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

5 files changed, 39 insertions, 100 deletions

diff --git a/arch/alpha/kernel/binfmt_loader.c b/arch/alpha/kernel/binfmt_loader.c
index d1f474d..9525660 100644
--- a/arch/alpha/kernel/binfmt_loader.c
+++ b/arch/alpha/kernel/binfmt_loader.c
@@ -5,7 +5,7 @@
 #include <linux/binfmts.h>
 #include <linux/a.out.h>
 
-static int load_binary(struct linux_binprm *bprm, struct pt_regs *regs)
+static int load_binary(struct linux_binprm *bprm)
 {
 	struct exec *eh = (struct exec *)bprm->buf;
 	unsigned long loader;
@@ -37,7 +37,7 @@ static int load_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	retval = prepare_binprm(bprm);
 	if (retval < 0)
 		return retval;
-	return search_binary_handler(bprm,regs);
+	return search_binary_handler(bprm);
 }
 
 static struct linux_binfmt loader_format = {
diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S
index a760783..f62a994 100644
--- a/arch/alpha/kernel/entry.S
+++ b/arch/alpha/kernel/entry.S
@@ -612,47 +612,24 @@ ret_from_kernel_thread:
  * Special system calls.  Most of these are special in that they either
  * have to play switch_stack games or in some way use the pt_regs struct.
  */
-	.align	4
-	.globl	sys_fork
-	.ent	sys_fork
-sys_fork:
-	.prologue 0
-	mov	$sp, $21
-	bsr	$1, do_switch_stack
-	bis	$31, SIGCHLD, $16
-	mov	$31, $17
-	mov	$31, $18
-	mov	$31, $19
-	mov	$31, $20
-	jsr	$26, alpha_clone
-	bsr	$1, undo_switch_stack
-	ret
-.end sys_fork
 
+.macro	fork_like name
 	.align	4
-	.globl	sys_clone
-	.ent	sys_clone
-sys_clone:
+	.globl	alpha_\name
+	.ent	alpha_\name
+alpha_\name:
 	.prologue 0
-	mov	$sp, $21
 	bsr	$1, do_switch_stack
-	/* $16, $17, $18, $19, $20 come from the user.  */
-	jsr	$26, alpha_clone
-	bsr	$1, undo_switch_stack
+	jsr	$26, sys_\name
+	ldq	$26, 56($sp)
+	lda	$sp, SWITCH_STACK_SIZE($sp)
 	ret
-.end sys_clone
+.end	alpha_\name
+.endm
 
-	.align	4
-	.globl	sys_vfork
-	.ent	sys_vfork
-sys_vfork:
-	.prologue 0
-	mov	$sp, $16
-	bsr	$1, do_switch_stack
-	jsr	$26, alpha_vfork
-	bsr	$1, undo_switch_stack
-	ret
-.end sys_vfork
+fork_like fork
+fork_like vfork
+fork_like clone
 
 	.align	4
 	.globl	sys_sigreturn
@@ -661,8 +638,6 @@ sys_sigreturn:
 	.prologue 0
 	lda	$9, ret_from_straced
 	cmpult	$26, $9, $9
-	mov	$sp, $17
-	lda	$18, -SWITCH_STACK_SIZE($sp)
 	lda	$sp, -SWITCH_STACK_SIZE($sp)
 	jsr	$26, do_sigreturn
 	bne	$9, 1f
@@ -678,8 +653,6 @@ sys_rt_sigreturn:
 	.prologue 0
 	lda	$9, ret_from_straced
 	cmpult	$26, $9, $9
-	mov	$sp, $17
-	lda	$18, -SWITCH_STACK_SIZE($sp)
 	lda	$sp, -SWITCH_STACK_SIZE($sp)
 	jsr	$26, do_rt_sigreturn
 	bne	$9, 1f
diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c
index 51987dc..b5d0d09 100644
--- a/arch/alpha/kernel/process.c
+++ b/arch/alpha/kernel/process.c
@@ -235,51 +235,28 @@ release_thread(struct task_struct *dead_task)
 }
 
