9 files changed, 154 insertions, 131 deletions
diff --git a/arch/ia64/include/asm/mca.h b/arch/ia64/include/asm/mca.h
index 44a0b53df900..c171cdf0a789 100644
--- a/arch/ia64/include/asm/mca.h
+++ b/arch/ia64/include/asm/mca.h
@@ -145,12 +145,14 @@ extern void ia64_mca_ucmc_handler(struct pt_regs *, struct ia64_sal_os_state *);
 extern void ia64_init_handler(struct pt_regs *,
                              struct switch_stack *,
                              struct ia64_sal_os_state *);
+extern void ia64_os_init_on_kdump(void);
 extern void ia64_monarch_init_handler(void);
 extern void ia64_slave_init_handler(void);
 extern void ia64_mca_cmc_vector_setup(void);
 extern int  ia64_reg_MCA_extension(int (*fn)(void *, struct ia64_sal_os_state *));
 extern void ia64_unreg_MCA_extension(void);
 extern unsigned long ia64_get_rnat(unsigned long *);
+extern void ia64_set_psr_mc(void);
 extern void ia64_mca_printk(const char * fmt, ...)
         __attribute__ ((format (printf, 1, 2)));
diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c
index f065093f8e9b..6631a9dfafdc 100644
--- a/arch/ia64/kernel/crash.c
+++ b/arch/ia64/kernel/crash.c
@@ -23,6 +23,7 @@
 int kdump_status[NR_CPUS];
 static atomic_t kdump_cpu_frozen;
 atomic_t kdump_in_progress;
+static int kdump_freeze_monarch;
 static int kdump_on_init = 1;
 static int kdump_on_fatal_mca = 1;
@@ -108,10 +109,38 @@ machine_crash_shutdown(struct pt_regs *pt)
         */
        kexec_disable_iosapic();
 #ifdef CONFIG_SMP
+        /*
+         * If kdump_on_init is set and an INIT is asserted here, kdump will
+         * be started again via INIT monarch.
+         */
+        local_irq_disable();
+        ia64_set_psr_mc();      /* mask MCA/INIT */
+        if (atomic_inc_return(&kdump_in_progress) != 1)
+                unw_init_running(kdump_cpu_freeze, NULL);
+        /*
+         * Now this cpu is ready for kdump.
+         * Stop all others by IPI or INIT.  They could receive INIT from
+         * outside and might be INIT monarch, but only thing they have to
+         * do is falling into kdump_cpu_freeze().
+         *
+         * If an INIT is asserted here:
+         * - All receivers might be slaves, since some of cpus could already
+         *   be frozen and INIT might be masked on monarch.  In this case,
+         *   all slaves will be frozen soon since kdump_in_progress will let
+         *   them into DIE_INIT_SLAVE_LEAVE.
+         * - One might be a monarch, but INIT rendezvous will fail since
+         *   at least this cpu already have INIT masked so it never join
+         *   to the rendezvous.  In this case, all slaves and monarch will
+         *   be frozen soon with no wait since the INIT rendezvous is skipped
+         *   by kdump_in_progress.
+         */
        kdump_smp_send_stop();
        /* not all cpu response to IPI, send INIT to freeze them */
-        if (kdump_wait_cpu_freeze() && kdump_on_init)   {
+        if (kdump_wait_cpu_freeze()) {
                kdump_smp_send_init();
+                /* wait again, don't go ahead if possible */
+                kdump_wait_cpu_freeze();
        }
 #endif
 }
@@ -129,17 +158,17 @@ void
 kdump_cpu_freeze(struct unw_frame_info *info, void *arg)
 {
        int cpuid;
        local_irq_disable();
        cpuid = smp_processor_id();
        crash_save_this_cpu();
        current->thread.ksp = (__u64)info->sw - 16;
+        ia64_set_psr_mc();      /* mask MCA/INIT and stop reentrance */
        atomic_inc(&kdump_cpu_frozen);
        kdump_status[cpuid] = 1;
        mb();
-#ifdef CONFIG_HOTPLUG_CPU
-        if (cpuid != 0)
-                ia64_jump_to_sal(&sal_boot_rendez_state[cpuid]);
-#endif
        for (;;)
                cpu_relax();
 }
@@ -150,6 +179,20 @@ kdump_init_notifier(struct notifier_block *self, unsigned long val, void *data)
        struct ia64_mca_notify_die *nd;
        struct die_args *args = data;
