Merge nuts.ninka.net:/disk1/davem/BK/network-2.5

into nuts.ninka.net:/disk1/davem/BK/net-2.5

39 files changed, 391 insertions, 249 deletions

diff --git a/Documentation/video4linux/meye.txt b/Documentation/video4linux/meye.txt
index 9d49f3b01b33..17c3dbf4f81a 100644
--- a/Documentation/video4linux/meye.txt
+++ b/Documentation/video4linux/meye.txt
@@ -33,6 +33,11 @@ to the developers under a NDA (which allows the develoment of a GPL
 driver however), but things are not moving very fast (see
 http://r-engine.sourceforge.net/) (PCI vendor/device is 0x10cf/0x2011).
 
+There is a forth model connected on the USB bus in TR1* Vaio laptops.
+This camera is not supported at all by the current driver, in fact
+little information if any is available for this camera
+(USB vendor/device is 0x054c/0x0107).
+
 Driver options:
 ---------------
 
diff --git a/Makefile b/Makefile
index c978097e6e90..2e15f0810476 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 0
-EXTRAVERSION = -test8
+EXTRAVERSION = -test9
 
 # *DOCUMENTATION*
 # To see a list of typical targets execute "make help"
diff --git a/arch/h8300/kernel/time.c b/arch/h8300/kernel/time.c
index 3637f91fa7a6..bd6dd0802816 100644
--- a/arch/h8300/kernel/time.c
+++ b/arch/h8300/kernel/time.c
@@ -143,3 +143,9 @@ int do_settimeofday(struct timespec *tv)
 }
 
 EXPORT_SYMBOL(do_settimeofday);
+
+unsigned long long sched_clock(void)
+{
+	return (unsigned long long)jiffies * (1000000000 / HZ);
+
+}
diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
index f9d37a739fe5..cd16781c5dda 100644
--- a/arch/ia64/kernel/efi.c
+++ b/arch/ia64/kernel/efi.c
@@ -711,6 +711,32 @@ efi_mem_attributes (unsigned long phys_addr)
 	return 0;
 }
 
+int
+valid_phys_addr_range (unsigned long phys_addr, unsigned long *size)
+{
+	void *efi_map_start, *efi_map_end, *p;
+	efi_memory_desc_t *md;
+	u64 efi_desc_size;
+
+	efi_map_start = __va(ia64_boot_param->efi_memmap);
+	efi_map_end   = efi_map_start + ia64_boot_param->efi_memmap_size;
+	efi_desc_size = ia64_boot_param->efi_memdesc_size;
+
+	for (p = efi_map_start; p < efi_map_end; p += efi_desc_size) {
+		md = p;
+
+		if (phys_addr - md->phys_addr < (md->num_pages << EFI_PAGE_SHIFT)) {
+			if (!(md->attribute & EFI_MEMORY_WB))
+				return 0;
+
+			if (*size > md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - phys_addr)
+				*size = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - phys_addr;
+			return 1;
+		}
+	}
+	return 0;
+}
+
 static void __exit
 efivars_exit (void)
 {
diff --git a/arch/ia64/kernel/gate-data.S b/arch/ia64/kernel/gate-data.S
index beb5234f7927..258c0a3238fb 100644
--- a/arch/ia64/kernel/gate-data.S
+++ b/arch/ia64/kernel/gate-data.S
@@ -1,3 +1,3 @@
-	.section .data.gate, "ax"
+	.section .data.gate, "aw"
 
 	.incbin "arch/ia64/kernel/gate.so"
diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c
index 163544a6b903..471bb733efbc 100644
--- a/arch/ia64/kernel/irq.c
+++ b/arch/ia64/kernel/irq.c
@@ -405,7 +405,7 @@ void enable_irq(unsigned int irq)
 	spin_lock_irqsave(&desc->lock, flags);
 	switch (desc->depth) {
 	case 1: {
-		unsigned int status = desc->status & ~(IRQ_DISABLED | IRQ_INPROGRESS);
+		unsigned int status = desc->status & ~IRQ_DISABLED;
 		desc->status = status;
 		if ((status & (IRQ_PENDING | IRQ_REPLAY)) == IRQ_PENDING) {
 			desc->status = status | IRQ_REPLAY;
diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c
index 0a7ad2a30565..e8ea942fa2a1 100644
--- a/arch/ia64/kernel/module.c
+++ b/arch/ia64/kernel/module.c
@@ -322,6 +322,10 @@ module_alloc (unsigned long size)
 void
 module_free (struct module *mod, void *module_region)
 {
+	if (mod->arch.init_unw_table && module_region == mod->module_init) {
+		unw_remove_unwind_table(mod->arch.init_unw_table);
+		mod->arch.init_unw_table = NULL;
+	}
 	vfree(module_region);
 }
 
@@ -843,28 +847,92 @@ apply_relocate (Elf64_Shdr *sechdrs, const char *strtab, unsigned int symindex,
 	return -ENOEXEC;
 }
 
+/*
+ * Modules contain a single unwind table which covers both the core and the init text
+ * sections but since the two are not contiguous, we need to split this table up such that
+ * we can register (and unregister) each "segment" seperately.  Fortunately, this sounds
+ * more complicated than it really is.
+ */
+static void
+register_unwind_table (struct module *mod)
+{
+	struct unw_table_entry *start = (void *) mod->arch.unwind->sh_addr;
+	struct unw_table_entry *end = start + mod->arch.unwind->sh_size / sizeof (*start);
+	struct unw_table_entry tmp, *e1, *e2, *core, *init;
+	unsigned long num_init = 0, num_core = 0;
+
+	/* First, count how many init and core unwind-table entries there are.  */
+	for (e1 = start; e1 < end; ++e1)
+		if (in_init(mod, e1->start_offset))
+			++num_init;
+		else
+			++num_core;
+	/*
+	 * Second, sort the table such that all unwind-table entries for the init and core
+	 * text sections are nicely separated.  We do this with a stupid bubble sort
+	 * (unwind tables don't get ridiculously huge).
+	 */
+	for (e1 = start; e1 < end; ++e1) {
+		for (e2 = e1 + 1; e2 < end; ++e2) {
+			if (e2->start_offset < e1->start_offset) {
+				tmp = *e1;
+				*e1 = *e2;
+				*e2 = tmp;
+			}
+		}
+	}
+	/*
+	 * Third, locate the init and core segments in the unwind table:
+	 */
+	if (in_init(mod, start->start_offset)) {
+		init = start;
+		core = start + num_init;
+	} else {
+		core = start;
+		init = start + num_core;
+	}
+
+	DEBUGP("%s: name=%s, gp=%lx, num_init=%lu, num_core=%lu\n", __FUNCTION__,
+	       mod->name, mod->arch.gp, num_init, num_core);
+
+	/*
+	 * Fourth, register both tables (if not empty).
+	 */
+	if (num_core > 0) {
+		mod->arch.core_unw_table = unw_add_unwind_table(mod->name, 0, mod->arch.gp,
+								core, core + num_core);
+		DEBUGP("%s:  core: handle=%p [%p-%p)\n", __FUNCTION__,
+		       mod->arch.core_unw_table, core, core + num_core);
+	}
+	if (num_init > 0) {
+		mod->arch.init_unw_table = unw_add_unwind_table(mod->name, 0, mod->arch.gp,
+								init, init + num_init);
+		DEBUGP("%s:  init: handle=%p [%p-%p)\n", __FUNCTION__,
+		       mod->arch.init_unw_table, init, init + num_init);
+	}
+}
+
 int
 module_finalize (const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *mod)
 {
 	DEBUGP("%s: init: entry=%p\n", __FUNCTION__, mod->init);
 	if (mod->arch.unwind)
-		mod->arch.unw_table = unw_add_unwind_table(mod->name, 0, mod->arch.gp,
-							   (void *) mod->arch.unwind->sh_addr,
-							   ((void *) mod->arch.unwind->sh_addr
-							    + mod->arch.unwind->sh_size));
+		register_unwind_table(mod);
 	return 0;
 }
 
 void
 module_arch_cleanup (struct module *mod)
 {
-	if (mod->arch.unwind)
-		unw_remove_unwind_table(mod->arch.unw_table);
+	if (mod->arch.init_unw_table)
+		unw_remove_unwind_table(mod->arch.init_unw_table);
+	if (mod->arch.core_unw_table)
+		unw_remove_unwind_table(mod->arch.core_unw_table);
 }
 
