Hi,

I'm using the gecko OS and the device called WGM160P

First, is it possible to creat an easy web site? I read the IDE was made with Eclipse (or something like that). With Eclipse it's possible to creat a web site so may be it's possible too.

Then is it possible to establish a connexion between the web site and the device through the main (IDE program)? For example on a click on a button on the web site switch on/off LED1 and LED2.

If it's possible how can I do this?

Thanks.

Hi,

I have a problem at the very beginning

reboot<CR><LF>
Rebooting<CR><LF>
SILABS-WGM160PK-4.0.7, Gecko_OS-4.0.7, WGM160P-W00001<CR><LF>
> get gpio.usage<CR><LF>
Unknown variable or option<CR><LF>

> get gp u<CR><LF>
Unknown variable or option<CR><LF>

>set gp i 5 out 1<CR><LF>
Unknown variable or option<CR><LF>

regards,

Krzysztof 

Hello,


For a custom designed board we use a WF111 module for WiFi on an ARM A5 (Atmel sama5) platform

With the latest ( 5.2.2-r3 ) driver and Linux kernel 4.4.26 the driver loads automaticly, Linux reconises the module and loads the unifi_sdio driver.

This seems to all work fine, but no Wifi connection comes available for a while. When it does it's still non functional.
The MMC driver gives me an error that a timeout expired while waiting for the hardware interrupt.
After that the unifi_sdio driver tries to intialize the WF111 again and fails. This cycle repeats itself forever.

A small log of this:

sdhci: Secure Digital Host Controller Interface driver
sdhci: Copyright(c) Pierre Ossman
sdhci-pltfm: SDHCI platform and OF driver helper
pinctrl-at91-pio4 fc038000.pinctrl: pin PA30 already requested by 80000000.nand; cannot claim for b0000000.sdio-host
pinctrl-at91-pio4 fc038000.pinctrl: pin-30 (b0000000.sdio-host) status -22
pinctrl-at91-pio4 fc038000.pinctrl: could not request pin 30 (PA30) from group PA30  on device atmel_pinctrl
sdhci-at91 b0000000.sdio-host: Error applying setting, reverse things back
sdhci-at91 b0000000.sdio-host: update clk mul to 39 as gck rate is 480000000 Hz
mmc0: SDHCI controller on b0000000.sdio-host [b0000000.sdio-host] using ADMA
mmc0: new high speed SDIO card at address 0001

UniFi SDIO Driver: 5.2.2-r3 Feb 16 2017 14:05:57
CSR SME with WEXT support
Split patch support
Kernel 4.4.26
UniFi: Using native Linux MMC driver for SDIO.

sdio bus_id:      mmc0:0001:1 - UniFi card 0x0 inserted

SDIO: Skip power on; card is already powered.
unifi0: Initialising UniFi, attempt 1
unifi0: Chip ID 0x07  Function 1  Block Size 512  Name UniFi-4(UF60xx)
Starting network: unifi0: Chip Version 0x3A22
unifi0: Calling CsrSdioHardReset
unifi0: Falling back to software hard reset
unifi0: Chip ID 0x07  Function 1  Block Size 512  Name UniFi-4(UF60xx)
unifi0: unifi_dl_patch c4629fe8 0100060e
OK

unifi0: SDIO block size 64 requires 8 padding chunks
unifi0: UniFi f/w protocol version 9.1 (driver 9.1)
unifi0: Firmware build 1089: 2010-10-05 14:50 cindr03_core_softmac_rom_sdio_gcc 1089 bfsw@eagle@630492

