf("Message status callback: (%s) subscribed\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_UNSUBSCRIBED: printf("Message status callback: (%s) subscription removed\r\n", object.uri_path()); break; case M2MBase::MESSAGE_STATUS_REJECTED: printf("Message status callback: (%s) server has rejected the message\r\n", object.uri_path()); break; default: break; } }
void sent_callback(const M2MBase& base, const M2MBase::MessageDeliveryStatus status, const M2MBase::MessageType /type/) { switch(status) { case M2MBase::MESSAGE_STATUS_DELIVERED: unregister(); break; default: break; } }
void unregister_triggered(void) { printf("Unregister resource triggered\r\n"); unregister_res->send_delayed_post_response(); }
void factory_reset_triggered(void*) { printf("Factory reset resource triggered\r\n");
// First send response, so server won't be left waiting.
// Factory reset resource is by default expecting explicit
// delayed response sending.
factory_reset_res->send_delayed_post_response();
// Then run potentially long-taking factory reset routines.
kcm_factory_reset();}
// This function is called when a POST request is received for resource 5000/0/1. void unregister(void) { printf("Unregister resource executed\r\n"); client->close(); }
void main_application(void)
{
printf("
// make sure the line buffering is on as non-trace builds do
// not produce enough output to fill the buffer
setlinebuf(stdout);printf("<test> main_application 1\r\n");
// Initialize trace-library first
if (application_init_mbed_trace() != 0) {
printf("Failed initializing mbed trace\r\n" );
return;
}
printf("<test> main_application 2\r\n");
// Initialize storage
if (mcc_platform_storage_init() != 0) {
printf("Failed to initialize storage\r\n" );
return;
}
printf("<test> main_application 3\r\n");
// Initialize platform-specific components
if(mcc_platform_init() != 0) {
printf("ERROR - platform_init() failed!\r\n");
return;
}
printf("<test> main_application 4\r\n");
// Print some statistics of the object sizes and their heap memory consumption.
// NOTE: This *must* be done before creating MbedCloudClient, as the statistic calculation
// creates and deletes M2MSecurity and M2MDevice singleton objects, which are also used by
// the MbedCloudClient.print_m2mobject_stats();// SimpleClient is used for registering and unregistering resources to a server.
SimpleM2MClient mbedClient;
// Save pointer to mbedClient so that other functions can access it.
client = &mbedClient;
printf("<test> main_application 5\r\n");
/*
* Pre-initialize network stack and client library.
*
* Specifically for nanostack mesh networks on Mbed OS platform it is important to initialize
* the components in correct order to avoid out-of-memory issues in Device Management Client initialization.
* The order for these use cases should be:
* 1. Initialize network stack using `nsapi_create_stack()` (Mbed OS only). // Implemented in `mcc_platform_interface_init()`.
* 2. Initialize Device Management Client using `init()`. // Implemented in `mbedClient.init()`.
* 3. Connect to network interface using 'connect()`. // Implemented in `mcc_platform_interface_connect()`.
* 4. Connect Device Management Client to service using `setup()`. // Implemented in `mbedClient.register_and_connect)`.
*/
(void) mcc_platform_interface_init();
mbedClient.init();
printf("<test> main_application 6\r\n");
// application_init() runs the following initializations:
// 1. platform initialization
// 2. print memory statistics if MBED_HEAP_STATS_ENABLED is defined
// 3. FCC initialization.
if (!application_init()) {
printf("Initialization failed, exiting application!\r\n");
return;
}
printf("<test> main_application 7\r\n");
// Print platform information
mcc_platform_sw_build_info();
printf("<test> main_application 8\r\n");
// Initialize network
if (!mcc_platform_interface_connect()) {
printf("Network initialized, registering...\r\n");
} else {
return;
}
printf("<test> main_application 9\r\n");printf("Client initialized\r\n");
print_heap_stats();print_stack_statistics();printf("<test> main_application 10\r\n");// Create resource for button count. Path of this resource will be: 3200/0/5501.
