Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6

* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:
  [IA64] Fix large MCA bootmem allocation
  [IA64] Simplify cpu_idle_wait
  [IA64] Synchronize RBS on PTRACE_ATTACH
  [IA64] Synchronize kernel RSE to user-space and back
  [IA64] Rename TIF_PERFMON_WORK back to TIF_NOTIFY_RESUME
  [IA64] Wire up timerfd_{create,settime,gettime} syscalls

5 files changed, 202 insertions, 77 deletions

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index f5d3efbfbeda..3c331c464b40 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1573,7 +1573,7 @@ sys_call_table:
 	data8 sys_fchmodat
 	data8 sys_faccessat
 	data8 sys_pselect6
-	data8 sys_ppoll
+	data8 sys_ppoll				// 1295
 	data8 sys_unshare
 	data8 sys_splice
 	data8 sys_set_robust_list
@@ -1588,5 +1588,8 @@ sys_call_table:
 	data8 sys_signalfd
 	data8 sys_ni_syscall
 	data8 sys_eventfd
+	data8 sys_timerfd_create		// 1310
+	data8 sys_timerfd_settime
+	data8 sys_timerfd_gettime
 
 	.org sys_call_table + 8*NR_syscalls	// guard against failures to increase NR_syscalls
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 846e7e036b13..6e17aed53135 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -17,7 +17,7 @@
  * Copyright (C) 2000 Intel
  * Copyright (C) Chuck Fleckenstein <cfleck@co.intel.com>
  *
- * Copyright (C) 1999, 2004 Silicon Graphics, Inc.
+ * Copyright (C) 1999, 2004-2008 Silicon Graphics, Inc.
  * Copyright (C) Vijay Chander <vijay@engr.sgi.com>
  *
  * Copyright (C) 2006 FUJITSU LIMITED
@@ -1762,11 +1762,8 @@ format_mca_init_stack(void *mca_data, unsigned long offset,
 /* Caller prevents this from being called after init */
 static void * __init_refok mca_bootmem(void)
 {
-	void *p;
-
-	p = alloc_bootmem(sizeof(struct ia64_mca_cpu) * NR_CPUS +
-	                  KERNEL_STACK_SIZE);
-	return (void *)ALIGN((unsigned long)p, KERNEL_STACK_SIZE);
+	return __alloc_bootmem(sizeof(struct ia64_mca_cpu),
+	                    KERNEL_STACK_SIZE, 0);
 }
 
 /* Do per-CPU MCA-related initialization.  */
@@ -1774,33 +1771,33 @@ void __cpuinit
 ia64_mca_cpu_init(void *cpu_data)
 {
 	void *pal_vaddr;
+	void *data;
+	long sz = sizeof(struct ia64_mca_cpu);
+	int cpu = smp_processor_id();
 	static int first_time = 1;
 
-	if (first_time) {
-		void *mca_data;
-		int cpu;
-
-		first_time = 0;
-		mca_data = mca_bootmem();
-		for (cpu = 0; cpu < NR_CPUS; cpu++) {
-			format_mca_init_stack(mca_data,
-					offsetof(struct ia64_mca_cpu, mca_stack),
-					"MCA", cpu);
-			format_mca_init_stack(mca_data,
-					offsetof(struct ia64_mca_cpu, init_stack),
-					"INIT", cpu);
-			__per_cpu_mca[cpu] = __pa(mca_data);
-			mca_data += sizeof(struct ia64_mca_cpu);
-		}
-	}
-
 	/*
-	 * The MCA info structure was allocated earlier and its
-	 * physical address saved in __per_cpu_mca[cpu].  Copy that
-	 * address * to ia64_mca_data so we can access it as a per-CPU
-	 * variable.
+	 * Structure will already be allocated if cpu has been online,
+	 * then offlined.
 	 */
-	__get_cpu_var(ia64_mca_data) = __per_cpu_mca[smp_processor_id()];
+	if (__per_cpu_mca[cpu]) {
+		data = __va(__per_cpu_mca[cpu]);
+	} else {
+		if (first_time) {
+			data = mca_bootmem();
+			first_time = 0;
+		} else
+			data = page_address(alloc_pages_node(numa_node_id(),
+					GFP_KERNEL, get_order(sz)));
+		if (!data)
+			panic("Could not allocate MCA memory for cpu %d\n",
+					cpu);
+	}
+	format_mca_init_stack(data, offsetof(struct ia64_mca_cpu, mca_stack),
+		"MCA", cpu);
+	format_mca_init_stack(data, offsetof(struct ia64_mca_cpu, init_stack),
+		"INIT", cpu);
+	__get_cpu_var(ia64_mca_data) = __per_cpu_mca[cpu] = __pa(data);
 
