modem_cellular: Add support for the simcom a76xx modem #90081

olalonde · 2025-05-16T21:49:30Z

Add support for the simcom a76xx modem which is similar to the simcom 7080 but has a few key differences. Tested with a simcom A7672SA module but as there is a single simcom A76XX AT commands manual, the driver should work with other modems of the series.

Tested with:

CONFIG_MODEM_CMUX_WORK_BUFFER_SIZE=1507

and

/ {
        aliases {
                modem = &modem;
        };
};

/* Tested with Simcom A7672SA module connected via UART
 * TX: P1.2
 * RX: P1.1
 * PWRKEY: P0.02
 */
&uart1 {
        compatible = "nordic,nrf-uarte";
        status = "okay";
        current-speed = <115200>;
        /* The module I tested with doesn't expose RCS/CTS pins
         * hw-flow-control;
         */

        pinctrl-0 = <&uart1_default>;

        modem: modem {
                compatible = "simcom,a76xx";
                status = "okay";
                mdm-power-gpios = <&gpio0 2 (GPIO_ACTIVE_HIGH)>;
        };
};

github-actions · 2025-05-16T21:50:09Z

Hello @olalonde, and thank you very much for your first pull request to the Zephyr project!
Our Continuous Integration pipeline will execute a series of checks on your Pull Request commit messages and code, and you are expected to address any failures by updating the PR. Please take a look at our commit message guidelines to find out how to format your commit messages, and at our contribution workflow to understand how to update your Pull Request. If you haven't already, please make sure to review the project's Contributor Expectations and update (by amending and force-pushing the commits) your pull request if necessary.
If you are stuck or need help please join us on Discord and ask your question there. Additionally, you can escalate the review when applicable. 😊

jukkar · 2025-05-22T06:01:24Z

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.overlay

@@ -0,0 +1,27 @@
+/ {


This does not look right. Now anyone having this DK will get this simcom modem added automatically even if he/she does not have one.

I used the nrf52840dk board for testing the modem as none of the supported Zephyr boards uses it. There's no reason to run the cellular modem sample with it since it doesn't have a modem, so the chance of confusing someone is low. Is there a better way to do this? Or should i just remove it from the sample? I don't mind either way it was mostly just to facilitate my testing.

The problem is that someone might be using the nRF52840 DK as the host MCU just as you are doing, but with another modem.
Maybe a solution would be to move and rename this overlay so that it's just for this particular modem and you can include it manually in your build.
Don't know what the others think of this.

@tomi-font Personally, I feel boards that don't have a modem on them shouldn't be included here at all. Perhaps instead we could include documentation on how to write a DTS for this sample instead.

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.conf

tomi-font · 2025-05-22T06:32:22Z

samples/net/cellular_modem/Kconfig

+config SAMPLE_CELLULAR_MODEM_ENDPOINT_HOSTNAME
+	string "Endpoint hostname"
+	default "test-endpoint.com"


Nice! @bjarki-andreasen something really should be done with regards to test-endpoint.com... cc @kartben

As a side note, in my own project, I recently added this to my CMakeLists.txt to avoid committing secrets to git:

if(NOT EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/local.conf) file(WRITE ${CMAKE_CURRENT_SOURCE_DIR}/local.conf "# This file is not version controlled, you can use it for local only configuration") endif() set(EXTRA_CONF_FILE ${CMAKE_CURRENT_SOURCE_DIR}/local.conf )

And I also added local.conf to .gitignore.

