Commit 0cc582ee authored by Dave Jones's avatar Dave Jones Committed by Jaroslav Kysela

[PATCH] x86-64 bluesmoke update

- Drop Intel P4 bits
- Update other parts in line with latest i386 bluesmoke.c
parent d6ba10fd
/*
* arch/x86_64/kernel/bluesmoke.c - x86-64 Machine Check Exception Reporting
*/
#include <linux/init.h> #include <linux/init.h>
#include <linux/types.h> #include <linux/types.h>
...@@ -13,151 +16,44 @@ ...@@ -13,151 +16,44 @@
#include <asm/pgtable.h> #include <asm/pgtable.h>
#include <asm/tlbflush.h> #include <asm/tlbflush.h>
#ifdef CONFIG_X86_MCE
static int mce_disabled __initdata = 0; static int mce_disabled __initdata = 0;
static int banks; static int banks;
/*
* If we get an MCE, we don't know what state the caches/TLB's are
* going to be in, so we throw them all away.
*/
static void inline flush_all (void)
{
__asm__ __volatile__ ("invd": : );
__flush_tlb();
}
/* /*
* P4/Xeon Thermal transition interrupt handler * Machine Check Handler For Hammer
*/ */
static void intel_thermal_interrupt(struct pt_regs *regs) static void hammer_machine_check(struct pt_regs * regs, long error_code)
{
#ifdef CONFIG_X86_LOCAL_APIC
u32 l, h;
unsigned int cpu = smp_processor_id();
ack_APIC_irq();
rdmsr(MSR_IA32_THERM_STATUS, l, h);
if (l & 1) {
printk(KERN_EMERG "CPU#%d: Temperature above threshold\n", cpu);
printk(KERN_EMERG "CPU#%d: Running in modulated clock mode\n", cpu);
} else {
printk(KERN_INFO "CPU#%d: Temperature/speed normal\n", cpu);
}
#endif
}
static void unexpected_thermal_interrupt(struct pt_regs *regs)
{
printk(KERN_ERR "CPU#%d: Unexpected LVT TMR interrupt!\n", smp_processor_id());
}
/*
* Thermal interrupt handler for this CPU setup
*/
static void (*vendor_thermal_interrupt)(struct pt_regs *regs) = unexpected_thermal_interrupt;
asmlinkage void smp_thermal_interrupt(struct pt_regs regs)
{
vendor_thermal_interrupt(&regs);
}
/* P4/Xeon Thermal regulation detect and init */
static void __init intel_init_thermal(struct cpuinfo_x86 *c)
{
#ifdef CONFIG_X86_LOCAL_APIC
u32 l, h;
unsigned int cpu = smp_processor_id();
/* Thermal monitoring */
if (!test_bit(X86_FEATURE_ACPI, &c->x86_capability))
return; /* -ENODEV */
/* Clock modulation */
if (!test_bit(X86_FEATURE_ACC, &c->x86_capability))
return; /* -ENODEV */
rdmsr(MSR_IA32_MISC_ENABLE, l, h);
/* first check if its enabled already, in which case there might
* be some SMM goo which handles it, so we can't even put a handler
* since it might be delivered via SMI already -zwanem.
*/
if (l & (1<<3)) {
printk(KERN_DEBUG "CPU#%d: Thermal monitoring already enabled\n", cpu);
} else {
wrmsr(MSR_IA32_MISC_ENABLE, l | (1<<3), h);
printk(KERN_INFO "CPU#%d: Thermal monitoring enabled\n", cpu);
}
/* check wether a vector already exists */
l = apic_read(APIC_LVTTHMR);
if (l & 0xff) {
printk(KERN_DEBUG "CPU#%d: Thermal LVT already handled\n", cpu);
return; /* -EBUSY */
}
wrmsr(MSR_IA32_MISC_ENABLE, l | (1<<3), h);
printk(KERN_INFO "CPU#%d: Thermal monitoring enabled\n", cpu);
/* The temperature transition interrupt handler setup */
l = THERMAL_APIC_VECTOR; /* our delivery vector */
l |= (APIC_DM_FIXED | APIC_LVT_MASKED); /* we'll mask till we're ready */
apic_write_around(APIC_LVTTHMR, l);
rdmsr(MSR_IA32_THERM_INTERRUPT, l, h);
wrmsr(MSR_IA32_THERM_INTERRUPT, l | 0x3 , h);
/* ok we're good to go... */
vendor_thermal_interrupt = intel_thermal_interrupt;
l = apic_read(APIC_LVTTHMR);
apic_write_around(APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
return;
#endif
}
/*
* Machine Check Handler For PII/PIII
*/
static void intel_machine_check(struct pt_regs * regs, long error_code)
{ {
int recover=1; int recover=1;
u32 alow, ahigh, high, low; u32 alow, ahigh, high, low;
u32 mcgstl, mcgsth; u32 mcgstl, mcgsth;
int i; int i;