 	/*
 	 * Stash away a copy of the PTE needed to map the per-CPU page.
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 78acd9fe97e9..f6b99719f10f 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -586,21 +586,6 @@ pfm_put_task(struct task_struct *task)
 }
 
 static inline void
-pfm_set_task_notify(struct task_struct *task)
-{
-	struct thread_info *info;
-
-	info = (struct thread_info *) ((char *) task + IA64_TASK_SIZE);
-	set_bit(TIF_PERFMON_WORK, &info->flags);
-}
-
-static inline void
-pfm_clear_task_notify(void)
-{
-	clear_thread_flag(TIF_PERFMON_WORK);
-}
-
-static inline void
 pfm_reserve_page(unsigned long a)
 {
 	SetPageReserved(vmalloc_to_page((void *)a));
@@ -3724,7 +3709,7 @@ pfm_restart(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 
 		PFM_SET_WORK_PENDING(task, 1);
 
-		pfm_set_task_notify(task);
+		tsk_set_notify_resume(task);
 
 		/*
 		 * XXX: send reschedule if task runs on another CPU
@@ -5082,7 +5067,7 @@ pfm_handle_work(void)
 
 	PFM_SET_WORK_PENDING(current, 0);
 
-	pfm_clear_task_notify();
+	tsk_clear_notify_resume(current);
 
 	regs = task_pt_regs(current);
 
@@ -5450,7 +5435,7 @@ pfm_overflow_handler(struct task_struct *task, pfm_context_t *ctx, u64 pmc0, str
 			 * when coming from ctxsw, current still points to the
 			 * previous task, therefore we must work with task and not current.
 			 */
-			pfm_set_task_notify(task);
+			tsk_set_notify_resume(task);
 		}
 		/*
 		 * defer until state is changed (shorten spin window). the context is locked
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index 7377d323131d..49937a383b23 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -52,7 +52,6 @@
 #include "sigframe.h"
 
 void (*ia64_mark_idle)(int);
-static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
 
 unsigned long boot_option_idle_override = 0;
 EXPORT_SYMBOL(boot_option_idle_override);
@@ -157,6 +156,17 @@ show_regs (struct pt_regs *regs)
 		show_stack(NULL, NULL);
 }
 
+void tsk_clear_notify_resume(struct task_struct *tsk)
+{
+#ifdef CONFIG_PERFMON
+	if (tsk->thread.pfm_needs_checking)
+		return;
+#endif
+	if (test_ti_thread_flag(task_thread_info(tsk), TIF_RESTORE_RSE))
+		return;
+	clear_ti_thread_flag(task_thread_info(tsk), TIF_NOTIFY_RESUME);
+}
+
 void
 do_notify_resume_user (sigset_t *unused, struct sigscratch *scr, long in_syscall)
 {
@@ -175,6 +185,10 @@ do_notify_resume_user (sigset_t *unused, struct sigscratch *scr, long in_syscall
 	/* deal with pending signal delivery */
 	if (test_thread_flag(TIF_SIGPENDING)||test_thread_flag(TIF_RESTORE_SIGMASK))
 		ia64_do_signal(scr, in_syscall);
+
+	/* copy user rbs to kernel rbs */
+	if (unlikely(test_thread_flag(TIF_RESTORE_RSE)))
+		ia64_sync_krbs();
 }
 
 static int pal_halt        = 1;
@@ -239,33 +253,23 @@ static inline void play_dead(void)
 }
 #endif /* CONFIG_HOTPLUG_CPU */
 
