Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/signal

Pull third pile of kernel_execve() patches from Al Viro:
 "The last bits of infrastructure for kernel_thread() et.al., with
  alpha/arm/x86 use of those.  Plus sanitizing the asm glue and
  do_notify_resume() on alpha, fixing the "disabled irq while running
  task_work stuff" breakage there.

  At that point the rest of kernel_thread/kernel_execve/sys_execve work
  can be done independently for different architectures.  The only
  pending bits that do depend on having all architectures converted are
  restrictred to fs/* and kernel/* - that'll obviously have to wait for
  the next cycle.

  I thought we'd have to wait for all of them done before we start
  eliminating the longjump-style insanity in kernel_execve(), but it
  turned out there's a very simple way to do that without flagday-style
  changes."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/signal:
  alpha: switch to saner kernel_execve() semantics
  arm: switch to saner kernel_execve() semantics
  x86, um: convert to saner kernel_execve() semantics
  infrastructure for saner ret_from_kernel_thread semantics
  make sure that kernel_thread() callbacks call do_exit() themselves
  make sure that we always have a return path from kernel_execve()
  ppc: eeh_event should just use kthread_run()
  don't bother with kernel_thread/kernel_execve for launching linuxrc
  alpha: get rid of switch_stack argument of do_work_pending()
  alpha: don't bother passing switch_stack separately from regs
  alpha: take SIGPENDING/NOTIFY_RESUME loop into signal.c
  alpha: simplify TIF_NEED_RESCHED handling

arch/Kconfig

@@ -274,6 +274,9 @@ config ARCH_WANT_OLD_COMPAT_IPC
 config GENERIC_KERNEL_THREAD
 	bool
 
+config GENERIC_KERNEL_EXECVE
+	bool
+
 config HAVE_ARCH_SECCOMP_FILTER
 	bool
 	help

arch/alpha/Kconfig

@@ -21,6 +21,7 @@ config ALPHA
 	select GENERIC_STRNCPY_FROM_USER
 	select GENERIC_STRNLEN_USER
 	select GENERIC_KERNEL_THREAD
+	select GENERIC_KERNEL_EXECVE
 	help
 	  The Alpha is a 64-bit general-purpose processor designed and
 	  marketed by the Digital Equipment Corporation of blessed memory,

arch/alpha/include/asm/unistd.h

@@ -482,7 +482,6 @@
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_RT_SIGSUSPEND
 #define __ARCH_WANT_SYS_EXECVE
-#define __ARCH_WANT_KERNEL_EXECVE
 
 /* "Conditional" syscalls.  What we want is
 

arch/alpha/kernel/entry.S

@@ -311,7 +311,7 @@ entSys:
 
 	.align	4
 ret_from_sys_call:
-	cmovne	$26, 0, $19		/* $19 = 0 => non-restartable */
+	cmovne	$26, 0, $18		/* $18 = 0 => non-restartable */
 	ldq	$0, SP_OFF($sp)
 	and	$0, 8, $0
 	beq	$0, ret_to_kernel
@@ -320,8 +320,8 @@ ret_to_user:
 		sampling and the rti.  */
 	lda	$16, 7
 	call_pal PAL_swpipl
-	ldl	$5, TI_FLAGS($8)
-	and	$5, _TIF_WORK_MASK, $2
+	ldl	$17, TI_FLAGS($8)
+	and	$17, _TIF_WORK_MASK, $2
 	bne	$2, work_pending
 restore_all:
 	RESTORE_ALL
@@ -341,10 +341,10 @@ $syscall_error:
 	 * frame to indicate that a negative return value wasn't an
 	 * error number..
 	 */
-	ldq	$19, 0($sp)	/* old syscall nr (zero if success) */
-	beq	$19, $ret_success
+	ldq	$18, 0($sp)	/* old syscall nr (zero if success) */
+	beq	$18, $ret_success
 