+        if (atomic_read(&kdump_in_progress)) {
+                switch (val) {
+                case DIE_INIT_MONARCH_LEAVE:
+                        if (!kdump_freeze_monarch)
+                                break;
+                        /* fall through */
+                case DIE_INIT_SLAVE_LEAVE:
+                case DIE_INIT_MONARCH_ENTER:
+                case DIE_MCA_RENDZVOUS_LEAVE:
+                        unw_init_running(kdump_cpu_freeze, NULL);
+                        break;
+                }
+        }
        if (!kdump_on_init && !kdump_on_fatal_mca)
                return NOTIFY_DONE;
@@ -162,43 +205,31 @@ kdump_init_notifier(struct notifier_block *self, unsigned long val, void *data)
        }
        if (val != DIE_INIT_MONARCH_LEAVE &&
-            val != DIE_INIT_SLAVE_LEAVE &&
            val != DIE_INIT_MONARCH_PROCESS &&
-            val != DIE_MCA_RENDZVOUS_LEAVE &&
            val != DIE_MCA_MONARCH_LEAVE)
                return NOTIFY_DONE;
        nd = (struct ia64_mca_notify_die *)args->err;
-        /* Reason code 1 means machine check rendezvous*/
-        if ((val == DIE_INIT_MONARCH_LEAVE || val == DIE_INIT_SLAVE_LEAVE
-            || val == DIE_INIT_MONARCH_PROCESS) && nd->sos->rv_rc == 1)
-                return NOTIFY_DONE;
        switch (val) {
        case DIE_INIT_MONARCH_PROCESS:
-                if (kdump_on_init) {
+                /* Reason code 1 means machine check rendezvous*/
-                        atomic_set(&kdump_in_progress, 1);
+                if (kdump_on_init && (nd->sos->rv_rc != 1)) {
-                        *(nd->monarch_cpu) = -1;
+                        if (atomic_inc_return(&kdump_in_progress) != 1)
+                                kdump_freeze_monarch = 1;
                }
                break;
        case DIE_INIT_MONARCH_LEAVE:
-                if (kdump_on_init)
+                /* Reason code 1 means machine check rendezvous*/
+                if (kdump_on_init && (nd->sos->rv_rc != 1))
                        machine_kdump_on_init();
                break;
-        case DIE_INIT_SLAVE_LEAVE:
-                if (atomic_read(&kdump_in_progress))
-                        unw_init_running(kdump_cpu_freeze, NULL);
-                break;
-        case DIE_MCA_RENDZVOUS_LEAVE:
-                if (atomic_read(&kdump_in_progress))
-                        unw_init_running(kdump_cpu_freeze, NULL);
-                break;
        case DIE_MCA_MONARCH_LEAVE:
                /* *(nd->data) indicate if MCA is recoverable */
                if (kdump_on_fatal_mca && !(*(nd->data))) {
-                        atomic_set(&kdump_in_progress, 1);
+                        if (atomic_inc_return(&kdump_in_progress) == 1)
-                        *(nd->monarch_cpu) = -1;
+                                machine_kdump_on_init();
-                        machine_kdump_on_init();
+                        /* We got fatal MCA while kdump!? No way!! */
                }
                break;
        }
diff --git a/arch/ia64/kernel/head.S b/arch/ia64/kernel/head.S
index 23f846de62d5..1a6e44515eb4 100644
--- a/arch/ia64/kernel/head.S
+++ b/arch/ia64/kernel/head.S
@@ -167,7 +167,7 @@ RestRR:											\
        mov _tmp2=((ia64_rid(IA64_REGION_ID_KERNEL, (num<<61)) << 8) | (pgsize << 2) | vhpt);; \
        mov rr[_tmp1]=_tmp2
-        .section __special_page_section,"ax"
+        __PAGE_ALIGNED_DATA
        .global empty_zero_page
 empty_zero_page:
@@ -181,7 +181,7 @@ swapper_pg_dir:
 halt_msg:
        stringz "Halting kernel\n"
-        .section .text.head,"ax"
+        __REF
        .global start_ap
@@ -1242,7 +1242,7 @@ GLOBAL_ENTRY(ia64_jump_to_sal)
        movl r16=SAL_PSR_BITS_TO_SET;;
        mov cr.ipsr=r16
        mov cr.ifs=r0;;
-        rfi;;
+        rfi;;                   // note: this unmask MCA/INIT (psr.mc)
 1:
        /*
