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

Pull big execve/kernel_thread/fork unification series from Al Viro:
 "All architectures are converted to new model.  Quite a bit of that
  stuff is actually shared with architecture trees; in such cases it's
  literally shared branch pulled by both, not a cherry-pick.

  A lot of ugliness and black magic is gone (-3KLoC total in this one):

   - kernel_thread()/kernel_execve()/sys_execve() redesign.

     We don't do syscalls from kernel anymore for either kernel_thread()
     or kernel_execve():

     kernel_thread() is essentially clone(2) with callback run before we
     return to userland, the callbacks either never return or do
     successful do_execve() before returning.

     kernel_execve() is a wrapper for do_execve() - it doesn't need to
     do transition to user mode anymore.

     As a result kernel_thread() and kernel_execve() are
     arch-independent now - they live in kernel/fork.c and fs/exec.c
     resp.  sys_execve() is also in fs/exec.c and it's completely
     architecture-independent.

   - daemonize() is gone, along with its parts in fs/*.c

   - struct pt_regs * is no longer passed to do_fork/copy_process/
     copy_thread/do_execve/search_binary_handler/->load_binary/do_coredump.

   - sys_fork()/sys_vfork()/sys_clone() unified; some architectures
     still need wrappers (ones with callee-saved registers not saved in
     pt_regs on syscall entry), but the main part of those suckers is in
     kernel/fork.c now."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/signal: (113 commits)
  do_coredump(): get rid of pt_regs argument
  print_fatal_signal(): get rid of pt_regs argument
  ptrace_signal(): get rid of unused arguments
  get rid of ptrace_signal_deliver() arguments
  new helper: signal_pt_regs()
  unify default ptrace_signal_deliver
  flagday: kill pt_regs argument of do_fork()
  death to idle_regs()
  don't pass regs to copy_process()
  flagday: don't pass regs to copy_thread()
  bfin: switch to generic vfork, get rid of pointless wrappers
  xtensa: switch to generic clone()
  openrisc: switch to use of generic fork and clone
  unicore32: switch to generic clone(2)
  score: switch to generic fork/vfork/clone
  c6x: sanitize copy_thread(), get rid of clone(2) wrapper, switch to generic clone()
  take sys_fork/sys_vfork/sys_clone prototypes to linux/syscalls.h
  mn10300: switch to generic fork/vfork/clone
  h8300: switch to generic fork/vfork/clone
  tile: switch to generic clone()
  ...

Conflicts:
	arch/microblaze/include/asm/Kbuild

4 files changed, 20 insertions, 95 deletions

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index ef46f66bc0d6..aa8f2ba6289b 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -330,40 +330,18 @@ ENTRY(ret_from_fork)
 	la	%r11,STACK_FRAME_OVERHEAD(%r15)
 	l	%r12,__LC_THREAD_INFO
 	l	%r13,__LC_SVC_NEW_PSW+4
-	tm	__PT_PSW+1(%r11),0x01	# forking a kernel thread ?
-	je	1f
 	l	%r1,BASED(.Lschedule_tail)
 	basr	%r14,%r1		# call schedule_tail
 	TRACE_IRQS_ON
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
-	j	sysc_tracenogo
-
-1:	# it's a kernel thread
-	st	%r15,__PT_R15(%r11)	# store stack pointer for new kthread
-	l	%r1,BASED(.Lschedule_tail)
-	basr	%r14,%r1		# call schedule_tail
-	TRACE_IRQS_ON
-	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
-	lm	%r9,%r11,__PT_R9(%r11)	# load gprs
+	tm	__PT_PSW+1(%r11),0x01	# forking a kernel thread ?
+	jne	sysc_tracenogo
+	# it's a kernel thread
+	lm	%r9,%r10,__PT_R9(%r11)	# load gprs
 ENTRY(kernel_thread_starter)
 	la	%r2,0(%r10)
 	basr	%r14,%r9
-	la	%r2,0
-	br	%r11			# do_exit
-
-#
-# kernel_execve function needs to deal with pt_regs that is not
-# at the usual place
-#
-ENTRY(ret_from_kernel_execve)
-	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
-	lr	%r15,%r2
-	lr	%r11,%r2
-	ahi	%r15,-STACK_FRAME_OVERHEAD
-	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
-	l	%r12,__LC_THREAD_INFO
-	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
-	j	sysc_return
+	j	sysc_tracenogo
 