 #ifdef CONFIG_SMP
 void
-percpu_modcopy (void  *pcpudst, const void *src, unsigned long size)
+percpu_modcopy (void *pcpudst, const void *src, unsigned long size)
 {
 	unsigned int i;
 	for (i = 0; i < NR_CPUS; i++)
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 606c9f9201f2..e0d9dba8cbd4 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -202,8 +202,8 @@
 #define GET_PMU_OWNER()		pfm_get_cpu_var(pmu_owner)
 #define GET_PMU_CTX()		pfm_get_cpu_var(pmu_ctx)
 
-#define LOCK_PFS()	    	spin_lock(&pfm_sessions.pfs_lock)
-#define UNLOCK_PFS()	    	spin_unlock(&pfm_sessions.pfs_lock)
+#define LOCK_PFS(g)	    	spin_lock_irqsave(&pfm_sessions.pfs_lock, g)
+#define UNLOCK_PFS(g)	    	spin_unlock_irqrestore(&pfm_sessions.pfs_lock, g)
 
 #define PFM_REG_RETFLAG_SET(flags, val)	do { flags &= ~PFM_REG_RETFL_MASK; flags |= (val); } while(0)
 
@@ -618,6 +618,7 @@ static struct file_system_type pfm_fs_type = {
 	.get_sb   = pfmfs_get_sb,
 	.kill_sb  = kill_anon_super,
 };
+
 DEFINE_PER_CPU(unsigned long, pfm_syst_info);
 DEFINE_PER_CPU(struct task_struct *, pmu_owner);
 DEFINE_PER_CPU(pfm_context_t  *, pmu_ctx);
@@ -634,6 +635,8 @@ static struct file_operations pfm_file_ops;
 static void pfm_lazy_save_regs (struct task_struct *ta);
 #endif
 
+void dump_pmu_state(const char *);
+
 /*
  * the HP simulator must be first because
  * CONFIG_IA64_HP_SIM is independent of CONFIG_MCKINLEY or CONFIG_ITANIUM
@@ -1283,10 +1286,11 @@ out:
 static int
 pfm_reserve_session(struct task_struct *task, int is_syswide, unsigned int cpu)
 {
+	unsigned long flags;
 	/*
 	 * validy checks on cpu_mask have been done upstream
 	 */
-	LOCK_PFS();
+	LOCK_PFS(flags);
 
 	DPRINT(("in sys_sessions=%u task_sessions=%u dbregs=%u syswide=%d cpu=%u\n",
 		pfm_sessions.pfs_sys_sessions,
@@ -1325,7 +1329,7 @@ pfm_reserve_session(struct task_struct *task, int is_syswide, unsigned int cpu)
 		is_syswide,
 		cpu));
 
-	UNLOCK_PFS();
+	UNLOCK_PFS(flags);
 
 	return 0;
 
@@ -1334,7 +1338,7 @@ error_conflict:
   		pfm_sessions.pfs_sys_session[cpu]->pid,
 		smp_processor_id()));
 abort:
-	UNLOCK_PFS();
+	UNLOCK_PFS(flags);
 
 	return -EBUSY;
 
@@ -1343,11 +1347,11 @@ abort:
 static int
 pfm_unreserve_session(pfm_context_t *ctx, int is_syswide, unsigned int cpu)
 {
-
+	unsigned long flags;
 	/*
 	 * validy checks on cpu_mask have been done upstream
 	 */
-	LOCK_PFS();
+	LOCK_PFS(flags);
 
 	DPRINT(("in sys_sessions=%u task_sessions=%u dbregs=%u syswide=%d cpu=%u\n",
 		pfm_sessions.pfs_sys_sessions,
@@ -1380,7 +1384,7 @@ pfm_unreserve_session(pfm_context_t *ctx, int is_syswide, unsigned int cpu)
 		is_syswide,
 		cpu));
 
-	UNLOCK_PFS();
+	UNLOCK_PFS(flags);
 
 	return 0;
 }
@@ -1655,7 +1659,7 @@ pfm_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned lon
 }
 
 /*
- * context is locked when coming here
+ * context is locked when coming here and interrupts are disabled
  */
 static inline int
 pfm_do_fasync(int fd, struct file *filp, pfm_context_t *ctx, int on)
@@ -1789,6 +1793,7 @@ pfm_syswide_cleanup_other_cpu(pfm_context_t *ctx)
  * even if the task itself is in the middle of being ctxsw out.
  */
 static int pfm_context_unload(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs);
+
 static int
 pfm_close(struct inode *inode, struct file *filp)
 {
@@ -1803,10 +1808,6 @@ pfm_close(struct inode *inode, struct file *filp)
 	int free_possible = 1;
 	int state, is_system;
 
-	{ u64 psr = pfm_get_psr();
-	  BUG_ON((psr & IA64_PSR_I) == 0UL);
-	}
-
 	DPRINT(("pfm_close called private=%p\n", filp->private_data));
 
 	if (!inode) {
@@ -1815,7 +1816,7 @@ pfm_close(struct inode *inode, struct file *filp)
 	}
 
 	if (PFM_IS_FILE(filp) == 0) {
-		printk(KERN_ERR "perfmon: pfm_close: bad magic [%d]\n", current->pid);
+		DPRINT(("bad magic for [%d]\n", current->pid));
 		return -EBADF;
 	}
 
@@ -1824,6 +1825,23 @@ pfm_close(struct inode *inode, struct file *filp)
 		printk(KERN_ERR "perfmon: pfm_close: NULL ctx [%d]\n", current->pid);
 		return -EBADF;
 	}
+	/*
+	 * remove our file from the async queue, if we use this mode.
+	 * This can be done without the context being protected. We come
+	 * here when the context has become unreacheable by other tasks.
+	 *
+	 * We may still have active monitoring at this point and we may
+	 * end up in pfm_overflow_handler(). However, fasync_helper()
+	 * operates with interrupts disabled and it cleans up the
+	 * queue. If the PMU handler is called prior to entering
+	 * fasync_helper() then it will send a signal. If it is
+	 * invoked after, it will find an empty queue and no
+	 * signal will be sent. In both case, we are safe
+	 */
+	if (filp->f_flags & FASYNC) {
+		DPRINT(("[%d] cleaning up async_queue=%p\n", current->pid, ctx->ctx_async_queue));
+		pfm_do_fasync (-1, filp, ctx, 0);
+	}
 
 	PROTECT_CTX(ctx, flags);
 
@@ -1832,24 +1850,17 @@ pfm_close(struct inode *inode, struct file *filp)
 
 	task = PFM_CTX_TASK(ctx);
 
-	/*
-	 * remove our file from the async queue, if we use it
-	 */
-	if (filp->f_flags & FASYNC) {
-		DPRINT(("[%d] before async_queue=%p\n", current->pid, ctx->ctx_async_queue));
-		pfm_do_fasync (-1, filp, ctx, 0);
-		DPRINT(("[%d] after async_queue=%p\n", current->pid, ctx->ctx_async_queue));
-	}
 
