22 files changed, 143 insertions, 99 deletions

diff --git a/arch/alpha/include/asm/Kbuild b/arch/alpha/include/asm/Kbuild
index b7c8f188b313..4dad27360576 100644
--- a/arch/alpha/include/asm/Kbuild
+++ b/arch/alpha/include/asm/Kbuild
@@ -9,3 +9,4 @@ unifdef-y += console.h
 unifdef-y += fpu.h
 unifdef-y += sysinfo.h
 unifdef-y += compiler.h
+unifdef-y += swab.h
diff --git a/arch/alpha/include/asm/atomic.h b/arch/alpha/include/asm/atomic.h
index ca88e54dec93..62b363584b2b 100644
--- a/arch/alpha/include/asm/atomic.h
+++ b/arch/alpha/include/asm/atomic.h
@@ -1,6 +1,7 @@
 #ifndef _ALPHA_ATOMIC_H
 #define _ALPHA_ATOMIC_H
 
+#include <linux/types.h>
 #include <asm/barrier.h>
 #include <asm/system.h>
 
@@ -13,14 +14,6 @@
  */
 
 
-/*
- * Counter is volatile to make sure gcc doesn't try to be clever
- * and move things around on us. We need to use _exactly_ the address
- * the user gave us, not some alias that contains the same information.
- */
-typedef struct { volatile int counter; } atomic_t;
-typedef struct { volatile long counter; } atomic64_t;
-
 #define ATOMIC_INIT(i)		( (atomic_t) { (i) } )
 #define ATOMIC64_INIT(i)	( (atomic64_t) { (i) } )
 
diff --git a/arch/alpha/include/asm/byteorder.h b/arch/alpha/include/asm/byteorder.h
index 58e958fc7f1b..6772f3168701 100644
--- a/arch/alpha/include/asm/byteorder.h
+++ b/arch/alpha/include/asm/byteorder.h
@@ -1,47 +1,7 @@
 #ifndef _ALPHA_BYTEORDER_H
 #define _ALPHA_BYTEORDER_H
 
-#include <asm/types.h>
-#include <linux/compiler.h>
-#include <asm/compiler.h>
-
-#ifdef __GNUC__
-
-static inline __attribute_const__ __u32 __arch__swab32(__u32 x)
-{
-	/*
-	 * Unfortunately, we can't use the 6 instruction sequence
-	 * on ev6 since the latency of the UNPKBW is 3, which is
-	 * pretty hard to hide.  Just in case a future implementation
-	 * has a lower latency, here's the sequence (also by Mike Burrows)
-	 *
-	 * UNPKBW a0, v0       v0: 00AA00BB00CC00DD
-	 * SLL v0, 24, a0      a0: BB00CC00DD000000
-	 * BIS v0, a0, a0      a0: BBAACCBBDDCC00DD
-	 * EXTWL a0, 6, v0     v0: 000000000000BBAA
-	 * ZAP a0, 0xf3, a0    a0: 00000000DDCC0000
-	 * ADDL a0, v0, v0     v0: ssssssssDDCCBBAA
-	 */
-
-	__u64 t0, t1, t2, t3;
-
-	t0 = __kernel_inslh(x, 7);	/* t0 : 0000000000AABBCC */
-	t1 = __kernel_inswl(x, 3);	/* t1 : 000000CCDD000000 */
-	t1 |= t0;			/* t1 : 000000CCDDAABBCC */
-	t2 = t1 >> 16;			/* t2 : 0000000000CCDDAA */
-	t0 = t1 & 0xFF00FF00;		/* t0 : 00000000DD00BB00 */
-	t3 = t2 & 0x00FF00FF;		/* t3 : 0000000000CC00AA */
-	t1 = t0 + t3;			/* t1 : ssssssssDDCCBBAA */
-
-	return t1;
-}
-
-#define __arch__swab32 __arch__swab32
-
-#endif /* __GNUC__ */
-
-#define __BYTEORDER_HAS_U64__
-
+#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif /* _ALPHA_BYTEORDER_H */
diff --git a/arch/alpha/include/asm/smp.h b/arch/alpha/include/asm/smp.h
index 544c69af8168..547e90951cec 100644
--- a/arch/alpha/include/asm/smp.h
+++ b/arch/alpha/include/asm/smp.h
@@ -45,7 +45,6 @@ extern struct cpuinfo_alpha cpu_data[NR_CPUS];
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 extern int smp_num_cpus;
-#define cpu_possible_map	cpu_present_map
 
