valgrind

I'm trying to compile a program with debugging symbols so that valgrind will give me line numbers. I have found that if I compile a simple test program in one go (with -g) then it contains the symbols. However, if I compile in two passes (i.e. compile then link) then it does not contain the debugging symbols. Here's the compile command for the single pass case: g++ -g file.c -o file And for two passes g++ -g -c file.c -o file.o g++ -g file.o -o file The actual program looks like this and contains a simple Invalid Write int main(){ int* x = new int[10]; x[10]=1; } If I compile with one pass

I am developing a statically-linked 64-bit C++ application on 64-bit CentOS 5.8 using the standard gcc 4.4 packages from the CentOS repositories. It appears to be using more memory than I expected, so I tried using massif to profile the memory usage. I have compiled with debug information and then run valgrind --tool=massif ./MyProg from the directory where MyProg resides. It never produces any results other than the following massif.out.XXXX example. desc: (none) cmd: ./MyProg time_unit: i #----------- snapshot=0 #----------- time=0 mem_heap_B=0 mem_heap_extra_B=0 mem_stacks_B=0 heap_tree

I have been trying to understand why valgrind complains about "Use of uninitialised value of size 8" for this small test program that uses ucontexts. It is basically a program that creates "n_ucs" ucontexts and switches over them for "max_switch" times. I understand the "Warning: client switching stacks?" (which is basically what the program all about), but I can't really make sense to all the "Use of uninitialised value of size 8" I would like to get some help understanding if Valgrind errors are false positive or if this program has something fundamentally wrong. (I see a lot of them on a

Out of curiosity I ran coded up several different versions of matrix Multiplication and ran cachegrind against it. In my results below, I was wondering which parts were L1,L2,L3 misses and references and what it all really means? Below is my code for the matrix multiplications also, in case anyone needs that. #define SLOWEST ==6933== Cachegrind, a cache and branch-prediction profiler ==6933== Copyright (C) 2002-2012, and GNU GPL'd, by Nicholas Nethercote et al. ==6933== Using Valgrind-3.8.1 and LibVEX; rerun with -h for copyright info ==6933== Command: ./a.out 500 ==6933== --6933-- warning: L3