+	regs = ia64_task_regs(task);
 
-	DPRINT(("[%d] ctx_state=%d\n", current->pid, state));
+	DPRINT(("[%d] ctx_state=%d is_current=%d\n", 
+		current->pid, state,
+		task == current ? 1 : 0));
 
 	if (state == PFM_CTX_UNLOADED || state == PFM_CTX_TERMINATED) {
 		goto doit;
 	}
 
-	regs = ia64_task_regs(task);
-
 	/*
 	 * context still loaded/masked and self monitoring,
 	 * we stop/unload and we destroy right here
@@ -1898,12 +1909,11 @@ pfm_close(struct inode *inode, struct file *filp)
 
 			ctx->ctx_state = PFM_CTX_TERMINATED;
 
-			DPRINT(("[%d] ctx_state=%d\n", current->pid, state));
+			DPRINT(("[%d] ctx_state=%d\n", current->pid, ctx->ctx_state));
 		}
 		goto doit;
 	}
 
-
 	/*
 	 * The task is currently blocked or will block after an overflow.
 	 * we must force it to wakeup to get out of the
@@ -3482,6 +3492,7 @@ int
 pfm_use_debug_registers(struct task_struct *task)
 {
 	pfm_context_t *ctx = task->thread.pfm_context;
+	unsigned long flags;
 	int ret = 0;
 
 	if (pmu_conf.use_rr_dbregs == 0) return 0;
@@ -3503,7 +3514,7 @@ pfm_use_debug_registers(struct task_struct *task)
 	 */
 	if (ctx && ctx->ctx_fl_using_dbreg == 1) return -1;
 
-	LOCK_PFS();
+	LOCK_PFS(flags);
 
 	/*
 	 * We cannot allow setting breakpoints when system wide monitoring
@@ -3519,7 +3530,7 @@ pfm_use_debug_registers(struct task_struct *task)
 		  pfm_sessions.pfs_sys_use_dbregs,
 		  task->pid, ret));
 
-	UNLOCK_PFS();
+	UNLOCK_PFS(flags);
 
 	return ret;
 }
@@ -3535,11 +3546,12 @@ pfm_use_debug_registers(struct task_struct *task)
 int
 pfm_release_debug_registers(struct task_struct *task)
 {
+	unsigned long flags;
 	int ret;
 
 	if (pmu_conf.use_rr_dbregs == 0) return 0;
 
-	LOCK_PFS();
+	LOCK_PFS(flags);
 	if (pfm_sessions.pfs_ptrace_use_dbregs == 0) {
 		printk(KERN_ERR "perfmon: invalid release for [%d] ptrace_use_dbregs=0\n", task->pid);
 		ret = -1;
@@ -3547,7 +3559,7 @@ pfm_release_debug_registers(struct task_struct *task)
 		pfm_sessions.pfs_ptrace_use_dbregs--;
 		ret = 0;
 	}
-	UNLOCK_PFS();
+	UNLOCK_PFS(flags);
 
 	return ret;
 }
@@ -3723,7 +3735,6 @@ pfm_debug(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 		memset(pfm_stats, 0, sizeof(pfm_stats));
 		for(m=0; m < NR_CPUS; m++) pfm_stats[m].pfm_ovfl_intr_cycles_min = ~0UL;
 	}
-
 	return 0;
 }
 
@@ -3735,6 +3746,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_
 {
 	struct thread_struct *thread = NULL;
 	pfarg_dbreg_t *req = (pfarg_dbreg_t *)arg;
+	unsigned long flags;
 	dbreg_t dbreg;
 	unsigned int rnum;
 	int first_time;
@@ -3793,7 +3805,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_
 	 * written after the context is loaded
 	 */
 	if (is_loaded) {
-		LOCK_PFS();
+		LOCK_PFS(flags);
 
 		if (first_time && is_system) {
 			if (pfm_sessions.pfs_ptrace_use_dbregs)
@@ -3801,7 +3813,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_
 			else
 				pfm_sessions.pfs_sys_use_dbregs++;
 		}
-		UNLOCK_PFS();
+		UNLOCK_PFS(flags);
 	}
 
 	if (ret != 0) return ret;
@@ -3902,11 +3914,11 @@ abort_mission:
 	 * in case it was our first attempt, we undo the global modifications
 	 */
 	if (first_time) {
-		LOCK_PFS();
+		LOCK_PFS(flags);
 		if (ctx->ctx_fl_system) {
 			pfm_sessions.pfs_sys_use_dbregs--;
 		}
-		UNLOCK_PFS();
+		UNLOCK_PFS(flags);
 		ctx->ctx_fl_using_dbreg = 0;
 	}
 	/*
@@ -3959,7 +3971,11 @@ pfm_stop(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 		DPRINT(("[%d] should be running on CPU%d\n", current->pid, ctx->ctx_cpu));
 		return -EBUSY;
 	}
-
+	DPRINT(("current [%d] task [%d] ctx_state=%d is_system=%d\n",
+		current->pid,
+		PFM_CTX_TASK(ctx)->pid,
+		state,
+		is_system));
 	/*
 	 * in system mode, we need to update the PMU directly
 	 * and the user level state of the caller, which may not
@@ -4157,6 +4173,7 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	struct task_struct *task;
 	struct thread_struct *thread;
 	struct pfm_context_t *old;
+	unsigned long flags;
 #ifndef CONFIG_SMP
 	struct task_struct *owner_task = NULL;
 #endif
@@ -4217,7 +4234,7 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			DPRINT(("load_pid [%d] task is debugged, cannot load range restrictions\n", req->load_pid));
 			goto error;
 		}
-		LOCK_PFS();
+		LOCK_PFS(flags);
 
 		if (is_system) {
 			if (pfm_sessions.pfs_ptrace_use_dbregs) {
@@ -4230,7 +4247,7 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			}
 		}
 
-		UNLOCK_PFS();
+		UNLOCK_PFS(flags);
 
 		if (ret) goto error;
 	}
@@ -4377,9 +4394,9 @@ error:
 	 * we must undo the dbregs setting (for system-wide)
 	 */
 	if (ret && set_dbregs) {
-		LOCK_PFS();
+		LOCK_PFS(flags);
 		pfm_sessions.pfs_sys_use_dbregs--;
-		UNLOCK_PFS();
+		UNLOCK_PFS(flags);
 	}
 	/*
 	 * release task, there is now a link with the context
@@ -4605,11 +4622,14 @@ pfm_exit_thread(struct task_struct *task)
 			printk(KERN_ERR "perfmon: pfm_exit_thread [%d] unexpected state=%d\n", task->pid, state);
 			break;
 	}
+	UNPROTECT_CTX(ctx, flags);
+
 	{ u64 psr = pfm_get_psr();
 	  BUG_ON(psr & (IA64_PSR_UP|IA64_PSR_PP));
 	  BUG_ON(GET_PMU_OWNER());
+	  BUG_ON(ia64_psr(regs)->up);
+	  BUG_ON(ia64_psr(regs)->pp);
 	}
-	UNPROTECT_CTX(ctx, flags);
 
 	/*
 	 * All memory free operations (especially for vmalloc'ed memory)
@@ -5488,7 +5508,7 @@ pfm_proc_info(char *page)
 	char *p = page;
 	struct list_head * pos;
 	pfm_buffer_fmt_t * entry;
-	unsigned long psr;
+	unsigned long psr, flags;
 	int online_cpus = 0;
 	int i;
 