         * Invalidate all TLB data/inst
diff --git a/arch/ia64/kernel/machine_kexec.c b/arch/ia64/kernel/machine_kexec.c
index 0823de1f6ebe..3d3aeef46947 100644
--- a/arch/ia64/kernel/machine_kexec.c
+++ b/arch/ia64/kernel/machine_kexec.c
@@ -24,6 +24,8 @@
 #include <asm/delay.h>
 #include <asm/meminit.h>
 #include <asm/processor.h>
+#include <asm/sal.h>
+#include <asm/mca.h>
 typedef NORET_TYPE void (*relocate_new_kernel_t)(
                                        unsigned long indirection_page,
@@ -85,13 +87,26 @@ static void ia64_machine_kexec(struct unw_frame_info *info, void *arg)
        void *pal_addr = efi_get_pal_addr();
        unsigned long code_addr = (unsigned long)page_address(image->control_code_page);
        int ii;
+        u64 fp, gp;
+        ia64_fptr_t *init_handler = (ia64_fptr_t *)ia64_os_init_on_kdump;
        BUG_ON(!image);
        if (image->type == KEXEC_TYPE_CRASH) {
                crash_save_this_cpu();
                current->thread.ksp = (__u64)info->sw - 16;
+                /* Register noop init handler */
+                fp = ia64_tpa(init_handler->fp);
+                gp = ia64_tpa(ia64_getreg(_IA64_REG_GP));
+                ia64_sal_set_vectors(SAL_VECTOR_OS_INIT, fp, gp, 0, fp, gp, 0);
+        } else {
+                /* Unregister init handlers of current kernel */
+                ia64_sal_set_vectors(SAL_VECTOR_OS_INIT, 0, 0, 0, 0, 0, 0);
        }
+        /* Unregister mca handler - No more recovery on current kernel */
+        ia64_sal_set_vectors(SAL_VECTOR_OS_MCA, 0, 0, 0, 0, 0, 0);
        /* Interrupts aren't acceptable while we reboot */
        local_irq_disable();
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 7b30d21c5190..d2877a7bfe2e 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -1682,14 +1682,25 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
        if (!sos->monarch) {
                ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_INIT;
+#ifdef CONFIG_KEXEC
+                while (monarch_cpu == -1 && !atomic_read(&kdump_in_progress))
+                        udelay(1000);
+#else
                while (monarch_cpu == -1)
-                       cpu_relax();     /* spin until monarch enters */
+                        cpu_relax();    /* spin until monarch enters */
+#endif
                NOTIFY_INIT(DIE_INIT_SLAVE_ENTER, regs, (long)&nd, 1);
                NOTIFY_INIT(DIE_INIT_SLAVE_PROCESS, regs, (long)&nd, 1);
+#ifdef CONFIG_KEXEC
+                while (monarch_cpu != -1 && !atomic_read(&kdump_in_progress))
+                        udelay(1000);
+#else
                while (monarch_cpu != -1)
-                       cpu_relax();     /* spin until monarch leaves */
+                        cpu_relax();    /* spin until monarch leaves */
+#endif
                NOTIFY_INIT(DIE_INIT_SLAVE_LEAVE, regs, (long)&nd, 1);
diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S
index a06d46548ff9..7461d2573d41 100644
--- a/arch/ia64/kernel/mca_asm.S
+++ b/arch/ia64/kernel/mca_asm.S
@@ -40,6 +40,7 @@
        .global ia64_do_tlb_purge
        .global ia64_os_mca_dispatch
+        .global ia64_os_init_on_kdump
        .global ia64_os_init_dispatch_monarch
        .global ia64_os_init_dispatch_slave
@@ -299,6 +300,25 @@ END(ia64_os_mca_virtual_begin)
 //StartMain////////////////////////////////////////////////////////////////////
 //
+// NOP init handler for kdump.  In panic situation, we may receive INIT
+// while kernel transition.  Since we initialize registers on leave from
+// current kernel, no longer monarch/slave handlers of current kernel in
+// virtual mode are called safely.
+// We can unregister these init handlers from SAL, however then the INIT
+// will result in warmboot by SAL and we cannot retrieve the crashdump.
+// Therefore register this NOP function to SAL, to prevent entering virtual
+// mode and resulting warmboot by SAL.
