25 files changed, 303 insertions, 270 deletions

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index ff7ae6b664e8..10b6b9e7716b 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -252,6 +252,15 @@ config NR_CPUS
           than 64 will cause the use of a CPU mask array, causing a small
           performance hit.
 
+config IA64_NR_NODES
+        int "Maximum number of NODEs (256-1024)" if (IA64_SGI_SN2 || IA64_GENERIC)
+        range 256 1024
+        depends on IA64_SGI_SN2 || IA64_GENERIC
+        default "256"
+        help
+          This option specifies the maximum number of nodes in your SSI system.
+          If in doubt, use the default.
+
 config HOTPLUG_CPU
         bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
         depends on SMP && EXPERIMENTAL
diff --git a/arch/ia64/configs/gensparse_defconfig b/arch/ia64/configs/gensparse_defconfig
index 744fd2f79f61..0d29aa2066b3 100644
--- a/arch/ia64/configs/gensparse_defconfig
+++ b/arch/ia64/configs/gensparse_defconfig
@@ -116,6 +116,7 @@ CONFIG_IOSAPIC=y
 CONFIG_FORCE_MAX_ZONEORDER=17
 CONFIG_SMP=y
 CONFIG_NR_CPUS=512
+CONFIG_IA64_NR_NODES=256
 CONFIG_HOTPLUG_CPU=y
 # CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index 8206752161bb..a718034d68d0 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -116,6 +116,7 @@ CONFIG_IA64_SGI_SN_XP=m
 CONFIG_FORCE_MAX_ZONEORDER=17
 CONFIG_SMP=y
 CONFIG_NR_CPUS=1024
+CONFIG_IA64_NR_NODES=256
 # CONFIG_HOTPLUG_CPU is not set
 CONFIG_SCHED_SMT=y
 CONFIG_PREEMPT=y
diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig
index 3e767288a745..6cba55da572a 100644
--- a/arch/ia64/defconfig
+++ b/arch/ia64/defconfig
@@ -116,6 +116,7 @@ CONFIG_IOSAPIC=y
 CONFIG_FORCE_MAX_ZONEORDER=17
 CONFIG_SMP=y
 CONFIG_NR_CPUS=512
+CONFIG_IA64_NR_NODES=256
 CONFIG_HOTPLUG_CPU=y
 # CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
diff --git a/arch/ia64/dig/setup.c b/arch/ia64/dig/setup.c
index c9104bfff667..38aa9c108857 100644
--- a/arch/ia64/dig/setup.c
+++ b/arch/ia64/dig/setup.c
@@ -69,8 +69,3 @@ dig_setup (char **cmdline_p)
         screen_info.orig_video_isVGA = 1;       /* XXX fake */
         screen_info.orig_video_ega_bx = 3;      /* XXX fake */
 }
-
-void __init
-dig_irq_init (void)
-{
-}
diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index 4722ec51c70c..a4e218ce2edb 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -420,6 +420,26 @@ int __devinitdata pxm_to_nid_map[MAX_PXM_DOMAINS];
 int __initdata nid_to_pxm_map[MAX_NUMNODES];
 static struct acpi_table_slit __initdata *slit_table;
 
+static int get_processor_proximity_domain(struct acpi_table_processor_affinity *pa)
+{
+        int pxm;
+
+        pxm = pa->proximity_domain;
+        if (ia64_platform_is("sn2"))
+                pxm += pa->reserved[0] << 8;
+        return pxm;
+}
+
+static int get_memory_proximity_domain(struct acpi_table_memory_affinity *ma)
+{
+        int pxm;
+
+        pxm = ma->proximity_domain;
+        if (ia64_platform_is("sn2"))
+                pxm += ma->reserved1[0] << 8;
+        return pxm;
+}
+
 /*
  * ACPI 2.0 SLIT (System Locality Information Table)
  * http://devresource.hp.com/devresource/Docs/TechPapers/IA64/slit.pdf
@@ -443,13 +463,20 @@ void __init acpi_numa_slit_init(struct acpi_table_slit *slit)
 void __init
 acpi_numa_processor_affinity_init(struct acpi_table_processor_affinity *pa)
 {
+        int pxm;
+
+        if (!pa->flags.enabled)
+                return;
+
+        pxm = get_processor_proximity_domain(pa);
+
         /* record this node in proximity bitmap */
-        pxm_bit_set(pa->proximity_domain);
+        pxm_bit_set(pxm);
 
         node_cpuid[srat_num_cpus].phys_id =
             (pa->apic_id << 8) | (pa->lsapic_eid);
         /* nid should be overridden as logical node id later */
-        node_cpuid[srat_num_cpus].nid = pa->proximity_domain;
+        node_cpuid[srat_num_cpus].nid = pxm;
         srat_num_cpus++;
 }
 
@@ -457,10 +484,10 @@ void __init
 acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma)
 {
         unsigned long paddr, size;
-        u8 pxm;
+        int pxm;
         struct node_memblk_s *p, *q, *pend;
 
-        pxm = ma->proximity_domain;
+        pxm = get_memory_proximity_domain(ma);
 
         /* fill node memory chunk structure */
         paddr = ma->base_addr_hi;
diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index dcd906fe5749..829a43cab797 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -865,6 +865,7 @@ ENTRY(interrupt)
         ;;
         SAVE_REST
         ;;
+        MCA_RECOVER_RANGE(interrupt)
         alloc r14=ar.pfs,0,0,2,0 // must be first in an insn group
         mov out0=cr.ivr         // pass cr.ivr as first arg
         add out1=16,sp          // pass pointer to pt_regs as second arg
diff --git a/arch/ia64/kernel/machvec.c b/arch/ia64/kernel/machvec.c
index c3a04ee7f4f6..4b0b71d5aef4 100644
--- a/arch/ia64/kernel/machvec.c
+++ b/arch/ia64/kernel/machvec.c
@@ -14,7 +14,15 @@
 struct ia64_machine_vector ia64_mv;
 EXPORT_SYMBOL(ia64_mv);
 