@@ -5528,7 +5548,7 @@ pfm_proc_info(char *page)
   		}
 	}
 
-	LOCK_PFS();
+	LOCK_PFS(flags);
 	p += sprintf(p, "proc_sessions             : %u\n"
 			"sys_sessions              : %u\n"
 			"sys_use_dbregs            : %u\n"
@@ -5537,7 +5557,7 @@ pfm_proc_info(char *page)
 			pfm_sessions.pfs_sys_sessions,
 			pfm_sessions.pfs_sys_use_dbregs,
 			pfm_sessions.pfs_ptrace_use_dbregs);
-	UNLOCK_PFS();
+	UNLOCK_PFS(flags);
 
 	spin_lock(&pfm_buffer_fmt_lock);
 
@@ -5712,10 +5732,6 @@ pfm_save_regs(struct task_struct *task)
 	 */
 	ctx->ctx_saved_psr_up = psr & IA64_PSR_UP;
 
-	{ u64 foo = pfm_get_psr();
-	  BUG_ON(foo & ((IA64_PSR_UP|IA64_PSR_PP)));
-	}
-
 	/*
 	 * release ownership of this PMU.
 	 * PM interrupts are masked, so nothing
@@ -5771,6 +5787,8 @@ pfm_save_regs(struct task_struct *task)
 	 */
 	psr = pfm_get_psr();
 
+	BUG_ON(foo & (IA64_PSR_I));
+
 	/*
 	 * stop monitoring:
 	 * This is the last instruction which may generate an overflow
@@ -5785,12 +5803,6 @@ pfm_save_regs(struct task_struct *task)
 	 */
 	ctx->ctx_saved_psr_up = psr & IA64_PSR_UP;
 
-#if 1
-	{ u64 foo = pfm_get_psr();
-	  BUG_ON(foo & (IA64_PSR_I));
-	  BUG_ON(foo & ((IA64_PSR_UP|IA64_PSR_PP)));
-	}
-#endif
 	return;
 save_error:
 	printk(KERN_ERR "perfmon: pfm_save_regs CPU%d [%d] NULL context PM_VALID=%ld\n",
@@ -5805,11 +5817,9 @@ pfm_lazy_save_regs (struct task_struct *task)
 	struct thread_struct *t;
 	unsigned long flags;
 
-#if 1
-	{ u64 foo  = pfm_get_psr();
-	  BUG_ON(foo & IA64_PSR_UP);
+	{ u64 psr  = pfm_get_psr();
+	  BUG_ON(psr & IA64_PSR_UP);
 	}
-#endif
 
 	ctx = PFM_GET_CTX(task);
 	t   = &task->thread;
@@ -5851,7 +5861,7 @@ pfm_lazy_save_regs (struct task_struct *task)
 	/*
 	 * unfreeze PMU if had pending overflows
 	 */
-	if (t->pmcs[0] & ~1UL) pfm_unfreeze_pmu();
+	if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
 
 	/*
 	 * now get can unmask PMU interrupts, they will
@@ -5900,10 +5910,8 @@ pfm_load_regs (struct task_struct *task)
 	flags = pfm_protect_ctx_ctxsw(ctx);
 	psr   = pfm_get_psr();
 
-#if 1
 	BUG_ON(psr & (IA64_PSR_UP|IA64_PSR_PP));
 	BUG_ON(psr & IA64_PSR_I);
-#endif
 
 	if (unlikely(ctx->ctx_state == PFM_CTX_ZOMBIE)) {
 		struct pt_regs *regs = ia64_task_regs(task);
@@ -6060,10 +6068,8 @@ pfm_load_regs (struct task_struct *task)
 	t     = &task->thread;
 	psr   = pfm_get_psr();
 
-#if 1
 	BUG_ON(psr & (IA64_PSR_UP|IA64_PSR_PP));
 	BUG_ON(psr & IA64_PSR_I);
-#endif
 
 	/*
 	 * we restore ALL the debug registers to avoid picking up
@@ -6218,7 +6224,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 		/*
 		 * clear whatever overflow status bits there were
 		 */
-		task->thread.pmcs[0] &= ~0x1;
+		task->thread.pmcs[0] = 0;
 	}
 	ovfl_val = pmu_conf.ovfl_val;
 	/*
@@ -6400,6 +6406,11 @@ pfm_init_percpu (void)
 	pfm_clear_psr_pp();
 	pfm_clear_psr_up();
 
+	/*
+	 * we run with the PMU not frozen at all times
+	 */
+	pfm_unfreeze_pmu();
+
 	if (smp_processor_id() == 0)
 		register_percpu_irq(IA64_PERFMON_VECTOR, &perfmon_irqaction);
 
@@ -6427,49 +6438,75 @@ pfm_init_percpu (void)
 		if (PMD_IS_IMPL(i) == 0) continue;
 		ia64_set_pmd(i, 0UL);
 	}
-
-	/*
-	 * we run with the PMU not frozen at all times
-	 */
-	pfm_unfreeze_pmu();
 }
 
 /*
  * used for debug purposes only
  */
 void
-dump_pmu_state(void)
+dump_pmu_state(const char *from)
 {
 	struct task_struct *task;
 	struct thread_struct *t;
+	struct pt_regs *regs;
 	pfm_context_t *ctx;
-	unsigned long psr;
-	int i;
+	unsigned long psr, dcr, info, flags;
+	int i, this_cpu;
+
+	local_irq_save(flags);
 
-	printk("current [%d] %s\n", current->pid, current->comm);
+	this_cpu = smp_processor_id();
+	regs     = ia64_task_regs(current);
+	info     = PFM_CPUINFO_GET();
+	dcr      = ia64_getreg(_IA64_REG_CR_DCR);
+
+	if (info == 0 && ia64_psr(regs)->pp == 0 && (dcr & IA64_DCR_PP) == 0) {
+		local_irq_restore(flags);
+		return;
+	}
+
+	printk("CPU%d from %s() current [%d] iip=0x%lx %s\n", 
+		this_cpu, 
+		from, 
+		current->pid, 
+		regs->cr_iip,
+		current->comm);
 
 	task = GET_PMU_OWNER();
 	ctx  = GET_PMU_CTX();
 
-	printk("owner [%d] ctx=%p\n", task ? task->pid : -1, ctx);
+	printk("->CPU%d owner [%d] ctx=%p\n", this_cpu, task ? task->pid : -1, ctx);
 
 	psr = pfm_get_psr();
 
-	printk("psr.pp=%ld psr.up=%ld\n", (psr >> IA64_PSR_PP_BIT) &0x1UL, (psr >> IA64_PSR_PP_BIT)&0x1UL);
+	printk("->CPU%d pmc0=0x%lx psr.pp=%d psr.up=%d dcr.pp=%d syst_info=0x%lx user_psr.up=%d user_psr.pp=%d\n", 
+		this_cpu,
+		ia64_get_pmc(0),
+		psr & IA64_PSR_PP ? 1 : 0,
+		psr & IA64_PSR_UP ? 1 : 0,
+		dcr & IA64_DCR_PP ? 1 : 0,
+		info,
+		ia64_psr(regs)->up,
+		ia64_psr(regs)->pp);
+
+	ia64_psr(regs)->up = 0;
+	ia64_psr(regs)->pp = 0;
 
 	t = &current->thread;
 
 	for (i=1; PMC_IS_LAST(i) == 0; i++) {
 		if (PMC_IS_IMPL(i) == 0) continue;
-		printk("pmc[%d]=0x%lx tpmc=0x%lx\n", i, ia64_get_pmc(i), t->pmcs[i]);
+		printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, t->pmcs[i]);
 	}
 
 	for (i=1; PMD_IS_LAST(i) == 0; i++) {
 		if (PMD_IS_IMPL(i) == 0) continue;
-		printk("pmd[%d]=0x%lx tpmd=0x%lx\n", i, ia64_get_pmd(i), t->pmds[i]);
+		printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, t->pmds[i]);
 	}
+
 	if (ctx) {
-		printk("ctx_state=%d vaddr=%p addr=%p fd=%d ctx_task=[%d] saved_psr_up=0x%lx\n",
+		printk("->CPU%d ctx_state=%d vaddr=%p addr=%p fd=%d ctx_task=[%d] saved_psr_up=0x%lx\n",
+				this_cpu,
 				ctx->ctx_state,
 				ctx->ctx_smpl_vaddr,
 				ctx->ctx_smpl_hdr,
@@ -6477,6 +6514,7 @@ dump_pmu_state(void)
 				ctx->ctx_msgq_tail,
 				ctx->ctx_saved_psr_up);
 	}
+	local_irq_restore(flags);
 }
 