-	ldq	$20, 72($sp)	/* .. and this a3 */
+	ldq	$19, 72($sp)	/* .. and this a3 */
 	subq	$31, $0, $0	/* with error in v0 */
 	addq	$31, 1, $1	/* set a3 for errno return */
 	stq	$0, 0($sp)
@@ -362,51 +362,35 @@ $ret_success:
  * Do all cleanup when returning from all interrupts and system calls.
  *
  * Arguments:
- *       $5: TI_FLAGS.
  *       $8: current.
- *      $19: The old syscall number, or zero if this is not a return
+ *      $17: TI_FLAGS.
+ *      $18: The old syscall number, or zero if this is not a return
  *           from a syscall that errored and is possibly restartable.
- *      $20: The old a3 value
+ *      $19: The old a3 value
  */
 
 	.align	4
 	.ent	work_pending
 work_pending:
-	and	$5, _TIF_NEED_RESCHED, $2
-	beq	$2, $work_notifysig
+	and	$17, _TIF_NOTIFY_RESUME | _TIF_SIGPENDING, $2
+	bne	$2, $work_notifysig
 
 $work_resched:
-	subq	$sp, 16, $sp
-	stq	$19, 0($sp)              /* save syscall nr */
-	stq	$20, 8($sp)              /* and error indication (a3) */
+	/*
+	 * We can get here only if we returned from syscall without SIGPENDING
+	 * or got through work_notifysig already.  Either case means no syscall
+	 * restarts for us, so let $18 and $19 burn.
+	 */
 	jsr	$26, schedule
-	ldq	$19, 0($sp)
-	ldq	$20, 8($sp)
-	addq	$sp, 16, $sp
-	/* Make sure need_resched and sigpending don't change between
-		sampling and the rti.  */
-	lda	$16, 7
-	call_pal PAL_swpipl
-	ldl	$5, TI_FLAGS($8)
-	and	$5, _TIF_WORK_MASK, $2
-	beq	$2, restore_all
-	and	$5, _TIF_NEED_RESCHED, $2
-	bne	$2, $work_resched
+	mov	0, $18
+	br	ret_to_user
 
 $work_notifysig:
 	mov	$sp, $16
 	bsr	$1, do_switch_stack
-	mov	$sp, $17
-	mov	$5, $18
-	mov	$19, $9		/* save old syscall number */
-	mov	$20, $10	/* save old a3 */
-	and	$5, _TIF_SIGPENDING, $2
-	cmovne	$2, 0, $9	/* we don't want double syscall restarts */
-	jsr	$26, do_notify_resume
-	mov	$9, $19
-	mov	$10, $20
+	jsr	$26, do_work_pending
 	bsr	$1, undo_switch_stack
-	br	ret_to_user
+	br	restore_all
 .end work_pending
 
 /*
@@ -454,9 +438,9 @@ $strace_success:
 
 	.align	3
 $strace_error:
-	ldq	$19, 0($sp)	/* old syscall nr (zero if success) */
-	beq	$19, $strace_success
-	ldq	$20, 72($sp)	/* .. and this a3 */
+	ldq	$18, 0($sp)	/* old syscall nr (zero if success) */
+	beq	$18, $strace_success
+	ldq	$19, 72($sp)	/* .. and this a3 */
 
 	subq	$31, $0, $0	/* with error in v0 */
 	addq	$31, 1, $1	/* set a3 for errno return */
@@ -464,11 +448,11 @@ $strace_error:
 	stq	$1, 72($sp)	/* a3 for return */
 
 	bsr	$1, do_switch_stack
-	mov	$19, $9		/* save old syscall number */
-	mov	$20, $10	/* save old a3 */
+	mov	$18, $9		/* save old syscall number */
+	mov	$19, $10	/* save old a3 */
 	jsr	$26, syscall_trace_leave
-	mov	$9, $19
-	mov	$10, $20
+	mov	$9, $18
+	mov	$10, $19
 	bsr	$1, undo_switch_stack
 
 	mov	$31, $26	/* tell "ret_from_sys_call" we can restart */
@@ -619,24 +603,9 @@ ret_from_kernel_thread:
 	mov	$9, $27
 	mov	$10, $16
 	jsr	$26, ($9)
-	ldgp	$gp, 0($26)
-	mov	$0, $16
-	mov	$31, $26
-	jmp	$31, sys_exit
-.end ret_from_kernel_thread
-
-	.globl	ret_from_kernel_execve
-	.align	4
-	.ent	ret_from_kernel_execve
-ret_from_kernel_execve:
-	mov	$16, $sp
-	/* Avoid the HAE being gratuitously wrong, to avoid restoring it.  */
-	ldq	$2, alpha_mv+HAE_CACHE
-	stq	$2, 152($sp)		/* HAE */
 	mov	$31, $19		/* to disable syscall restarts */
 	br	$31, ret_to_user
-
-.end	ret_from_kernel_execve
+.end ret_from_kernel_thread
 
