Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6

[linux-2.6-omap-h63xx.git] / arch / ia64 / kernel / mca_drv.c

diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c
index eb39bc9c133bd05dbf5adba4017dc286d56e473a..37c88eb55873792250d74ac22455ee5cdcab93af 100644 (file)
--- 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>
@@ -121,10 +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_DEBUG "OS_MCA: process [pid: %d](%s) encounters MCA.\n",
-               current->pid, current->comm);
+       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)) {
@@ -132,7 +135,7 @@ mca_handler_bh(unsigned long paddr)
                printk(KERN_DEBUG "Page isolation: ( %lx ) success.\n", paddr);
                break;
        case ISOLATE_NG:
-               printk(KERN_DEBUG "Page isolation: ( %lx ) failure.\n", paddr);
+               printk(KERN_CRIT "Page isolation: ( %lx ) failure.\n", paddr);
                break;
        default:
                break;
@@ -437,22 +440,30 @@ recover_from_read_error(slidx_table_t *slidx,
         *    the process not have any locks of kernel.
         */
 
+       /* Is minstate valid? */
+       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 */
@@ -462,6 +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;
@@ -547,16 +559,32 @@ recover_from_processor_error(int platform, slidx_table_t *slidx,
                (pal_processor_state_info_t*)peidx_psp(peidx);
 
        /*
-        * We cannot recover errors with other than bus_check.
+        * Processor recovery status must key off of the PAL recovery
+        * status in the Processor State Parameter.
+        */
+
+       /*
+        * The machine check is corrected.
+        */
+       if (psp->cm == 1)
+               return 1;
+
+       /*
+        * The error was not contained.  Software must be reset.
         */
-       if (psp->cc || psp->rc || psp->uc)
+       if (psp->us || psp->ci == 0)
                return 0;
 
        /*
-        * If there is no bus error, record is weird but we need not to recover.
+        * 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
         */
        if (psp->bc == 0 || pbci == NULL)
-               return 1;
+               return 0;
 
        /*
         * Sorry, we cannot handle so many.
@@ -570,8 +598,6 @@ recover_from_processor_error(int platform, slidx_table_t *slidx,
                return 0;
        if (pbci->eb && pbci->bsi > 0)
                return 0;
-       if (psp->ci == 0)
-               return 0;
 
        /*
         * This is a local MCA and estimated as recoverble external bus error.