-static struct ia64_machine_vector *
+static __initdata const char *mvec_name;
+static __init int setup_mvec(char *s)
+{
+        mvec_name = s;
+        return 0;
+}
+early_param("machvec", setup_mvec);
+
+static struct ia64_machine_vector * __init
 lookup_machvec (const char *name)
 {
         extern struct ia64_machine_vector machvec_start[];
@@ -33,10 +41,13 @@ machvec_init (const char *name)
 {
         struct ia64_machine_vector *mv;
 
+        if (!name)
+                name = mvec_name ? mvec_name : acpi_get_sysname();
         mv = lookup_machvec(name);
-        if (!mv) {
-                panic("generic kernel failed to find machine vector for platform %s!", name);
-        }
+        if (!mv)
+                panic("generic kernel failed to find machine vector for"
+                      " platform %s!", name);
+
         ia64_mv = *mv;
         printk(KERN_INFO "booting generic kernel on platform %s\n", name);
 }
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index b57e723f194c..87ff7fe33cfb 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -83,6 +83,7 @@
 #include <asm/irq.h>
 #include <asm/hw_irq.h>
 
+#include "mca_drv.h"
 #include "entry.h"
 
 #if defined(IA64_MCA_DEBUG_INFO)
@@ -133,7 +134,7 @@ static int cpe_poll_enabled = 1;
 
 extern void salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe);
 
-static int mca_init;
+static int mca_init __initdata;
 
 
 static void inline
@@ -184,7 +185,7 @@ static ia64_state_log_t ia64_state_log[IA64_MAX_LOG_TYPES];
  * Inputs   :   info_type   (SAL_INFO_TYPE_{MCA,INIT,CMC,CPE})
  * Outputs      :       None
  */
-static void
+static void __init
 ia64_log_init(int sal_info_type)
 {
         u64     max_size = 0;
@@ -281,6 +282,50 @@ ia64_mca_log_sal_error_record(int sal_info_type)
                 ia64_sal_clear_state_info(sal_info_type);
 }
 
+/*
+ * search_mca_table
+ *  See if the MCA surfaced in an instruction range
+ *  that has been tagged as recoverable.
+ *
+ *  Inputs
+ *      first   First address range to check
+ *      last    Last address range to check
+ *      ip      Instruction pointer, address we are looking for
+ *
+ * Return value:
+ *      1 on Success (in the table)/ 0 on Failure (not in the  table)
+ */
+int
+search_mca_table (const struct mca_table_entry *first,
+                const struct mca_table_entry *last,
+                unsigned long ip)
+{
+        const struct mca_table_entry *curr;
+        u64 curr_start, curr_end;
+
+        curr = first;
+        while (curr <= last) {
+                curr_start = (u64) &curr->start_addr + curr->start_addr;
+                curr_end = (u64) &curr->end_addr + curr->end_addr;
+
+                if ((ip >= curr_start) && (ip <= curr_end)) {
+                        return 1;
+                }
+                curr++;
+        }
+        return 0;
+}
+
+/* Given an address, look for it in the mca tables. */
+int mca_recover_range(unsigned long addr)
+{
+        extern struct mca_table_entry __start___mca_table[];
+        extern struct mca_table_entry __stop___mca_table[];
+
+        return search_mca_table(__start___mca_table, __stop___mca_table-1, addr);
+}
+EXPORT_SYMBOL_GPL(mca_recover_range);
+
 #ifdef CONFIG_ACPI
 
 int cpe_vector = -1;
@@ -355,7 +400,7 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs)
  *  Outputs
  *      None
  */
-static void
+static void __init
 ia64_mca_register_cpev (int cpev)
 {
         /* Register the CPE interrupt vector with SAL */
@@ -386,7 +431,7 @@ ia64_mca_register_cpev (int cpev)
  * Outputs
  *      None
  */
-void
+void __cpuinit
 ia64_mca_cmc_vector_setup (void)
 {
         cmcv_reg_t      cmcv;
@@ -747,31 +792,34 @@ ia64_mca_modify_original_stack(struct pt_regs *regs,
                 ia64_mca_modify_comm(previous_current);
                 goto no_mod;
         }
-        if (r13 != sos->prev_IA64_KR_CURRENT) {
-                msg = "inconsistent previous current and r13";
-                goto no_mod;
-        }
-        if ((r12 - r13) >= KERNEL_STACK_SIZE) {
-                msg = "inconsistent r12 and r13";
-                goto no_mod;
-        }
-        if ((ar_bspstore - r13) >= KERNEL_STACK_SIZE) {
-                msg = "inconsistent ar.bspstore and r13";
-                goto no_mod;
-        }
-        va.p = old_bspstore;
-        if (va.f.reg < 5) {
-                msg = "old_bspstore is in the wrong region";
-                goto no_mod;
-        }
-        if ((ar_bsp - r13) >= KERNEL_STACK_SIZE) {
-                msg = "inconsistent ar.bsp and r13";
-                goto no_mod;
-        }
-        size += (ia64_rse_skip_regs(old_bspstore, slots) - old_bspstore) * 8;
-        if (ar_bspstore + size > r12) {
-                msg = "no room for blocked state";
-                goto no_mod;
+
+        if (!mca_recover_range(ms->pmsa_iip)) {
+                if (r13 != sos->prev_IA64_KR_CURRENT) {
+                        msg = "inconsistent previous current and r13";
+                        goto no_mod;
+                }
+                if ((r12 - r13) >= KERNEL_STACK_SIZE) {
+                        msg = "inconsistent r12 and r13";
+                        goto no_mod;
+                }
+                if ((ar_bspstore - r13) >= KERNEL_STACK_SIZE) {
+                        msg = "inconsistent ar.bspstore and r13";
+                        goto no_mod;
+                }
+                va.p = old_bspstore;
+                if (va.f.reg < 5) {
+                        msg = "old_bspstore is in the wrong region";
+                        goto no_mod;
+                }
+                if ((ar_bsp - r13) >= KERNEL_STACK_SIZE) {
+                        msg = "inconsistent ar.bsp and r13";
+                        goto no_mod;
+                }
+                size += (ia64_rse_skip_regs(old_bspstore, slots) - old_bspstore) * 8;
+                if (ar_bspstore + size > r12) {
+                        msg = "no room for blocked state";
+                        goto no_mod;
+                }
         }
 