+//
+ia64_os_init_on_kdump:
+        mov             r8=r0           // IA64_INIT_RESUME
+        mov             r9=r10          // SAL_GP
+        mov             r22=r17         // *minstate
+        ;;
+        mov             r10=r0          // return to same context
+        mov             b0=r12          // SAL_CHECK return address
+        br              b0
+//
 // SAL to OS entry point for INIT on all processors.  This has been defined for
 // registration purposes with SAL as a part of ia64_mca_init.  Monarch and
 // slave INIT have identical processing, except for the value of the
@@ -1073,3 +1093,30 @@ GLOBAL_ENTRY(ia64_get_rnat)
        mov ar.rsc=3
        br.ret.sptk.many rp
 END(ia64_get_rnat)
+// void ia64_set_psr_mc(void)
+//
+// Set psr.mc bit to mask MCA/INIT.
+GLOBAL_ENTRY(ia64_set_psr_mc)
+        rsm psr.i | psr.ic              // disable interrupts
+        ;;
+        srlz.d
+        ;;
+        mov r14 = psr                   // get psr{36:35,31:0}
+        movl r15 = 1f
+        ;;
+        dep r14 = -1, r14, PSR_MC, 1    // set psr.mc
+        ;;
+        dep r14 = -1, r14, PSR_IC, 1    // set psr.ic
+        ;;
+        dep r14 = -1, r14, PSR_BN, 1    // keep bank1 in use
+        ;;
+        mov cr.ipsr = r14
+        mov cr.ifs = r0
+        mov cr.iip = r15
+        ;;
+        rfi
+1:
+        br.ret.sptk.many rp
+END(ia64_set_psr_mc)
diff --git a/arch/ia64/kernel/relocate_kernel.S b/arch/ia64/kernel/relocate_kernel.S
index 903babd22d62..32f6fc131fbe 100644
--- a/arch/ia64/kernel/relocate_kernel.S
+++ b/arch/ia64/kernel/relocate_kernel.S
@@ -52,7 +52,7 @@ GLOBAL_ENTRY(relocate_new_kernel)
        srlz.i
        ;;
        mov ar.rnat=r18
-        rfi
+        rfi                             // note: this unmask MCA/INIT (psr.mc)
        ;;
 1:
        //physical mode code begin
diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index eb4214d1c5af..0a0c77b2c988 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -51,8 +51,6 @@ SECTIONS
        KPROBES_TEXT
        *(.gnu.linkonce.t*)
    }
-  .text.head : AT(ADDR(.text.head) - LOAD_OFFSET)
-        { *(.text.head) }
  .text2 : AT(ADDR(.text2) - LOAD_OFFSET)
        { *(.text2) }
 #ifdef CONFIG_SMP
@@ -66,14 +64,7 @@ SECTIONS
  NOTES :code :note             /* put .notes in text and mark in PT_NOTE  */
  code_continues : {} :code     /* switch back to regular program...  */
-  /* Exception table */
+  EXCEPTION_TABLE(16)
-  . = ALIGN(16);
-  __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET)
-        {
-          __start___ex_table = .;
-          *(__ex_table)
-          __stop___ex_table = .;
-        }
  /* MCA table */
  . = ALIGN(16);
@@ -115,38 +106,9 @@ SECTIONS
  . = ALIGN(PAGE_SIZE);
  __init_begin = .;
-  .init.text : AT(ADDR(.init.text) - LOAD_OFFSET)
-        {
-          _sinittext = .;
-          INIT_TEXT
-          _einittext = .;
-        }
-  .init.data : AT(ADDR(.init.data) - LOAD_OFFSET)
-        { INIT_DATA }
-#ifdef CONFIG_BLK_DEV_INITRD
+  INIT_TEXT_SECTION(PAGE_SIZE)
-  .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET)
+  INIT_DATA_SECTION(16)
-        {
-          __initramfs_start = .;
-          *(.init.ramfs)
-          __initramfs_end = .;
-        }
-#endif
-   . = ALIGN(16);
-  .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET)
-        {
-          __setup_start = .;
-          *(.init.setup)
-          __setup_end = .;
-        }
-  .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET)
-        {
-          __initcall_start = .;
-        INITCALLS
-          __initcall_end = .;
-        }
  .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET)
        {
@@ -204,24 +166,13 @@ SECTIONS
        }
 #endif
-  . = ALIGN(8);
-   __con_initcall_start = .;
-  .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET)
