Re: "net arp" command displays "ARP cache is empty" even after a series of arpings from peer devices
#nrf52840
Jukka Rissanen
Hi, from the attached .config file, the zephyr IP address is 10.128.8.81, and if you ping it from your network, you get a ICMP reply correctly but still no entry in arp cache in zephyr, is this correct? There definitely should be entry in arp cache as otherwise zephyr would not be able to send the reply back. Are you sure that you are pinging the correct device in this case? Cheers, Jukka
On Fri, 2019-05-31 at 04:00 -0700, giriprasad@... wrote: Hi,
|
|
Re: [Zephyr-announce] Zephyr SDK 0.10.1-rc1 available
Kumar Gala
On Jun 6, 2019, at 6:26 AM, Paul Sokolovsky <paul.sokolovsky@linaro.org> wrote:Is there a newer Newlib for Xtensa available? - k
|
|
Upcoming Event: Zephyr Project: Dev Meeting - Thu, 06/06/2019 8:00am-9:00am, Please RSVP
#cal-reminder
devel@lists.zephyrproject.org Calendar <devel@...>
Reminder: Zephyr Project: Dev Meeting When: Thursday, 6 June 2019, 8:00am to 9:00am, (GMT-07:00) America/Los Angeles Where:https://zoom.us/j/993312203 An RSVP is requested. Click here to RSVP Organizer: devel@... Description: Join Zoom Meeting
|
|
Re: [Zephyr-announce] Zephyr SDK 0.10.1-rc1 available
Paul Sokolovsky
Hello Kumar,
On Thu, 6 Jun 2019 06:01:27 -0500 "Kumar Gala" <kumar.gala@linaro.org> wrote: Hi, Can we please before the release do something about very outdated Newlib in Xtensa SDK toolchain: https://github.com/zephyrproject-rtos/sdk-ng/issues/64 ? That complicates most of POSIX subsys work, and requires landing ugly workarounds in the main codebase. Thanks, Paul Linaro.org | Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog
|
|
Zephyr SDK 0.10.1-rc1 available
Kumar Gala
Hi,
Latest version of the SDK can be found here: https://github.com/zephyrproject-rtos/sdk-ng/releases/tag/v0.10.1-rc1 Please download and try things out and report any issues. Changes since the last release: • --enable-initfini-array to gcc configure • Fix bossa (revert back to 1.9.1) • Enable multilibs on RISC-V • Fix inclusion of MIPS toolchain • Add x86_64 toolchain • Fix assembler issue for Cortex M-23 • Fixes for openocd on ARC • Updated TI CC32xx openocd scripts Thanks to all that contributed fixes and enhancements to this version of the SDK. - k
|
|
DEV meeting topic
Puzdrowski, Andrzej
HI all,
As you know, the Zephyr Dev-review meeting takes place on Thursday 9AM-10AM (PDT) (18h-19h (CEST)).
We want to cover topics related to the settings pattern load PR: https://github.com/zephyrproject-rtos/zephyr/pull/16609/
We want to be sure that direction in which we want enhance settings will fit for the needs and can be accepted by the majority.
Thanks,
Andrzej
|
|
Re: Soft serial 9k6 for debugging on free GPIO?
Piotr Mienkowski
Hi Martijn,
On 31.05.19 15:53, martijn@oblivion.nl wrote: We have a prototype utilising the nRF52840 BME340 SoC module. Since weSince nRF52840 is a Cortex-M4 device yet another option that you have is to use Serial Wire Output (SWO). ARM added it specifically to support trace/logging. It is a high speed interface with low code footprint / resource usage. Zephyr logger has SWO backend but you would need to prepare a PR to add support for Nordic device(s). It shouldn't be too difficult. At minimum you would need to select 'HAS_SWO' Kconfig option and - looking quickly at the documentation, there may be things I missed - set TRACECONFIG.TRACEMUX = 1. Regards, Piotr
|
|
Upcoming Event: Zephyr Project: APIs - Tue, 06/04/2019 9:00am-10:00am, Please RSVP
#cal-reminder
devel@lists.zephyrproject.org Calendar <devel@...>
Reminder: Zephyr Project: APIs When: Tuesday, 4 June 2019, 9:00am to 10:00am, (GMT-07:00) America/Los Angeles Where:https://zoom.us/j/177647878 An RSVP is requested. Click here to RSVP Organizer: devel@... Description: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/177647878 Live meeting minutes: https://docs.google.com/
|
|
Re: Soft serial 9k6 for debugging on free GPIO?
