Takumi Ando

@masayuki2009 I(actually our team) develop two editions of the IoT devices(same SoC) running NuttX. The version A uses internal RTC and the version B should use external one. But I...

@xiaoxiang781216 Thank you for your kind explanation! I understand your idea. But many internal RTC lower-half drivers are enabled with each ```up_rtc_xxx()``` implementations at the same time. For example: ```/arch/arm/src/stm32/Make.defs```...

@xiaoxiang781216 Thank you, I will try them! By the way, - Why do lower-half drivers depend on ```up_rtc_xxx()``` implementations in ```Make.defs```? We cannot enable only lower-half in current codes. -...

>arch_rtc.c is introduced very later. Before that, all RTC driver need not only implement rtc_lowerhalf_s, but also up_rtc_xxx. So, lower-half drivers for external RTCs(under ```/drivers/timers/```) are JUST NOT implemented yet,...

I couldn't find the issue form of https://github.com/sonydevworld/spresense-nuttx. So I posted here.

@SPRESENSE I understand. Thank you.

This bug is similar to https://github.com/libcsp/libcsp/issues/695. The timing of the execution of ```csp_id_strip()```(this means ```timestamp_rx = 0```) differs between CSP V1 and V2. So ```cfpid``` will not be broken in...

@yashi ### Apply this patch to use CSP v1 ```diff diff --git a/examples/csp_client.c b/examples/csp_client.c index 37625a3..b59a804 100644 --- a/examples/csp_client.c +++ b/examples/csp_client.c @@ -139,6 +139,8 @@ int main(int argc, char *...

Yes. Currently, this bug happens only with CAN interface with CSP v1. However, I think we need to choose one of the following options. - Remove union from ```csp_packet_t```(https://github.com/libcsp/libcsp/pull/710) -...

Sorry, my explanation was wrong. >OTOH, a USART interface seems to work: - A USART interface doesn't use ```cfpid```, so this bug doesn't happen. - A CAN interface with CSP...