It would be useful here to allow developers to set a private CONFIG_SAMPLE_CELLULAR_MODEM_ENDPOINT_HOSTNAME without having to edit a version controlled file. And for other configurations (e.g. logging) that are useful for testing but that you don't necessarily want to commit to git.

samples/net/cellular_modem/README.rst

drivers/modem/Kconfig.cellular

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.conf

samples/net/cellular_modem/README.rst

tomi-font · 2025-05-27T12:53:33Z

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.overlay

@@ -0,0 +1,27 @@
+/ {


The problem is that someone might be using the nRF52840 DK as the host MCU just as you are doing, but with another modem.
Maybe a solution would be to move and rename this overlay so that it's just for this particular modem and you can include it manually in your build.
Don't know what the others think of this.

tomi-font · 2025-05-27T12:56:17Z

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.conf

@@ -0,0 +1,3 @@
+# Otherwise we run into DLCI buffer overruns (maybe due to lack of UART hardware flow control?)
+CONFIG_MODEM_CMUX_WORK_BUFFER_SIZE=1507
+CONFIG_NET_IPV6=n


Explanation for this? (as a comment)

tomi-font · 2025-05-27T12:57:38Z

samples/net/cellular_modem/prj.conf

@@ -1,5 +1,6 @@
 # Copyright (c) 2023 Bjarki Arge Andreasen
 # SPDX-License-Identifier: Apache-2.0
+CONFIG_SAMPLE_CELLULAR_MODEM_ENDPOINT_HOSTNAME=""


@olalonde I guess you include the hostname by passing an environment variable (in CMake or when building with west), but there is no reason not to have a default here.

Is there a default that works? I just thought it would make it more obvious that this option needs to be set.

Then in that case maybe you would change the default of the Kconfig option where it's defined, I think it would make more sense.
But it's true that given test-endpoint.com is no longer functional it might be better to completely remove it.

tomi-font · 2025-05-27T12:58:33Z

drivers/modem/modem_cellular.c

+
+MODEM_CHAT_SCRIPT_CMDS_DEFINE(simcom_a76xx_shutdown_chat_script_cmds,
+			      MODEM_CHAT_SCRIPT_CMD_RESP("AT+CPOF", ok_match));
+
+MODEM_CHAT_SCRIPT_DEFINE(simcom_a76xx_shutdown_chat_script,
+			 simcom_a76xx_shutdown_chat_script_cmds, abort_matches,
+			 modem_cellular_chat_callback_handler, 15);
+


Not that important but this commit could be squashed into the first one.

tomi-font · 2025-05-27T13:00:04Z

drivers/modem/modem_cellular.c

+			      #ifdef CONFIG_MODEM_CELLULAR_POWER_ON_GNSS
+			      /* Power on the GNSS module.
+			       * We need to do this early, otherwise it does not work when
+			       * doing it later (e.g. from a user pipe).
+			       */
+			      MODEM_CHAT_SCRIPT_CMD_RESP_MULT("AT+CGNSSPWR=1", allow_match),
+			      #endif


Suggested change

#ifdef CONFIG_MODEM_CELLULAR_POWER_ON_GNSS

/* Power on the GNSS module.

* We need to do this early, otherwise it does not work when

* doing it later (e.g. from a user pipe).

*/

MODEM_CHAT_SCRIPT_CMD_RESP_MULT("AT+CGNSSPWR=1", allow_match),

#endif

#ifdef CONFIG_MODEM_CELLULAR_POWER_ON_GNSS

/* Power on the GNSS module.

* We need to do this early, otherwise it does not work when

* doing it later (e.g. from a user pipe).

*/

MODEM_CHAT_SCRIPT_CMD_RESP_MULT("AT+CGNSSPWR=1", allow_match),

#endif

bjarki-andreasen

The board overlay for the cellular sample must be removed. In zephyr, we have the option of boards and shields for common portable build targets. overlays specific to a local setup is not allowed upstream (outside of common vendor specific fixtures)

bjarki-andreasen · 2025-05-27T13:21:35Z

drivers/modem/Kconfig.cellular

+config MODEM_CELLULAR_POWER_ON_GNSS
+	bool "Power on the modem's GNSS module"
+	depends on DT_HAS_SIMCOM_A76XX_ENABLED


Please remove this config, the modem_cellular.c driver is an MVP for cellular networking, its not meant for power optimization of GNSS. Just enable it unconditionally. For finer control, introduce a device driver specific driver.