Martijn Smit
Hi Henrik,
toggle quoted messageShow quoted text
RTT is the way to move forward as it seems, since it will provide a clean interface both on hardware and config. Thanks for pointing me there, and thanks for the soft-serial pointers from others in this thread as it might come in handy. So, I've tried for the last two days to set RTT up, using the blinky or logging examples using a PCA10059 nRF52840 dongle just to experiment. I'm not connecting the dongle using USB but with a tag-connect TC2050-IDC-NL on the Segger J-Link with the pogo-pins directly on the board. If I only use the CONFIG_RTT_CONSOLE=y nothing happens and the JLinkRTTLogger keeps searching for the RTT Control Block. Nothing happens. If I take the log sample from the Zephyr project and use the prj_rtt.conf setting provided I run into different issues: CONFIG_LOG=y CONFIG_LOG_RUNTIME_FILTERING=y CONFIG_LOG_BUFFER_SIZE=2048 CONFIG_LOG_PRINTK=y CONFIG_LOG_PROCESS_TRIGGER_THRESHOLD=0 CONFIG_LOG_BACKEND_RTT=y CONFIG_USE_SEGGER_RTT=y CONFIG_COVERAGE=n When using the Nordic SDK I'm getting into a world of hurt getting the Segger to compile in as this is triggered by the CONFIG_USE_SEGGER_RTT directive which can be fixed adding more include directories. The RTT tarball from Segger J-Link install is more forgiven, but I cannot get my project to compile since it eventually boils down to a missing include file on the ninja ccache step ... [109/116] Building C object zephyr/CMakeFiles/zephyr.dir/drivers/console/rtt_console.c.obj FAILED: zephyr/CMakeFiles/zephyr.dir/drivers/console/rtt_console.c.obj ccache /usr/local/bin/arm-none-eabi-gcc -DBUILD_VERSION=zephyr-v1.14.0-1238-gbd001081fce1 -DKERNEL -DNRF52840_XXAA -D_FORTIFY_SOURCE=2 -D__ZEPHYR__=1 -I../../../../../kernel/include -I../../../../../arch/arm/include -I../../../../../include -I../../../../../include/drivers -Izephyr/include/generated -I../../../../../soc/arm/nordic_nrf/nrf52 -I../../../../../soc/arm/nordic_nrf/include -I../../../../../lib/libc/minimal/include -I../../../../../ext/hal/cmsis/Include -I../../../../../ext/hal/nordic/nrfx -I../../../../../ext/hal/nordic/nrfx/drivers/include -I../../../../../ext/hal/nordic/nrfx/hal -I../../../../../ext/hal/nordic/nrfx/mdk -I../../../../../ext/hal/nordic/. -isystem /usr/local/Cellar/arm-none-eabi-gcc/8-2018-q4-major/gcc/bin/../lib/gcc/arm-none-eabi/8.2.1/include -isystem /usr/local/Cellar/arm-none-eabi-gcc/8-2018-q4-major/gcc/bin/../lib/gcc/arm-none-eabi/8.2.1/include-fixed -Os -nostdinc -g -imacros /Users/smeedy/code/zephyrproject/zephyr/samples/subsys/logging/logger/build/zephyr/include/generated/autoconf.h -ffreestanding -fno-common -mthumb -mcpu=cortex-m4 -Wall -Wformat -Wformat-security -Wno-format-zero-length -Wno-main -Wno-pointer-sign -Wpointer-arith -Wno-unused-but-set-variable -Werror=implicit-int -fno-asynchronous-unwind-tables -fno-pie -fno-pic -fno-strict-overflow -fno-reorder-functions -fno-defer-pop -fmacro-prefix-map=/Users/smeedy/code/zephyrproject/zephyr=. -ffunction-sections -fdata-sections -mabi=aapcs -march=armv7e-m -std=c99 -MD -MT zephyr/CMakeFiles/zephyr.dir/drivers/console/rtt_console.c.obj -MF zephyr/CMakeFiles/zephyr.dir/drivers/console/rtt_console.c.obj.d -o zephyr/CMakeFiles/zephyr.dir/drivers/console/rtt_console.c.obj -c /Users/smeedy/code/zephyrproject/zephyr/drivers/console/rtt_console.c /Users/smeedy/code/zephyrproject/zephyr/drivers/console/rtt_console.c:16:10: fatal error: SEGGER_RTT.h: No such file or directory