11 files changed, 206 insertions, 55 deletions

diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c
index 75a2a2c12258..2236fabbb3c6 100644
--- a/arch/ia64/kernel/asm-offsets.c
+++ b/arch/ia64/kernel/asm-offsets.c
@@ -35,6 +35,7 @@ void foo(void)
         BLANK();
 
         DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
+        DEFINE(TI_CPU, offsetof(struct thread_info, cpu));
         DEFINE(TI_PRE_COUNT, offsetof(struct thread_info, preempt_count));
 
         BLANK();
diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c
index 5cdd2f5fa064..80a94e707827 100644
--- a/arch/ia64/kernel/crash.c
+++ b/arch/ia64/kernel/crash.c
@@ -21,9 +21,9 @@
 #include <asm/mca.h>
 
 int kdump_status[NR_CPUS];
-atomic_t kdump_cpu_freezed;
+static atomic_t kdump_cpu_frozen;
 atomic_t kdump_in_progress;
-int kdump_on_init = 1;
+static int kdump_on_init = 1;
 
 static inline Elf64_Word
 *append_elf_note(Elf64_Word *buf, char *name, unsigned type, void *data,
@@ -86,7 +86,7 @@ kdump_wait_cpu_freeze(void)
         int cpu_num = num_online_cpus() - 1;
         int timeout = 1000;
         while(timeout-- > 0) {
-                if (atomic_read(&kdump_cpu_freezed) == cpu_num)
+                if (atomic_read(&kdump_cpu_frozen) == cpu_num)
                         return 0;
                 udelay(1000);
         }
@@ -108,8 +108,8 @@ machine_crash_shutdown(struct pt_regs *pt)
         kexec_disable_iosapic();
 #ifdef CONFIG_SMP
         kdump_smp_send_stop();
+        /* not all cpu response to IPI, send INIT to freeze them */
         if (kdump_wait_cpu_freeze() && kdump_on_init)   {
-                //not all cpu response to IPI, send INIT to freeze them
                 kdump_smp_send_init();
         }
 #endif
@@ -136,7 +136,7 @@ kdump_cpu_freeze(struct unw_frame_info *info, void *arg)
         cpuid = smp_processor_id();
         crash_save_this_cpu();
         current->thread.ksp = (__u64)info->sw - 16;
-        atomic_inc(&kdump_cpu_freezed);
+        atomic_inc(&kdump_cpu_frozen);
         kdump_status[cpuid] = 1;
         mb();
 #ifdef CONFIG_HOTPLUG_CPU
@@ -164,7 +164,7 @@ kdump_init_notifier(struct notifier_block *self, unsigned long val, void *data)
 
         nd = (struct ia64_mca_notify_die *)args->err;
         /* Reason code 1 means machine check rendezous*/
-        if ((val == DIE_INIT_MONARCH_ENTER || DIE_INIT_SLAVE_ENTER) &&
+        if ((val == DIE_INIT_MONARCH_ENTER || val == DIE_INIT_SLAVE_ENTER) &&
                  nd->sos->rv_rc == 1)
                 return NOTIFY_DONE;
 
diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c
index 4061593e5b17..f45f91d38cab 100644
--- a/arch/ia64/kernel/efi.c
+++ b/arch/ia64/kernel/efi.c
@@ -971,6 +971,11 @@ efi_memmap_init(unsigned long *s, unsigned long *e)
                 if (!is_memory_available(md))
                         continue;
 
+#ifdef CONFIG_CRASH_DUMP
+                /* saved_max_pfn should ignore max_addr= command line arg */
+                if (saved_max_pfn < (efi_md_end(md) >> PAGE_SHIFT))
+                        saved_max_pfn = (efi_md_end(md) >> PAGE_SHIFT);
+#endif
                 /*
                  * Round ends inward to granule boundaries
                  * Give trimmings to uncached allocator
@@ -1010,11 +1015,6 @@ efi_memmap_init(unsigned long *s, unsigned long *e)
                 } else
                         ae = efi_md_end(md);
 
