SECCOMP-BPF: Refactor the BPF sandbox API to use fewer "static" fields and methods.

sandbox/linux/seccomp-bpf/util.cc

Index: sandbox/linux/seccomp-bpf/util.cc
diff --git a/sandbox/linux/seccomp-bpf/util.cc b/sandbox/linux/seccomp-bpf/util.cc
index 77a92d9091439127c2fce90dcaea98ef937a1da7..e84e1900dc5a5bcb865338f4a04e3057f3855c06 100644
--- a/sandbox/linux/seccomp-bpf/util.cc
+++ b/sandbox/linux/seccomp-bpf/util.cc
@@ -115,10 +115,12 @@ bool Util::GetFds(int transport, void *buf, size_t *len, ...) {
   return true;
 }
 
-void Util::CloseAllBut(int fd, ...) {
-  int proc_fd;
+void Util::CloseAllBut(int proc_fd, int fd, ...) {
   int fdir;
-  if ((proc_fd = Sandbox::proc_fd()) < 0 ||
+  if (proc_fd < 0) {
+    proc_fd = open("/proc/", O_RDONLY|O_DIRECTORY);

[jln (very slow on Chromium), 2013/02/20 01:35:49]

It really scares me to do that. We absolutely don't want to keep an open fd to
/proc.

It looks like this whole function could be deleted anyways and is not used by
demo.cc  ?

If you really want to do that, please use a ScopedFD to make sure it gets closed
before we return.

+  }
+  if (proc_fd < 0 ||
       (fdir = openat(proc_fd, "self/fd", O_RDONLY|O_DIRECTORY)) < 0) {
     SANDBOX_DIE("Cannot access \"/proc/self/fd\"");
   }