         ia64_mca_modify_comm(previous_current);
@@ -1443,7 +1491,7 @@ static struct irqaction mca_cpep_irqaction = {
  * format most of the fields.
  */
 
-static void
+static void __cpuinit
 format_mca_init_stack(void *mca_data, unsigned long offset,
                 const char *type, int cpu)
 {
@@ -1467,7 +1515,7 @@ format_mca_init_stack(void *mca_data, unsigned long offset,
 
 /* Do per-CPU MCA-related initialization.  */
 
-void __devinit
+void __cpuinit
 ia64_mca_cpu_init(void *cpu_data)
 {
         void *pal_vaddr;
diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c
index e883d85906db..37c88eb55873 100644
--- a/arch/ia64/kernel/mca_drv.c
+++ b/arch/ia64/kernel/mca_drv.c
@@ -6,6 +6,7 @@
  * Copyright (C) Hidetoshi Seto (seto.hidetoshi@jp.fujitsu.com)
  * Copyright (C) 2005 Silicon Graphics, Inc
  * Copyright (C) 2005 Keith Owens <kaos@sgi.com>
+ * Copyright (C) 2006 Russ Anderson <rja@sgi.com>
  */
 #include <linux/config.h>
 #include <linux/types.h>
@@ -121,11 +122,12 @@ mca_page_isolate(unsigned long paddr)
  */
 
 void
-mca_handler_bh(unsigned long paddr)
+mca_handler_bh(unsigned long paddr, void *iip, unsigned long ipsr)
 {
-        printk(KERN_ERR
-                "OS_MCA: process [pid: %d](%s) encounters MCA (paddr=%lx)\n",
-                current->pid, current->comm, paddr);
+        printk(KERN_ERR "OS_MCA: process [cpu %d, pid: %d, uid: %d, "
+                "iip: %p, psr: 0x%lx,paddr: 0x%lx](%s) encounters MCA.\n",
+                raw_smp_processor_id(), current->pid, current->uid,
+                iip, ipsr, paddr, current->comm);
 
         spin_lock(&mca_bh_lock);
         switch (mca_page_isolate(paddr)) {
@@ -442,21 +444,26 @@ recover_from_read_error(slidx_table_t *slidx,
         if (!peidx_bottom(peidx) || !(peidx_bottom(peidx)->valid.minstate))
                 return 0;
         psr1 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_ipsr);
+        psr2 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_xpsr);
 
         /*
          *  Check the privilege level of interrupted context.
          *   If it is user-mode, then terminate affected process.
          */
-        if (psr1->cpl != 0) {
+
+        pmsa = sos->pal_min_state;
+        if (psr1->cpl != 0 ||
+           ((psr2->cpl != 0) && mca_recover_range(pmsa->pmsa_iip))) {
                 smei = peidx_bus_check(peidx, 0);
                 if (smei->valid.target_identifier) {
                         /*
                          *  setup for resume to bottom half of MCA,
                          * "mca_handler_bhhook"
                          */
-                        pmsa = sos->pal_min_state;
-                        /* pass to bhhook as 1st argument (gr8) */
+                        /* pass to bhhook as argument (gr8, ...) */
                         pmsa->pmsa_gr[8-1] = smei->target_identifier;
+                        pmsa->pmsa_gr[9-1] = pmsa->pmsa_iip;
+                        pmsa->pmsa_gr[10-1] = pmsa->pmsa_ipsr;
                         /* set interrupted return address (but no use) */
                         pmsa->pmsa_br0 = pmsa->pmsa_iip;
                         /* change resume address to bottom half */
@@ -466,6 +473,7 @@ recover_from_read_error(slidx_table_t *slidx,
                         psr2 = (struct ia64_psr *)&pmsa->pmsa_ipsr;
                         psr2->cpl = 0;
                         psr2->ri  = 0;
+                        psr2->bn  = 1;
                         psr2->i  = 0;
 
                         return 1;
diff --git a/arch/ia64/kernel/mca_drv.h b/arch/ia64/kernel/mca_drv.h
index e2f6fa1e0ef6..31a2e52bb16f 100644
--- a/arch/ia64/kernel/mca_drv.h
+++ b/arch/ia64/kernel/mca_drv.h
@@ -111,3 +111,10 @@ typedef struct slidx_table {
         slidx_foreach_entry(__pos, &((slidx)->sec)) { __count++; }\
         __count; })
 