-        { *(.con_initcall.init) }
-  __con_initcall_end = .;
-  __security_initcall_start = .;
-  .security_initcall.init : AT(ADDR(.security_initcall.init) - LOAD_OFFSET)
-        { *(.security_initcall.init) }
-  __security_initcall_end = .;
  . = ALIGN(PAGE_SIZE);
  __init_end = .;
-  /* The initial task and kernel stack */
-  .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET)
-        { *(.data.init_task) }
  .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET)
-        { *(__special_page_section)
+        {
+        PAGE_ALIGNED_DATA(PAGE_SIZE)
+          . = ALIGN(PAGE_SIZE);
          __start_gate_section = .;
          *(.data.gate)
          __stop_gate_section = .;
@@ -236,12 +187,6 @@ SECTIONS
                                 * kernel data
                                 */
-  .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET)
-        { *(.data.read_mostly) }
-  .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET)
-        { *(.data.cacheline_aligned) }
  /* Per-cpu data: */
  . = ALIGN(PERCPU_PAGE_SIZE);
  PERCPU_VADDR(PERCPU_ADDR, :percpu)
@@ -258,6 +203,9 @@ SECTIONS
                __cpu0_per_cpu = .;
  . = . + PERCPU_PAGE_SIZE;     /* cpu0 per-cpu space */
 #endif
+                INIT_TASK_DATA(PAGE_SIZE)
+                CACHELINE_ALIGNED_DATA(SMP_CACHE_BYTES)
+                READ_MOSTLY_DATA(SMP_CACHE_BYTES)
                DATA_DATA
                *(.data1)
                *(.gnu.linkonce.d*)
@@ -274,48 +222,15 @@ SECTIONS
  .sdata : AT(ADDR(.sdata) - LOAD_OFFSET)
        { *(.sdata) *(.sdata1) *(.srdata) }
  _edata  =  .;
-  __bss_start = .;
-  .sbss : AT(ADDR(.sbss) - LOAD_OFFSET)
+  BSS_SECTION(0, 0, 0)
-        { *(.sbss) *(.scommon) }
-  .bss : AT(ADDR(.bss) - LOAD_OFFSET)
-        { *(.bss) *(COMMON) }
-  __bss_stop = .;
  _end = .;
  code : { } :code
-  /* Stabs debugging sections.  */
-  .stab 0 : { *(.stab) }
+  STABS_DEBUG
-  .stabstr 0 : { *(.stabstr) }
+  DWARF_DEBUG
-  .stab.excl 0 : { *(.stab.excl) }
-  .stab.exclstr 0 : { *(.stab.exclstr) }
-  .stab.index 0 : { *(.stab.index) }
-  .stab.indexstr 0 : { *(.stab.indexstr) }
-  /* DWARF debug sections.
-     Symbols in the DWARF debugging sections are relative to the beginning
-     of the section so we begin them at 0.  */
-  /* DWARF 1 */
-  .debug          0 : { *(.debug) }
-  .line           0 : { *(.line) }
-  /* GNU DWARF 1 extensions */
-  .debug_srcinfo  0 : { *(.debug_srcinfo) }
-  .debug_sfnames  0 : { *(.debug_sfnames) }
-  /* DWARF 1.1 and DWARF 2 */
-  .debug_aranges  0 : { *(.debug_aranges) }
-  .debug_pubnames 0 : { *(.debug_pubnames) }
-  /* DWARF 2 */
-  .debug_info     0 : { *(.debug_info) }
-  .debug_abbrev   0 : { *(.debug_abbrev) }
-  .debug_line     0 : { *(.debug_line) }
-  .debug_frame    0 : { *(.debug_frame) }
-  .debug_str      0 : { *(.debug_str) }
-  .debug_loc      0 : { *(.debug_loc) }
-  .debug_macinfo  0 : { *(.debug_macinfo) }
-  /* SGI/MIPS DWARF 2 extensions */
-  .debug_weaknames 0 : { *(.debug_weaknames) }
-  .debug_funcnames 0 : { *(.debug_funcnames) }
-  .debug_typenames 0 : { *(.debug_typenames) }
-  .debug_varnames  0 : { *(.debug_varnames) }
  /* Default discards */
  DISCARDS
diff --git a/arch/ia64/sn/pci/pcibr/pcibr_ate.c b/arch/ia64/sn/pci/pcibr/pcibr_ate.c
index 239b3cedcf2b..5bc34eac9e01 100644
--- a/arch/ia64/sn/pci/pcibr/pcibr_ate.c
+++ b/arch/ia64/sn/pci/pcibr/pcibr_ate.c
@@ -54,6 +54,8 @@ static int find_free_ate(struct ate_resource *ate_resource, int start,
                                        break;
                                }
                        }
+                        if (i >= ate_resource->num_ate)
+                                return -1;
                } else
                        index++;        /* Try next ate */
        }