 extern void arch_send_call_function_single_ipi(int cpu);
 extern void arch_send_call_function_ipi(cpumask_t mask);
diff --git a/arch/alpha/include/asm/swab.h b/arch/alpha/include/asm/swab.h
new file mode 100644
index 000000000000..68e7089e02d5
--- /dev/null
+++ b/arch/alpha/include/asm/swab.h
@@ -0,0 +1,42 @@
+#ifndef _ALPHA_SWAB_H
+#define _ALPHA_SWAB_H
+
+#include <asm/types.h>
+#include <linux/compiler.h>
+#include <asm/compiler.h>
+
+#ifdef __GNUC__
+
+static inline __attribute_const__ __u32 __arch_swab32(__u32 x)
+{
+	/*
+	 * Unfortunately, we can't use the 6 instruction sequence
+	 * on ev6 since the latency of the UNPKBW is 3, which is
+	 * pretty hard to hide.  Just in case a future implementation
+	 * has a lower latency, here's the sequence (also by Mike Burrows)
+	 *
+	 * UNPKBW a0, v0       v0: 00AA00BB00CC00DD
+	 * SLL v0, 24, a0      a0: BB00CC00DD000000
+	 * BIS v0, a0, a0      a0: BBAACCBBDDCC00DD
+	 * EXTWL a0, 6, v0     v0: 000000000000BBAA
+	 * ZAP a0, 0xf3, a0    a0: 00000000DDCC0000
+	 * ADDL a0, v0, v0     v0: ssssssssDDCCBBAA
+	 */
+
+	__u64 t0, t1, t2, t3;
+
+	t0 = __kernel_inslh(x, 7);	/* t0 : 0000000000AABBCC */
+	t1 = __kernel_inswl(x, 3);	/* t1 : 000000CCDD000000 */
+	t1 |= t0;			/* t1 : 000000CCDDAABBCC */
+	t2 = t1 >> 16;			/* t2 : 0000000000CCDDAA */
+	t0 = t1 & 0xFF00FF00;		/* t0 : 00000000DD00BB00 */
+	t3 = t2 & 0x00FF00FF;		/* t3 : 0000000000CC00AA */
+	t1 = t0 + t3;			/* t1 : ssssssssDDCCBBAA */
+
+	return t1;
+}
+#define __arch_swab32 __arch_swab32
+
+#endif /* __GNUC__ */
+
+#endif /* _ALPHA_SWAB_H */
diff --git a/arch/alpha/include/asm/topology.h b/arch/alpha/include/asm/topology.h
index 149532e162c4..b4f284c72ff3 100644
--- a/arch/alpha/include/asm/topology.h
+++ b/arch/alpha/include/asm/topology.h
@@ -39,7 +39,24 @@ static inline cpumask_t node_to_cpumask(int node)
 	return node_cpu_mask;
 }
 
+extern struct cpumask node_to_cpumask_map[];
+/* FIXME: This is dumb, recalculating every time.  But simple. */
+static const struct cpumask *cpumask_of_node(int node)
+{
+	int cpu;
+
+	cpumask_clear(&node_to_cpumask_map[node]);
+
+	for_each_online_cpu(cpu) {
+		if (cpu_to_node(cpu) == node)
+			cpumask_set_cpu(cpu, node_to_cpumask_map[node]);
+	}
+
+	return &node_to_cpumask_map[node];
+}
+
 #define pcibus_to_cpumask(bus)	(cpu_online_map)
+#define cpumask_of_pcibus(bus)	(cpu_online_mask)
 
 #endif /* !CONFIG_NUMA */
 # include <asm-generic/topology.h>
diff --git a/arch/alpha/kernel/Makefile b/arch/alpha/kernel/Makefile
index ac706c1d7ada..b4697759a123 100644
--- a/arch/alpha/kernel/Makefile
+++ b/arch/alpha/kernel/Makefile
@@ -8,7 +8,7 @@ EXTRA_CFLAGS	:= -Werror -Wno-sign-compare
 
 obj-y    := entry.o traps.o process.o init_task.o osf_sys.o irq.o \
 	    irq_alpha.o signal.o setup.o ptrace.o time.o \
-	    alpha_ksyms.o systbls.o err_common.o io.o
+	    alpha_ksyms.o systbls.o err_common.o io.o binfmt_loader.o
 