 /*
@@ -6499,10 +6537,8 @@ pfm_inherit(struct task_struct *task, struct pt_regs *regs)
 	PFM_SET_WORK_PENDING(task, 0);
 
 	/*
-	 * restore default psr settings
+	 * the psr bits are already set properly in copy_threads()
 	 */
-	ia64_psr(regs)->pp = ia64_psr(regs)->up = 0;
-	ia64_psr(regs)->sp = 1;
 }
 #else  /* !CONFIG_PERFMON */
 asmlinkage long
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index ff17b9b87a14..ffa4cdd21633 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -353,9 +353,13 @@ copy_thread (int nr, unsigned long clone_flags,
 	/* copy parts of thread_struct: */
 	p->thread.ksp = (unsigned long) child_stack - 16;
 
-	/* stop some PSR bits from being inherited: */
+	/* stop some PSR bits from being inherited.
+	 * the psr.up/psr.pp bits must be cleared on fork but inherited on execve()
+	 * therefore we must specify them explicitly here and not include them in
+	 * IA64_PSR_BITS_TO_CLEAR.
+	 */
 	child_ptregs->cr_ipsr =  ((child_ptregs->cr_ipsr | IA64_PSR_BITS_TO_SET)
-				  & ~IA64_PSR_BITS_TO_CLEAR);
+			      & ~(IA64_PSR_BITS_TO_CLEAR | IA64_PSR_PP | IA64_PSR_UP));
 
 	/*
 	 * NOTE: The calling convention considers all floating point
diff --git a/arch/ia64/kernel/unwind_i.h b/arch/ia64/kernel/unwind_i.h
index ac5b3afd5c6c..1853dc45073a 100644
--- a/arch/ia64/kernel/unwind_i.h
+++ b/arch/ia64/kernel/unwind_i.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2000, 2002 Hewlett-Packard Co
+ * Copyright (C) 2000, 2002-2003 Hewlett-Packard Co
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  *
  * Kernel unwind support.
@@ -45,12 +45,6 @@ struct unw_info_block {
 	/* personality routine and language-specific data follow behind descriptors */
 };
 
-struct unw_table_entry {
-	u64 start_offset;
-	u64 end_offset;
-	u64 info_offset;
-};
-
 struct unw_table {
 	struct unw_table *next;		/* must be first member! */
 	const char *name;
diff --git a/arch/x86_64/kernel/acpi/boot.c b/arch/x86_64/kernel/acpi/boot.c
index 56aaf7859b64..f3094f0b7f17 100644
--- a/arch/x86_64/kernel/acpi/boot.c
+++ b/arch/x86_64/kernel/acpi/boot.c
@@ -251,6 +251,33 @@ acpi_parse_hpet (
 } 
 #endif
 
+#ifdef CONFIG_ACPI_BUS
+/*
+ * Set specified PIC IRQ to level triggered mode.
+ *
+ * Port 0x4d0-4d1 are ECLR1 and ECLR2, the Edge/Level Control Registers
+ * for the 8259 PIC.  bit[n] = 1 means irq[n] is Level, otherwise Edge.
+ * ECLR1 is IRQ's 0-7 (IRQ 0, 1, 2 must be 0)
+ * ECLR2 is IRQ's 8-15 (IRQ 8, 13 must be 0)
+ *
+ * As the BIOS should have done this for us,
+ * print a warning if the IRQ wasn't already set to level.
+ */
+
+void acpi_pic_set_level_irq(unsigned int irq)
+{
+	unsigned char mask = 1 << (irq & 7);
+	unsigned int port = 0x4d0 + (irq >> 3);
+	unsigned char val = inb(port);
+
+	if (!(val & mask)) {
+		printk(KERN_WARNING PREFIX "IRQ %d was Edge Triggered, "
+			"setting to Level Triggerd\n", irq);
+		outb(val | mask, port);
+	}
+}
+#endif /* CONFIG_ACPI_BUS */
+
 static unsigned long __init
 acpi_scan_rsdp (
 	unsigned long		start,
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index a97ab7acc18f..fcfa1af913c7 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -566,8 +566,14 @@ error_kernelspace:
 	incl %ebx
        /* There are two places in the kernel that can potentially fault with
           usergs. Handle them here. The exception handlers after
-	  iret run with kernel gs again, so don't set the user space flag. */
-	cmpq $iret_label,RIP(%rsp) 
+	   iret run with kernel gs again, so don't set the user space flag.
+	   B stepping K8s sometimes report an truncated RIP for IRET 
+	   exceptions returning to compat mode. Check for these here too. */
+	leaq iret_label(%rip),%rbp
+	cmpq %rbp,RIP(%rsp) 
+	je   error_swapgs
+	movl %ebp,%ebp	/* zero extend */
+	cmpq %rbp,RIP(%rsp) 
 	je   error_swapgs
 	cmpq $gs_change,RIP(%rsp)
         je   error_swapgs
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 65c7acadef46..1148ddec35a3 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -622,11 +622,13 @@ static inline int IO_APIC_irq_trigger(int irq)
 	return 0;
 }
 
-int irq_vector[NR_IRQS] = { FIRST_DEVICE_VECTOR , 0 };
+/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */
+u8 irq_vector[NR_IRQ_VECTORS] = { FIRST_DEVICE_VECTOR , 0 };
 
 static int __init assign_irq_vector(int irq)
 {
 	static int current_vector = FIRST_DEVICE_VECTOR, offset = 0;
+	BUG_ON(irq >= NR_IRQ_VECTORS);
 	if (IO_APIC_VECTOR(irq) > 0)
 		return IO_APIC_VECTOR(irq);
 next:
diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 19379a38687e..4a0804165984 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -189,8 +189,7 @@ void pda_init(int cpu)
 	pda->irqstackptr += IRQSTACKSIZE-64;
 } 
 
-#define EXCEPTION_STK_ORDER 0 /* >= N_EXCEPTION_STACKS*EXCEPTION_STKSZ */
-char boot_exception_stacks[N_EXCEPTION_STACKS*EXCEPTION_STKSZ];
+char boot_exception_stacks[N_EXCEPTION_STACKS * EXCEPTION_STKSZ];
 
 void syscall_init(void)
 {
@@ -226,15 +225,12 @@ void __init cpu_init (void)
 #endif
 	struct tss_struct * t = &init_tss[cpu];
 	unsigned long v, efer; 
-	char *estacks; 
+	char *estacks = NULL; 
 	struct task_struct *me;
 
 	/* CPU 0 is initialised in head64.c */
 	if (cpu != 0) {
 		pda_init(cpu);
-		estacks = (char *)__get_free_pages(GFP_ATOMIC, 0); 
-		if (!estacks)
-			panic("Can't allocate exception stacks for CPU %d\n",cpu);
 	} else 
 		estacks = boot_exception_stacks; 
 
