Merge tag 'threads-v5.5' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner...

Pull thread management updates from Christian Brauner:

 - A pidfd's fdinfo file currently contains the field "Pid:\t<pid>"
   where <pid> is the pid of the process in the pid namespace of the
   procfs instance the fdinfo file for the pidfd was opened in.

   The fdinfo file has now gained a new "NSpid:\t<ns-pid1>[\t<ns-pid2>[...]]"
   field which lists the pids of the process in all child pid namespaces
   provided the pid namespace of the procfs instance it is looked up
   under has an ancestoral relationship with the pid namespace of the
   process. If it does not 0 will be shown and no further pid namespaces
   will be listed. Tests included. (Christian Kellner)

 - If the process the pidfd references has already exited, print -1 for
   the Pid and NSpid fields in the pidfd's fdinfo file. Tests included.
   (me)

 - Add CLONE_CLEAR_SIGHAND. This lets callers clear all signal handler
   that are not SIG_DFL or SIG_IGN at process creation time. This
   originated as a feature request from glibc to improve performance and
   elimate races in their posix_spawn() implementation. Tests included.
   (me)

 - Add support for choosing a specific pid for a process with clone3().
   This is the feature which was part of the thread update for v5.4 but
   after a discussion at LPC in Lisbon we decided to delay it for one
   more cycle in order to make the interface more generic. This has now
   done. It is now possible to choose a specific pid in a whole pid
   namespaces (sub)hierarchy instead of just one pid namespace. In order
   to choose a specific pid the caller must have CAP_SYS_ADMIN in all
   owning user namespaces of the target pid namespaces. Tests included.
   (Adrian Reber)

 - Test improvements and extensions. (Andrei Vagin, me)

* tag 'threads-v5.5' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux:
  selftests/clone3: skip if clone3() is ENOSYS
  selftests/clone3: check that all pids are released on error paths
  selftests/clone3: report a correct number of fails
  selftests/clone3: flush stdout and stderr before clone3() and _exit()
  selftests: add tests for clone3() with *set_tid
  fork: extend clone3() to support setting a PID
  selftests: add tests for clone3()
  tests: test CLONE_CLEAR_SIGHAND
  clone3: add CLONE_CLEAR_SIGHAND
  pid: use pid_has_task() in pidfd_open()
  exit: use pid_has_task() in do_wait()
  pid: use pid_has_task() in __change_pid()
  test: verify fdinfo for pidfd of reaped process
  pidfd: check pid has attached task in fdinfo
  pidfd: add tests for NSpid info in fdinfo
  pidfd: add NSpid entries to fdinfo