+struct mca_table_entry {
+        int start_addr; /* location-relative starting address of MCA recoverable range */
+        int end_addr;   /* location-relative ending address of MCA recoverable range */
+};
+
+extern const struct mca_table_entry *search_mca_tables (unsigned long addr);
+extern int mca_recover_range(unsigned long);
diff --git a/arch/ia64/kernel/mca_drv_asm.S b/arch/ia64/kernel/mca_drv_asm.S
index 3f298ee4d00c..e6a580d354b9 100644
--- a/arch/ia64/kernel/mca_drv_asm.S
+++ b/arch/ia64/kernel/mca_drv_asm.S
@@ -14,15 +14,12 @@
 
 GLOBAL_ENTRY(mca_handler_bhhook)
         invala                          // clear RSE ?
-        ;;
         cover
         ;;
         clrrrb
         ;;                                              
-        alloc   r16=ar.pfs,0,2,1,0      // make a new frame
-        ;;
+        alloc   r16=ar.pfs,0,2,3,0      // make a new frame
         mov     ar.rsc=0
-        ;;
         mov     r13=IA64_KR(CURRENT)    // current task pointer
         ;;
         mov     r2=r13
@@ -30,7 +27,6 @@ GLOBAL_ENTRY(mca_handler_bhhook)
         addl    r22=IA64_RBS_OFFSET,r2
         ;;
         mov     ar.bspstore=r22
-        ;;
         addl    sp=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2
         ;;
         adds    r2=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
@@ -40,12 +36,12 @@ GLOBAL_ENTRY(mca_handler_bhhook)
         movl    loc1=mca_handler_bh     // recovery C function
         ;;
         mov     out0=r8                 // poisoned address
+        mov     out1=r9                 // iip
+        mov     out2=r10                // psr
         mov     b6=loc1
         ;;
         mov     loc1=rp
-        ;;
-        ssm     psr.i
-        ;;
+        ssm     psr.i | psr.ic
         br.call.sptk.many rp=b6         // does not return ...
         ;;
         mov     ar.pfs=loc0
@@ -53,5 +49,4 @@ GLOBAL_ENTRY(mca_handler_bhhook)
         ;;
         mov     r8=r0
         br.ret.sptk.many rp
-        ;;
 END(mca_handler_bhhook)
diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c
index a68ce6678092..0766493d4d00 100644
--- a/arch/ia64/kernel/numa.c
+++ b/arch/ia64/kernel/numa.c
@@ -25,7 +25,7 @@
 #include <asm/processor.h>
 #include <asm/smp.h>
 
-u8 cpu_to_node_map[NR_CPUS] __cacheline_aligned;
+u16 cpu_to_node_map[NR_CPUS] __cacheline_aligned;
 EXPORT_SYMBOL(cpu_to_node_map);
 
 cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned;
diff --git a/arch/ia64/kernel/patch.c b/arch/ia64/kernel/patch.c
index 6a4ac7d70b35..bc11bb096f58 100644
--- a/arch/ia64/kernel/patch.c
+++ b/arch/ia64/kernel/patch.c
@@ -115,7 +115,7 @@ ia64_patch_vtop (unsigned long start, unsigned long end)
         ia64_srlz_i();
 }
 
-void
+void __init
 ia64_patch_mckinley_e9 (unsigned long start, unsigned long end)
 {
         static int first_time = 1;
@@ -149,7 +149,7 @@ ia64_patch_mckinley_e9 (unsigned long start, unsigned long end)
         ia64_srlz_i();
 }
 
-static void
+static void __init
 patch_fsyscall_table (unsigned long start, unsigned long end)
 {
         extern unsigned long fsyscall_table[NR_syscalls];
@@ -166,7 +166,7 @@ patch_fsyscall_table (unsigned long start, unsigned long end)
         ia64_srlz_i();
 }
 
-static void
+static void __init
 patch_brl_fsys_bubble_down (unsigned long start, unsigned long end)
 {
         extern char fsys_bubble_down[];
@@ -184,7 +184,7 @@ patch_brl_fsys_bubble_down (unsigned long start, unsigned long end)
         ia64_srlz_i();
 }
 
-void
+void __init
 ia64_patch_gate (void)
 {
 #       define START(name)      ((unsigned long) __start_gate_##name##_patchlist)
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 958c1508036f..eb388e271b2b 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -130,8 +130,8 @@ EXPORT_SYMBOL(ia64_max_iommu_merge_mask);
 /*
  * We use a special marker for the end of memory and it uses the extra (+1) slot
  */
-struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1];
-int num_rsvd_regions;
+struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1] __initdata;
+int num_rsvd_regions __initdata;
 
 
 /*
@@ -140,7 +140,7 @@ int num_rsvd_regions;
  * caller-specified function is called with the memory ranges that remain after filtering.
  * This routine does not assume the incoming segments are sorted.
  */
-int
+int __init
 filter_rsvd_memory (unsigned long start, unsigned long end, void *arg)
 {
         unsigned long range_start, range_end, prev_start;
@@ -176,7 +176,7 @@ filter_rsvd_memory (unsigned long start, unsigned long end, void *arg)
         return 0;
 }
 
-static void
+static void __init
 sort_regions (struct rsvd_region *rsvd_region, int max)
 {
         int j;
@@ -217,7 +217,7 @@ __initcall(register_memory);
  * initrd, etc.  There are currently %IA64_MAX_RSVD_REGIONS defined,
  * see include/asm-ia64/meminit.h if you need to define more.
  */
-void
+void __init
 reserve_memory (void)
 {
         int n = 0;
@@ -269,7 +269,7 @@ reserve_memory (void)
  * Grab the initrd start and end from the boot parameter struct given us by
  * the boot loader.
  */
-void
+void __init
 find_initrd (void)
 {
 #ifdef CONFIG_BLK_DEV_INITRD
@@ -361,7 +361,7 @@ mark_bsp_online (void)
 }
 