-#ifdef CONFIG_CRASH_DUMP
-                /* saved_max_pfn should ignore max_addr= command line arg */
-                if (saved_max_pfn < (ae >> PAGE_SHIFT))
-                        saved_max_pfn = (ae >> PAGE_SHIFT);
-#endif
                 /* keep within max_addr= and min_addr= command line arg */
                 as = max(as, min_addr);
                 ae = min(ae, max_addr);
@@ -1142,7 +1142,7 @@ efi_initialize_iomem_resources(struct resource *code_resource,
 /* find a block of memory aligned to 64M exclude reserved regions
    rsvd_regions are sorted
  */
-unsigned long
+unsigned long __init
 kdump_find_rsvd_region (unsigned long size,
                 struct rsvd_region *r, int n)
 {
diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 7a05b1cb2ad5..8589e84a27c6 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -10,6 +10,8 @@
  *                      probably broke it along the way... ;-)
  * 13-Jul-04 clameter   Implement fsys_clock_gettime and revise fsys_gettimeofday to make
  *                      it capable of using memory based clocks without falling back to C code.
+ * 08-Feb-07 Fenghua Yu Implement fsys_getcpu.
+ *
  */
 
 #include <asm/asmmacro.h>
@@ -505,6 +507,59 @@ EX(.fail_efault, (p15) st8 [r34]=r3)
 #endif
 END(fsys_rt_sigprocmask)
 
+/*
+ * fsys_getcpu doesn't use the third parameter in this implementation. It reads
+ * current_thread_info()->cpu and corresponding node in cpu_to_node_map.
+ */
+ENTRY(fsys_getcpu)
+        .prologue
+        .altrp b6
+        .body
+        ;;
+        add r2=TI_FLAGS+IA64_TASK_SIZE,r16
+        tnat.nz p6,p0 = r32                     // guard against NaT argument
+        add r3=TI_CPU+IA64_TASK_SIZE,r16
+        ;;
+        ld4 r3=[r3]                             // M r3 = thread_info->cpu
+        ld4 r2=[r2]                             // M r2 = thread_info->flags
+(p6)    br.cond.spnt.few .fail_einval           // B
+        ;;
+        tnat.nz p7,p0 = r33                     // I guard against NaT argument
+(p7)    br.cond.spnt.few .fail_einval           // B
+#ifdef CONFIG_NUMA
+        movl r17=cpu_to_node_map
+        ;;
+EX(.fail_efault, probe.w.fault r32, 3)          // M This takes 5 cycles
+EX(.fail_efault, probe.w.fault r33, 3)          // M This takes 5 cycles
+        shladd r18=r3,1,r17
+        ;;
+        ld2 r20=[r18]                           // r20 = cpu_to_node_map[cpu]
+        and r2 = TIF_ALLWORK_MASK,r2
+        ;;
+        cmp.ne p8,p0=0,r2
+(p8)    br.spnt.many fsys_fallback_syscall
+        ;;
+        ;;
+EX(.fail_efault, st4 [r32] = r3)
+EX(.fail_efault, st2 [r33] = r20)
+        mov r8=0
+        ;;
+#else
+EX(.fail_efault, probe.w.fault r32, 3)          // M This takes 5 cycles
+EX(.fail_efault, probe.w.fault r33, 3)          // M This takes 5 cycles
+        and r2 = TIF_ALLWORK_MASK,r2
+        ;;
+        cmp.ne p8,p0=0,r2
+(p8)    br.spnt.many fsys_fallback_syscall
+        ;;
+EX(.fail_efault, st4 [r32] = r3)
+EX(.fail_efault, st2 [r33] = r0)
+        mov r8=0
+        ;;
+#endif
+        FSYS_RETURN
+END(fsys_getcpu)
+
 ENTRY(fsys_fallback_syscall)
         .prologue
         .altrp b6
@@ -878,6 +933,56 @@ fsyscall_table:
         data8 0                         // timer_delete
         data8 0                         // clock_settime
         data8 fsys_clock_gettime        // clock_gettime
+        data8 0                         // clock_getres         // 1255
+        data8 0                         // clock_nanosleep
+        data8 0                         // fstatfs64
+        data8 0                         // statfs64
+        data8 0                         // mbind
+        data8 0                         // get_mempolicy        // 1260
+        data8 0                         // set_mempolicy
+        data8 0                         // mq_open
+        data8 0                         // mq_unlink
+        data8 0                         // mq_timedsend
+        data8 0                         // mq_timedreceive      // 1265
+        data8 0                         // mq_notify
+        data8 0                         // mq_getsetattr
+        data8 0                         // kexec_load
+        data8 0                         // vserver
+        data8 0                         // waitid               // 1270
+        data8 0                         // add_key
+        data8 0                         // request_key
+        data8 0                         // keyctl
+        data8 0                         // ioprio_set
+        data8 0                         // ioprio_get           // 1275
+        data8 0                         // move_pages
+        data8 0                         // inotify_init
+        data8 0                         // inotify_add_watch
+        data8 0                         // inotify_rm_watch
+        data8 0                         // migrate_pages        // 1280
+        data8 0                         // openat
+        data8 0                         // mkdirat
+        data8 0                         // mknodat
+        data8 0                         // fchownat
+        data8 0                         // futimesat            // 1285
+        data8 0                         // newfstatat
+        data8 0                         // unlinkat
+        data8 0                         // renameat
+        data8 0                         // linkat
+        data8 0                         // symlinkat            // 1290
+        data8 0                         // readlinkat
+        data8 0                         // fchmodat
+        data8 0                         // faccessat
+        data8 0
+        data8 0                                                 // 1295
+        data8 0                         // unshare
+        data8 0                         // splice
+        data8 0                         // set_robust_list
+        data8 0                         // get_robust_list
+        data8 0                         // sync_file_range      // 1300
+        data8 0                         // tee
+        data8 0                         // vmsplice
+        data8 0
+        data8 fsys_getcpu               // getcpu               // 1304
 