Could be this is going to be a cmake question (sorry for that). I added stuff like include_directories(...) and others into the CMakeLists.txt file, but I cannot seem to get this going since the extra include dirs keep on missing. Any pointers are welcome, Thanks again for your time, Martijn On 01/06/2019, 14:11, "devel@lists.zephyrproject.org on behalf of Henrik Brix Andersen" <devel@lists.zephyrproject.org on behalf of henrik@brixandersen.dk> wrote: Hi Martijn, If you are using a Segger J-Link probe you can use the RAM-based RTT console by setting CONFIG_RTT_CONSOLE=y See https://docs.zephyrproject.org/latest/guides/debugging/host-tools.html#jlink-debug-host-tools for more information. ./Brix -- Henrik Brix Andersen
On 31 May 2019, at 15.53, martijn@oblivion.nl wrote:> > Hi there, > > We have a prototype utilising the nRF52840 BME340 SoC module. Since we use a LoRaWAN module and a MAX3485 we used the two 840 uarts for connecting to these subsystems. But I do have some free GPIOs on the prototype PCB which I'd like to use as a console when running the project in debug mode. So I'm looking for a way to have a low-speed soft-serial taking care of logging. Any pointers, or perhaps other ways to get logging output (we are using BLE)? > > Looked at some discussions on the Nordic devzone, but not a lot going for it. > > Thanks, > Martijn >
|
|
IBM Watson IoT Platform ( Bluemix)
Hi All,
I thought I would have a look at the IBM Watson IoT Platform (Bluemix) but I can't seem to connect. I followed the guides but I must be missing something. I have connected to the Quick start test systems yet haven't been able to make any headway with the real version. Any help would be greatly appropriated. Billy..
|
|
Re: API meeting: agenda
alpi@...
Thank you very much for the agenda Carles.
toggle quoted messageShow quoted text
With it, it should be much clearer to all what to expect from the meeting, and can easily decide about joining or not BR Alberto
-----Original Message-----
From: devel@lists.zephyrproject.org [mailto:devel@lists.zephyrproject.org] On Behalf Of Cufi, Carles via Lists.Zephyrproject.Org Sent: Tuesday 4 June 2019 11:42 To: devel@lists.zephyrproject.org; users@lists.zephyrproject.org Cc: devel@lists.zephyrproject.org Subject: [Zephyr-devel] API meeting: agenda Agenda: - Flash: write protection and read alignment unification - GPIO: Proposal review Additional items in the "Triage" column in the GitHub project may be discussed if time permits. If you want an item included in the meeting, please add it to the GitHub project. https://github.com/zephyrproject-rtos/zephyr/wiki/Zephyr-Committee-and-Working-Group-Meetings#zephyr-api-discussion https://github.com/zephyrproject-rtos/zephyr/projects/18 https://docs.google.com/document/d/1lv-8B5QE2m4FjBcvfqAXFIgQfW5oz6306zJ7GIZIWCk/edit Regards, Carles
|
|
API meeting: agenda
Carles Cufi
Agenda:
- Flash: write protection and read alignment unification - GPIO: Proposal review Additional items in the "Triage" column in the GitHub project may be discussed if time permits. If you want an item included in the meeting, please add it to the GitHub project. https://github.com/zephyrproject-rtos/zephyr/wiki/Zephyr-Committee-and-Working-Group-Meetings#zephyr-api-discussion https://github.com/zephyrproject-rtos/zephyr/projects/18 https://docs.google.com/document/d/1lv-8B5QE2m4FjBcvfqAXFIgQfW5oz6306zJ7GIZIWCk/edit Regards, Carles
|
|
Re: Soft serial 9k6 for debugging on free GPIO?