 #ifdef CONFIG_SMP
-static void
+static void __init
 check_for_logical_procs (void)
 {
         pal_logical_to_physical_t info;
@@ -388,6 +388,14 @@ check_for_logical_procs (void)
 }
 #endif
 
+static __initdata int nomca;
+static __init int setup_nomca(char *s)
+{
+        nomca = 1;
+        return 0;
+}
+early_param("nomca", setup_nomca);
+
 void __init
 setup_arch (char **cmdline_p)
 {
@@ -401,35 +409,15 @@ setup_arch (char **cmdline_p)
         efi_init();
         io_port_init();
 
+        parse_early_param();
+
 #ifdef CONFIG_IA64_GENERIC
-        {
-                const char *mvec_name = strstr (*cmdline_p, "machvec=");
-                char str[64];
-
-                if (mvec_name) {
-                        const char *end;
-                        size_t len;
-
-                        mvec_name += 8;
-                        end = strchr (mvec_name, ' ');
-                        if (end)
-                                len = end - mvec_name;
-                        else
-                                len = strlen (mvec_name);
-                        len = min(len, sizeof (str) - 1);
-                        strncpy (str, mvec_name, len);
-                        str[len] = '\0';
-                        mvec_name = str;
-                } else
-                        mvec_name = acpi_get_sysname();
-                machvec_init(mvec_name);
-        }
+        machvec_init(NULL);
 #endif
 
         if (early_console_setup(*cmdline_p) == 0)
                 mark_bsp_online();
 
-        parse_early_param();
 #ifdef CONFIG_ACPI
         /* Initialize the ACPI boot-time table parser */
         acpi_table_init();
@@ -492,7 +480,7 @@ setup_arch (char **cmdline_p)
 #endif
 
         /* enable IA-64 Machine Check Abort Handling unless disabled */
-        if (!strstr(saved_command_line, "nomca"))
+        if (!nomca)
                 ia64_mca_init();
 
         platform_setup(cmdline_p);
@@ -622,7 +610,7 @@ struct seq_operations cpuinfo_op = {
         .show =         show_cpuinfo
 };
 
-void
+static void __cpuinit
 identify_cpu (struct cpuinfo_ia64 *c)
 {
         union {
@@ -699,7 +687,7 @@ setup_per_cpu_areas (void)
  * In addition, the minimum of the i-cache stride sizes is calculated for
  * "flush_icache_range()".
  */
-static void
+static void __cpuinit
 get_max_cacheline_size (void)
 {
         unsigned long line_size, max = 1;
@@ -762,10 +750,10 @@ get_max_cacheline_size (void)
  * cpu_init() initializes state that is per-CPU.  This function acts
  * as a 'CPU state barrier', nothing should get across.
  */
-void
+void __cpuinit
 cpu_init (void)
 {
-        extern void __devinit ia64_mmu_init (void *);
+        extern void __cpuinit ia64_mmu_init (void *);
         unsigned long num_phys_stacked;
         pal_vm_info_2_u_t vmi;
         unsigned int max_ctx;
@@ -893,7 +881,7 @@ void sched_cacheflush(void)
         ia64_sal_cache_flush(3);
 }
 
-void
+void __init
 check_bugs (void)
 {
         ia64_patch_mckinley_e9((unsigned long) __start___mckinley_e9_bundles,
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index c4b633b36dab..44e9547878ac 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -624,32 +624,8 @@ void __devinit smp_prepare_boot_cpu(void)
         per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 }
 
-/*
- * mt_info[] is a temporary store for all info returned by
- * PAL_LOGICAL_TO_PHYSICAL, to be copied into cpuinfo_ia64 when the
- * specific cpu comes.
- */
-static struct {
-        __u32   socket_id;
-        __u16   core_id;
-        __u16   thread_id;
-        __u16   proc_fixed_addr;
-        __u8    valid;
-} mt_info[NR_CPUS] __devinitdata;
-
 #ifdef CONFIG_HOTPLUG_CPU
 static inline void
-remove_from_mtinfo(int cpu)
-{
-        int i;
-
-        for_each_cpu(i)
-                if (mt_info[i].valid &&  mt_info[i].socket_id ==
-                                                cpu_data(cpu)->socket_id)
-                        mt_info[i].valid = 0;
-}
-
-static inline void
 clear_cpu_sibling_map(int cpu)
 {
         int i;
@@ -678,12 +654,6 @@ remove_siblinginfo(int cpu)
 
         /* remove it from all sibling map's */
         clear_cpu_sibling_map(cpu);
-
-        /* if this cpu is the last in the core group, remove all its info 
-         * from mt_info structure
-         */
-        if (last)
-                remove_from_mtinfo(cpu);
 }
 
 extern void fixup_irqs(void);
@@ -878,40 +848,6 @@ init_smp_config(void)
                        ia64_sal_strerror(sal_ret));
 }
 