 /*
- * "alpha_clone()".. By the time we get here, the
- * non-volatile registers have also been saved on the
- * stack. We do some ugly pointer stuff here.. (see
- * also copy_thread)
- *
- * Notice that "fork()" is implemented in terms of clone,
- * with parameters (SIGCHLD, 0).
- */
-int
-alpha_clone(unsigned long clone_flags, unsigned long usp,
-	    int __user *parent_tid, int __user *child_tid,
-	    unsigned long tls_value, struct pt_regs *regs)
-{
-	if (!usp)
-		usp = rdusp();
-
-	return do_fork(clone_flags, usp, regs, 0, parent_tid, child_tid);
-}
-
-int
-alpha_vfork(struct pt_regs *regs)
-{
-	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, rdusp(),
-		       regs, 0, NULL, NULL);
-}
-
-/*
  * Copy an alpha thread..
  */
 
 int
 copy_thread(unsigned long clone_flags, unsigned long usp,
 	    unsigned long arg,
-	    struct task_struct * p, struct pt_regs * regs)
+	    struct task_struct *p)
 {
 	extern void ret_from_fork(void);
 	extern void ret_from_kernel_thread(void);
 
 	struct thread_info *childti = task_thread_info(p);
 	struct pt_regs *childregs = task_pt_regs(p);
+	struct pt_regs *regs = current_pt_regs();
 	struct switch_stack *childstack, *stack;
 	unsigned long settls;
 
 	childstack = ((struct switch_stack *) childregs) - 1;
-	if (unlikely(!regs)) {
+	childti->pcb.ksp = (unsigned long) childstack;
+	childti->pcb.flags = 1;	/* set FEN, clear everything else */
+
+	if (unlikely(p->flags & PF_KTHREAD)) {
 		/* kernel thread */
 		memset(childstack, 0,
 			sizeof(struct switch_stack) + sizeof(struct pt_regs));
@@ -288,12 +265,17 @@ copy_thread(unsigned long clone_flags, unsigned long usp,
 		childstack->r10 = arg;
 		childregs->hae = alpha_mv.hae_cache,
 		childti->pcb.usp = 0;
-		childti->pcb.ksp = (unsigned long) childstack;
-		childti->pcb.flags = 1;	/* set FEN, clear everything else */
 		return 0;
 	}
+	/* Note: if CLONE_SETTLS is not set, then we must inherit the
+	   value from the parent, which will have been set by the block
+	   copy in dup_task_struct.  This is non-intuitive, but is
+	   required for proper operation in the case of a threaded
+	   application calling fork.  */
+	if (clone_flags & CLONE_SETTLS)
+		childti->pcb.unique = regs->r20;
+	childti->pcb.usp = usp ?: rdusp();
 	*childregs = *regs;
-	settls = regs->r20;
 	childregs->r0 = 0;
 	childregs->r19 = 0;
 	childregs->r20 = 1;	/* OSF/1 has some strange fork() semantics.  */
@@ -301,22 +283,6 @@ copy_thread(unsigned long clone_flags, unsigned long usp,
 	stack = ((struct switch_stack *) regs) - 1;
 	*childstack = *stack;
 	childstack->r26 = (unsigned long) ret_from_fork;
-	childti->pcb.usp = usp;
-	childti->pcb.ksp = (unsigned long) childstack;
-	childti->pcb.flags = 1;	/* set FEN, clear everything else */
-
-	/* Set a new TLS for the child thread?  Peek back into the
-	   syscall arguments that we saved on syscall entry.  Oops,
-	   except we'd have clobbered it with the parent/child set
-	   of r20.  Read the saved copy.  */
-	/* Note: if CLONE_SETTLS is not set, then we must inherit the
-	   value from the parent, which will have been set by the block
-	   copy in dup_task_struct.  This is non-intuitive, but is
-	   required for proper operation in the case of a threaded
-	   application calling fork.  */
-	if (clone_flags & CLONE_SETTLS)
-		childti->pcb.unique = settls;
-
 	return 0;
 }
 