unifi0: CsrWifiRouterCtrlWifiOnResHandler : WIFI ON failed : 65535
unifi0: SME client close (unifi0)
unifi0: Attempt to set mode to TORPID when lower power mode is disabled
unifi0: Message for the SME dropped, SME has gone away
mmc0: Timeout waiting for hardware interrupt.
mmc0: Timeout waiting for hardware interrupt.
mmc0: Got command interrupt 0x00000001 even though no command operation was in progress.
unifi0: Initialising UniFi, attempt 1
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: Chip Version 0x3A22
unifi0: Calling CsrSdioHardReset
unifi0: Falling back to software hard reset
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: unifi_dl_patch c4629fe8 0100060e
unifi0: SDIO block size 64 requires 8 padding chunks
unifi0: UniFi f/w protocol version 9.1 (driver 9.1)
unifi0: Firmware build 1089: 2010-10-05 14:50 cindr03_core_softmac_rom_sdio_gcc 1089 bfsw@eagle@630492
unifi0: CsrWifiRouterCtrlWifiOnResHandler : WIFI ON failed : 65535
unifi0: SME client close (unifi0)
unifi0: Attempt to set mode to TORPID when lower power mode is disabled
unifi0: Message for the SME dropped, SME has gone away
unifi0: Initialising UniFi, attempt 1
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
success
unifi0: Chip Version 0x3A22
unifi0: Calling CsrSdioHardReset
unifi0: Falling back to software hard reset
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: unifi_dl_patch c4629fe8 0100060e
unifi0: SDIO block size 64 requires 8 padding chunks
unifi0: UniFi f/w protocol version 9.1 (driver 9.1)
unifi0: Firmware build 1089: 2010-10-05 14:50 cindr03_core_softmac_rom_sdio_gcc 1089 bfsw@eagle@630492
unifi0: unifi0 is wlan0
unifi0: UniFi ready
unifi0: CsrWifiRouterCtrlWifiOnResHandler : WIFI ON failed : 65535
unifi0: SME client close (unifi0)
unifi0: Message for the SME dropped, SME has gone away
unifi0: Attempt to set mode to TORPID when lower power mode is disabled
unifi0: Message for the SME dropped, SME has gone away
mmc0: Timeout waiting for hardware interrupt.
mmc0: Timeout waiting for hardware interrupt.
mmc0: Got command interrupt 0x00000001 even though no command operation was in progress.
unifi0: Initialising UniFi, attempt 1
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: Chip Version 0x3A22
unifi0: Calling CsrSdioHardReset
unifi0: Falling back to software hard reset
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: unifi_dl_patch c4629fe8 0100060e
unifi0: SDIO block size 64 requires 8 padding chunks
unifi0: UniFi f/w protocol version 9.1 (driver 9.1)
unifi0: Firmware build 1089: 2010-10-05 14:50 cindr03_core_softmac_rom_sdio_gcc 1089 bfsw@eagle@630492
unifi0: CsrWifiRouterCtrlWifiOnResHandler : WIFI ON failed : 65535
unifi0: SME client close (unifi0)
unifi0: Attempt to set mode to TORPID when lower power mode is disabled
unifi0: Message for the SME dropped, SME has gone away
unifi0: Initialising UniFi, attempt 1
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: Chip Version 0x3A22
unifi0: Calling CsrSdioHardReset
unifi0: Falling back to software hard reset
unifi0: Chip ID 0x07  Function 1  Block Size 64  Name UniFi-4(UF60xx)
unifi0: unifi_dl_patch c4629fe8 0100060e
unifi0: SDIO block size 64 requires 8 padding chunks
unifi0: UniFi f/w protocol version 9.1 (driver 9.1)
unifi0: Firmware build 1089: 2010-10-05 14:50 cindr03_core_softmac_rom_sdio_gcc 1089 bfsw@eagle@630492
unifi0: CsrWifiRouterCtrlWifiOnResHandler : WIFI ON failed : 65535
unifi0: SME client close (unifi0)
unifi0: Attempt to set mode to TORPID when lower power mode is disabled
unifi0: Message for the SME dropped, SME has gone away

I am hoping someone can point me in the right direction where I have to start looking for this problem.
Is it related to the driver or is there something wrong in the device tree config or somewhere else in Linux.

Best regards,

Rik

Hi,

i started using the module WGM160P with the Gecko OS Studio and the IDE, for testing the peripherals inside the module.

For my new project (WiFi Gateway) i need to interface the wireless module WGM160P with other three MCU module connected through the UART peripheral.
Reading the documents "UG384: WGM160P Hardware Design User’s Guide" i realized that the SDK Gecko OS actually manage only two UART (GOS_UART_0, GOS_UART_1),
but in the datasheet of the module WGM160P there are four UART and two LUART peripherals.

First question:
The GOS_UART_0 is dedicated at the Gecko OS only, or it is possible to use with a custom protocol?

Second question:
It's possible using more than two UART inside the Gecko OS project?

Thank's.
Best regards,
Mauro.

I'm attempting to use vsprintf but it causes a hard fault every time.  Everything builds fine.  Is vsprintf supported? 

ex:

void Function(const char *p_string, ...)

{
    char buffer[256];

    va_list args;
    va_start(args, p_string);
    
    vsprintf(buffer, p_string, args);

    va_end(args);

}

The end goal is to add some additional logging information (time, calling module) to the logging string before sending buffer to gos_write_log.

Alternatively is there a way to get the functionality of gos_log without a new line being automatically added at the end?