-static inline int __devinit
-check_for_mtinfo_index(void)
-{
-        int i;
-        
-        for_each_cpu(i)
-                if (!mt_info[i].valid)
-                        return i;
-
-        return -1;
-}
-
-/*
- * Search the mt_info to find out if this socket's cid/tid information is
- * cached or not. If the socket exists, fill in the core_id and thread_id 
- * in cpuinfo
- */
-static int __devinit
-check_for_new_socket(__u16 logical_address, struct cpuinfo_ia64 *c)
-{
-        int i;
-        __u32 sid = c->socket_id;
-
-        for_each_cpu(i) {
-                if (mt_info[i].valid && mt_info[i].proc_fixed_addr == logical_address
-                    && mt_info[i].socket_id == sid) {
-                        c->core_id = mt_info[i].core_id;
-                        c->thread_id = mt_info[i].thread_id;
-                        return 1; /* not a new socket */
-                }
-        }
-        return 0;
-}
-
 /*
  * identify_siblings(cpu) gets called from identify_cpu. This populates the 
  * information related to logical execution units in per_cpu_data structure.
@@ -921,14 +857,12 @@ identify_siblings(struct cpuinfo_ia64 *c)
 {
         s64 status;
         u16 pltid;
-        u64 proc_fixed_addr;
-        int count, i;
         pal_logical_to_physical_t info;
 
         if (smp_num_cpucores == 1 && smp_num_siblings == 1)
                 return;
 
-        if ((status = ia64_pal_logical_to_phys(0, &info)) != PAL_STATUS_SUCCESS) {
+        if ((status = ia64_pal_logical_to_phys(-1, &info)) != PAL_STATUS_SUCCESS) {
                 printk(KERN_ERR "ia64_pal_logical_to_phys failed with %ld\n",
                        status);
                 return;
@@ -937,47 +871,12 @@ identify_siblings(struct cpuinfo_ia64 *c)
                 printk(KERN_ERR "ia64_sal_pltid failed with %ld\n", status);
                 return;
         }
-        if ((status = ia64_pal_fixed_addr(&proc_fixed_addr)) != PAL_STATUS_SUCCESS) {
-                printk(KERN_ERR "ia64_pal_fixed_addr failed with %ld\n", status);
-                return;
-        }
 
         c->socket_id =  (pltid << 8) | info.overview_ppid;
         c->cores_per_socket = info.overview_cpp;
         c->threads_per_core = info.overview_tpc;
-        count = c->num_log = info.overview_num_log;
+        c->num_log = info.overview_num_log;
 
-        /* If the thread and core id information is already cached, then
-         * we will simply update cpu_info and return. Otherwise, we will
-         * do the PAL calls and cache core and thread id's of all the siblings.
-         */
-        if (check_for_new_socket(proc_fixed_addr, c))
-                return;
-
-        for (i = 0; i < count; i++) {
-                int index;
-
-                if (i && (status = ia64_pal_logical_to_phys(i, &info))
-                          != PAL_STATUS_SUCCESS) {
-                        printk(KERN_ERR "ia64_pal_logical_to_phys failed"
-                                        " with %ld\n", status);
-                        return;
-                }
-                if (info.log2_la == proc_fixed_addr) {
-                        c->core_id = info.log1_cid;
-                        c->thread_id = info.log1_tid;
-                }
-
-                index = check_for_mtinfo_index();
-                /* We will not do the mt_info caching optimization in this case.
-                 */
-                if (index < 0)
-                        continue;
-
-                mt_info[index].valid = 1;
-                mt_info[index].socket_id = c->socket_id;
-                mt_info[index].core_id = info.log1_cid;
-                mt_info[index].thread_id = info.log1_tid;
-                mt_info[index].proc_fixed_addr = info.log2_la;
-        }
+        c->core_id = info.log1_cid;
+        c->thread_id = info.log1_tid;
 }
diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index 73af6267d2ef..0b9e56dd7f05 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -70,34 +70,9 @@ SECTIONS
           __stop___ex_table = .;
         }
 
-  .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET)
-        {
-          __start___vtop_patchlist = .;
-          *(.data.patch.vtop)
-          __end___vtop_patchlist = .;
-        }
-
-  .data.patch.mckinley_e9 : AT(ADDR(.data.patch.mckinley_e9) - LOAD_OFFSET)
-        {
-          __start___mckinley_e9_bundles = .;
-          *(.data.patch.mckinley_e9)
-          __end___mckinley_e9_bundles = .;
-        }
-
   /* Global data */
   _data = .;
 