vytvir · 2025-05-19T09:43:30Z

drivers/modem/modem_cellular.c

+MODEM_CHAT_SCRIPT_CMDS_DEFINE(simcom_a76xx_periodic_chat_script_cmds,
+			      MODEM_CHAT_SCRIPT_CMD_RESP("AT+CREG?", ok_match),
+			      MODEM_CHAT_SCRIPT_CMD_RESP("AT+CEREG?", ok_match),
+			      MODEM_CHAT_SCRIPT_CMD_RESP("AT+CGREG?", ok_match));


The AT+CREG? command is used to query circuit-switched network (2G/3G) registration status. The AT+CGREG? command is used to query GPRS (2G) registration status, and the AT+CEREG? is used to query EPS registration status (LTE, essentially). Not all A76XX series modems support all of these technologies, so it is not unexpected to receive a +CGREG: 0,0 or +CREG: 0,0 while you are connected to an LTE network.
The way the current state machine works is it will start some mechanism to recover from a loss of registration. I suspect running these commands in an LTE-only module is going to potentially introduce issues. In general, this area needs to be improved because for multi-RAT modems the logic is finicky at best.

vytvir · 2025-05-19T11:42:38Z

drivers/modem/modem_cellular.c

@@ -616,6 +618,8 @@ static int modem_cellular_on_idle_state_enter(struct modem_cellular_data *data)
 	}

 	modem_cellular_notify_user_pipes_disconnected(data);
+	net_if_carrier_off(modem_ppp_get_iface(data->ppp));
+	net_if_dormant_off(modem_ppp_get_iface(data->ppp));


The net_if_dormant_off() should do nothing, because we never call net_if_dormant_on(...). While net_if_carrier_off(...) should be called when leaving CARRIER_ON_STATE in modem_cellular_carrier_on_state_leave(...). Basically, in the same state it was previously turned on.
~~Could you maybe turn on CONFIG_MODEM_LOG_LEVEL_DBG=y as it will allow us to track these state changes.~~
Seems that notify_iface_up(...) in net_if.c is never called. Could you please enable CONFIG_NET_IF_LOG_LEVEL_DBG=y so we understand the cause why? Because we should be calling carrier_on.

We covered this here #90084

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.conf

vytvir · 2025-05-27T20:19:47Z

samples/net/cellular_modem/boards/nrf52840dk_nrf52840.overlay

@@ -0,0 +1,27 @@
+/ {


@tomi-font Personally, I feel boards that don't have a modem on them shouldn't be included here at all. Perhaps instead we could include documentation on how to write a DTS for this sample instead.

vytvir · 2025-05-27T20:21:04Z

samples/net/cellular_modem/prj.conf

@@ -1,5 +1,6 @@
 # Copyright (c) 2023 Bjarki Arge Andreasen
 # SPDX-License-Identifier: Apache-2.0
+CONFIG_SAMPLE_CELLULAR_MODEM_ENDPOINT_HOSTNAME=""


@olalonde I guess you include the hostname by passing an environment variable (in CMake or when building with west), but there is no reason not to have a default here.

olalonde · 2025-05-28T02:04:09Z

@bjarki-andreasen I think I addressed all the issues.

Add support for the simcom a76xx modem which is similar to the simcom 7080 but has a few key differences. Tested with a simcom A7672SA module but as there is a single simcom A76XX AT commands manual, the driver should work with other modems of the series. Signed-off-by: Olivier Lalonde <[email protected]>

sonarqubecloud · 2025-05-28T09:34:58Z

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

tomi-font · 2025-05-28T11:36:12Z

samples/net/cellular_modem/Kconfig

+
+config SAMPLE_CELLULAR_MODEM_ENDPOINT_HOSTNAME
+	string "Endpoint hostname"
+	default "test-endpoint.com"


Actually I think it makes sense to stop pretending test-endpoint.com should be functional:

Suggested change

default "test-endpoint.com"

I agree, it's really not ideal.
The sample/instructions should be updated to not depend it anymore, or the doc should be updated to indicate how one may host their own server (with the caveat that users might struggle to make their endpoint publicly accessible)

IMO the sample should be a really minimal showcase of printing modem-info, showing that we can ping 8.8.8.8 or whatever, and that's it. Things like testing throughput etc. are really something that's left for a test IMO, and probably don't belong to the sample.

