Merge tag 'ptrace_stop-cleanup-for-v5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace

Pull ptrace_stop cleanups from Eric Biederman:
 "While looking at the ptrace problems with PREEMPT_RT and the problems
  Peter Zijlstra was encountering with ptrace in his freezer rewrite I
  identified some cleanups to ptrace_stop that make sense on their own
  and move make resolving the other problems much simpler.

  The biggest issue is the habit of the ptrace code to change
  task->__state from the tracer to suppress TASK_WAKEKILL from waking up
  the tracee. No other code in the kernel does that and it is straight
  forward to update signal_wake_up and friends to make that unnecessary.

  Peter's task freezer sets frozen tasks to a new state TASK_FROZEN and
  then it stores them by calling "wake_up_state(t, TASK_FROZEN)" relying
  on the fact that all stopped states except the special stop states can
  tolerate spurious wake up and recover their state.

  The state of stopped and traced tasked is changed to be stored in
  task->jobctl as well as in task->__state. This makes it possible for
  the freezer to recover tasks in these special states, as well as
  serving as a general cleanup. With a little more work in that
  direction I believe TASK_STOPPED can learn to tolerate spurious wake
  ups and become an ordinary stop state.

  The TASK_TRACED state has to remain a special state as the registers
  for a process are only reliably available when the process is stopped
  in the scheduler. Fundamentally ptrace needs acess to the saved
  register values of a task.

  There are bunch of semi-random ptrace related cleanups that were found
  while looking at these issues.

  One cleanup that deserves to be called out is from commit 57b6de08b5f6
  ("ptrace: Admit ptrace_stop can generate spuriuos SIGTRAPs"). This
  makes a change that is technically user space visible, in the handling
  of what happens to a tracee when a tracer dies unexpectedly. According
  to our testing and our understanding of userspace nothing cares that
  spurious SIGTRAPs can be generated in that case"

* tag 'ptrace_stop-cleanup-for-v5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
  sched,signal,ptrace: Rework TASK_TRACED, TASK_STOPPED state
  ptrace: Always take siglock in ptrace_resume
  ptrace: Don't change __state
  ptrace: Admit ptrace_stop can generate spuriuos SIGTRAPs
  ptrace: Document that wait_task_inactive can't fail
  ptrace: Reimplement PTRACE_KILL by always sending SIGKILL
  signal: Use lockdep_assert_held instead of assert_spin_locked
  ptrace: Remove arch_ptrace_attach
  ptrace/xtensa: Replace PT_SINGLESTEP with TIF_SINGLESTEP
  ptrace/um: Replace PT_DTRACE with TIF_SINGLESTEP
  signal: Replace __group_send_sig_info with send_signal_locked
  signal: Rename send_signal send_signal_locked

10 files changed, 15 insertions, 75 deletions

diff --git a/arch/ia64/include/asm/ptrace.h b/arch/ia64/include/asm/ptrace.h
index a10a498eede1..402874489890 100644
--- a/arch/ia64/include/asm/ptrace.h
+++ b/arch/ia64/include/asm/ptrace.h
@@ -139,10 +139,6 @@ static inline long regs_return_value(struct pt_regs *regs)
   #define arch_ptrace_stop_needed() \
 	(!test_thread_flag(TIF_RESTORE_RSE))
 
-  extern void ptrace_attach_sync_user_rbs (struct task_struct *);
-  #define arch_ptrace_attach(child) \
-	ptrace_attach_sync_user_rbs(child)
-
   #define arch_has_single_step()  (1)
   #define arch_has_block_step()   (1)
 
diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index 4fc6e38a8459..ab8aeb34d1d9 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -618,63 +618,6 @@ void ia64_sync_krbs(void)
 }
 