While incorporating custom commands into my app, I found that returning GOS_CMD_BAD_ARGS from my command function caused my registered periodic callback event to stop triggering.  It does not cause a call to the deinit function.

In reading the documentation, I didn't find any references to what value should be returned from custom commands and what behavior would ensue based on different values.  Does anyone know of any documentation that describes this?

I created a simple test app to demonstrate the behavior.  If you execute the command "good", it will continue to generate events as demonstrated by a call to GOS_LOG.  However, if you execute the command "err", the callback function no longer sends messages to the log.  The only substantial difference between good and bad is the value that is returned (gos_cmd_result_t) GOS_CMD_BAD_ARGS.

#include "gos.h"

GOS_CMD_CREATE_COMMAND      (command, err, "command_err", "err", 0, 1, true);
GOS_CMD_CREATE_COMMAND      (command, good, "command_good", "good", 0, 1, true);

gos_cmd_result_t command_err_command(int argc, char **argv) {
	// Format command response
	gos_cmd_format_response(GOS_CMD_SUCCESS, "Err response");

	return GOS_CMD_BAD_ARGS;
}

gos_cmd_result_t command_good_command(int argc, char **argv) {
	// Format command response
	gos_cmd_format_response(GOS_CMD_SUCCESS, "Good response");

	return GOS_CMD_SUCCESS;
}

void my_callback(void *arg) {
	GOS_LOG("Callback Executed");
}

/*************************************************************************************************/
void gos_app_init(void)
{
    // Print to the serial terminal
    GOS_LOG("Hello world!");

    // Register recurring routine "my_callback" to run every 5 seconds
    gos_event_register_periodic(my_callback, NULL, 5000, GOS_EVENT_FLAG_RUN_NOW);
}

/*************************************************************************************************/
void gos_app_deinit(void)
{
    GOS_LOG("Application being de-initializing");
}

/*************************************************************************************************/
void gos_app_idle(void)
{
    // This is called when the event thread has no more pending events.
    // This API is optional.
}

Here is a capture of the serial traffic showing the commands interlaced with the logging:

[Ready]
LOCAL> Hello world!
Callback Executed
Callback Executed
good
Good response
LOCAL> Callback Executed
Callback Executed
err
Err response
LOCAL> good
Good response
LOCAL> err
Err response
LOCAL> 

As demonstrated in the above, once the "err" command is run, the callback handler either stops being executed (or possibly is unable go log via GOS_LOG. However, further execution of a command demonstrates that they are still able to write to GOS_LOG).

Hi, 

I use the AMW007 in a product to replace a Wi-Fi module that was end of life. The GUI uses the term Link quality and Link strength.

I read out the RSSI from the AMW007.

Is it possible to convert this value to signal strength/quality in %?

Is it OK to say that -100 dBm is 0% and -50 dBm is 100% as link quality? And how does Link strength fit in?

I now the RSSI is a better indication, but changing the GUI is a difficult task..

Thomas

I can't seem to claim my device to have OTA updates as it keeps on returning "invalid_credentials". I logged into the DMS using my silicon labs account which I believe may be causing the issue. Has anyone had this issue with claiming their device when using a DMS account that is linked to your Silicon Labs account?

I’ve been looking through your web site trying to understand the different Wi-Fi solutions you offer.  As a point of reference I have also been looking at Particle products, specifically their Argon modules and their Device OS and Device Cloud.  I have a few issues which I hope you can clarify for me.  I want an integrated module that contains in addition to the Wi-Fi processor an ARM core where you can write application code and run peripherals from the ARM.  I also want the module to have support to get data back and forth from the cloud, and eventually to and from a service like Azure.

1. a.  I believe your ZenrtriOS, now superseded by Gecko OS, handles the business of cloud support?

2. So Gecko OS seems equivalent to Particle Device OS/Device Cloud?

2. It seems like your Wi-Fi express modules (AMWxx7) don’t have an accessible ARM core.  The evaluation kit for them need to plug into an EFM8 main board.  The AMWxx7 cannot run as an independent radio board attached to a WSTK board.

3. a. I don’t understand the AMWxx6 modules.  They have an integrated ARM core but there are no modules available to attach to a WSTK board.  How do you develop application code for them?  It doesn’t seem like Simplicity Studio recognizes them.  The development kits for them simply provide a PC serial terminal connection, no real development environment?

b. Okay, I see there is a ZentriOS SDK to do app development.  It just doesn’t seem as robust as a full blown development environment like Simplicity Studio.  But I guess it does let you access the ARM peripherals (like SPI)? 

