This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

check_stack during thread context switch


Hello,

in my ecos application (powerpc, timesliced, nultithread) the routine check_stack is asserting telling me:

Stack base corrupt - i: 2
007E7550: 00 7E 75 50 00 00 00 20  00 7E 75 68 03 7D 96 7B  |.~uP... .~uh.}.{|
007E7560: 04 7A 91 7C 00 7E 75 68  00 7E 75 A8 00 19 89 C4  |.z.|.~uh.~u.....|
ASSERT FAIL: <8>thread.inl[119]void Cyg_HardwareThread::check_stack() Stack base corrupt


I checked the stack size but it seams to be enough.

I feel, any body is writing in memory unpredictable. Maybe with an uninitiasilsed or wrong pointer. Or it is writing over the end of an array alocated in the stack.

To catch the originator I want that ecos checks the stack whenever it switches the thread context.

How can I say to eCos it should do it?

Best Regars
Adolfo


--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]