         // fill in zeros for the remaining entries
         .zero:
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index d6aab40c6416..dcfbf3e7a9ef 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -446,7 +446,7 @@ iosapic_end_level_irq (unsigned int irq)
 #define iosapic_disable_level_irq       mask_irq
 #define iosapic_ack_level_irq           nop
 
-struct hw_interrupt_type irq_type_iosapic_level = {
+struct irq_chip irq_type_iosapic_level = {
         .name =         "IO-SAPIC-level",
         .startup =      iosapic_startup_level_irq,
         .shutdown =     iosapic_shutdown_level_irq,
@@ -454,6 +454,8 @@ struct hw_interrupt_type irq_type_iosapic_level = {
         .disable =      iosapic_disable_level_irq,
         .ack =          iosapic_ack_level_irq,
         .end =          iosapic_end_level_irq,
+        .mask =         mask_irq,
+        .unmask =       unmask_irq,
         .set_affinity = iosapic_set_affinity
 };
 
@@ -493,7 +495,7 @@ iosapic_ack_edge_irq (unsigned int irq)
 #define iosapic_disable_edge_irq        nop
 #define iosapic_end_edge_irq            nop
 
-struct hw_interrupt_type irq_type_iosapic_edge = {
+struct irq_chip irq_type_iosapic_edge = {
         .name =         "IO-SAPIC-edge",
         .startup =      iosapic_startup_edge_irq,
         .shutdown =     iosapic_disable_edge_irq,
@@ -501,6 +503,8 @@ struct hw_interrupt_type irq_type_iosapic_edge = {
         .disable =      iosapic_disable_edge_irq,
         .ack =          iosapic_ack_edge_irq,
         .end =          iosapic_end_edge_irq,
+        .mask =         mask_irq,
+        .unmask =       unmask_irq,
         .set_affinity = iosapic_set_affinity
 };
 
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index a76add3e76a2..491687f84fb5 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -1192,8 +1192,6 @@ void
 ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw,
                  struct ia64_sal_os_state *sos)
 {
-        pal_processor_state_info_t *psp = (pal_processor_state_info_t *)
-                &sos->proc_state_param;
         int recover, cpu = smp_processor_id();
         struct task_struct *previous_current;
         struct ia64_mca_notify_die nd =
@@ -1223,10 +1221,8 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw,
         /* Get the MCA error record and log it */
         ia64_mca_log_sal_error_record(SAL_INFO_TYPE_MCA);
 
-        /* TLB error is only exist in this SAL error record */
-        recover = (psp->tc && !(psp->cc || psp->bc || psp->rc || psp->uc))
-        /* other error recovery */
-           || (ia64_mca_ucmc_extension
+        /* MCA error recovery */
+        recover = (ia64_mca_ucmc_extension
                 && ia64_mca_ucmc_extension(
                         IA64_LOG_CURR_BUFFER(SAL_INFO_TYPE_MCA),
                         sos));
diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c
index afc1403799c9..832cf1e647e8 100644
--- a/arch/ia64/kernel/mca_drv.c
+++ b/arch/ia64/kernel/mca_drv.c
@@ -602,11 +602,40 @@ recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx,
                 default:
                         break;
                 }
+        } else if (psp->cc && !psp->bc) {       /* Cache error */
+                status = recover_from_read_error(slidx, peidx, pbci, sos);
         }
 