 /*
- * After PTRACE_ATTACH, a thread's register backing store area in user
- * space is assumed to contain correct data whenever the thread is
- * stopped.  arch_ptrace_stop takes care of this on tracing stops.
- * But if the child was already stopped for job control when we attach
- * to it, then it might not ever get into ptrace_stop by the time we
- * want to examine the user memory containing the RBS.
- */
-void
-ptrace_attach_sync_user_rbs (struct task_struct *child)
-{
-	int stopped = 0;
-	struct unw_frame_info info;
-
-	/*
-	 * If the child is in TASK_STOPPED, we need to change that to
-	 * TASK_TRACED momentarily while we operate on it.  This ensures
-	 * that the child won't be woken up and return to user mode while
-	 * we are doing the sync.  (It can only be woken up for SIGKILL.)
-	 */
-
-	read_lock(&tasklist_lock);
-	if (child->sighand) {
-		spin_lock_irq(&child->sighand->siglock);
-		if (READ_ONCE(child->__state) == TASK_STOPPED &&
-		    !test_and_set_tsk_thread_flag(child, TIF_RESTORE_RSE)) {
-			set_notify_resume(child);
-
-			WRITE_ONCE(child->__state, TASK_TRACED);
-			stopped = 1;
-		}
-		spin_unlock_irq(&child->sighand->siglock);
-	}
-	read_unlock(&tasklist_lock);
-
-	if (!stopped)
-		return;
-
-	unw_init_from_blocked_task(&info, child);
-	do_sync_rbs(&info, ia64_sync_user_rbs);
-
-	/*
-	 * Now move the child back into TASK_STOPPED if it should be in a
-	 * job control stop, so that SIGCONT can be used to wake it up.
-	 */
-	read_lock(&tasklist_lock);
-	if (child->sighand) {
-		spin_lock_irq(&child->sighand->siglock);
-		if (READ_ONCE(child->__state) == TASK_TRACED &&
-		    (child->signal->flags & SIGNAL_STOP_STOPPED)) {
-			WRITE_ONCE(child->__state, TASK_STOPPED);
-		}
-		spin_unlock_irq(&child->sighand->siglock);
-	}
-	read_unlock(&tasklist_lock);
-}
-
-/*
  * Write f32-f127 back to task->thread.fph if it has been modified.
  */
 inline void
diff --git a/arch/um/include/asm/thread_info.h b/arch/um/include/asm/thread_info.h
index 1395cbd7e340..c7b4b49826a2 100644
--- a/arch/um/include/asm/thread_info.h
+++ b/arch/um/include/asm/thread_info.h
@@ -60,6 +60,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_RESTORE_SIGMASK	7
 #define TIF_NOTIFY_RESUME	8
 #define TIF_SECCOMP		9	/* secure computing */
+#define TIF_SINGLESTEP		10	/* single stepping userspace */
 
 #define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
@@ -68,5 +69,6 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_MEMDIE		(1 << TIF_MEMDIE)
 #define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1 << TIF_SECCOMP)
+#define _TIF_SINGLESTEP		(1 << TIF_SINGLESTEP)
 
 #endif
diff --git a/arch/um/kernel/exec.c b/arch/um/kernel/exec.c
index c85e40c72779..58938d75871a 100644
--- a/arch/um/kernel/exec.c
+++ b/arch/um/kernel/exec.c
@@ -43,7 +43,7 @@ void start_thread(struct pt_regs *regs, unsigned long eip, unsigned long esp)
 {
 	PT_REGS_IP(regs) = eip;
 	PT_REGS_SP(regs) = esp;
-	current->ptrace &= ~PT_DTRACE;
+	clear_thread_flag(TIF_SINGLESTEP);
 #ifdef SUBARCH_EXECVE1
 	SUBARCH_EXECVE1(regs->regs);
 #endif
diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c
index 181cc9aafb25..80b90b1276a1 100644
--- a/arch/um/kernel/process.c
+++ b/arch/um/kernel/process.c
@@ -336,7 +336,7 @@ int singlestepping(void * t)
 {
 	struct task_struct *task = t ? t : current;
 
-	if (!(task->ptrace & PT_DTRACE))
+	if (!test_thread_flag(TIF_SINGLESTEP))
 		return 0;
 
 	if (task->thread.singlestep_syscall)
diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c
index bfaf6ab1ac03..5154b27de580 100644
--- a/arch/um/kernel/ptrace.c
+++ b/arch/um/kernel/ptrace.c
@@ -11,7 +11,7 @@
 
 void user_enable_single_step(struct task_struct *child)
 {
-	child->ptrace |= PT_DTRACE;
+	set_tsk_thread_flag(child, TIF_SINGLESTEP);
 	child->thread.singlestep_syscall = 0;
 
 #ifdef SUBARCH_SET_SINGLESTEPPING
@@ -21,7 +21,7 @@ void user_enable_single_step(struct task_struct *child)
 
 void user_disable_single_step(struct task_struct *child)
 {
-	child->ptrace &= ~PT_DTRACE;
+	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
 	child->thread.singlestep_syscall = 0;
 
 #ifdef SUBARCH_SET_SINGLESTEPPING
@@ -120,7 +120,7 @@ static void send_sigtrap(struct uml_pt_regs *regs, int error_code)
 }
 