@@ -282,10 +278,15 @@ void __init cpu_init (void)
 	/*
 	 * set up and load the per-CPU TSS
 	 */
-	estacks += EXCEPTION_STKSZ;
 	for (v = 0; v < N_EXCEPTION_STACKS; v++) {
-		t->ist[v] = (unsigned long)estacks;
+		if (cpu) {
+			estacks = (char *)__get_free_pages(GFP_ATOMIC, 0);
+			if (!estacks)
+				panic("Cannot allocate exception stack %ld %d\n",
+				      v, cpu); 
+		}
 		estacks += EXCEPTION_STKSZ;
+		t->ist[v] = (unsigned long)estacks;
 	}
 
 	t->io_bitmap_base = INVALID_IO_BITMAP_OFFSET;
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 299b3d50e7df..7ee5cea794dc 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -487,25 +487,3 @@ asmlinkage void smp_call_function_interrupt(void)
 		atomic_inc(&call_data->finished);
 	}
 }
-
-/* Slow. Should be only used for debugging. */
-int slow_smp_processor_id(void)
-{ 
-	int stack_location;
-	unsigned long sp = (unsigned long)&stack_location; 
-	int offset = 0, cpu;
-
-	for (offset = 0; next_cpu(offset, cpu_online_map) < NR_CPUS; offset = cpu + 1) {
-		cpu = next_cpu(offset, cpu_online_map);
-
-		if (sp >= (u64)cpu_pda[cpu].irqstackptr - IRQSTACKSIZE && 
-		    sp <= (u64)cpu_pda[cpu].irqstackptr)
-			return cpu;
-
-		unsigned long estack = init_tss[cpu].ist[0] - EXCEPTION_STKSZ;
-		if (sp >= estack && sp <= estack+(1<<(PAGE_SHIFT+EXCEPTION_STK_ORDER)))
-			return cpu;			
-	}
-
-	return stack_smp_processor_id();
-} 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index a57945080f41..8b0791621425 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -111,6 +111,14 @@ void do_gettimeofday(struct timeval *tv)
 		sec = xtime.tv_sec;
 		usec = xtime.tv_nsec / 1000;
 
+		/*
+		 * If time_adjust is negative then NTP is slowing the clock
+		 * so make sure not to go into next possible interval.
+		 * Better to lose some accuracy than have time go backwards..
+		 */
+		if (unlikely(time_adjust < 0) && usec > tickadj)
+			usec = tickadj;
+
 		t = (jiffies - wall_jiffies) * (1000000L / HZ) +
 			do_gettimeoffset();
 		usec += t;
@@ -477,22 +485,28 @@ unsigned long get_cmos_time(void)
 static unsigned int  ref_freq = 0;
 static unsigned long loops_per_jiffy_ref = 0;
 
-//static unsigned long fast_gettimeoffset_ref = 0;
 static unsigned long cpu_khz_ref = 0;
 
 static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
 				 void *data)
 {
         struct cpufreq_freqs *freq = data;
+	unsigned long *lpj;
+
+#ifdef CONFIG_SMP
+	lpj = &cpu_data[freq->cpu].loops_per_jiffy;
+#else
+	lpj = &boot_cpu_data.loops_per_jiffy;
+#endif
 
 	if (!ref_freq) {
 		ref_freq = freq->old;
-		loops_per_jiffy_ref = cpu_data[freq->cpu].loops_per_jiffy;
+		loops_per_jiffy_ref = *lpj;
 		cpu_khz_ref = cpu_khz;
 	}
         if ((val == CPUFREQ_PRECHANGE  && freq->old < freq->new) ||
             (val == CPUFREQ_POSTCHANGE && freq->old > freq->new)) {
-                cpu_data[freq->cpu].loops_per_jiffy =
+                *lpj =
 		cpufreq_scale(loops_per_jiffy_ref, ref_freq, freq->new);
 
 		cpu_khz = cpufreq_scale(cpu_khz_ref, ref_freq, freq->new);
diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c
index 2042d9295796..3208ad9c98f3 100644
--- a/arch/x86_64/kernel/x8664_ksyms.c
+++ b/arch/x86_64/kernel/x8664_ksyms.c
@@ -153,7 +153,7 @@ EXPORT_SYMBOL_GPL(unset_nmi_callback);
 
 extern void * memset(void *,int,__kernel_size_t);
 extern size_t strlen(const char *);
-extern char * bcopy(const char * src, char * dest, int count);
+extern void bcopy(const char * src, char * dest, int count);
 extern void * memmove(void * dest,const void *src,size_t count);
 extern char * strcpy(char * dest,const char *src);
 extern int strcmp(const char * cs,const char * ct);
diff --git a/arch/x86_64/mm/extable.c b/arch/x86_64/mm/extable.c
index 710e9372a9d6..8b62676f3e73 100644
--- a/arch/x86_64/mm/extable.c
+++ b/arch/x86_64/mm/extable.c
@@ -14,6 +14,10 @@ search_extable(const struct exception_table_entry *first,
 	       const struct exception_table_entry *last,
 	       unsigned long value)
 {
+	/* Work around a B stepping K8 bug */
+	if ((value >> 32) == 0)
+		value |= 0xffffffffUL << 32; 
+
         while (first <= last) {
 		const struct exception_table_entry *mid;
 		long diff;
diff --git a/arch/x86_64/mm/k8topology.c b/arch/x86_64/mm/k8topology.c
index 43df86e4f3ca..8df7e3f276b8 100644
--- a/arch/x86_64/mm/k8topology.c
+++ b/arch/x86_64/mm/k8topology.c
@@ -164,5 +164,8 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 		rr++; 
 	}
 
+	if (found == 1) 
+		fake_node = 1;
+
 	return 0;
 } 
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index c9ee3ee2b491..b2e05d04c566 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -104,6 +104,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 	if (nodeid + 1 > numnodes)
 		numnodes = nodeid + 1;
 	nodes_present |= (1UL << nodeid); 
+	node_set_online(nodeid);
 } 
 
 /* Initialize final allocator for a zone */
diff --git a/drivers/char/sonypi.h b/drivers/char/sonypi.h
index 81221cd44af6..a15a789e317b 100644
--- a/drivers/char/sonypi.h
+++ b/drivers/char/sonypi.h
@@ -37,7 +37,7 @@
 #ifdef __KERNEL__
 
 #define SONYPI_DRIVER_MAJORVERSION	 1
-#define SONYPI_DRIVER_MINORVERSION	20
+#define SONYPI_DRIVER_MINORVERSION	21
 
 #define SONYPI_DEVICE_MODEL_TYPE1	1
 #define SONYPI_DEVICE_MODEL_TYPE2	2
@@ -329,8 +329,8 @@ struct sonypi_eventtypes {
 	{ SONYPI_DEVICE_MODEL_TYPE2, 0x08, SONYPI_PKEY_MASK, sonypi_pkeyev },
 	{ SONYPI_DEVICE_MODEL_TYPE2, 0x11, SONYPI_BACK_MASK, sonypi_backev },
 	{ SONYPI_DEVICE_MODEL_TYPE2, 0x08, SONYPI_HELP_MASK, sonypi_helpev },
-	{ SONYPI_DEVICE_MODEL_TYPE2, 0x08, SONYPI_ZOOM_MASK, sonypi_zoomev },
-	{ SONYPI_DEVICE_MODEL_TYPE2, 0x08, SONYPI_THUMBPHRASE_MASK, sonypi_thumbphraseev },
+	{ SONYPI_DEVICE_MODEL_TYPE2, 0x21, SONYPI_ZOOM_MASK, sonypi_zoomev },
+	{ SONYPI_DEVICE_MODEL_TYPE2, 0x20, SONYPI_THUMBPHRASE_MASK, sonypi_thumbphraseev },
 	{ SONYPI_DEVICE_MODEL_TYPE2, 0x31, SONYPI_MEMORYSTICK_MASK, sonypi_memorystickev },
 	{ SONYPI_DEVICE_MODEL_TYPE2, 0x41, SONYPI_BATTERY_MASK, sonypi_batteryev },
 
diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c
index d63939c82407..9c9dd3e3d2b9 100644
--- a/drivers/input/keyboard/atkbd.c
+++ b/drivers/input/keyboard/atkbd.c
@@ -184,11 +184,19 @@ static irqreturn_t atkbd_interrupt(struct serio *serio, unsigned char data,
 		atkbd->resend = 0;
 #endif
 
