Hi David, Mmap with PROT_NONE does not affected the core dump size.
Here is a simple test prog: #include <sys/mman.h> #include <time.h> #include <stdint.h> #include <unistd.h> #include <stdlib.h> #include <stdio.h> int main(int argc, char** argv) { // FIXME(fengli): XXXXX uint64_t gb = atoi(argv[1]); void* ptr = mmap(0, gb << 30, PROT_NONE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); if (ptr == (void*)-1) { perror("[-] mmap failed with MAP_PRIVATE | MAP_ANONYMOUS"); exit(1); } while(1) sleep(1); return 0; } Thanks. David Marchand <david.march...@redhat.com> 于2020年4月21日周二 下午3:41写道: > > On Mon, Apr 20, 2020 at 11:41 AM Feng Li <lifeng1...@gmail.com> wrote: > > > > Thank you, Marchand, > > David is fine. > > > > > I have just tested your patch, and it doesn't work. > > The core dump file is still very very large, the same to virtual memory > > size. > > Please double check that the patch is in. > I remember checking coredump size with my patch. > > This needs more investigation but I don't have time atm. > Anatoly, opinion? > > > -- > David Marchand >