Merge tag 'execve-v6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux

Pull execve updates from Kees Cook:
 "Most are small refactorings and bug fixes, but three things stand out:
  switching timens (which got reverted before) looks solid now,
  FOLL_FORCE has been removed (no failures seen yet across several weeks
  in -next), and some whitespace cleanups (which are long overdue).

   - Add timens support (when switching mm). This version has survived
     in -next for the entire cycle (Andrei Vagin)

   - Various small bug fixes, refactoring, and readability improvements
     (Bernd Edlinger, Rolf Eike Beer, Bo Liu, Li Zetao Liu Shixin)

   - Remove FOLL_FORCE for stack setup (Kees Cook)

   - Whitespace cleanups (Rolf Eike Beer, Kees Cook)"

* tag 'execve-v6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  binfmt_misc: fix shift-out-of-bounds in check_special_flags
  binfmt: Fix error return code in load_elf_fdpic_binary()
  exec: Remove FOLL_FORCE for stack setup
  binfmt_elf: replace IS_ERR() with IS_ERR_VALUE()
  binfmt_elf: simplify error handling in load_elf_phdrs()
  binfmt_elf: fix documented return value for load_elf_phdrs()
  exec: simplify initial stack size expansion
  binfmt: Fix whitespace issues
  exec: Add comments on check_unsafe_exec() fs counting
  ELF uapi: add spaces before '{'
  selftests/timens: add a test for vfork+exit
  fs/exec: switch timens when a task gets a new mm

1 files changed, 24 insertions, 10 deletions

diff --git a/fs/exec.c b/fs/exec.c
index a0b1f0337a62..089a743f636b 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -64,6 +64,7 @@
 #include <linux/io_uring.h>
 #include <linux/syscall_user_dispatch.h>
 #include <linux/coredump.h>
+#include <linux/time_namespace.h>
 
 #include <linux/uaccess.h>
 #include <asm/mmu_context.h>
@@ -171,7 +172,7 @@ SYSCALL_DEFINE1(uselib, const char __user *, library)
 exit:
 	fput(file);
 out:
-  	return error;
+	return error;
 }
 #endif /* #ifdef CONFIG_USELIB */
 
@@ -199,7 +200,7 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
 {
 	struct page *page;
 	int ret;
-	unsigned int gup_flags = FOLL_FORCE;
+	unsigned int gup_flags = 0;
 
 #ifdef CONFIG_STACK_GROWSUP
 	if (write) {
@@ -842,16 +843,13 @@ int setup_arg_pages(struct linux_binprm *bprm,
 	 * will align it up.
 	 */
 	rlim_stack = bprm->rlim_stack.rlim_cur & PAGE_MASK;
+
+	stack_expand = min(rlim_stack, stack_size + stack_expand);
+
 #ifdef CONFIG_STACK_GROWSUP
-	if (stack_size + stack_expand > rlim_stack)
-		stack_base = vma->vm_start + rlim_stack;
-	else
-		stack_base = vma->vm_end + stack_expand;
+	stack_base = vma->vm_start + stack_expand;
 #else
-	if (stack_size + stack_expand > rlim_stack)
-		stack_base = vma->vm_end - rlim_stack;
-	else
-		stack_base = vma->vm_start - stack_expand;
+	stack_base = vma->vm_end - stack_expand;
 #endif
 	current->mm->start_stack = bprm->p;
 	ret = expand_stack(vma, stack_base);
@@ -1297,6 +1295,10 @@ int begin_new_exec(struct linux_binprm * bprm)
 
 	bprm->mm = NULL;
 
+	retval = exec_task_namespaces();
+	if (retval)
+		goto out_unlock;
+
 #ifdef CONFIG_POSIX_TIMERS
 	spin_lock_irq(&me->sighand->siglock);
 	posix_cpu_timers_exit(me);
@@ -1568,6 +1570,12 @@ static void check_unsafe_exec(struct linux_binprm *bprm)
 	if (task_no_new_privs(current))
 		bprm->unsafe |= LSM_UNSAFE_NO_NEW_PRIVS;
 
+	/*
+	 * If another task is sharing our fs, we cannot safely
+	 * suid exec because the differently privileged task
+	 * will be able to manipulate the current directory, etc.
+	 * It would be nice to force an unshare instead...
+	 */
 	t = p;
 	n_fs = 1;
 	spin_lock(&p->fs->lock);
@@ -1748,6 +1756,7 @@ static int search_binary_handler(struct linux_binprm *bprm)
 	return retval;
 }
 
+/* binfmt handlers will call back into begin_new_exec() on success. */
 static int exec_binprm(struct linux_binprm *bprm)
 {
 	pid_t old_pid, old_vpid;
@@ -1806,6 +1815,11 @@ static int bprm_execve(struct linux_binprm *bprm,
 	if (retval)
 		return retval;
 
+	/*
+	 * Check for unsafe execution states before exec_binprm(), which
+	 * will call back into begin_new_exec(), into bprm_creds_from_file(),
+	 * where setuid-ness is evaluated.
+	 */
 	check_unsafe_exec(bprm);
 	current->in_execve = 1;