Hi Martinj,
This Zephyr example does UART over GPIO: https://github.com/NXPmicro/zephyr/tree/soft_periph/samples/soft_periph Here's a picture with the system overview: https://www.zephyrproject.org/elc-openiot-nxp-demo/ However it receives the data to send via RPMsg from a second core running Linux and vice versa. It could be possible to remove the RPMsg related part and rewrite the code into a driver and select that peripheral as a debug console in board's DTS file. The code requires a hardware timer and supports just 8N1 format. Best regards, Stana
|
|
Re: Soft serial 9k6 for debugging on free GPIO?
Henrik Brix Andersen
Hi Martijn,
toggle quoted messageShow quoted text
If you are using a Segger J-Link probe you can use the RAM-based RTT console by setting CONFIG_RTT_CONSOLE=y See https://docs.zephyrproject.org/latest/guides/debugging/host-tools.html#jlink-debug-host-tools for more information. ./Brix -- Henrik Brix Andersen
On 31 May 2019, at 15.53, martijn@oblivion.nl wrote:
|
|
Re: Soft serial 9k6 for debugging on free GPIO?
Jian Zhang
Hi Martijn,
Not sure about the exact situation you are facing, listed below please find some of the suggestions I can think of: 1. There are soft uart implementations on github, which you might be able to port to zephyr; 2. If you got SPI interface on your module, there are bridge modules available to convert i2c/spi to uart; // sc16is740 etc 3. If you only care about the application level log, probably you can use bluetooth to build a virtual serial port for logging purpose; Best Regards, Jian
|
|
Soft serial 9k6 for debugging on free GPIO?
Martijn Smit
Hi there,
We have a prototype utilising the nRF52840 BME340 SoC module. Since we use a LoRaWAN module and a MAX3485 we used the two 840 uarts for connecting to these subsystems. But I do have some free GPIOs on the prototype PCB which I'd like to use as a console when running the project in debug mode. So I'm looking for a way to have a low-speed soft-serial taking care of logging. Any pointers, or perhaps other ways to get logging output (we are using BLE)? Looked at some discussions on the Nordic devzone, but not a lot going for it. Thanks, Martijn
|
|
"net arp" command displays "ARP cache is empty" even after a series of arpings from peer devices
#nrf52840
giriprasad@...
Hi,
I have interfaced "ENC28J60" to "PCA10056"(NRF52840) through SPI. Made necessary configurations in order to reflect hardware changes in software. Flashed sample application, "dumb_http_server" to the board. Enabled "ARP" and "ICMP". I am able to ping the board from peer devices in network and vice versa. Also, I am able to "arping" the board from peer devices in network. After this, I have issued "net arp" command in the serial console of the board. By this, I am expecting a list of peer devices to be displayed on the console. But console throws a message saying "ARP cache is empty". Can I know the reason for this behavior? Please let me know, if I was wrong in the process. Also, let me know if more information is needed. Attached configuration file for reference. Thanks & Regards, Giri Prasad N.
