On 3/31/2011 11:56 AM, Thomas De Schampheleire wrote: > On Wed, Feb 23, 2011 at 6:03 AM, ANDY KENNEDY <[email protected]> wrote: >>>> >>>> So, initially it looks like this would (at least) work for mips32r2 >>>> >>> >>> Fine, please let us to know your outcomes. >> >> >> Doh! Stupid question: How would I test that this _actually_ works? >> >> I mean, I tried this: >> >> >> #include <stdio.h> >> #include <stdlib.h> >> #include <signal.h> >> #include <execinfo.h> >> >> void *barray[128]; >> >> static void segfault(int sgn) { >> size_t size = backtrace(barray, sizeof(barray)); >> fprintf(stderr, "Backtrace: \n"); >> backtrace_symbols_fd(barray, size, 2); >> abort(); >> } >> >> int f1(int *ptr) >> { >> printf("Pointer = %i\n", *ptr); >> return 0; >> } >> >> int f2(int *ptr) >> { >> f1((int *)(ptr + 0x5555aaaa)); >> return 0; >> } >> >> int main(void) >> { >> signal(11, segfault); >> f2((int *)109324); >> return 0; >> } >> >> but I only got this: >> >> Backtrace: >> Aborted >> >> >> When I compile I have to specify -lubacktrace; however, when linking >> against >> glibc, the backtrace family is apparently included by default. >> >> FYI: I get no linker errors, and no warnings. >> >> Please advise. >> >> Andy > > Hi Andy, > > Did you make any progress with backtrace in uClibc since this thread? > > Thanks, > Thomas >
Hello, I'm finally back to work on uClibc. I'm preparing some patches to make backtrace available for all archs, asking you guys to test on your platforms. Thanks, Carmelo _______________________________________________ uClibc mailing list [email protected] http://lists.busybox.net/mailman/listinfo/uclibc