 /*
- * XXX Check PT_DTRACE vs TIF_SINGLESTEP for singlestepping check and
+ * XXX Check TIF_SINGLESTEP for singlestepping check and
  * PT_PTRACED vs TIF_SYSCALL_TRACE for syscall tracing check
  */
 int syscall_trace_enter(struct pt_regs *regs)
@@ -144,7 +144,7 @@ void syscall_trace_leave(struct pt_regs *regs)
 	audit_syscall_exit(regs);
 
 	/* Fake a debug trap */
-	if (ptraced & PT_DTRACE)
+	if (test_thread_flag(TIF_SINGLESTEP))
 		send_sigtrap(&regs->regs, 0);
 
 	if (!test_thread_flag(TIF_SYSCALL_TRACE))
diff --git a/arch/um/kernel/signal.c b/arch/um/kernel/signal.c
index 88cd9b5c1b74..ae4658f576ab 100644
--- a/arch/um/kernel/signal.c
+++ b/arch/um/kernel/signal.c
@@ -53,7 +53,7 @@ static void handle_signal(struct ksignal *ksig, struct pt_regs *regs)
 	unsigned long sp;
 	int err;
 
-	if ((current->ptrace & PT_DTRACE) && (current->ptrace & PT_PTRACED))
+	if (test_thread_flag(TIF_SINGLESTEP) && (current->ptrace & PT_PTRACED))
 		singlestep = 1;
 
 	/* Did we come from a system call? */
@@ -128,7 +128,7 @@ void do_signal(struct pt_regs *regs)
 	 * on the host.  The tracing thread will check this flag and
 	 * PTRACE_SYSCALL if necessary.
 	 */
-	if (current->ptrace & PT_DTRACE)
+	if (test_thread_flag(TIF_SINGLESTEP))
 		current->thread.singlestep_syscall =
 			is_syscall(PT_REGS_IP(&current->thread.regs));
 
diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index 0f3c307b37b3..8e2b2552b5ee 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -180,8 +180,7 @@ void set_task_blockstep(struct task_struct *task, bool on)
 	 *
 	 * NOTE: this means that set/clear TIF_BLOCKSTEP is only safe if
 	 * task is current or it can't be running, otherwise we can race
-	 * with __switch_to_xtra(). We rely on ptrace_freeze_traced() but
-	 * PTRACE_KILL is not safe.
+	 * with __switch_to_xtra(). We rely on ptrace_freeze_traced().
 	 */
 	local_irq_disable();
 	debugctl = get_debugctlmsr();
diff --git a/arch/xtensa/kernel/ptrace.c b/arch/xtensa/kernel/ptrace.c
index 22cdaa6729d3..f29477162ede 100644
--- a/arch/xtensa/kernel/ptrace.c
+++ b/arch/xtensa/kernel/ptrace.c
@@ -224,12 +224,12 @@ const struct user_regset_view *task_user_regset_view(struct task_struct *task)
 
 void user_enable_single_step(struct task_struct *child)
 {
-	child->ptrace |= PT_SINGLESTEP;
+	set_tsk_thread_flag(child, TIF_SINGLESTEP);
 }
 
 void user_disable_single_step(struct task_struct *child)
 {
-	child->ptrace &= ~PT_SINGLESTEP;
+	clear_tsk_thread_flag(child, TIF_SINGLESTEP);
 }
 
 /*
diff --git a/arch/xtensa/kernel/signal.c b/arch/xtensa/kernel/signal.c
index c9ffd42db873..876d5df157ed 100644
--- a/arch/xtensa/kernel/signal.c
+++ b/arch/xtensa/kernel/signal.c
@@ -472,7 +472,7 @@ static void do_signal(struct pt_regs *regs)
 		/* Set up the stack frame */
 		ret = setup_frame(&ksig, sigmask_to_save(), regs);
 		signal_setup_done(ret, &ksig, 0);
-		if (current->ptrace & PT_SINGLESTEP)
+		if (test_thread_flag(TIF_SINGLESTEP))
 			task_pt_regs(current)->icountlevel = 1;
 
 		return;
@@ -498,7 +498,7 @@ static void do_signal(struct pt_regs *regs)
 	/* If there's no signal to deliver, we just restore the saved mask.  */
 	restore_saved_sigmask();
 
-	if (current->ptrace & PT_SINGLESTEP)
+	if (test_thread_flag(TIF_SINGLESTEP))
 		task_pt_regs(current)->icountlevel = 1;
 	return;
 }