Yeah, I agree as well :) With the state of the modem subsystem and drivers, we should be looking to networking samples to test networking, the cellular part really should not make a difference :)

bjarki-andreasen

Nice work!

rerickson1

Looks good. I think it would be better to split this into at least two commits. One for driver changes, and the other for the sample changes.

github-actions · 2025-05-28T14:38:32Z

Hi @olalonde!
Congratulations on getting your very first Zephyr pull request merged 🎉🥳. This is a fantastic achievement, and we're thrilled to have you as part of our community!

To celebrate this milestone and showcase your contribution, we'd love to award you the Zephyr Technical Contributor badge. If you're interested, please claim your badge by filling out this form: Claim Your Zephyr Badge.

Thank you for your valuable input, and we look forward to seeing more of your contributions in the future! 🪁

github-actions bot added area: Modem Drivers area: Networking area: Modem area: Samples Samples labels May 16, 2025

github-actions bot requested review from bjarki-andreasen, jukkar, kartben, nashif, pdgendt, rerickson1, rlubos, ssharks, tbursztyka and tomi-font May 16, 2025 21:50

github-actions bot assigned bjarki-andreasen May 16, 2025

olalonde force-pushed the simcom-a76xx branch from b6c97ba to ef2ea96 Compare May 16, 2025 22:21

olalonde mentioned this pull request May 16, 2025

Cellular sample fails to bring back L4 connection after modem restart #90084

Closed

olalonde force-pushed the simcom-a76xx branch 9 times, most recently from 43d0b7c to 41fa7b7 Compare May 21, 2025 22:55

jukkar reviewed May 22, 2025

View reviewed changes

tomi-font reviewed May 22, 2025

View reviewed changes

olalonde force-pushed the simcom-a76xx branch from 41fa7b7 to b121947 Compare May 23, 2025 21:14

olalonde force-pushed the simcom-a76xx branch from b121947 to 6a62527 Compare May 23, 2025 21:17

tomi-font requested changes May 27, 2025

View reviewed changes

bjarki-andreasen requested changes May 27, 2025

View reviewed changes

vytvir reviewed May 27, 2025

View reviewed changes

olalonde force-pushed the simcom-a76xx branch 2 times, most recently from 6f7d1e4 to 442d575 Compare May 28, 2025 02:01

olalonde force-pushed the simcom-a76xx branch from 442d575 to 63ac9ba Compare May 28, 2025 08:58

olalonde force-pushed the simcom-a76xx branch from 63ac9ba to e83a4a0 Compare May 28, 2025 09:04

tomi-font approved these changes May 28, 2025

View reviewed changes

bjarki-andreasen approved these changes May 28, 2025

View reviewed changes

rerickson1 reviewed May 28, 2025

View reviewed changes

kartben merged commit afc481a into zephyrproject-rtos:main May 28, 2025
26 checks passed

modem_cellular: Add support for the simcom a76xx modem #90081

modem_cellular: Add support for the simcom a76xx modem #90081

Uh oh!

Conversation

olalonde commented May 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented May 16, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

olalonde May 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

olalonde May 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bjarki-andreasen left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

olalonde commented May 28, 2025

Uh oh!

sonarqubecloud bot commented May 28, 2025

Quality Gate passed

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bjarki-andreasen left a comment

Choose a reason for hiding this comment

Uh oh!

rerickson1 left a comment

Choose a reason for hiding this comment

Uh oh!

olalonde commented May 16, 2025 •

edited

Loading

olalonde May 23, 2025 •

edited

Loading

olalonde May 23, 2025 •

edited

Loading

bjarki-andreasen left a comment •

edited

Loading