         return status;
 }
 
+/*
+ * recover_from_tlb_check
+ * @peidx:      pointer of index of processor error section
+ *
+ * Return value:
+ *      1 on Success / 0 on Failure
+ */
+static int
+recover_from_tlb_check(peidx_table_t *peidx)
+{
+        sal_log_mod_error_info_t *smei;
+        pal_tlb_check_info_t *ptci;
+
+        smei = (sal_log_mod_error_info_t *)peidx_tlb_check(peidx, 0);
+        ptci = (pal_tlb_check_info_t *)&(smei->check_info);
+
+        /*
+         * Look for signature of a duplicate TLB DTC entry, which is
+         * a SW bug and always fatal.
+         */
+        if (ptci->op == PAL_TLB_CHECK_OP_PURGE
+            && !(ptci->itr || ptci->dtc || ptci->itc))
+                return fatal_mca("Duplicate TLB entry");
+
+        return mca_recovered("TLB check recovered");
+}
+
 /**
  * recover_from_processor_error
  * @platform:   whether there are some platform error section or not
@@ -618,13 +647,6 @@ recover_from_platform_error(slidx_table_t *slidx, peidx_table_t *peidx,
  * Return value:
  *      1 on Success / 0 on Failure
  */
-/*
- *  Later we try to recover when below all conditions are satisfied.
- *   1. Only one processor error section is exist.
- *   2. BUS_CHECK is exist and the others are not exist.(Except TLB_CHECK)
- *   3. The entry of BUS_CHECK_INFO is 1.
- *   4. "External bus error" flag is set and the others are not set.
- */
 
 static int
 recover_from_processor_error(int platform, slidx_table_t *slidx,
@@ -652,38 +674,39 @@ recover_from_processor_error(int platform, slidx_table_t *slidx,
                 return fatal_mca("error not contained");
 
         /*
+         * Look for recoverable TLB check
+         */
+        if (psp->tc && !(psp->cc || psp->bc || psp->rc || psp->uc))
+                return recover_from_tlb_check(peidx);
+
+        /*
          * The cache check and bus check bits have four possible states
          *   cc bc
-         *    0  0      Weird record, not recovered
-         *    1  0      Cache error, not recovered
-         *    0  1      I/O error, attempt recovery
          *    1  1      Memory error, attempt recovery
+         *    1  0      Cache error, attempt recovery
+         *    0  1      I/O error, attempt recovery
+         *    0  0      Other error type, not recovered
          */
-        if (psp->bc == 0 || pbci == NULL)
-                return fatal_mca("No bus check");
+        if (psp->cc == 0 && (psp->bc == 0 || pbci == NULL))
+                return fatal_mca("No cache or bus check");
 
         /*
-         * Sorry, we cannot handle so many.
+         * Cannot handle more than one bus check.
          */
         if (peidx_bus_check_num(peidx) > 1)
                 return fatal_mca("Too many bus checks");
-        /*
-         * Well, here is only one bus error.
-         */
+
         if (pbci->ib)
                 return fatal_mca("Internal Bus error");
-        if (pbci->cc)
-                return fatal_mca("Cache-cache error");
         if (pbci->eb && pbci->bsi > 0)
                 return fatal_mca("External bus check fatal status");
 
         /*
-         * This is a local MCA and estimated as recoverble external bus error.
-         * (e.g. a load from poisoned memory)
-         * This means "there are some platform errors".
+         * This is a local MCA and estimated as a recoverble error.
          */
         if (platform)
                 return recover_from_platform_error(slidx, peidx, pbci, sos);
+
         /*
          * On account of strange SAL error record, we cannot recover.
          */
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index e7220900ea14..c81080df70df 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -68,7 +68,7 @@ int ia64_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
 {
         struct msi_msg  msg;
         unsigned long   dest_phys_id;
-        unsigned int    irq, vector;
+        int     irq, vector;
 
         irq = create_irq();
         if (irq < 0)
@@ -76,7 +76,7 @@ int ia64_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
 
         set_irq_msi(irq, desc);
         dest_phys_id = cpu_physical_id(first_cpu(cpu_online_map));
-        vector = irq;
+        vector = irq_to_vector(irq);
 
         msg.address_hi = 0;
         msg.address_lo =
@@ -110,7 +110,7 @@ static void ia64_ack_msi_irq(unsigned int irq)
 
 static int ia64_msi_retrigger_irq(unsigned int irq)
 {
-        unsigned int vector = irq;
+        unsigned int vector = irq_to_vector(irq);
         ia64_resend_irq(vector);
 
         return 1;
diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index 3f8918782e0c..00f803246948 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -1573,6 +1573,7 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data)
 
               case PTRACE_DETACH:
                 /* detach a process that was attached. */
+                clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
                 ret = ptrace_detach(child, data);
                 goto out_tsk;
 