 
 /*

arch/alpha/kernel/signal.c

@@ -298,8 +298,9 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
 
 static long
 setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, 
-		 struct switch_stack *sw, unsigned long mask, unsigned long sp)
+		 unsigned long mask, unsigned long sp)
 {
+	struct switch_stack *sw = (struct switch_stack *)regs - 1;
 	long i, err = 0;
 
 	err |= __put_user(on_sig_stack((unsigned long)sc), &sc->sc_onstack);
@@ -354,7 +355,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
 
 static int
 setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
-	    struct pt_regs *regs, struct switch_stack * sw)
+	    struct pt_regs *regs)
 {
 	unsigned long oldsp, r26, err = 0;
 	struct sigframe __user *frame;
@@ -364,7 +365,7 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		return -EFAULT;
 
-	err |= setup_sigcontext(&frame->sc, regs, sw, set->sig[0], oldsp);
+	err |= setup_sigcontext(&frame->sc, regs, set->sig[0], oldsp);
 	if (err)
 		return -EFAULT;
 
@@ -401,7 +402,7 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 
 static int
 setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-	       sigset_t *set, struct pt_regs *regs, struct switch_stack * sw)
+	       sigset_t *set, struct pt_regs *regs)
 {
 	unsigned long oldsp, r26, err = 0;
 	struct rt_sigframe __user *frame;
@@ -420,7 +421,7 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
 	err |= __put_user(sas_ss_flags(oldsp), &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, sw,
+	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, 
 				set->sig[0], oldsp);
 	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
 	if (err)
@@ -464,15 +465,15 @@ setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
  */
 static inline void
 handle_signal(int sig, struct k_sigaction *ka, siginfo_t *info,
-	      struct pt_regs * regs, struct switch_stack *sw)
+	      struct pt_regs * regs)
 {
 	sigset_t *oldset = sigmask_to_save();
 	int ret;
 
 	if (ka->sa.sa_flags & SA_SIGINFO)
-		ret = setup_rt_frame(sig, ka, info, oldset, regs, sw);
+		ret = setup_rt_frame(sig, ka, info, oldset, regs);
 	else
-		ret = setup_frame(sig, ka, oldset, regs, sw);
+		ret = setup_frame(sig, ka, oldset, regs);
 
 	if (ret) {
 		force_sigsegv(sig, current);
@@ -519,8 +520,7 @@ syscall_restart(unsigned long r0, unsigned long r19,
  * all (if we get here from anything but a syscall return, it will be 0)
  */
 static void
-do_signal(struct pt_regs * regs, struct switch_stack * sw,
-	  unsigned long r0, unsigned long r19)
+do_signal(struct pt_regs *regs, unsigned long r0, unsigned long r19)
 {
 	siginfo_t info;
 	int signr;
@@ -537,7 +537,7 @@ do_signal(struct pt_regs * regs, struct switch_stack * sw,
 		/* Whee!  Actually deliver the signal.  */
 		if (r0)
 			syscall_restart(r0, r19, regs, &ka);
-		handle_signal(signr, &ka, &info, regs, sw);
+		handle_signal(signr, &ka, &info, regs);
 		if (single_stepping) 
 			ptrace_set_bpt(current); /* re-set bpt */
 		return;
@@ -568,15 +568,23 @@ do_signal(struct pt_regs * regs, struct switch_stack * sw,
 }
 