 obj-$(CONFIG_VGA_HOSE)	+= console.o
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/alpha/kernel/binfmt_loader.c b/arch/alpha/kernel/binfmt_loader.c
new file mode 100644
index 000000000000..4a0af906b00a
--- /dev/null
+++ b/arch/alpha/kernel/binfmt_loader.c
@@ -0,0 +1,51 @@
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mm_types.h>
+#include <linux/binfmts.h>
+#include <linux/a.out.h>
+
+static int load_binary(struct linux_binprm *bprm, struct pt_regs *regs)
+{
+	struct exec *eh = (struct exec *)bprm->buf;
+	unsigned long loader;
+	struct file *file;
+	int retval;
+
+	if (eh->fh.f_magic != 0x183 || (eh->fh.f_flags & 0x3000) != 0x3000)
+		return -ENOEXEC;
+
+	if (bprm->loader)
+		return -ENOEXEC;
+
+	allow_write_access(bprm->file);
+	fput(bprm->file);
+	bprm->file = NULL;
+
+	loader = bprm->vma->vm_end - sizeof(void *);
+
+	file = open_exec("/sbin/loader");
+	retval = PTR_ERR(file);
+	if (IS_ERR(file))
+		return retval;
+
+	/* Remember if the application is TASO.  */
+	bprm->taso = eh->ah.entry < 0x100000000UL;
+
+	bprm->file = file;
+	bprm->loader = loader;
+	retval = prepare_binprm(bprm);
+	if (retval < 0)
+		return retval;
+	return search_binary_handler(bprm,regs);
+}
+
+static struct linux_binfmt loader_format = {
+	.load_binary	= load_binary,
+};
+
+static int __init init_loader_binfmt(void)
+{
+	return register_binfmt(&loader_format);
+}
+arch_initcall(init_loader_binfmt);
diff --git a/arch/alpha/kernel/init_task.c b/arch/alpha/kernel/init_task.c
index 1f762189fa64..c2938e574a40 100644
--- a/arch/alpha/kernel/init_task.c
+++ b/arch/alpha/kernel/init_task.c
@@ -8,7 +8,6 @@
 #include <asm/uaccess.h>
 
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c
index c626a821cdcb..7bc7489223f3 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -50,12 +50,13 @@ int irq_select_affinity(unsigned int irq)
 	if (!irq_desc[irq].chip->set_affinity || irq_user_affinity[irq])
 		return 1;
 
-	while (!cpu_possible(cpu) || !cpu_isset(cpu, irq_default_affinity))
+	while (!cpu_possible(cpu) ||
+	       !cpumask_test_cpu(cpu, irq_default_affinity))
 		cpu = (cpu < (NR_CPUS-1) ? cpu + 1 : 0);
 	last_cpu = cpu;
 
-	irq_desc[irq].affinity = cpumask_of_cpu(cpu);
-	irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu));
+	cpumask_copy(irq_desc[irq].affinity, cpumask_of(cpu));
+	irq_desc[irq].chip->set_affinity(irq, cpumask_of(cpu));
 	return 0;
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/alpha/kernel/pci.c b/arch/alpha/kernel/pci.c
index ff8cb638472e..a3b938811400 100644
--- a/arch/alpha/kernel/pci.c
+++ b/arch/alpha/kernel/pci.c
@@ -320,24 +320,6 @@ pcibios_update_irq(struct pci_dev *dev, int irq)
 	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
 }
 
-/* Most Alphas have straight-forward swizzling needs.  */
-
-u8 __init
-common_swizzle(struct pci_dev *dev, u8 *pinp)
-{
-	u8 pin = *pinp;
-
-	while (dev->bus->parent) {
-		pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
-		/* Move up the chain of bridges. */
-		dev = dev->bus->self;
-        }
-	*pinp = pin;
-
-	/* The slot is the slot of the last bridge. */
-	return PCI_SLOT(dev->devfn);
-}
-
 void
 pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
 			 struct resource *res)
diff --git a/arch/alpha/kernel/pci_impl.h b/arch/alpha/kernel/pci_impl.h
index f8b74995a002..00edd04b585e 100644
--- a/arch/alpha/kernel/pci_impl.h
+++ b/arch/alpha/kernel/pci_impl.h
@@ -106,16 +106,11 @@ struct pci_iommu_arena;
  *   Where A = pin 1, B = pin 2 and so on and pin=0 = default = A.
  *   Thus, each swizzle is ((pin-1) + (device#-4)) % 4
  *
- *   The following code swizzles for exactly one bridge.  The routine
- *   common_swizzle below handles multiple bridges.  But there are a
- *   couple boards that do strange things, so we define this here.
+ *   pci_swizzle_interrupt_pin() swizzles for exactly one bridge.  The routine
+ *   pci_common_swizzle() handles multiple bridges.  But there are a
+ *   couple boards that do strange things.
  */
 
