Part of code only executed if I toggle breakpoints

[jmf1]

Posted on September 10, 2016 at 08:21

Hello,

Trying to debug my application in [DEAD LINK /public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Ausio%20async%20USB%20attempt%20not%20working%20-%20feedback%20not%20read%20andor%20no%20effect&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=53]Audio async USB attempt not working - feedback not read and/or no effect I have a very strange behaviour that I don't know how to interpret. Basically, if I run my code, some parts of my code are not executed. I have LEDs that are set to ON at call of functions, and LEDs stay OFF. But at some point, if I toggle some breakpoint, even in some other parts of the code, then the function will be called ! This is about Audio USB and the TX of feedback to the Host. DataIn and ISOIncomplete are not called when I prime the feedback, but If I toggle the breakpoints, then they are at least one time. If I have set breakpoints at SOF, DataIn and ISOInc, then I see the actuation of the breakpoints ''in some sequence''. It looks like if there was some masking of the interrupts below those calls, and if slowing down the execution was allowing the call. But I don't know where to look... I use After further investigation, using STM32Cube_FW_F4_V1.13.0 and a modified version of usbd_audio.c MCD Application Team V2.4.2 11-December-2015, and I have modified the usbd_core.c to have the IsoIncomplete callback effective. Thanks a lot for heading me in some direction... JM

[waclawek.jan]

Posted on September 11, 2016 at 09:46

> It looks like if there was some masking of the interrupts below those calls, and if slowing down the execution was allowing the call.

I am not attempting to understand your application, but generally if you stop program execution and hardware continues to run, you start getting things like overflows. JW