 void
-do_notify_resume(struct pt_regs *regs, struct switch_stack *sw,
-		 unsigned long thread_info_flags,
+do_work_pending(struct pt_regs *regs, unsigned long thread_flags,
 		 unsigned long r0, unsigned long r19)
 {
-	if (thread_info_flags & _TIF_SIGPENDING)
-		do_signal(regs, sw, r0, r19);
-
-	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
-		clear_thread_flag(TIF_NOTIFY_RESUME);
-		tracehook_notify_resume(regs);
-	}
+	do {
+		if (thread_flags & _TIF_NEED_RESCHED) {
+			schedule();
+		} else {
+			local_irq_enable();
+			if (thread_flags & _TIF_SIGPENDING) {
+				do_signal(regs, r0, r19);
+				r0 = 0;
+			} else {
+				clear_thread_flag(TIF_NOTIFY_RESUME);
+				tracehook_notify_resume(regs);
+			}
+		}
+		local_irq_disable();
+		thread_flags = current_thread_info()->flags;
+	} while (thread_flags & _TIF_WORK_MASK);
 }

arch/arm/Kconfig

@@ -53,6 +53,7 @@ config ARM
 	select GENERIC_STRNLEN_USER
 	select DCACHE_WORD_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && !CPU_BIG_ENDIAN
 	select GENERIC_KERNEL_THREAD
+	select GENERIC_KERNEL_EXECVE
 	help
 	  The ARM series is a line of low-power-consumption RISC chip designs
 	  licensed by ARM Ltd and targeted at embedded applications and

arch/arm/include/asm/unistd.h

@@ -479,7 +479,6 @@
 #define __ARCH_WANT_SYS_SOCKETCALL
 #endif
 #define __ARCH_WANT_SYS_EXECVE
-#define __ARCH_WANT_KERNEL_EXECVE
 
 /*
  * "Conditional" syscalls

arch/arm/kernel/entry-common.S

@@ -86,35 +86,14 @@ ENDPROC(ret_to_user)
  */
 ENTRY(ret_from_fork)
 	bl	schedule_tail
+	cmp	r5, #0
+	movne	r0, r4
+	movne	lr, pc
+	movne	pc, r5
 	get_thread_info tsk
-	mov	why, #1
 	b	ret_slow_syscall
 ENDPROC(ret_from_fork)
 
-ENTRY(ret_from_kernel_thread)
- UNWIND(.fnstart)
- UNWIND(.cantunwind)
-	bl	schedule_tail
-	mov	r0, r4
-	adr	lr, BSYM(1f)	@ kernel threads should not exit
-	mov	pc, r5
-1:	bl	do_exit
-	nop
- UNWIND(.fnend)
-ENDPROC(ret_from_kernel_thread)
-
-/*
- * turn a kernel thread into userland process
- * use: ret_from_kernel_execve(struct pt_regs *normal)
- */
-ENTRY(ret_from_kernel_execve)
-	mov	why, #0			@ not a syscall
-	str	why, [r0, #S_R0]	@ ... and we want 0 in ->ARM_r0 as well
-	get_thread_info tsk		@ thread structure
-	mov	sp, r0			@ stack pointer just under pt_regs
-	b	ret_slow_syscall
-ENDPROC(ret_from_kernel_execve)
-
 	.equ NR_syscalls,0
 #define CALL(x) .equ NR_syscalls,NR_syscalls+1
 #include "calls.S"

arch/arm/kernel/process.c

@@ -373,7 +373,6 @@ void release_thread(struct task_struct *dead_task)
 }
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
-asmlinkage void ret_from_kernel_thread(void) __asm__("ret_from_kernel_thread");
 
 int
 copy_thread(unsigned long clone_flags, unsigned long stack_start,
@@ -388,13 +387,13 @@ copy_thread(unsigned long clone_flags, unsigned long stack_start,
 		*childregs = *regs;
 		childregs->ARM_r0 = 0;
 		childregs->ARM_sp = stack_start;
-		thread->cpu_context.pc = (unsigned long)ret_from_fork;
 	} else {
+		memset(childregs, 0, sizeof(struct pt_regs));
 		thread->cpu_context.r4 = stk_sz;
 		thread->cpu_context.r5 = stack_start;
-		thread->cpu_context.pc = (unsigned long)ret_from_kernel_thread;
 		childregs->ARM_cpsr = SVC_MODE;
 	}
+	thread->cpu_context.pc = (unsigned long)ret_from_fork;
 	thread->cpu_context.sp = (unsigned long)childregs;
 
 	clear_ptrace_hw_breakpoint(p);

arch/powerpc/platforms/pseries/eeh_event.c

@@ -23,6 +23,7 @@
 #include <linux/pci.h>
 #include <linux/slab.h>
 #include <linux/workqueue.h>
+#include <linux/kthread.h>
 #include <asm/eeh_event.h>
 #include <asm/ppc-pci.h>
 