4. That leaves the WGM110/160P.  It seems the 110 is older (and strangely more expensive even though it has less Flash and less features, so maybe only used if you use BGScript).  The WGM160P has a radio board module (SLWRB4321A) that can be used with WSTK board for application development. 

My conclusions:  The AMWxx7 is not good for me since it is not a standalone solution. 

Unless there is a way to do (robust) application development on them (ZentriOS SDK seems to be the only solution), the AMWxx6 might not be the right choice.  Maybe I am selling ZentriOS SDK short.

The best solution might be the WGM160P (SLWRB4321A for development, and the bare module for mass production).

I appreciate any answers/opinions/comments on the above.

Hi Ayman,

I use the AMW006 module in stream mode, the baud rate is 38400.

Here is the problem that the streamed communication shows dropouts. These dropouts do not occur with direct cable connection.

I use the $ character for telegram synchronization, but this should not be bothering because the stream break sequence never occurs.

Are there any restrictions on the data content in stream mode that could disrupt the telegram traffic ?

Best regards

Holger

Hello

is there any posibility to run the ADC of the WGM160 module with more than 1ksample per second? I need a sample rate of 22ks/s or more. This is to convert an audio signal into a continuous data stream transfering via TCP.

For example: in GOS the ADC converts the input signal with a sample rate of 22kSamples / s. After a configured count of conversions GOS initiates a interrupt and the application can read this data chunk. Internal in the GOS the ADC converter works again for the next chunk until application stops this.

An simple example would be nice...

Many thanks.

Hello,

I have an issue trying to work with this board. How can I start to program apps for it ? The ZentriOS SDK/IDE seem not compatible with it...

I want to use the chip as a buffer. Its aim is to send data to another chip through wifi. The data are brought by SPI to the Zentri chip and directly send by Wifi.

Thank you for your help.

As I begin to immerse myself into Gecko I'm finding a few expected:

- Documentation inconsistencies w/ implementation
- Inconsistencies with how the code is implemented

Is there an appropriate bug tracker/product feature request forum for these types of things?  Or is here where it should be noted?

Examples...

Documentation

Application Development->Developing Applications->Structure->App Functions
  - gos_app_idle: Documentation states that it returns a boolean that determines if the execution of the app should continue.
  - However, the prototype and examples are: void gos_app_idle (void);

Code Implementations

gos_i2c_device_t 
  - Within the structure, most elements that are based on a typedef'd enum are types as such.  For example: "gos_i2c_t port;"
  - However, the variable "speed" has a type of uint32_t where there is otherwise a typedef'd enum of "gos_i2c_speed_t" that could have been used and would be helpful during development as it would improve Gecko OS IDE suggestions to include the appropriate options and documentation.
  - The "flags" variable in the same structure should have a type of gos_i2c_flag_t (I assume, as the documentation is mute on the topic of appropriate flag values)

Please note that I am not trying to be at all critical, but rather helpful in improving the overall product and developer experience.

I've previously been able to build projects, but now have the following problem.

Creating a project from the Gecko OS Studio or when cloning an existing project (blinky), without making any changes to the project I receive an immediate build error.

Steps to reproduce

1. Launch Gecko OS Studio
2. Projects->Clone Project
3. Select blinky and click Clone
4. Studio provisions the necessary files
5. Click Open in IDE
6. Right Click on the blinky project->Build/Download/Run
7. Receive the following build output

__________________________________________________________________

15:44:17 **** Build of configuration Standard Build for project blinky ****

/Users/<removed>/gecko_os_workspace/blinky/make.sh download run all 

tools/build_system/gecko_os/makefiles/parse_makegoals.mk:123: *** missing separator.  Stop.

15:44:17 Build Finished (took 187ms)

________________________________________________________

I then tried Gecko OS Studio->Settings->Clean Tools and SDK, deleted the ~/Library/Application\ Support/Gecko\ OS\ Studio, and deleted the /Applications/Gecko OS Studio

I then reinstalled and had the same issue.

OSX 10.13.6
Gecko OS Studio Version 0.9.24 (0.9.24.78)

By the way, the same steps above work correctly on the same version of Gecko OS Studio on a Windows 10 machine

I set up the AMW106 in wlan.powersafe.mode 1, system.powersafe.mode 1, automatic UDPS. After reset the modul comes up with appox. 6-8mA average current consumption. That works very nice. But after network_restart the current goes up to approx 70mA, as it has without powersafe modes activated.  Is there any method to set it back without rebooting ?

best regards

Thomas