-static inline u8 bridge_swizzle(u8 pin, u8 slot) 
-{
-	return (((pin-1) + slot) % 4) + 1;
-}
-
 
 /* The following macro is used to implement the table-based irq mapping
    function for all single-bus Alphas.  */
@@ -184,7 +179,7 @@ extern int pci_probe_only;
 extern unsigned long alpha_agpgart_size;
 
 extern void common_init_pci(void);
-extern u8 common_swizzle(struct pci_dev *, u8 *);
+#define common_swizzle pci_common_swizzle
 extern struct pci_controller *alloc_pci_controller(void);
 extern struct resource *alloc_resource(void);
 
diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c
index 351407e07e71..f238370c907d 100644
--- a/arch/alpha/kernel/process.c
+++ b/arch/alpha/kernel/process.c
@@ -94,6 +94,7 @@ common_shutdown_1(void *generic_ptr)
 		flags |= 0x00040000UL; /* "remain halted" */
 		*pflags = flags;
 		cpu_clear(cpuid, cpu_present_map);
+		cpu_clear(cpuid, cpu_possible_map);
 		halt();
 	}
 #endif
@@ -120,6 +121,7 @@ common_shutdown_1(void *generic_ptr)
 #ifdef CONFIG_SMP
 	/* Wait for the secondaries to halt. */
 	cpu_clear(boot_cpuid, cpu_present_map);
+	cpu_clear(boot_cpuid, cpu_possible_map);
 	while (cpus_weight(cpu_present_map))
 		barrier();
 #endif
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index a449e999027c..02bee6983ce2 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -79,6 +79,11 @@ int alpha_l3_cacheshape;
 unsigned long alpha_verbose_mcheck = CONFIG_VERBOSE_MCHECK_ON;
 #endif
 
+#ifdef CONFIG_NUMA
+struct cpumask node_to_cpumask_map[MAX_NUMNODES] __read_mostly;
+EXPORT_SYMBOL(node_to_cpumask_map);
+#endif
+
 /* Which processor we booted from.  */
 int boot_cpuid;
 
diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c
index cf7da10097bb..d953e510f68d 100644
--- a/arch/alpha/kernel/smp.c
+++ b/arch/alpha/kernel/smp.c
@@ -70,11 +70,6 @@ enum ipi_message_type {
 /* Set to a secondary's cpuid when it comes online.  */
 static int smp_secondary_alive __devinitdata = 0;
 
-/* Which cpus ids came online.  */
-cpumask_t cpu_online_map;
-
-EXPORT_SYMBOL(cpu_online_map);
-
 int smp_num_probed;		/* Internal processor count */
 int smp_num_cpus = 1;		/* Number that came online.  */
 EXPORT_SYMBOL(smp_num_cpus);
@@ -440,6 +435,7 @@ setup_smp(void)
 				((char *)cpubase + i*hwrpb->processor_size);
 			if ((cpu->flags & 0x1cc) == 0x1cc) {
 				smp_num_probed++;
+				cpu_set(i, cpu_possible_map);
 				cpu_set(i, cpu_present_map);
 				cpu->pal_revision = boot_cpu_palrev;
 			}
@@ -473,6 +469,7 @@ smp_prepare_cpus(unsigned int max_cpus)
 
 	/* Nothing to do on a UP box, or when told not to.  */
 	if (smp_num_probed == 1 || max_cpus == 0) {
+		cpu_possible_map = cpumask_of_cpu(boot_cpuid);
 		cpu_present_map = cpumask_of_cpu(boot_cpuid);
 		printk(KERN_INFO "SMP mode deactivated.\n");
 		return;
diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c
index c71b0fd7a61f..9c9d1fd4155f 100644
--- a/arch/alpha/kernel/sys_dp264.c
+++ b/arch/alpha/kernel/sys_dp264.c
@@ -177,19 +177,19 @@ cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 }
 
 static void
-dp264_set_affinity(unsigned int irq, cpumask_t affinity)
+dp264_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&dp264_irq_lock);
-	cpu_set_irq_affinity(irq, affinity);
+	cpu_set_irq_affinity(irq, *affinity);
 	tsunami_update_irq_hw(cached_irq_mask);
 	spin_unlock(&dp264_irq_lock);
 }
 
 static void
-clipper_set_affinity(unsigned int irq, cpumask_t affinity)
+clipper_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&dp264_irq_lock);
-	cpu_set_irq_affinity(irq - 16, affinity);
+	cpu_set_irq_affinity(irq - 16, *affinity);
 	tsunami_update_irq_hw(cached_irq_mask);
 	spin_unlock(&dp264_irq_lock);
 }
@@ -481,7 +481,7 @@ monet_swizzle(struct pci_dev *dev, u8 *pinp)
 				slot = PCI_SLOT(dev->devfn);
 				break;
 			}
-			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn)) ;
+			pin = pci_swizzle_interrupt_pin(dev, pin);
 