@@ -59,8 +60,6 @@ static int eeh_event_handler(void * dummy)
 	struct eeh_event *event;
 	struct eeh_pe *pe;
 
-	set_task_comm(current, "eehd");
-
 	spin_lock_irqsave(&eeh_eventlist_lock, flags);
 	event = NULL;
 
@@ -108,7 +107,7 @@ static int eeh_event_handler(void * dummy)
  */
 static void eeh_thread_launcher(struct work_struct *dummy)
 {
-	if (kernel_thread(eeh_event_handler, NULL, CLONE_KERNEL) < 0)
+	if (IS_ERR(kthread_run(eeh_event_handler, NULL, "eehd")))
 		printk(KERN_ERR "Failed to start EEH daemon\n");
 }
 

arch/um/include/asm/processor-generic.h

@@ -26,7 +26,6 @@ struct thread_struct {
 	jmp_buf *fault_catcher;
 	struct task_struct *prev_sched;
 	unsigned long temp_stack;
-	jmp_buf *exec_buf;
 	struct arch_thread arch;
 	jmp_buf switch_buf;
 	int mm_count;
@@ -54,7 +53,6 @@ struct thread_struct {
 	.fault_addr		= NULL, \
 	.prev_sched		= NULL, \
 	.temp_stack		= 0, \
-	.exec_buf		= NULL, \
 	.arch			= INIT_ARCH_THREAD, \
 	.request		= { 0 } \
 }

arch/um/include/shared/os.h

@@ -191,7 +191,6 @@ extern int os_getpid(void);
 extern int os_getpgrp(void);
 
 extern void init_new_thread_signals(void);
-extern int run_kernel_thread(int (*fn)(void *), void *arg, jmp_buf **jmp_ptr);
 
 extern int os_map_memory(void *virt, int fd, unsigned long long off,
 			 unsigned long len, int r, int w, int x);

arch/um/kernel/exec.c

@@ -47,8 +47,3 @@ void start_thread(struct pt_regs *regs, unsigned long eip, unsigned long esp)
 #endif
 }
 EXPORT_SYMBOL(start_thread);
-
-void __noreturn ret_from_kernel_execve(struct pt_regs *unused)
-{
-	UML_LONGJMP(current->thread.exec_buf, 1);
-}

arch/um/kernel/process.c

@@ -135,14 +135,10 @@ void new_thread_handler(void)
 	arg = current->thread.request.u.thread.arg;
 
 	/*
-	 * The return value is 1 if the kernel thread execs a process,
-	 * 0 if it just exits
+	 * callback returns only if the kernel thread execs a process
 	 */
-	n = run_kernel_thread(fn, arg, &current->thread.exec_buf);
-	if (n == 1)
-		userspace(&current->thread.regs.regs);
-	else
-		do_exit(0);
+	n = fn(arg);
+	userspace(&current->thread.regs.regs);
 }
 
 /* Called magically, see new_thread_handler above */

arch/um/os-Linux/process.c

@@ -244,16 +244,3 @@ void init_new_thread_signals(void)
 	signal(SIGWINCH, SIG_IGN);
 	signal(SIGTERM, SIG_DFL);
 }
-
-int run_kernel_thread(int (*fn)(void *), void *arg, jmp_buf **jmp_ptr)
-{
-	jmp_buf buf;
-	int n;
-
-	*jmp_ptr = &buf;
-	n = UML_SETJMP(&buf);
-	if (n != 0)
-		return n;
-	(*fn)(arg);
-	return 0;
-}

arch/x86/Kconfig

@@ -109,6 +109,7 @@ config X86
 	select HAVE_RCU_USER_QS if X86_64
 	select HAVE_IRQ_TIME_ACCOUNTING
 	select GENERIC_KERNEL_THREAD
+	select GENERIC_KERNEL_EXECVE
 
 config INSTRUCTION_DECODER
 	def_bool y

arch/x86/include/asm/unistd.h

@@ -51,7 +51,6 @@
 # define __ARCH_WANT_SYS_UTIME
 # define __ARCH_WANT_SYS_WAITPID
 # define __ARCH_WANT_SYS_EXECVE
-# define __ARCH_WANT_KERNEL_EXECVE
 