-void cpu_idle_wait(void)
+static void do_nothing(void *unused)
 {
-	unsigned int cpu, this_cpu = get_cpu();
-	cpumask_t map;
-	cpumask_t tmp = current->cpus_allowed;
-
-	set_cpus_allowed(current, cpumask_of_cpu(this_cpu));
-	put_cpu();
-
-	cpus_clear(map);
-	for_each_online_cpu(cpu) {
-		per_cpu(cpu_idle_state, cpu) = 1;
-		cpu_set(cpu, map);
-	}
-
-	__get_cpu_var(cpu_idle_state) = 0;
+}
 
-	wmb();
-	do {
-		ssleep(1);
-		for_each_online_cpu(cpu) {
-			if (cpu_isset(cpu, map) && !per_cpu(cpu_idle_state, cpu))
-				cpu_clear(cpu, map);
-		}
-		cpus_and(map, map, cpu_online_map);
-	} while (!cpus_empty(map));
-	set_cpus_allowed(current, tmp);
+/*
+ * cpu_idle_wait - Used to ensure that all the CPUs discard old value of
+ * pm_idle and update to new pm_idle value. Required while changing pm_idle
+ * handler on SMP systems.
+ *
+ * Caller must have changed pm_idle to the new value before the call. Old
+ * pm_idle value will not be used by any CPU after the return of this function.
+ */
+void cpu_idle_wait(void)
+{
+	smp_mb();
+	/* kick all the CPUs so that they exit out of pm_idle */
+	smp_call_function(do_nothing, NULL, 0, 1);
 }
 EXPORT_SYMBOL_GPL(cpu_idle_wait);
 
@@ -293,9 +297,6 @@ cpu_idle (void)
 #ifdef CONFIG_SMP
 			min_xtp();
 #endif
-			if (__get_cpu_var(cpu_idle_state))
-				__get_cpu_var(cpu_idle_state) = 0;
-
 			rmb();
 			if (mark_idle)
 				(*mark_idle)(1);
diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index 2e96f17b2f3b..331d6768b5d5 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -547,6 +547,129 @@ ia64_sync_user_rbs (struct task_struct *child, struct switch_stack *sw,
 	return 0;
 }
 