diff --cc MAINTAINERS
Simple merge

diff --cc include/uapi/linux/sched.h
index 25b4fa00bad1f4ccc22f7dfe708a95807415b151,a0b1c224c72b556a86d8e7f92fdffca86c622e13..4a02178324641f555336164316ba2e30805c3719
--- 1/include/uapi/linux/sched.h
--- 2/include/uapi/linux/sched.h
+++ b/include/uapi/linux/sched.h
@@@ -36,28 -39,38 +39,42 @@@
  #ifndef __ASSEMBLY__
  /**
   * struct clone_args - arguments for the clone3 syscall
-  * @flags:       Flags for the new process as listed above.
-  *               All flags are valid except for CSIGNAL and
-  *               CLONE_DETACHED.
-  * @pidfd:       If CLONE_PIDFD is set, a pidfd will be
-  *               returned in this argument.
-  * @child_tid:   If CLONE_CHILD_SETTID is set, the TID of the
-  *               child process will be returned in the child's
-  *               memory.
-  * @parent_tid:  If CLONE_PARENT_SETTID is set, the TID of
-  *               the child process will be returned in the
-  *               parent's memory.
-  * @exit_signal: The exit_signal the parent process will be
-  *               sent when the child exits.
-  * @stack:       Specify the location of the stack for the
-  *               child process.
-  *               Note, @stack is expected to point to the
-  *               lowest address. The stack direction will be
-  *               determined by the kernel and set up
-  *               appropriately based on @stack_size.
-  * @stack_size:  The size of the stack for the child process.
-  * @tls:         If CLONE_SETTLS is set, the tls descriptor
-  *               is set to tls.
+  * @flags:        Flags for the new process as listed above.
+  *                All flags are valid except for CSIGNAL and
+  *                CLONE_DETACHED.
+  * @pidfd:        If CLONE_PIDFD is set, a pidfd will be
+  *                returned in this argument.
+  * @child_tid:    If CLONE_CHILD_SETTID is set, the TID of the
+  *                child process will be returned in the child's
+  *                memory.
+  * @parent_tid:   If CLONE_PARENT_SETTID is set, the TID of
+  *                the child process will be returned in the
+  *                parent's memory.
+  * @exit_signal:  The exit_signal the parent process will be
+  *                sent when the child exits.
+  * @stack:        Specify the location of the stack for the
+  *                child process.
++ *                Note, @stack is expected to point to the
++ *                lowest address. The stack direction will be
++ *                determined by the kernel and set up
++ *                appropriately based on @stack_size.
+  * @stack_size:   The size of the stack for the child process.
+  * @tls:          If CLONE_SETTLS is set, the tls descriptor
+  *                is set to tls.
+  * @set_tid:      Pointer to an array of type *pid_t. The size
+  *                of the array is defined using @set_tid_size.
+  *                This array is used to select PIDs/TIDs for
+  *                newly created processes. The first element in
+  *                this defines the PID in the most nested PID
+  *                namespace. Each additional element in the array
+  *                defines the PID in the parent PID namespace of
+  *                the original PID namespace. If the array has
+  *                less entries than the number of currently
+  *                nested PID namespaces only the PIDs in the
+  *                corresponding namespaces are set.
+  * @set_tid_size: This defines the size of the array referenced
+  *                in @set_tid. This cannot be larger than the
+  *                kernel's limit of nested PID namespaces.
   *
   * The structure is versioned by size and thus extensible.
   * New struct members must go at the end of the struct and

diff --cc kernel/fork.c
index 13b38794efb56940c18592d0ef26904d5cacb100,417570263f1fb80b50976121b751cf32b06873b3..35f91ee9105772d7653ef21cd5f7af2bcd060fb1
--- 1/kernel/fork.c
--- 2/kernel/fork.c
+++ b/kernel/fork.c
@@@ -2561,41 -2641,10 +2641,38 @@@ noinline static int copy_clone_args_fro
        return 0;
  }
  
 -static bool clone3_args_valid(const struct kernel_clone_args *kargs)
 +/**
 + * clone3_stack_valid - check and prepare stack
 + * @kargs: kernel clone args
 + *
 + * Verify that the stack arguments userspace gave us are sane.
 + * In addition, set the stack direction for userspace since it's easy for us to
 + * determine.
 + */
 +static inline bool clone3_stack_valid(struct kernel_clone_args *kargs)
 +{
 +      if (kargs->stack == 0) {
 +              if (kargs->stack_size > 0)
 +                      return false;
 +      } else {
 +              if (kargs->stack_size == 0)
 +                      return false;
 +
 +              if (!access_ok((void __user *)kargs->stack, kargs->stack_size))
 +                      return false;
 +
 +#if !defined(CONFIG_STACK_GROWSUP) && !defined(CONFIG_IA64)
 +              kargs->stack += kargs->stack_size;
 +#endif
 +      }
 +
 +      return true;
 +}
 +
 +static bool clone3_args_valid(struct kernel_clone_args *kargs)
  {
-       /*
-        * All lower bits of the flag word are taken.
-        * Verify that no other unknown flags are passed along.
-        */
-       if (kargs->flags & ~CLONE_LEGACY_FLAGS)
+       /* Verify that no unknown flags are passed along. */
+       if (kargs->flags & ~(CLONE_LEGACY_FLAGS | CLONE_CLEAR_SIGHAND))
                return false;
  
        /*

diff --cc tools/testing/selftests/Makefile
Simple merge