 /*
  * "Conditional" syscalls

arch/x86/kernel/entry_32.S

@@ -299,12 +299,20 @@ ENTRY(ret_from_fork)
 	CFI_ENDPROC
 END(ret_from_fork)
 
-ENTRY(ret_from_kernel_execve)
-	movl %eax, %esp
-	movl $0,PT_EAX(%esp)
+ENTRY(ret_from_kernel_thread)
+	CFI_STARTPROC
+	pushl_cfi %eax
+	call schedule_tail
 	GET_THREAD_INFO(%ebp)
+	popl_cfi %eax
+	pushl_cfi $0x0202		# Reset kernel eflags
+	popfl_cfi
+	movl PT_EBP(%esp),%eax
+	call *PT_EBX(%esp)
+	movl $0,PT_EAX(%esp)
 	jmp syscall_exit
-END(ret_from_kernel_execve)
+	CFI_ENDPROC
+ENDPROC(ret_from_kernel_thread)
 
 /*
  * Interrupt exit functions should be protected against kprobes
@@ -1015,21 +1023,6 @@ END(spurious_interrupt_bug)
  */
 	.popsection
 
-ENTRY(ret_from_kernel_thread)
-	CFI_STARTPROC
-	pushl_cfi %eax
-	call schedule_tail
-	GET_THREAD_INFO(%ebp)
-	popl_cfi %eax
-	pushl_cfi $0x0202		# Reset kernel eflags
-	popfl_cfi
-	movl PT_EBP(%esp),%eax
-	call *PT_EBX(%esp)
-	call do_exit
-	ud2			# padding for call trace
-	CFI_ENDPROC
-ENDPROC(ret_from_kernel_thread)
-
 #ifdef CONFIG_XEN
 /* Xen doesn't set %esp to be precisely what the normal sysenter
    entrypoint expects, so fix it up before using the normal path. */

arch/x86/kernel/entry_64.S

@@ -563,15 +563,13 @@ ENTRY(ret_from_fork)
 	jmp ret_from_sys_call			# go to the SYSRET fastpath
 
 1:
-	subq $REST_SKIP, %rsp	# move the stack pointer back
+	subq $REST_SKIP, %rsp	# leave space for volatiles
 	CFI_ADJUST_CFA_OFFSET	REST_SKIP
 	movq %rbp, %rdi
 	call *%rbx
-	# exit
-	mov %eax, %edi
-	call do_exit
-	ud2			# padding for call trace
-
+	movl $0, RAX(%rsp)
+	RESTORE_REST
+	jmp int_ret_from_sys_call
 	CFI_ENDPROC
 END(ret_from_fork)
 
@@ -1326,20 +1324,6 @@ bad_gs:
 	jmp  2b
 	.previous
 
-ENTRY(ret_from_kernel_execve)
-	movq %rdi, %rsp
-	movl $0, RAX(%rsp)
-	// RESTORE_REST
-	movq 0*8(%rsp), %r15
-	movq 1*8(%rsp), %r14
-	movq 2*8(%rsp), %r13
-	movq 3*8(%rsp), %r12
-	movq 4*8(%rsp), %rbp
-	movq 5*8(%rsp), %rbx
-	addq $(6*8), %rsp
-	jmp int_ret_from_sys_call
-END(ret_from_kernel_execve)
-
 /* Call softirq on interrupt stack. Interrupts are off. */
 ENTRY(call_softirq)
 	CFI_STARTPROC

arch/x86/um/Kconfig

@@ -14,6 +14,7 @@ config UML_X86
 	def_bool y
 	select GENERIC_FIND_FIRST_BIT
 	select GENERIC_KERNEL_THREAD
+	select GENERIC_KERNEL_EXECVE
 
 config 64BIT
 	bool "64-bit kernel" if SUBARCH = "x86"

include/linux/syscalls.h

@@ -827,7 +827,15 @@ asmlinkage long sys_fanotify_mark(int fanotify_fd, unsigned int flags,
 				  const char  __user *pathname);
 asmlinkage long sys_syncfs(int fd);
 