-#if defined(CONFIG_IA64_GENERIC)
-  /* Machine Vector */
-  . = ALIGN(16);
-  .machvec : AT(ADDR(.machvec) - LOAD_OFFSET)
-        {
-          machvec_start = .;
-          *(.machvec)
-          machvec_end = .;
-        }
-#endif
-
   /* Unwind info & table: */
   . = ALIGN(8);
   .IA_64.unwind_info : AT(ADDR(.IA_64.unwind_info) - LOAD_OFFSET)
@@ -154,6 +129,41 @@ SECTIONS
           *(.initcall7.init)
           __initcall_end = .;
         }
+
+  /* MCA table */
+  . = ALIGN(16);
+  __mca_table : AT(ADDR(__mca_table) - LOAD_OFFSET)
+        {
+          __start___mca_table = .;
+          *(__mca_table)
+          __stop___mca_table = .;
+        }
+
+  .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET)
+        {
+          __start___vtop_patchlist = .;
+          *(.data.patch.vtop)
+          __end___vtop_patchlist = .;
+        }
+
+  .data.patch.mckinley_e9 : AT(ADDR(.data.patch.mckinley_e9) - LOAD_OFFSET)
+        {
+          __start___mckinley_e9_bundles = .;
+          *(.data.patch.mckinley_e9)
+          __end___mckinley_e9_bundles = .;
+        }
+
+#if defined(CONFIG_IA64_GENERIC)
+  /* Machine Vector */
+  . = ALIGN(16);
+  .machvec : AT(ADDR(.machvec) - LOAD_OFFSET)
+        {
+          machvec_start = .;
+          *(.machvec)
+          machvec_end = .;
+        }
+#endif
+
    __con_initcall_start = .;
   .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET)
         { *(.con_initcall.init) }
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index 9855ba318094..84fd1c14c8a9 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -97,7 +97,7 @@ find_max_pfn (unsigned long start, unsigned long end, void *arg)
  * Find a place to put the bootmap and return its starting address in
  * bootmap_start.  This address must be page-aligned.
  */
-int
+static int __init
 find_bootmap_location (unsigned long start, unsigned long end, void *arg)
 {
         unsigned long needed = *(unsigned long *)arg;
@@ -141,7 +141,7 @@ find_bootmap_location (unsigned long start, unsigned long end, void *arg)
  * Walk the EFI memory map and find usable memory for the system, taking
  * into account reserved areas.
  */
-void
+void __init
 find_memory (void)
 {
         unsigned long bootmap_size;
@@ -176,7 +176,7 @@ find_memory (void)
  *
  * Allocate and setup per-cpu data areas.
  */
-void *
+void * __cpuinit
 per_cpu_init (void)
 {
         void *cpu_data;
@@ -228,7 +228,7 @@ count_dma_pages (u64 start, u64 end, void *arg)
  * Set up the page tables.
  */
 
-void
+void __init
 paging_init (void)
 {
         unsigned long max_dma;
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 573d5cc63e2b..2f5e44862e91 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -525,7 +525,7 @@ void __init find_memory(void)
  * find_pernode_space() does most of this already, we just need to set
  * local_per_cpu_offset
  */
-void *per_cpu_init(void)
+void __cpuinit *per_cpu_init(void)
 {
         int cpu;
         static int first_time = 1;
diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index 9dbc7dadd165..8d506710fdbd 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -113,8 +113,7 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb,
                         unsigned long floor, unsigned long ceiling)
 {
         /*
-         * This is called only when is_hugepage_only_range(addr,),
-         * and it follows that is_hugepage_only_range(end,) also.
+         * This is called to free hugetlb page tables.
          *
          * The offset of these addresses from the base of the hugetlb
          * region must be scaled down by HPAGE_SIZE/PAGE_SIZE so that
@@ -126,9 +125,9 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb,
 
         addr = htlbpage_to_page(addr);
         end  = htlbpage_to_page(end);
-        if (is_hugepage_only_range(tlb->mm, floor, HPAGE_SIZE))
+        if (REGION_NUMBER(floor) == RGN_HPAGE)
                 floor = htlbpage_to_page(floor);
-        if (is_hugepage_only_range(tlb->mm, ceiling, HPAGE_SIZE))
+        if (REGION_NUMBER(ceiling) == RGN_HPAGE)
                 ceiling = htlbpage_to_page(ceiling);
 
         free_pgd_range(tlb, addr, end, floor, ceiling);
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index 08d94e6bfa18..ff4f31fcd330 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -206,7 +206,7 @@ free_initmem (void)
                (__init_end - __init_begin) >> 10);
 }
 
-void
+void __init
 free_initrd_mem (unsigned long start, unsigned long end)
 {
         struct page *page;
@@ -261,7 +261,7 @@ free_initrd_mem (unsigned long start, unsigned long end)
 /*
  * This installs a clean page in the kernel's page table.
  */
-struct page *
+static struct page * __init
 put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot)
 {
         pgd_t *pgd;
@@ -294,7 +294,7 @@ put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot)
         return page;
 }
 
-static void
+static void __init
 setup_gate (void)
 {
         struct page *page;
@@ -411,7 +411,7 @@ ia64_mmu_init (void *my_cpu_data)
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
 
-int
+int __init
 create_mem_map_page_table (u64 start, u64 end, void *arg)
 {
         unsigned long address, start_page, end_page;
@@ -519,7 +519,7 @@ ia64_pfn_valid (unsigned long pfn)
 }
 EXPORT_SYMBOL(ia64_pfn_valid);
 
-int
+int __init
 find_largest_hole (u64 start, u64 end, void *arg)
 {
         u64 *max_gap = arg;
@@ -535,7 +535,7 @@ find_largest_hole (u64 start, u64 end, void *arg)
 }
 #endif /* CONFIG_VIRTUAL_MEM_MAP */
 
-static int
+static int __init
 count_reserved_pages (u64 start, u64 end, void *arg)
 {
         unsigned long num_reserved = 0;
@@ -556,7 +556,7 @@ count_reserved_pages (u64 start, u64 end, void *arg)
  * purposes.
  */
 
-static int nolwsys;
+static int nolwsys __initdata;
 
 static int __init
 nolwsys_setup (char *s)
@@ -567,7 +567,7 @@ nolwsys_setup (char *s)
 
 __setup("nolwsys", nolwsys_setup);
 
-void
+void __init
 mem_init (void)
 {
         long reserved_pages, codesize, datasize, initsize;
diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index dfb3f2902379..5101ac462643 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -13,6 +13,8 @@
 #include <asm/sn/sn_feature_sets.h>
 #include <asm/sn/geo.h>
 #include <asm/sn/io.h>
+#include <asm/sn/l1.h>
+#include <asm/sn/module.h>
 #include <asm/sn/pcibr_provider.h>
 #include <asm/sn/pcibus_provider_defs.h>
 #include <asm/sn/pcidev.h>
@@ -710,9 +712,36 @@ cnodeid_get_geoid(cnodeid_t cnode)
         return hubdev->hdi_geoid;
 }
 
