On Wed, Sep 24, 2014 at 7:50 PM, Vladimir Nikulichev <[email protected]> wrote:
> More details from README: > > At first it examines process' memory layout reading /proc/pid/maps. When the > process is frozen (PTRACE_ATTACH to the main thread + SIGSTOP to start > freezing > other threads) it copies all threads' general-purpose registers and contents > of > stack from %rsp to end of memory region [..] How does this compare to the approach taken by perf tool included in the linux kernel sources? It does something similar, but doesn't use ptrace or signals. I'm guessing one difference is that you sample all the threads in the process at the same time, where as perf samples stacks of individual threads at different times depending on when they generate an event. -Arun _______________________________________________ Libunwind-devel mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/libunwind-devel