+#ifndef CONFIG_GENERIC_KERNEL_EXECVE
 int kernel_execve(const char *filename, const char *const argv[], const char *const envp[]);
+#else
+#define kernel_execve(filename, argv, envp) \
+	do_execve(filename, \
+		(const char __user *const __user *)argv, \
+		(const char __user *const __user *)envp, \
+		current_pt_regs())
+#endif
 
 
 asmlinkage long sys_perf_event_open(

init/do_mounts_initrd.c

@@ -16,13 +16,13 @@
 #include <linux/initrd.h>
 #include <linux/sched.h>
 #include <linux/freezer.h>
+#include <linux/kmod.h>
 
 #include "do_mounts.h"
 
 unsigned long initrd_start, initrd_end;
 int initrd_below_start_ok;
 unsigned int real_root_dev;	/* do_proc_dointvec cannot handle kdev_t */
-static int __initdata old_fd, root_fd;
 static int __initdata mount_initrd = 1;
 
 static int __init no_initrd(char *str)
@@ -33,33 +33,29 @@ static int __init no_initrd(char *str)
 
 __setup("noinitrd", no_initrd);
 
-static int __init do_linuxrc(void *_shell)
+static int init_linuxrc(struct subprocess_info *info, struct cred *new)
 {
-	static const char *argv[] = { "linuxrc", NULL, };
-	extern const char *envp_init[];
-	const char *shell = _shell;
-
-	sys_close(old_fd);sys_close(root_fd);
+	sys_unshare(CLONE_FS | CLONE_FILES);
+	/* move initrd over / and chdir/chroot in initrd root */
+	sys_chdir("/root");
+	sys_mount(".", "/", NULL, MS_MOVE, NULL);
+	sys_chroot(".");
 	sys_setsid();
-	return kernel_execve(shell, argv, envp_init);
+	return 0;
 }
 
 static void __init handle_initrd(void)
 {
+	static char *argv[] = { "linuxrc", NULL, };
+	extern char *envp_init[];
 	int error;
-	int pid;
 
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
 	mount_block_root("/dev/root.old", root_mountflags & ~MS_RDONLY);
 	sys_mkdir("/old", 0700);
-	root_fd = sys_open("/", 0, 0);
-	old_fd = sys_open("/old", 0, 0);
-	/* move initrd over / and chdir/chroot in initrd root */
-	sys_chdir("/root");
-	sys_mount(".", "/", NULL, MS_MOVE, NULL);
-	sys_chroot(".");
+	sys_chdir("/old");
 
 	/*
 	 * In case that a resume from disk is carried out by linuxrc or one of
@@ -67,27 +63,22 @@ static void __init handle_initrd(void)
 	 */
 	current->flags |= PF_FREEZER_SKIP;
 
-	pid = kernel_thread(do_linuxrc, "/linuxrc", SIGCHLD);
-	if (pid > 0)
-		while (pid != sys_wait4(-1, NULL, 0, NULL))
-			yield();
+	call_usermodehelper_fns("/linuxrc", argv, envp_init, UMH_WAIT_PROC,
+			init_linuxrc, NULL, NULL);
 
 	current->flags &= ~PF_FREEZER_SKIP;
 
 	/* move initrd to rootfs' /old */
-	sys_fchdir(old_fd);
-	sys_mount("/", ".", NULL, MS_MOVE, NULL);
+	sys_mount("..", ".", NULL, MS_MOVE, NULL);
 	/* switch root and cwd back to / of rootfs */
-	sys_fchdir(root_fd);
-	sys_chroot(".");
-	sys_close(old_fd);
-	sys_close(root_fd);
+	sys_chroot("..");
 
 	if (new_decode_dev(real_root_dev) == Root_RAM0) {
 		sys_chdir("/old");
 		return;
 	}
 