flush_all();
rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth);
if(mcgstl&(1<<0)) /* Recoverable ? */ if(mcgstl&(1<<0)) /* Recoverable ? */
recover=0; recover=0;
printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n", smp_processor_id(), mcgsth, mcgstl); printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n", smp_processor_id(), mcgsth, mcgstl);
for(i=0;i<banks;i++) for (i=0;i<banks;i++) {
{
rdmsr(MSR_IA32_MC0_STATUS+i*4,low, high); rdmsr(MSR_IA32_MC0_STATUS+i*4,low, high);
if(high&(1<<31)) if(high&(1<<31)) {
{
if(high&(1<<29)) if(high&(1<<29))
recover|=1; recover|=1;
if(high&(1<<25)) if(high&(1<<25))
recover|=2; recover|=2;
printk(KERN_EMERG "Bank %d: %08x%08x", i, high, low); printk(KERN_EMERG "Bank %d: %08x%08x", i, high, low);
high&=~(1<<31); high&=~(1<<31);
if(high&(1<<27)) if(high&(1<<27)) {
{
rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh); rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh);
printk("[%08x%08x]", ahigh, alow); printk("[%08x%08x]", ahigh, alow);
} }
if(high&(1<<26)) if(high&(1<<26)) {
{
rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh); rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh);
printk(" at %08x%08x", ahigh, alow); printk(" at %08x%08x", ahigh, alow);
} }
...@@ -178,6 +74,7 @@ static void intel_machine_check(struct pt_regs * regs, long error_code) ...@@ -178,6 +74,7 @@ static void intel_machine_check(struct pt_regs * regs, long error_code)
wrmsr(MSR_IA32_MCG_STATUS,mcgstl, mcgsth); wrmsr(MSR_IA32_MCG_STATUS,mcgstl, mcgsth);
} }
/* /*
* Handle unconfigured int18 (should never happen) * Handle unconfigured int18 (should never happen)
*/ */
...@@ -200,47 +97,50 @@ asmlinkage void do_machine_check(struct pt_regs * regs, long error_code) ...@@ -200,47 +97,50 @@ asmlinkage void do_machine_check(struct pt_regs * regs, long error_code)
#ifdef CONFIG_X86_MCE_NONFATAL #ifdef CONFIG_X86_MCE_NONFATAL
struct timer_list mce_timer; static struct timer_list mce_timer;
static int timerset = 0;
static void mce_checkregs (unsigned int cpu) #define MCE_RATE 15*HZ /* timer rate is 15s */
static void mce_checkregs (void *info)
{ {
u32 low, high; u32 low, high;
int i; int i;
unsigned int *cpu = info;
if (cpu!=smp_processor_id()) BUG_ON (*cpu != smp_processor_id());
BUG();
for (i=0; i<banks; i++) { for (i=0; i<banks; i++) {
rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high);
if ((low | high) != 0) { if ((low | high) != 0) {
flush_all();
printk (KERN_EMERG "MCE: The hardware reports a non fatal, correctable incident occured on CPU %d.\n", smp_processor_id()); printk (KERN_EMERG "MCE: The hardware reports a non fatal, correctable incident occured on CPU %d.\n", smp_processor_id());
printk (KERN_EMERG "Bank %d: %08x%08x\n", i, high, low); printk (KERN_EMERG "Bank %d: %08x%08x\n", i, high, low);
/* Scrub the error so we don't pick it up in 5 seconds time. */ /* Scrub the error so we don't pick it up in MCE_RATE seconds time. */
wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL);
/* Serialize */ /* Serialize */
wmb(); wmb();
} }
} }
/* Refresh the timer. */
mce_timer.expires = jiffies + 5 * HZ;
add_timer (&mce_timer);
} }
static void mce_timerfunc (unsigned long data) static void mce_timerfunc (unsigned long data)
{ {
int i; unsigned int i;
for (i=0; i<smp_num_cpus; i++) { for (i=0; i<smp_num_cpus; i++) {
if (i == smp_processor_id()) if (i == smp_processor_id())
mce_checkregs(i); mce_checkregs(&i);
else else
smp_call_function (mce_checkregs, i, 1, 1); smp_call_function (mce_checkregs, &i, 1, 1);
} }
/* Refresh the timer. */
mce_timer.expires = jiffies + MCE_RATE;
add_timer (&mce_timer);
} }
#endif #endif
...@@ -249,7 +149,7 @@ static void mce_timerfunc (unsigned long data) ...@@ -249,7 +149,7 @@ static void mce_timerfunc (unsigned long data)