button_res = mbedClient.add_cloud_resource(3200, 0, 5501, "button_resource", M2MResourceInstance::INTEGER,
M2MBase::GET_PUT_ALLOWED, 0, true, (void*)counter_updated, (void*)notification_status_callback);
button_res->set_value(0);
// Create resource for led blinking pattern. Path of this resource will be: 3201/0/5853.
pattern_res = mbedClient.add_cloud_resource(3201, 0, 5853, "pattern_resource", M2MResourceInstance::STRING,
M2MBase::GET_PUT_ALLOWED, "500:500:500:500", true, (void*)pattern_updated, (void*)notification_status_callback);
// Create resource for starting the led blinking. Path of this resource will be: 3201/0/5850.
blink_res = mbedClient.add_cloud_resource(3201, 0, 5850, "blink_resource", M2MResourceInstance::STRING,
M2MBase::POST_ALLOWED, "", false, (void*)blink_callback, (void*)notification_status_callback);
// Use delayed response
blink_res->set_delayed_response(true);
// Create resource for unregistering the device. Path of this resource will be: 5000/0/1.
unregister_res = mbedClient.add_cloud_resource(5000, 0, 1, "unregister", M2MResourceInstance::STRING,
M2MBase::POST_ALLOWED, NULL, false, (void*)unregister_triggered, (void*)sent_callback);
unregister_res->set_delayed_response(true);
// Create optional Device resource for running factory reset for the device. Path of this resource will be: 3/0/5.
factory_reset_res = M2MInterfaceFactory::create_device()->create_resource(M2MDevice::FactoryReset);
if (factory_reset_res) {
factory_reset_res->set_execute_function(factory_reset_triggered);
}printf("<test> main_application 11\r\n");// For high-latency networks with limited total bandwidth combined with large number // of endpoints, it helps to stabilize the network when Device Management Client has // delayed registration to Device Management after the network formation. // This is applicable in large Wi-SUN networks.
wait_application_startup_delay();printf("<test> main_application 12\r\n");
mbedClient.register_and_connect();
printf("<test> main_application 13\r\n");blinky.init(mbedClient, button_res);
blinky.request_next_loop_event();
blinky.request_automatic_increment_event();printf("<test> main_application 14\r\n");// Add certificate renewal callback
mbedClient.get_cloud_client().on_certificate_renewal(certificate_renewal_cb);printf("<test> main_application 15\r\n");(MBED_CONF_NANOSTACK_HAL_EVENT_LOOP_USE_MBED_EVENTS == 1) && \ defined(MBED_CONF_EVENTS_SHARED_DISPATCH_FROM_APPLICATION) && \ (MBED_CONF_EVENTS_SHARED_DISPATCH_FROM_APPLICATION == 1) printf("Starting mbed eventloop...\r\n");
eventOS_scheduler_mutex_wait();
EventQueue *queue = mbed::mbed_event_queue();
queue->dispatch_forever();printf("<test> main_application 16\r\n");
// Check if client is registering or registered, if true sleep and repeat.
while (mbedClient.is_register_called()) {
mcc_platform_do_wait(100);
}
printf("<test> main_application 17\r\n");
// Client unregistered, disconnect and exit program.
mcc_platform_interface_close();printf("<test> main_application end\r\n");} `
/*#if defined(SXOS)
#define MBED_CLOUD_CLIENT_TRANSPORT_MODE_TCP*/
#define SN_COAP_MAX_BLOCKWISE_PAYLOAD_SIZE MBED_CONF_MBED_CLIENT_SN_COAP_MAX_BLOCKWISE_PAYLOAD_SIZE#define SN_COAP_MAX_BLOCKWISE_PAYLOAD_SIZE 512/ Sets the flag to enable update support in Pelion Device Management Client /
#define MBED_CLOUD_CLIENT_SUPPORT_UPDATE/* Sets the download buffer for update client in bytes (min. 2048 bytes).
/ Developer flags for Update feature /
#define MBED_CLOUD_DEV_UPDATE_CERT
#define MBED_CLOUD_DEV_UPDATE_ID`
It just keeps rebooting so the log I have is below, I didn't get any AT log
Mbed Bootloader No Update image [DBG ] Active firmware up-to-date booting...