diff --git a/arch/alpha/kernel/signal.c b/arch/alpha/kernel/signal.c
index 32575f8..336393c 100644
--- a/arch/alpha/kernel/signal.c
+++ b/arch/alpha/kernel/signal.c
@@ -160,10 +160,10 @@ extern char compile_time_assert
 #define INSN_CALLSYS		0x00000083
 
 static long
-restore_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
-		   struct switch_stack *sw)
+restore_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs)
 {
 	unsigned long usp;
+	struct switch_stack *sw = (struct switch_stack *)regs - 1;
 	long i, err = __get_user(regs->pc, &sc->sc_pc);
 
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
@@ -215,9 +215,9 @@ restore_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
    registers and transfer control from userland.  */
 
 asmlinkage void
-do_sigreturn(struct sigcontext __user *sc, struct pt_regs *regs,
-	     struct switch_stack *sw)
+do_sigreturn(struct sigcontext __user *sc)
 {
+	struct pt_regs *regs = current_pt_regs();
 	sigset_t set;
 
 	/* Verify that it's a good sigcontext before using it */
@@ -228,7 +228,7 @@ do_sigreturn(struct sigcontext __user *sc, struct pt_regs *regs,
 
 	set_current_blocked(&set);
 
-	if (restore_sigcontext(sc, regs, sw))
+	if (restore_sigcontext(sc, regs))
 		goto give_sigsegv;
 
 	/* Send SIGTRAP if we're single-stepping: */
@@ -249,9 +249,9 @@ give_sigsegv:
 }
 
 asmlinkage void
-do_rt_sigreturn(struct rt_sigframe __user *frame, struct pt_regs *regs,
-		struct switch_stack *sw)
+do_rt_sigreturn(struct rt_sigframe __user *frame)
 {
+	struct pt_regs *regs = current_pt_regs();
 	sigset_t set;
 
 	/* Verify that it's a good ucontext_t before using it */
@@ -262,7 +262,7 @@ do_rt_sigreturn(struct rt_sigframe __user *frame, struct pt_regs *regs,
 
 	set_current_blocked(&set);
 
-	if (restore_sigcontext(&frame->uc.uc_mcontext, regs, sw))
+	if (restore_sigcontext(&frame->uc.uc_mcontext, regs))
 		goto give_sigsegv;
 
 	/* Send SIGTRAP if we're single-stepping: */
diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S
index 2ac6b45..4284ec7 100644
--- a/arch/alpha/kernel/systbls.S
+++ b/arch/alpha/kernel/systbls.S
@@ -12,7 +12,7 @@
 sys_call_table:
 	.quad alpha_ni_syscall			/* 0 */
 	.quad sys_exit
-	.quad sys_fork
+	.quad alpha_fork
 	.quad sys_read
 	.quad sys_write
 	.quad alpha_ni_syscall			/* 5 */
@@ -76,7 +76,7 @@ sys_call_table:
 	.quad sys_getpgrp
 	.quad sys_getpagesize
 	.quad alpha_ni_syscall			/* 65 */
-	.quad sys_vfork
+	.quad alpha_vfork
 	.quad sys_newstat
 	.quad sys_newlstat
 	.quad alpha_ni_syscall
@@ -330,7 +330,7 @@ sys_call_table:
 	.quad sys_ni_syscall			/* 309: old get_kernel_syms */
 	.quad sys_syslog			/* 310 */
 	.quad sys_reboot
-	.quad sys_clone
+	.quad alpha_clone
 	.quad sys_uselib
 	.quad sys_mlock
 	.quad sys_munlock			/* 315 */