+	switch (code) {
+		case ATKBD_RET_ACK:
+			atkbd->ack = 1;
+			goto out;
+		case ATKBD_RET_NAK:
+			atkbd->ack = -1;
+			goto out;
+	}
+
 	if (atkbd->translated) do {
 
 		if (atkbd->emul != 1) {
-			if (code == ATKBD_RET_EMUL0 || code == ATKBD_RET_EMUL1 ||
-			    code == ATKBD_RET_ACK || code == ATKBD_RET_NAK)
+			if (code == ATKBD_RET_EMUL0 || code == ATKBD_RET_EMUL1)
 				break;
 			if (code == ATKBD_RET_BAT) {
 				if (!atkbd->bat_xl)
@@ -212,15 +220,6 @@ static irqreturn_t atkbd_interrupt(struct serio *serio, unsigned char data,
 
 	} while (0);
 
-	switch (code) {
-		case ATKBD_RET_ACK:
-			atkbd->ack = 1;
-			goto out;
-		case ATKBD_RET_NAK:
-			atkbd->ack = -1;
-			goto out;
-	}
-
 	if (atkbd->cmdcnt) {
 		atkbd->cmdbuf[--atkbd->cmdcnt] = code;
 		goto out;
diff --git a/drivers/media/video/meye.h b/drivers/media/video/meye.h
index 283121482784..5e1f6589bc54 100644
--- a/drivers/media/video/meye.h
+++ b/drivers/media/video/meye.h
@@ -31,7 +31,7 @@
 #define _MEYE_PRIV_H_
 
 #define MEYE_DRIVER_MAJORVERSION	1
-#define MEYE_DRIVER_MINORVERSION	7
+#define MEYE_DRIVER_MINORVERSION	8
 
 #include <linux/config.h>
 #include <linux/types.h>
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 72ed4f97a8ae..11f64401f1bd 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -964,13 +964,17 @@ int fat_fill_super(struct super_block *sb, void *data, int silent,
 		error = first;
 		goto out_fail;
 	}
-	if (FAT_FIRST_ENT(sb, media) != first
-	    && (media != 0xf8 || FAT_FIRST_ENT(sb, 0xfe) != first)) {
-		if (!silent) {
+	if (FAT_FIRST_ENT(sb, media) == first) {
+		/* all is as it should be */
+	} else if (media == 0xf8 && FAT_FIRST_ENT(sb, 0xfe) == first) {
+		/* bad, reported on pc9800 */
+	} else if (first == 0) {
+		/* bad, reported with a SmartMedia card */
+	} else {
+		if (!silent)
 			printk(KERN_ERR "FAT: invalid first entry of FAT "
 			       "(0x%x != 0x%x)\n",
 			       FAT_FIRST_ENT(sb, media), first);
-		}
 		goto out_invalid;
 	}
 
diff --git a/include/asm-h8300/smplock.h b/include/asm-h8300/smplock.h
deleted file mode 100644
index 96565069c988..000000000000
--- a/include/asm-h8300/smplock.h
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * <asm/smplock.h>
- *
- * Default SMP lock implementation
- */
-#include <linux/interrupt.h>
-#include <linux/spinlock.h>
-
-extern spinlock_t kernel_flag;
-
-#define kernel_locked()		spin_is_locked(&kernel_flag)
-
-/*
- * Release global kernel lock and global interrupt lock
- */
-#define release_kernel_lock(task, cpu) \
-do { \
-	if (task->lock_depth >= 0) \
-		spin_unlock(&kernel_flag); \
-	release_irqlock(cpu); \
-	__sti(); \
-} while (0)
-
-/*
- * Re-acquire the kernel lock
- */
-#define reacquire_kernel_lock(task) \
-do { \
-	if (task->lock_depth >= 0) \
-		spin_lock(&kernel_flag); \
-} while (0)
-
-
-/*
- * Getting the big kernel lock.
- *
- * This cannot happen asynchronously,
- * so we only need to worry about other
- * CPU's.
- */
-extern __inline__ void lock_kernel(void)
-{
-	if (!++current->lock_depth)
-		spin_lock(&kernel_flag);
-}
-
-extern __inline__ void unlock_kernel(void)
-{
-	if (--current->lock_depth < 0)
-		spin_unlock(&kernel_flag);
-}
diff --git a/include/asm-ia64/io.h b/include/asm-ia64/io.h
index 297efb06c347..a4c79e29438c 100644
--- a/include/asm-ia64/io.h
+++ b/include/asm-ia64/io.h
@@ -72,6 +72,9 @@ phys_to_virt (unsigned long address)
 	return (void *) (address + PAGE_OFFSET);
 }
 
+#define ARCH_HAS_VALID_PHYS_ADDR_RANGE
+extern int valid_phys_addr_range (unsigned long addr, size_t *count); /* efi.c */
+
 /*
  * The following two macros are deprecated and scheduled for removal.
  * Please use the PCI-DMA interface defined in <asm/pci.h> instead.
diff --git a/include/asm-ia64/module.h b/include/asm-ia64/module.h
index d5b6ec9a1cec..85c82bd819f2 100644
--- a/include/asm-ia64/module.h
+++ b/include/asm-ia64/module.h
@@ -18,7 +18,8 @@ struct mod_arch_specific {
 	struct elf64_shdr *unwind;	/* unwind-table section */
 	unsigned long gp;		/* global-pointer for module */
 
-	void *unw_table;		/* unwind-table cookie returned by unwinder */
+	void *core_unw_table;		/* core unwind-table cookie returned by unwinder */
+	void *init_unw_table;		/* init unwind-table cookie returned by unwinder */
 	unsigned int next_got_entry;	/* index of next available got entry */
 };
 
diff --git a/include/asm-ia64/namei.h b/include/asm-ia64/namei.h
index b35693c6a0c3..a22af247726d 100644
--- a/include/asm-ia64/namei.h
+++ b/include/asm-ia64/namei.h
@@ -9,7 +9,7 @@
 #include <asm/ptrace.h>
 #include <asm/system.h>
 
-#define EMUL_PREFIX_LINUX_IA32 "emul/ia32-linux/"
+#define EMUL_PREFIX_LINUX_IA32 "/emul/ia32-linux/"
 
 static inline char *
 __emul_prefix (void)
diff --git a/include/asm-ia64/unwind.h b/include/asm-ia64/unwind.h
index 3f7624a10e9b..61426ad3ecdb 100644
--- a/include/asm-ia64/unwind.h
+++ b/include/asm-ia64/unwind.h
@@ -93,6 +93,12 @@ struct unw_frame_info {
  * The official API follows below:
  */
 
+struct unw_table_entry {
+	u64 start_offset;
+	u64 end_offset;
+	u64 info_offset;
+};
+
 /*
  * Initialize unwind support.
  */
diff --git a/include/asm-x86_64/hw_irq.h b/include/asm-x86_64/hw_irq.h
index aa7037368274..e1423be20796 100644
--- a/include/asm-x86_64/hw_irq.h
+++ b/include/asm-x86_64/hw_irq.h
@@ -76,8 +76,8 @@ struct hw_interrupt_type;
 
 
 #ifndef __ASSEMBLY__
-extern int irq_vector[NR_IRQS];
-#define IO_APIC_VECTOR(irq)	irq_vector[irq]
+extern u8 irq_vector[NR_IRQ_VECTORS];
+#define IO_APIC_VECTOR(irq)	((int)irq_vector[irq])
 