++ MbedOS Error Info ++ Error Status: 0x80010133 Code: 307 Module: 1 Error Message: Mutex: 0x20002170, Unknown Location: 0x8051127 Error Value: 0x20002170 Current Thread: main Id: 0x20009AF0 Entry: 0x8051051 StackSize: 0x1400 StackMem: 0x2000ADD8 SP: 0x2002FF38 For more info, visit: https://mbed.com/s/error?error=0x80010133&tgt=NUCLEO_F429ZI -- MbedOS Error Info --
= System will be rebooted due to a fatal error = = Reboot count(=89) reached maximum, system will halt after rebooting = Mbed Bootloader No Update image [DBG ] Active firmware up-to-date booting...
++ MbedOS Error Info ++ Error Status: 0x80010133 Code: 307 Module: 1 Error Message: Mutex: 0x20002170, Unknown Location: 0x8051127 Error Value: 0x20002170 Current Thread: main Id: 0x20009AF0 Entry: 0x8051051 StackSize: 0x1400 StackMem: 0x2000ADD8 SP: 0x2002FF38 For more info, visit: https://mbed.com/s/error?error=0x80010133&tgt=NUCLEO_F429ZI -- MbedOS Error Info --
= System will be rebooted due to a fatal error = = Reboot count(=90) reached maximum, system will halt after rebooting = Mbed Bootloader No Update image [DBG ] Active firmware up-to-date booting...
ciarmcom
commented
4 years ago Thank you for raising this detailed GitHub issue. I am now notifying our internal issue triagers. Internal Jira reference: https://jira.arm.com/browse/IOTOSM-2238
pan-
commented
4 years ago @dqdqdq31 Have you seen any improvements with Mbed OS 6.0 or the latest release ?
ciarmcom
commented
3 years ago We closed this issue because it has been inactive for quite some time and we believe it to be low priority. If you think that the priority should be higher, then please reopen with your justification for increasing the priority.
Run mbed-cloud-client-example project with mbed-os-6.0.0-beta-1 and it crashed. But it works with mbed-os-6.0.0-alpha-3. (all the steps and settings are the same)
Target: STM32F429 with Gemalto ELS-61
Here are the steps:
"target.macros_add": [ "CELLULAR_DEVICE=GEMALTO_CINTERION" ], "target.components_add" : ["GEMALTO_CINTERION"], "GEMALTO_CINTERION.tx" : "PD_5", "GEMALTO_CINTERION.rx" : "PD_6", "GEMALTO_CINTERION.rts" : "PD_4", "GEMALTO_CINTERION.cts" : "PD_3", "GEMALTO_CINTERION.provide-default" : "true", "mbed-client-pal.pal-max-frag-len" : 1, "mbed-client.sn-coap-max-blockwise-payload-size": 256, 3-4. add some settings: "mbed-trace.enable" : true, "cellular.debug-at" : true, "cellular.use-apn-lookup" : false, "lwip.ppp-enabled" : true, 3-5. and trace level "trace-level": { "help": "Options are TRACE_LEVEL_ERROR,TRACE_LEVEL_WARN,TRACE_LEVEL_INFO,TRACE_LEVEL_DEBUG, TRACE_LEVEL_DEBUG->TRACE_LEVEL_WARN", "macro_name": "MBED_TRACE_MAX_LEVEL", "value": "TRACE_LEVEL_DEBUG" }
Reproduce: 100%
Crash log like below and it keeps crashing and rebooting:
Mbed Bootloader No Update image [DBG ] Active firmware up-to-date booting...
++ MbedOS Error Info ++ Error Status: 0x80010133 Code: 307 Module: 1 Error Message: Mutex: 0x20002178, Unknown Location: 0x805E26B Error Value: 0x20002178 Current Thread: main Id: 0x20009F10 Entry: 0x805E195 StackSize: 0x1E00 StackMem: 0x2000DA60 SP: 0x2002FF38 For more info, visit: https://mbed.com/s/error?error=0x80010133&tgt=NUCLEO_F429ZI -- MbedOS Error Info --
= System will be rebooted due to a fatal error = = Reboot count(=1) reached maximum, system will halt after rebooting =