|
|
Upcoming Event: Zephyr Project: Dev Meeting - Thu, 05/30/2019 8:00am-9:00am, Please RSVP
#cal-reminder
devel@lists.zephyrproject.org Calendar <devel@...>
Reminder: Zephyr Project: Dev Meeting When: Thursday, 30 May 2019, 8:00am to 9:00am, (GMT-07:00) America/Los Angeles Where:https://zoom.us/j/993312203 An RSVP is requested. Click here to RSVP Organizer: devel@... Description: Join Zoom Meeting
|
|
Where is the proper place to update the RPL HBHO
Jian Zhang
Hi all,
I am working on bring back RPL to Zephyr, which includes migrating RPL to use net_pkt API. The old RPL implementation was using net_pkt_insert() to insert RPL HBHO header when finalizing the udp packet. However, I can see that the net_pkt_insert() was removed already. I am not very familiar with the Zephyr network architecture. One way I can think about is to update the RPL HBHO in net_udp_create(), though not sure if this is the right way to go ahead. Please help. Thanks and Best Regards, Jian
|
|
Re: Logging with a string via %s
Chruściński, Krzysztof
Hi,
toggle quoted messageShow quoted text
Logger is now able to detect lack of log_strdup() wrapping when %s is used with string coming from read write memory. Logger will assert if asserts are enabled or log error message if asserts are disabled. Feature is optional, enabled by default: https://github.com/zephyrproject-rtos/zephyr/pull/16310 Krzysztof
-----Original Message-----
From: devel@lists.zephyrproject.org <devel@lists.zephyrproject.org> On Behalf Of Chruscinski, Krzysztof via Lists.Zephyrproject.Org Sent: Tuesday, May 21, 2019 7:37 AM To: marc.herbert@intel.com; Dunaj, Pawel <Pawel.Dunaj@nordicsemi.no> Cc: devel@lists.zephyrproject.org Subject: Re: [Zephyr-devel] Logging with a string via %s Hi, Logger was redesigned to work in deferred mode (string processing and transport deferred to idle stage) to be capable of handling multiple backends and to be less intrusive than standard in-place approach. That is because Zephyr is targeting embedded systems where logger may be extensively used from interrupt context. This approach reduces logging time from hundreds of microseconds to just few, making it non-intrusive (compared to in-place logging). Additionally, deferred logging allows using multiple backends, including some which are slow (e.g. flashlog) or more complex (e.g. Bluetooth). In those cases in place logging wouldn't be possible at all. In order to achieve that, only arguments are stored when log message is created. %s is the victim of that but, as already mentioned, there are some countermeasures to take care of that: immediate mode (blocking) and log_strdup for duplicating transient string. For native_posix board immediate mode is the default one, maybe it should be the default for qemu boards as well but making it default for boards with SoC would soon lead to similar complaining email (like "I've enabled logging and my system crashed"). There are countermeasures that could be added that could help with that: - detection of lack of log_strdup during log processing. Based on the string address it could be determined if it's in .rodata or log_strdup buffer pool and if not warning/assert could occur - adding custom format specifiers for address (network, Bluetooth) since in many cases %s was used for locally rendered address. Krzysztof -----Original Message----- From: devel@lists.zephyrproject.org <devel@lists.zephyrproject.org> On Behalf Of Marc Herbert via Lists.Zephyrproject.Org Sent: Monday, May 20, 2019 7:00 PM To: Dunaj, Pawel <Pawel.Dunaj@nordicsemi.no> Cc: devel@lists.zephyrproject.org Subject: Re: [Zephyr-devel] Logging with a string via %s On 20 May 2019, at 00:20, pawel.dunaj@nordicsemi.no wrote: There's a violation of the "Least Astonishment" principle but it's not with the '%s'. The '%s' traditionally refers to the type of the argument only, it says nothing about memory management. I don't think LOG_ASYNC('%s', ...) or LOG(O_NONBLOCK, '%s', ... ) would surprise anyone. Confusing memory semantics with format specifiers would break not just user expectations but also tools like this: https://gcc.gnu.org/onlinedocs/gcc-9.1.0/gcc/Common-Function-Attributes.html#index-format-function-attribute (I found an insane number of bugs with this attribute) + other static checkers trying to make C a bit more safe. BTW the current API is neither "asynchronous" nor "non-blocking" in the usual API meanings. It's not asynchronous in the usual sense because there's no callback or any other second step to know when the memory can be freed or re-used. It's not non-blocking because it always succeeds, never asks to retry with EAGAIN/EWOULDBLOCK etc. It's a "globals-only", "UFO" API like nothing else.
|
|