 /*
  * Various low-level irq details needed by irq.c, process.c,
diff --git a/include/asm-x86_64/irq.h b/include/asm-x86_64/irq.h
index 7415660ffb69..f7512e095d3d 100644
--- a/include/asm-x86_64/irq.h
+++ b/include/asm-x86_64/irq.h
@@ -22,6 +22,7 @@
  * the usable vector space is 0x20-0xff (224 vectors)
  */
 #define NR_IRQS 224
+#define NR_IRQ_VECTORS NR_IRQS
 
 static __inline__ int irq_canonicalize(int irq)
 {
diff --git a/include/asm-x86_64/pci.h b/include/asm-x86_64/pci.h
index b072364de2bd..b64f5bd97a2d 100644
--- a/include/asm-x86_64/pci.h
+++ b/include/asm-x86_64/pci.h
@@ -24,6 +24,8 @@ extern unsigned long pci_mem_start;
 #define PCIBIOS_MIN_IO		0x1000
 #define PCIBIOS_MIN_MEM		(pci_mem_start)
 
+#define PCIBIOS_MIN_CARDBUS_IO	0x4000
+
 void pcibios_config_init(void);
 struct pci_bus * pcibios_scan_root(int bus);
 extern int (*pci_config_read)(int seg, int bus, int dev, int fn, int reg, int len, u32 *value);
diff --git a/include/asm-x86_64/processor.h b/include/asm-x86_64/processor.h
index 805b7c03dbbc..f468b68e3839 100644
--- a/include/asm-x86_64/processor.h
+++ b/include/asm-x86_64/processor.h
@@ -263,8 +263,8 @@ struct thread_struct {
 #define DOUBLEFAULT_STACK 2 
 #define NMI_STACK 3 
 #define N_EXCEPTION_STACKS 3  /* hw limit: 7 */
-#define EXCEPTION_STKSZ 1024
-#define EXCEPTION_STK_ORDER 0
+#define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
+#define EXCEPTION_STACK_ORDER 0 
 
 #define start_thread(regs,new_rip,new_rsp) do { \
 	asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0));	 \
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index ba4e4ad1a189..8f425a3acdf6 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -74,15 +74,7 @@ extern __inline int hard_smp_processor_id(void)
 	return GET_APIC_ID(*(unsigned int *)(APIC_BASE+APIC_ID));
 }
 
-extern int slow_smp_processor_id(void);
-
-extern inline int safe_smp_processor_id(void)
-{ 
-	if (disable_apic)
-		return slow_smp_processor_id(); 
-	else
-		return hard_smp_processor_id();
-} 
+#define safe_smp_processor_id() (cpuid_ebx(1) >> 24) 
 
 #define cpu_online(cpu) cpu_isset(cpu, cpu_online_map)
 #endif /* !ASSEMBLY */
diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h
index d4548e5ed6dc..1a614e8c0b18 100644
--- a/include/asm-x86_64/topology.h
+++ b/include/asm-x86_64/topology.h
@@ -10,13 +10,15 @@
 /* Map the K8 CPU local memory controllers to a simple 1:1 CPU:NODE topology */
 
 extern int fake_node;
+/* This is actually a cpumask_t, but doesn't matter because we don't have
+   >BITS_PER_LONG CPUs */
 extern unsigned long cpu_online_map;
 
 #define cpu_to_node(cpu)		(fake_node ? 0 : (cpu))
 #define memblk_to_node(memblk) 	(fake_node ? 0 : (memblk))
 #define parent_node(node)		(node)
 #define node_to_first_cpu(node) 	(fake_node ? 0 : (node))
-#define node_to_cpu_mask(node)	(fake_node ? cpu_online_map : (1UL << (node)))
+#define node_to_cpumask(node)	(fake_node ? cpu_online_map : (1UL << (node)))
 #define node_to_memblk(node)		(node)
 
 static inline unsigned long pcibus_to_cpumask(int bus)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4241be4f518f..031579e7e3fc 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -483,6 +483,7 @@ do { if (atomic_dec_and_test(&(tsk)->usage)) __put_task_struct(tsk); } while(0)
 					/* Not implemented yet, only for 486*/
 #define PF_STARTING	0x00000002	/* being created */
 #define PF_EXITING	0x00000004	/* getting shut down */
+#define PF_DEAD		0x00000008	/* Dead */
 #define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
 #define PF_SUPERPRIV	0x00000100	/* used super-user privileges */
 #define PF_DUMPCORE	0x00000200	/* dumped core */
diff --git a/kernel/exit.c b/kernel/exit.c
index 25b62f5971be..32f739f782d1 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -688,6 +688,7 @@ static void exit_notify(struct task_struct *tsk)
 	}
 
 	tsk->state = TASK_ZOMBIE;
+	tsk->flags |= PF_DEAD;
 	/*
 	 * In the preemption case it must be impossible for the task
 	 * to get runnable again, so use "_raw_" unlock to keep
diff --git a/kernel/sched.c b/kernel/sched.c
index 7a13a492c62d..f4e5d8e04b9c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -742,7 +742,7 @@ static inline void finish_task_switch(task_t *prev)
 {
 	runqueue_t *rq = this_rq();
 	struct mm_struct *mm = rq->prev_mm;
-	int drop_task_ref;
+	unsigned long prev_task_flags;
 
 	rq->prev_mm = NULL;
 
@@ -757,14 +757,11 @@ static inline void finish_task_switch(task_t *prev)
 	 * be dropped twice.
 	 * 		Manfred Spraul <manfred@colorfullife.com>
 	 */
-	drop_task_ref = 0;
-	if (unlikely(prev->state & (TASK_DEAD | TASK_ZOMBIE)))
-		drop_task_ref = 1;
-
+	prev_task_flags = prev->flags;
 	finish_arch_switch(rq, prev);
 	if (mm)
 		mmdrop(mm);
-	if (drop_task_ref)
+	if (unlikely(prev_task_flags & PF_DEAD))
 		put_task_struct(prev);
 }
 
diff --git a/lib/div64.c b/lib/div64.c
index d8c699e1f70e..365719f84832 100644
--- a/lib/div64.c
+++ b/lib/div64.c
@@ -25,25 +25,34 @@
 
 uint32_t __div64_32(uint64_t *n, uint32_t base)
 {
-	uint32_t low, low2, high, rem;
-
-	low   = *n   & 0xffffffff;
-	high  = *n  >> 32;
-	rem   = high % (uint32_t)base;
-	high  = high / (uint32_t)base;
-	low2  = low >> 16;
-	low2 += rem << 16;
-	rem   = low2 % (uint32_t)base;
-	low2  = low2 / (uint32_t)base;
-	low   = low  & 0xffff;
-	low  += rem << 16;
-	rem   = low  % (uint32_t)base;
-	low   = low  / (uint32_t)base;
-
-	*n = low +
-		((uint64_t)low2 << 16) +
-		((uint64_t)high << 32);
-
+	uint64_t rem = *n;
+	uint64_t b = base;
+	uint64_t res, d = 1;
+	uint32_t high = rem >> 32;
+
+	/* Reduce the thing a bit first */
+	res = 0;
+	if (high >= base) {
+		high /= base;
+		res = (uint64_t) high << 32;
+		rem -= (uint64_t) (high*base) << 32;
+	}
+
+	while ((int64_t)b > 0 && b < rem) {
+		b = b+b;
+		d = d+d;
+	}
+
+	do {
+		if (rem >= b) {
+			rem -= b;
+			res += d;
+		}
+		b >>= 1;
+		d >>= 1;
+	} while (d);
+
+	*n = res;
 	return rem;
 }