[PATCH] x86: Fix boot hang due to nmi watchdog init code

2.6.19  stopped booting (or booted based on build/config) on our x86_64
systems due to a bug introduced in 2.6.19.  check_nmi_watchdog schedules an
IPI on all cpus to  busy wait on a flag, but fails to set the busywait
flag if NMI functionality is disabled.  This causes the secondary cpus
to spin in an endless loop, causing the kernel bootup to hang.
Depending upon the build, the  busywait flag got overwritten (stack variable)
and caused  the kernel to bootup on certain builds.  Following patch fixes
the bug by setting the busywait flag before returning from check_nmi_watchdog.
I guess using a stack variable is not good here as the calling function could
potentially return while the busy wait loop is still spinning on the flag.

AK: I redid the patch significantly to be cleaner

Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andi Kleen <ak@suse.de>

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index f5bc7e1..a5e34d6 100644 (file)
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -195,6 +195,8 @@ static __cpuinit inline int nmi_known_cpu(void)
        return 0;
 }
 
+static int endflag __initdata = 0;
+
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -202,7 +204,6 @@ static __cpuinit inline int nmi_known_cpu(void)
  */
 static __init void nmi_cpu_busy(void *data)
 {
-       volatile int *endflag = data;
        local_irq_enable_in_hardirq();
        /* Intentionally don't use cpu_relax here. This is
           to make sure that the performance counter really ticks,
@@ -210,14 +211,13 @@ static __init void nmi_cpu_busy(void *data)
           pause instruction. On a real HT machine this is fine because
           all other CPUs are busy with "useless" delay loops and don't
           care if they get somewhat less cycles. */
-       while (*endflag == 0)
-               barrier();
+       while (endflag == 0)
+               mb();
 }
 #endif
 
 static int __init check_nmi_watchdog(void)
 {
-       volatile int endflag = 0;
        unsigned int *prev_nmi_count;
        int cpu;
 

diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 27e95e7..186aebb 100644 (file)
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -193,6 +193,8 @@ void nmi_watchdog_default(void)
                nmi_watchdog = NMI_IO_APIC;
 }
 
+static int endflag __initdata = 0;
+
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -200,7 +202,6 @@ void nmi_watchdog_default(void)
  */
 static __init void nmi_cpu_busy(void *data)
 {
-       volatile int *endflag = data;
        local_irq_enable_in_hardirq();
        /* Intentionally don't use cpu_relax here. This is
           to make sure that the performance counter really ticks,
@@ -208,14 +209,13 @@ static __init void nmi_cpu_busy(void *data)
           pause instruction. On a real HT machine this is fine because
           all other CPUs are busy with "useless" delay loops and don't
           care if they get somewhat less cycles. */
-       while (*endflag == 0)
-               barrier();
+       while (endflag == 0)
+               mb();
 }
 #endif
 
 int __init check_nmi_watchdog (void)
 {
-       volatile int endflag = 0;
        int *counts;
        int cpu;
 
@@ -256,6 +256,7 @@ int __init check_nmi_watchdog (void)
        if (!atomic_read(&nmi_active)) {
                kfree(counts);
                atomic_set(&nmi_active, -1);
+               endflag = 1;
                return -1;
        }
        endflag = 1;