On Sun, 2010-12-12 at 21:53 +0000, Alan Bowman wrote: > > hal/cortexm/arch/current/include/hal_arch.h:294 > > > > #ifndef HAL_IDLE_THREAD_ACTION > > #define HAL_IDLE_THREAD_ACTION(__count) __asm__ volatile ( "wfi\n" ) > > #endif > > If your issue is with the WFI instruction (and it sounds like it is), > then the STM32 chips (or is it the Cortex itself?) provide support for > maintaining a clock to keep JTAG operating when the chip is in its low > power modes - the default is to stop the clock and then the JTAG > connection is lost. You need to write to the DBGMCU_CR register, as > detailed in 30.16 of the RM0008 Reference Manual for the chips. I > certainly found I had issues unless I wrote this at the point where my > debugger connected to the target. I think that this is better done as > an openocd/jtag script item than as a modification to the eCos idle > thread handling, although probably a note in the documentation to that > effect would be worthwhile.
Thanks for your answers! I was already suspecting something like that but didn't look into the header-files, only in the C files and didn't find anything. I'll try that out tonight and report back if it didn't solve it, but I guess this is the problem. Cheers, Manuel -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
