SPWF01 Repeatable Hard Fault/crash

markb · ‎2015-10-01

Posted on October 01, 2015 at 09:51

Hi,

I am broadcasting a small UDP packet once a second through the SPWF01 using

AT+S.SOCKON=192.168.1.255,55555,u and AT+S.SOCKW and that appears to work just fine but

if I access the SPWF01 web server at the same time to reload some web pages it often dies with a hard fault like this

+WIND:8:Hard Fault:TCPIP: r0 2C303035, r1 00000300, r2 00000000, r3 0000FFFF, r12 2000867C, lr 08022E25, pc 08022E3A, psr 61000000

Or the module may just crash, reboot, etc.

If I don't do the UDP broadcast, then the module appears reliable.

This is very bad as I need to be able to do the UDP broadcast so clients can find the device.

Cheers,

Mark

markb · ‎2015-10-06

Posted on October 06, 2015 at 12:07

Sorry, setting ip_wait_timeout to 1000 didn't help. The module is still crashing when I reload index.html.

Gerardo GALLUCCI · ‎2015-10-06

Posted on October 06, 2015 at 12:14

Yes, confirm.

I saw after half an hour (SOCKW / F5).

SOCKD is ok.

I'll try to push this issue solving for next FW release.

Sorry for this

jerry

Gerardo GALLUCCI · ‎2015-10-06

Posted on October 06, 2015 at 13:44

Hi Mark,

are you waiting for ''OK'' response, before proceeding to next SOCKW?

Just to be sure, and safe...

ciao

j

markb · ‎2015-10-06

Posted on October 06, 2015 at 14:20

Ha, bingo!

That's the problem, I was waiting for the OK but timed out too soon and then another SOCKW was being set before the

first had completed and it crashed the module.

So, my bad, I don't mind admitting it but it would be better if your code code could detect the situation where a SOCKW is sent

before the result of the previous SOCKW has been received and either just ignore the second SOCKW or return ERROR.

Having the module crash on concurrent SOCKWs is not actually very helpful!

Cheers,

Mark

Gerardo GALLUCCI · ‎2015-10-06

Posted on October 06, 2015 at 16:26

Hi Mark,

glad to hear!

I agree.

Thanks

j