* Set up machine check reporting for processors with Intel style MCE * Set up machine check reporting for processors with Intel style MCE
*/ */
static void __init intel_mcheck_init(struct cpuinfo_x86 *c) static void __init hammer_mcheck_init(struct cpuinfo_x86 *c)
{ {
u32 l, h; u32 l, h;
int i; int i;
...@@ -262,41 +162,29 @@ static void __init intel_mcheck_init(struct cpuinfo_x86 *c) ...@@ -262,41 +162,29 @@ static void __init intel_mcheck_init(struct cpuinfo_x86 *c)
if( !test_bit(X86_FEATURE_MCE, c->x86_capability) ) if( !test_bit(X86_FEATURE_MCE, c->x86_capability) )
return; return;
/* /* Check for PPro style MCA */
* Check for PPro style MCA
*/
if( !test_bit(X86_FEATURE_MCA, c->x86_capability) ) if( !test_bit(X86_FEATURE_MCA, c->x86_capability) )
return; return;
/* Ok machine check is available */ /* Ok machine check is available */
machine_check_vector = hammer_machine_check;
machine_check_vector = intel_machine_check;
wmb(); wmb();
if(done==0) if(done==0)
printk(KERN_INFO "Intel machine check architecture supported.\n"); printk(KERN_INFO "Machine check architecture supported.\n");
rdmsr(MSR_IA32_MCG_CAP, l, h); rdmsr(MSR_IA32_MCG_CAP, l, h);
if(l&(1<<8)) /* Control register present ? */ if(l&(1<<8)) /* Control register present ? */
wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff);
banks = l&0xff; banks = l&0xff;
/* Don't enable bank 0 on intel P6 cores, it goes bang quickly. */
if (c->x86_vendor == X86_VENDOR_INTEL && c->x86 == 6) {
for(i=1; i<banks; i++)
wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff);
} else {
for(i=0; i<banks; i++) for(i=0; i<banks; i++)
wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff);
}
for(i=0; i<banks; i++) for(i=0; i<banks; i++)
wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0);
set_in_cr4(X86_CR4_MCE); set_in_cr4(X86_CR4_MCE);
printk(KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n", smp_processor_id()); printk(KERN_INFO "Machine check reporting enabled on CPU#%d.\n", smp_processor_id());
intel_init_thermal(c);
done=1; done=1;
} }
...@@ -307,27 +195,27 @@ static void __init intel_mcheck_init(struct cpuinfo_x86 *c) ...@@ -307,27 +195,27 @@ static void __init intel_mcheck_init(struct cpuinfo_x86 *c)
void __init mcheck_init(struct cpuinfo_x86 *c) void __init mcheck_init(struct cpuinfo_x86 *c)
{ {
if(mce_disabled==1) if(mce_disabled==1)
return; return;
switch(c->x86_vendor) switch(c->x86_vendor)
{ {
case X86_VENDOR_AMD: case X86_VENDOR_AMD:
if(c->x86 == 6 || c->x86 == 15) { hammer_mcheck_init(c);
intel_mcheck_init(c);
#ifdef CONFIG_X86_MCE_NONFATAL #ifdef CONFIG_X86_MCE_NONFATAL
/* Set the timer to check for non-fatal errors every 5 seconds */ if (timerset == 0) {
/* Set the timer to check for non-fatal
errors every MCE_RATE seconds */
init_timer (&mce_timer); init_timer (&mce_timer);
mce_timer.expires = jiffies + 5 * HZ; mce_timer.expires = jiffies + MCE_RATE;
mce_timer.data = 0; mce_timer.data = 0;
mce_timer.function = &mce_timerfunc; mce_timer.function = &mce_timerfunc;
add_timer (&mce_timer); add_timer (&mce_timer);
#endif timerset = 1;
printk(KERN_INFO "Machine check exception polling timer started.\n");
} }
break; #endif
case X86_VENDOR_INTEL:
intel_mcheck_init(c);
break; break;
default: default:
...@@ -349,3 +237,8 @@ static int __init mcheck_enable(char *str) ...@@ -349,3 +237,8 @@ static int __init mcheck_enable(char *str)
__setup("nomce", mcheck_disable); __setup("nomce", mcheck_disable);
__setup("mce", mcheck_enable); __setup("mce", mcheck_enable);
#else
asmlinkage void do_machine_check(struct pt_regs * regs, long error_code) {}
void __init mcheck_init(struct cpuinfo_x86 *c) {}
#endif
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment