|
From: John M. <me...@ge...> - 2010-07-22 17:31:05
|
On Thu, Jul 22, 2010 at 12:06 AM, Alessandro Biasci < a.b...@ev...> wrote: > Yes this is output from console of my board. Instead, this is the output > of Valgrind on initialization for debug my program: > > --611-- WARNING: Serious error when reading debug info > --611-- When reading debug info from /home/valgrind/pcc_test: > --611-- Can't make sense of .sdata section mapping > --611-- WARNING: Serious error when reading debug info > --611-- When reading debug info > from /home/valgrind/ppcInstall/lib/valgrind/vgpreload_core-ppc32-linux.so: > --611-- Can't make sense of .sdata section mapping > --611-- WARNING: Serious error when reading debug info > --611-- When reading debug info > from > /home/valgrind/ppcInstall/lib/valgrind/vgpreload_memcheck-ppc32-linux.so: > --611-- Can't make sense of .sdata section mapping > > Valgrind seems not working well on my system but I don't know why :( > > Some help?? > You don't say what your operating system is, nor what tools you used to compile your application. You might try running valgrind on a simple command, such as "ls" (on Linux), to see if it has any issues. You might try compiling your app with gcc, if you are not using it already, with the "-g" flag. I suspect that the errors valgrind are finding are real, but it is limited as to the information it can give back in stack traces due to the warnings at the beginning. HTH! -mehaf > > Il giorno mer, 21/07/2010 alle 16.34 -0700, John Mehaffey ha scritto: > > > > > > Hi, I'm trying to run Valgrind on PowerPC (mpc885) but I have some > > > problems. > > > > > > When I start valgrind I received this message: > > > > > > [ 741.302816] Bad emulation memcheck-ppc32-/987 > > > [ 741.302925] NIP: 38030e28 instruction: 10000484 opcode: 4 A: 0 B: 0 > > > C: 12 code: 242 rc: 0 > > > [ 741.310302] pte @ 0x38030e28: > > > (0xc39b1380)->(0xc39630c0)->0x02efe889 > > > [ 741.310302] RPN: 02efe PP: 2 SPS: 1 SH: 0 CI: 0 v: 1 > > > [ 741.310302] Kernel VA for NIP c2efee28 pte @ 0xc2efee28: no pmd > > > > > These don't look like messages from Valgrind. Is this from the > > system console? > > > > The opcode 0x10000484 is the prototypical Altivec instruction > > (vor 0,0,0) that Valgrind uses to check if your system supports > > Altivec instructions. > > > > Most likely you can ignore this, Valgrind is expecting either > > a SIGILL (or not), and should get the SIGILL in this case. > > > > -mehaf > > > These are the steps for compile it: > > > > > > svn co svn://svn.valgrind.org/valgrind/branches/VALGRIND_3_5_BRANCH > > > . /usr/local/angstrom/powerpc/environment-setup > > > ./autogen.sh > > > ./configure --prefix=/home/alessandro/Siemens/Valgrind/ppc_install > > > --host=powerpc-angstrom-linux --enable-only32bit --disable-tls > > > > > > Modify the line: > > > in m_transtab.c > > > vg_assert(cls == 16 || cls == 32 || cls == 64 || cls == 128); > > > in m_machine.c > > > vg_assert(szB == 16 || szB == 32 || szB == 64 || szB == 128); > > > in guest_ppc_toIR.c > > > vassert(lineszB == 16 || lineszB == 32 || lineszB == 64 || lineszB > == > > > 128); > > > > > > > > > make > > > make install > > > > > > I have also try to run a program but no memory leak was found (there > > > were instead). > > > > > > Some Help ?? > > > > > > > > > > > > ------------------------------------------------------------------------------ > > This SF.net email is sponsored by Sprint > > What will you do first with EVO, the first 4G phone? > > Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first > > _______________________________________________ > > Valgrind-users mailing list > > Val...@li... > > https://lists.sourceforge.net/lists/listinfo/valgrind-users > > > |