On 7/1/2014 3:37 PM, Peter Dufault wrote:
> On Jun 30, 2014, at 14:41 , Joel Sherrill wrote:
>
>> Is your "struct handler_arguments" declared on the stack (e.g. an
>> automatic variable)? If so, then the memory is likely clobbered by
>> the time the ISR runs.
> Not necessarily. A common method wo
Hi
I think the RSB is missing a patch for newlib for 4.11 or the newlib
version needs to be bumped.
I consistently get this in SMP builds:
sparc-rtems4.11-gcc --pipe -DHAVE_CONFIG_H -I..
-I../../cpukit/../../../sis/lib/include -mcpu=cypress -O2 -g
-ffunction-sections -fdata-sections -Wall -W
So far our testing shows this addresses the scenario
we were seeing. More testing has to occur on our side
because we need to clean up the code and see if we
added any logic that is now unneeded.
But ignoring any odd style comments, this seems to
fix the issue on the SPARC.
--joel
On 7/1/2014 9:
Hello everybody,
we have stubled on next problem with GCC-4.9.0 and Newlib-2.1 when linking
RTEMS 4.11 code (but issue is generic for Newlib use for each Cortex-R
optimized multilib on any OS)
COLLECT_GCC_OPTIONS='-v' '-qrtems' '-B'
'../tms570ls3137_hdk_sdram/lib/' '-specs=bsp_specs'
'-march=