 /*
  * Program check handler routine
diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
index d0d3f69a7346..d8251b98f17a 100644
--- a/arch/s390/kernel/entry.h
+++ b/arch/s390/kernel/entry.h
@@ -54,10 +54,6 @@ long sys_s390_fadvise64(int fd, u32 offset_high, u32 offset_low,
 long sys_s390_fadvise64_64(struct fadvise64_64_args __user *args);
 long sys_s390_fallocate(int fd, int mode, loff_t offset, u32 len_high,
 			u32 len_low);
-long sys_fork(void);
-long sys_clone(unsigned long newsp, unsigned long clone_flags,
-	       int __user *parent_tidptr, int __user *child_tidptr);
-long sys_vfork(void);
 long sys_sigsuspend(int history0, int history1, old_sigset_t mask);
 long sys_sigaction(int sig, const struct old_sigaction __user *act,
 		   struct old_sigaction __user *oact);
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index 07d8de353984..499e95e90f38 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -352,33 +352,17 @@ sysc_tracenogo:
 ENTRY(ret_from_fork)
 	la	%r11,STACK_FRAME_OVERHEAD(%r15)
 	lg	%r12,__LC_THREAD_INFO
-	tm	__PT_PSW+1(%r11),0x01	# forking a kernel thread ?
-	je	1f
 	brasl	%r14,schedule_tail
 	TRACE_IRQS_ON
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
-	j	sysc_tracenogo
-1:	# it's a kernel thread
-	stg	%r15,__PT_R15(%r11)	# store stack pointer for new kthread
-	brasl	%r14,schedule_tail
-	TRACE_IRQS_ON
-	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
-	lmg	%r9,%r11,__PT_R9(%r11)	# load gprs
+	tm	__PT_PSW+1(%r11),0x01	# forking a kernel thread ?
+	jne	sysc_tracenogo
+	# it's a kernel thread
+	lmg	%r9,%r10,__PT_R9(%r11)	# load gprs
 ENTRY(kernel_thread_starter)
 	la	%r2,0(%r10)
 	basr	%r14,%r9
-	la	%r2,0
-	br	%r11			# do_exit
-
-ENTRY(ret_from_kernel_execve)
-	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
-	lgr	%r15,%r2
-	lgr	%r11,%r2
-	aghi	%r15,-STACK_FRAME_OVERHEAD
-	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
-	lg	%r12,__LC_THREAD_INFO
-	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
-	j	sysc_return
+	j	sysc_tracenogo
 
 /*
  * Program check handler routine
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index cd31ad457a9b..536d64579d9a 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -117,8 +117,7 @@ void release_thread(struct task_struct *dead_task)
 }
 
 int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
-		unsigned long arg,
-		struct task_struct *p, struct pt_regs *regs)
+		unsigned long arg, struct task_struct *p)
 {
 	struct thread_info *ti;
 	struct fake_frame
@@ -150,7 +149,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 	frame->sf.gprs[9] = (unsigned long) frame;
 
 	/* Store access registers to kernel stack of new process. */
-	if (unlikely(!regs)) {
+	if (unlikely(p->flags & PF_KTHREAD)) {
 		/* kernel thread */
 		memset(&frame->childregs, 0, sizeof(struct pt_regs));
 		frame->childregs.psw.mask = psw_kernel_bits | PSW_MASK_DAT |
@@ -164,9 +163,10 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 
 		return 0;
 	}
-	frame->childregs = *regs;
+	frame->childregs = *current_pt_regs();
 	frame->childregs.gprs[2] = 0;	/* child returns 0 on fork. */
-	frame->childregs.gprs[15] = new_stackp;
+	if (new_stackp)
+		frame->childregs.gprs[15] = new_stackp;
 
 	/* Don't copy runtime instrumentation info */
 	p->thread.ri_cb = NULL;
@@ -183,57 +183,24 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 	       sizeof(s390_fp_regs));
 	/* Set a new TLS ?  */
 	if (clone_flags & CLONE_SETTLS)
-		p->thread.acrs[0] = regs->gprs[6];
+		p->thread.acrs[0] = frame->childregs.gprs[6];
 #else /* CONFIG_64BIT */
 	/* Save the fpu registers to new thread structure. */
 	save_fp_regs(&p->thread.fp_regs);
 	/* Set a new TLS ?  */
 	if (clone_flags & CLONE_SETTLS) {
+		unsigned long tls = frame->childregs.gprs[6];
 		if (is_compat_task()) {
-			p->thread.acrs[0] = (unsigned int) regs->gprs[6];
+			p->thread.acrs[0] = (unsigned int)tls;
 		} else {
-			p->thread.acrs[0] = (unsigned int)(regs->gprs[6] >> 32);
-			p->thread.acrs[1] = (unsigned int) regs->gprs[6];
+			p->thread.acrs[0] = (unsigned int)(tls >> 32);
+			p->thread.acrs[1] = (unsigned int)tls;
 		}
 	}
 #endif /* CONFIG_64BIT */
 	return 0;
 }
 
-SYSCALL_DEFINE0(fork)
-{
-	struct pt_regs *regs = task_pt_regs(current);
-	return do_fork(SIGCHLD, regs->gprs[15], regs, 0, NULL, NULL);
-}
-
-SYSCALL_DEFINE4(clone, unsigned long, newsp, unsigned long, clone_flags,
-		int __user *, parent_tidptr, int __user *, child_tidptr)
-{
-	struct pt_regs *regs = task_pt_regs(current);
-
-	if (!newsp)
-		newsp = regs->gprs[15];
-	return do_fork(clone_flags, newsp, regs, 0,
-		       parent_tidptr, child_tidptr);
-}
-
-/*
- * This is trivial, and on the face of it looks like it
- * could equally well be done in user mode.
- *
- * Not so, for quite unobvious reasons - register pressure.
- * In user mode vfork() cannot have a stack frame, and if
- * done by calling the "clone()" system call directly, you
- * do not have enough call-clobbered registers to hold all
- * the information you need.
- */
-SYSCALL_DEFINE0(vfork)
-{
-	struct pt_regs *regs = task_pt_regs(current);
-	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD,
-		       regs->gprs[15], regs, 0, NULL, NULL);
-}
-
 asmlinkage void execve_tail(void)
 {
 	current->thread.fp_regs.fpc = 0;