+static long
+ia64_sync_kernel_rbs (struct task_struct *child, struct switch_stack *sw,
+		unsigned long user_rbs_start, unsigned long user_rbs_end)
+{
+	unsigned long addr, val;
+	long ret;
+
+	/* now copy word for word from user rbs to kernel rbs: */
+	for (addr = user_rbs_start; addr < user_rbs_end; addr += 8) {
+		if (access_process_vm(child, addr, &val, sizeof(val), 0)
+				!= sizeof(val))
+			return -EIO;
+
+		ret = ia64_poke(child, sw, user_rbs_end, addr, val);
+		if (ret < 0)
+			return ret;
+	}
+	return 0;
+}
+
+typedef long (*syncfunc_t)(struct task_struct *, struct switch_stack *,
+			    unsigned long, unsigned long);
+
+static void do_sync_rbs(struct unw_frame_info *info, void *arg)
+{
+	struct pt_regs *pt;
+	unsigned long urbs_end;
+	syncfunc_t fn = arg;
+
+	if (unw_unwind_to_user(info) < 0)
+		return;
+	pt = task_pt_regs(info->task);
+	urbs_end = ia64_get_user_rbs_end(info->task, pt, NULL);
+
+	fn(info->task, info->sw, pt->ar_bspstore, urbs_end);
+}
+
+/*
+ * when a thread is stopped (ptraced), debugger might change thread's user
+ * stack (change memory directly), and we must avoid the RSE stored in kernel
+ * to override user stack (user space's RSE is newer than kernel's in the
+ * case). To workaround the issue, we copy kernel RSE to user RSE before the
+ * task is stopped, so user RSE has updated data.  we then copy user RSE to
+ * kernel after the task is resummed from traced stop and kernel will use the
+ * newer RSE to return to user. TIF_RESTORE_RSE is the flag to indicate we need
+ * synchronize user RSE to kernel.
+ */
+void ia64_ptrace_stop(void)
+{
+	if (test_and_set_tsk_thread_flag(current, TIF_RESTORE_RSE))
+		return;
+	tsk_set_notify_resume(current);
+	unw_init_running(do_sync_rbs, ia64_sync_user_rbs);
+}
+
+/*
+ * This is called to read back the register backing store.
+ */
+void ia64_sync_krbs(void)
+{
+	clear_tsk_thread_flag(current, TIF_RESTORE_RSE);
+	tsk_clear_notify_resume(current);
+
+	unw_init_running(do_sync_rbs, ia64_sync_kernel_rbs);
+}
+
+/*
+ * 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->signal) {
+		spin_lock_irq(&child->sighand->siglock);
+		if (child->state == TASK_STOPPED &&
+		    !test_and_set_tsk_thread_flag(child, TIF_RESTORE_RSE)) {
+			tsk_set_notify_resume(child);
+
+			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->signal) {
+		spin_lock_irq(&child->sighand->siglock);
+		if (child->state == TASK_TRACED &&
+		    (child->signal->flags & SIGNAL_STOP_STOPPED)) {
+			child->state = TASK_STOPPED;
+		}
+		spin_unlock_irq(&child->sighand->siglock);
+	}
+	read_unlock(&tasklist_lock);
+}
+
 static inline int
 thread_matches (struct task_struct *thread, unsigned long addr)
 {
@@ -1422,6 +1545,7 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data)
 	struct task_struct *child;
 	struct switch_stack *sw;
 	long ret;
+	struct unw_frame_info info;
 
 	lock_kernel();
 	ret = -EPERM;
@@ -1453,6 +1577,8 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data)
 
 	if (request == PTRACE_ATTACH) {
 		ret = ptrace_attach(child);
+		if (!ret)
+			arch_ptrace_attach(child);
 		goto out_tsk;
 	}
 
@@ -1481,6 +1607,11 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data)
 		/* write the word at location addr */
 		urbs_end = ia64_get_user_rbs_end(child, pt, NULL);
 		ret = ia64_poke(child, sw, urbs_end, addr, data);
+
+		/* Make sure user RBS has the latest data */
+		unw_init_from_blocked_task(&info, child);
+		do_sync_rbs(&info, ia64_sync_user_rbs);
+
 		goto out_tsk;
 
 	      case PTRACE_PEEKUSR:
@@ -1634,6 +1765,10 @@ syscall_trace_enter (long arg0, long arg1, long arg2, long arg3,
 	    && (current->ptrace & PT_PTRACED))
 		syscall_trace();
 
+	/* copy user rbs to kernel rbs */
+	if (test_thread_flag(TIF_RESTORE_RSE))
+		ia64_sync_krbs();
+
 	if (unlikely(current->audit_context)) {
 		long syscall;
 		int arch;
@@ -1671,4 +1806,8 @@ syscall_trace_leave (long arg0, long arg1, long arg2, long arg3,
 	    || test_thread_flag(TIF_SINGLESTEP))
 	    && (current->ptrace & PT_PTRACED))
 		syscall_trace();
+
+	/* copy user rbs to kernel rbs */
+	if (test_thread_flag(TIF_RESTORE_RSE))
+		ia64_sync_krbs();
 }