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 7d6fe65c93f4..dc7dd7648ec5 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -91,8 +91,6 @@ static struct resource code_resource = {
         .name   = "Kernel code",
         .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
 };
-extern void efi_initialize_iomem_resources(struct resource *,
-                struct resource *);
 extern char _text[], _end[], _etext[];
 
 unsigned long ia64_max_cacheline_size;
@@ -642,7 +640,7 @@ show_cpuinfo (struct seq_file *m, void *v)
                    "features   : %s\n"
                    "cpu number : %lu\n"
                    "cpu regs   : %u\n"
-                   "cpu MHz    : %lu.%06lu\n"
+                   "cpu MHz    : %lu.%03lu\n"
                    "itc MHz    : %lu.%06lu\n"
                    "BogoMIPS   : %lu.%02lu\n",
                    cpunum, c->vendor, c->family, c->model,
@@ -694,12 +692,15 @@ struct seq_operations cpuinfo_op = {
         .show =         show_cpuinfo
 };
 
-static char brandname[128];
+#define MAX_BRANDS      8
+static char brandname[MAX_BRANDS][128];
 
 static char * __cpuinit
 get_model_name(__u8 family, __u8 model)
 {
+        static int overflow;
         char brand[128];
+        int i;
 
         memcpy(brand, "Unknown", 8);
         if (ia64_pal_get_brand_info(brand)) {
@@ -711,12 +712,17 @@ get_model_name(__u8 family, __u8 model)
                         case 2: memcpy(brand, "Madison up to 9M cache", 23); break;
                 }
         }
-        if (brandname[0] == '\0')
-                return strcpy(brandname, brand);
-        else if (strcmp(brandname, brand) == 0)
-                return brandname;
-        else
-                return kstrdup(brand, GFP_KERNEL);
+        for (i = 0; i < MAX_BRANDS; i++)
+                if (strcmp(brandname[i], brand) == 0)
+                        return brandname[i];
+        for (i = 0; i < MAX_BRANDS; i++)
+                if (brandname[i][0] == '\0')
+                        return strcpy(brandname[i], brand);
+        if (overflow++ == 0)
+                printk(KERN_ERR
+                       "%s: Table overflow. Some processor model information will be missing\n",
+                       __FUNCTION__);
+        return "Unknown";
 }
 
 static void __cpuinit
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index b21ddecea943..ff7df439da6d 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -375,6 +375,7 @@ static void __devinit
 smp_callin (void)
 {
         int cpuid, phys_id, itc_master;
+        struct cpuinfo_ia64 *last_cpuinfo, *this_cpuinfo;
         extern void ia64_init_itm(void);
         extern volatile int time_keeper_id;
 
@@ -424,7 +425,21 @@ smp_callin (void)
          * Get our bogomips.
          */
         ia64_init_itm();
-        calibrate_delay();
+
+        /*
+         * Delay calibration can be skipped if new processor is identical to the
+         * previous processor.
+         */
+        last_cpuinfo = cpu_data(cpuid - 1);
+        this_cpuinfo = local_cpu_data;
+        if (last_cpuinfo->itc_freq != this_cpuinfo->itc_freq ||
+            last_cpuinfo->proc_freq != this_cpuinfo->proc_freq ||
+            last_cpuinfo->features != this_cpuinfo->features ||
+            last_cpuinfo->revision != this_cpuinfo->revision ||
+            last_cpuinfo->family != this_cpuinfo->family ||
+            last_cpuinfo->archrev != this_cpuinfo->archrev ||
+            last_cpuinfo->model != this_cpuinfo->model)
+                calibrate_delay();
         local_cpu_data->loops_per_jiffy = loops_per_jiffy;
 
 #ifdef CONFIG_IA32_SUPPORT