+	sys_chdir("/");
 	ROOT_DEV = new_decode_dev(real_root_dev);
 	mount_root();
 

init/main.c

@@ -69,6 +69,7 @@
 #include <linux/slab.h>
 #include <linux/perf_event.h>
 #include <linux/file.h>
+#include <linux/ptrace.h>
 
 #include <asm/io.h>
 #include <asm/bugs.h>
@@ -791,17 +792,17 @@ static void __init do_pre_smp_initcalls(void)
 		do_one_initcall(*fn);
 }
 
-static void run_init_process(const char *init_filename)
+static int run_init_process(const char *init_filename)
 {
 	argv_init[0] = init_filename;
-	kernel_execve(init_filename, argv_init, envp_init);
+	return kernel_execve(init_filename, argv_init, envp_init);
 }
 
-/* This is a non __init function. Force it to be noinline otherwise gcc
- * makes it inline to init() and it becomes part of init.text section
- */
-static noinline int init_post(void)
+static void __init kernel_init_freeable(void);
+
+static int __ref kernel_init(void *unused)
 {
+	kernel_init_freeable();
 	/* need to finish all async __init code before freeing the memory */
 	async_synchronize_full();
 	free_initmem();
@@ -813,7 +814,8 @@ static noinline int init_post(void)
 	flush_delayed_fput();
 
 	if (ramdisk_execute_command) {
-		run_init_process(ramdisk_execute_command);
+		if (!run_init_process(ramdisk_execute_command))
+			return 0;
 		printk(KERN_WARNING "Failed to execute %s\n",
 				ramdisk_execute_command);
 	}
@@ -825,20 +827,22 @@ static noinline int init_post(void)
 	 * trying to recover a really broken machine.
 	 */
 	if (execute_command) {
-		run_init_process(execute_command);
+		if (!run_init_process(execute_command))
+			return 0;
 		printk(KERN_WARNING "Failed to execute %s.  Attempting "
 					"defaults...\n", execute_command);
 	}
-	run_init_process("/sbin/init");
-	run_init_process("/etc/init");
-	run_init_process("/bin/init");
-	run_init_process("/bin/sh");
+	if (!run_init_process("/sbin/init") ||
+	    !run_init_process("/etc/init") ||
+	    !run_init_process("/bin/init") ||
+	    !run_init_process("/bin/sh"))
+		return 0;
 
 	panic("No init found.  Try passing init= option to kernel. "
 	      "See Linux Documentation/init.txt for guidance.");
 }
 
-static int __init kernel_init(void * unused)
+static void __init kernel_init_freeable(void)
 {
 	/*
 	 * Wait until kthreadd is all set-up.
@@ -893,7 +897,4 @@ static int __init kernel_init(void * unused)
 	 * we're essentially up and running. Get rid of the
 	 * initmem segments and start the user-mode stuff..
 	 */
-
-	init_post();
-	return 0;
 }

kernel/kmod.c

@@ -37,6 +37,7 @@
 #include <linux/notifier.h>
 #include <linux/suspend.h>
 #include <linux/rwsem.h>
+#include <linux/ptrace.h>
 #include <asm/uaccess.h>
 
 #include <trace/events/module.h>
@@ -221,11 +222,13 @@ static int ____call_usermodehelper(void *data)
 	retval = kernel_execve(sub_info->path,
 			       (const char *const *)sub_info->argv,
 			       (const char *const *)sub_info->envp);
+	if (!retval)
+		return 0;
 
 	/* Exec failed? */
 fail:
 	sub_info->retval = retval;
-	return 0;
+	do_exit(0);
 }
 
 static int call_helper(void *data)
@@ -292,7 +295,7 @@ static int wait_for_helper(void *data)
 	}
 
 	umh_complete(sub_info);
-	return 0;
+	do_exit(0);
 }
 
 /* This is run by khelper thread  */

kernel/kthread.c

@@ -16,6 +16,7 @@
 #include <linux/mutex.h>
 #include <linux/slab.h>
 #include <linux/freezer.h>
+#include <linux/ptrace.h>
 #include <trace/events/sched.h>
 
 static DEFINE_SPINLOCK(kthread_create_lock);