 			/* Move up the chain of bridges.  */
 			dev = dev->bus->self;
diff --git a/arch/alpha/kernel/sys_eiger.c b/arch/alpha/kernel/sys_eiger.c
index 7ef3b6fb3700..baf60f36cbd7 100644
--- a/arch/alpha/kernel/sys_eiger.c
+++ b/arch/alpha/kernel/sys_eiger.c
@@ -204,7 +204,7 @@ eiger_swizzle(struct pci_dev *dev, u8 *pinp)
 			break;
 		}
 		/* Must be a card-based bridge.  */
-		pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
+		pin = pci_swizzle_interrupt_pin(dev, pin);
 
 		/* Move up the chain of bridges.  */
 		dev = dev->bus->self;
diff --git a/arch/alpha/kernel/sys_miata.c b/arch/alpha/kernel/sys_miata.c
index 910b43cd63e8..61ccd95579ec 100644
--- a/arch/alpha/kernel/sys_miata.c
+++ b/arch/alpha/kernel/sys_miata.c
@@ -219,7 +219,7 @@ miata_swizzle(struct pci_dev *dev, u8 *pinp)
 				slot = PCI_SLOT(dev->devfn) + 9;
 				break;
 			}
-			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
+			pin = pci_swizzle_interrupt_pin(dev, pin);
 
 			/* Move up the chain of bridges.  */
 			dev = dev->bus->self;
diff --git a/arch/alpha/kernel/sys_noritake.c b/arch/alpha/kernel/sys_noritake.c
index eb2a1d63f484..538876b62449 100644
--- a/arch/alpha/kernel/sys_noritake.c
+++ b/arch/alpha/kernel/sys_noritake.c
@@ -257,7 +257,7 @@ noritake_swizzle(struct pci_dev *dev, u8 *pinp)
 				slot = PCI_SLOT(dev->devfn) + 15;
 				break;
 			}
-			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn)) ;
+			pin = pci_swizzle_interrupt_pin(dev, pin);
 
 			/* Move up the chain of bridges.  */
 			dev = dev->bus->self;
diff --git a/arch/alpha/kernel/sys_ruffian.c b/arch/alpha/kernel/sys_ruffian.c
index 5b99cf3cd69c..f15a329b6011 100644
--- a/arch/alpha/kernel/sys_ruffian.c
+++ b/arch/alpha/kernel/sys_ruffian.c
@@ -160,7 +160,7 @@ ruffian_swizzle(struct pci_dev *dev, u8 *pinp)
 				slot = PCI_SLOT(dev->devfn) + 10;
 				break;
 			}
-			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
+			pin = pci_swizzle_interrupt_pin(dev, pin);
 
 			/* Move up the chain of bridges.  */
 			dev = dev->bus->self;
diff --git a/arch/alpha/kernel/sys_sable.c b/arch/alpha/kernel/sys_sable.c
index a4555f497639..d232e42be018 100644
--- a/arch/alpha/kernel/sys_sable.c
+++ b/arch/alpha/kernel/sys_sable.c
@@ -425,7 +425,7 @@ lynx_swizzle(struct pci_dev *dev, u8 *pinp)
 				slot = PCI_SLOT(dev->devfn) + 11;
 				break;
 			}
-			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn)) ;
+			pin = pci_swizzle_interrupt_pin(dev, pin);
 
 			/* Move up the chain of bridges.  */
 			dev = dev->bus->self;
diff --git a/arch/alpha/kernel/sys_titan.c b/arch/alpha/kernel/sys_titan.c
index 52c91ccc1648..27f840a4ad3d 100644
--- a/arch/alpha/kernel/sys_titan.c
+++ b/arch/alpha/kernel/sys_titan.c
@@ -158,10 +158,10 @@ titan_cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 }
 
 static void
-titan_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+titan_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&titan_irq_lock);
-	titan_cpu_set_irq_affinity(irq - 16, affinity);
+	titan_cpu_set_irq_affinity(irq - 16, *affinity);
 	titan_update_irq_hw(titan_cached_irq_mask);
 	spin_unlock(&titan_irq_lock);
 }