+void sn_generate_path(struct pci_bus *pci_bus, char *address)
+{
+        nasid_t nasid;
+        cnodeid_t cnode;
+        geoid_t geoid;
+        moduleid_t moduleid;
+        u16 bricktype;
+
+        nasid = NASID_GET(SN_PCIBUS_BUSSOFT(pci_bus)->bs_base);
+        cnode = nasid_to_cnodeid(nasid);
+        geoid = cnodeid_get_geoid(cnode);
+        moduleid = geo_module(geoid);
+
+        sprintf(address, "module_%c%c%c%c%.2d",
+                '0'+RACK_GET_CLASS(MODULE_GET_RACK(moduleid)),
+                '0'+RACK_GET_GROUP(MODULE_GET_RACK(moduleid)),
+                '0'+RACK_GET_NUM(MODULE_GET_RACK(moduleid)),
+                MODULE_GET_BTCHAR(moduleid), MODULE_GET_BPOS(moduleid));
+
+        /* Tollhouse requires slot id to be displayed */
+        bricktype = MODULE_GET_BTYPE(moduleid);
+        if ((bricktype == L1_BRICKTYPE_191010) ||
+            (bricktype == L1_BRICKTYPE_1932))
+                        sprintf(address, "%s^%d", address, geo_slot(geoid));
+}
+
 subsys_initcall(sn_pci_init);
 EXPORT_SYMBOL(sn_pci_fixup_slot);
 EXPORT_SYMBOL(sn_pci_unfixup_slot);
 EXPORT_SYMBOL(sn_pci_controller_fixup);
 EXPORT_SYMBOL(sn_bus_store_sysdata);
 EXPORT_SYMBOL(sn_bus_free_sysdata);
+EXPORT_SYMBOL(sn_generate_path);
diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c
index c373113d073a..c265e02f5036 100644
--- a/arch/ia64/sn/kernel/irq.c
+++ b/arch/ia64/sn/kernel/irq.c
@@ -350,9 +350,6 @@ static void force_interrupt(int irq)
 static void sn_check_intr(int irq, struct sn_irq_info *sn_irq_info)
 {
         u64 regval;
-        int irr_reg_num;
-        int irr_bit;
-        u64 irr_reg;
         struct pcidev_info *pcidev_info;
         struct pcibus_info *pcibus_info;
 
@@ -373,23 +370,7 @@ static void sn_check_intr(int irq, struct sn_irq_info *sn_irq_info)
             pdi_pcibus_info;
         regval = pcireg_intr_status_get(pcibus_info);
 
-        irr_reg_num = irq_to_vector(irq) / 64;
-        irr_bit = irq_to_vector(irq) % 64;
-        switch (irr_reg_num) {
-        case 0:
-                irr_reg = ia64_getreg(_IA64_REG_CR_IRR0);
-                break;
-        case 1:
-                irr_reg = ia64_getreg(_IA64_REG_CR_IRR1);
-                break;
-        case 2:
-                irr_reg = ia64_getreg(_IA64_REG_CR_IRR2);
-                break;
-        case 3:
-                irr_reg = ia64_getreg(_IA64_REG_CR_IRR3);
-                break;
-        }
-        if (!test_bit(irr_bit, &irr_reg)) {
+        if (!ia64_get_irr(irq_to_vector(irq))) {
                 if (!test_bit(irq, pda->sn_in_service_ivecs)) {
                         regval &= 0xff;
                         if (sn_irq_info->irq_int_bit & regval &
diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index 99cb28e74295..feaf1a6e8101 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -369,9 +369,15 @@ static void tio_corelet_reset(nasid_t nasid, int corelet)
 
 static int is_fpga_tio(int nasid, int *bt)
 {
-        int ioboard_type;
+        u16 ioboard_type;
+        s64 rc;
 
-        ioboard_type = ia64_sn_sysctl_ioboard_get(nasid);
+        rc = ia64_sn_sysctl_ioboard_get(nasid, &ioboard_type);
+        if (rc) {
+                printk(KERN_WARNING "ia64_sn_sysctl_ioboard_get failed: %ld\n",
+                       rc);
+                return 0;
+        }
 
         switch (ioboard_type) {
         case L1_BRICKTYPE_SA:
diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c
index 98f716bd92f0..ab1211ef0176 100644
--- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c
+++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c
@@ -74,6 +74,22 @@ static int sal_pcibr_error_interrupt(struct pcibus_info *soft)
         return (int)ret_stuff.v0;
 }
 
+u16 sn_ioboard_to_pci_bus(struct pci_bus *pci_bus)
+{
+        s64 rc;
+        u16 ioboard;
+        nasid_t nasid = NASID_GET(SN_PCIBUS_BUSSOFT(pci_bus)->bs_base);
+
+        rc = ia64_sn_sysctl_ioboard_get(nasid, &ioboard);
+        if (rc) {
+                printk(KERN_WARNING "ia64_sn_sysctl_ioboard_get failed: %ld\n",
+                       rc);
+                return 0;
+        }
+
+        return ioboard;
+}
+
 /* 
  * PCI Bridge Error interrupt handler.  Gets invoked whenever a PCI 
  * bridge sends an error interrupt.
@@ -255,3 +271,4 @@ pcibr_init_provider(void)
 
 EXPORT_SYMBOL_GPL(sal_pcibr_slot_enable);
 EXPORT_SYMBOL_GPL(sal_pcibr_slot_disable);
+EXPORT_SYMBOL_GPL(sn_ioboard_to_pci_bus);