bool
select GENERIC_ALLOCATOR
+config ARCH_USES_PG_UNCACHED
+ def_bool y
+ depends on IA64_UNCACHED_ALLOCATOR
+
config AUDIT_ARCH
bool
default y
If unsure, say Y.
+config ARCH_USES_PG_UNCACHED
+ def_bool y
+ depends on X86_PAT
+
config EFI
bool "EFI runtime service support"
depends on ACPI
memcpy(dst, src, len);
}
-#define PG_non_WB PG_arch_1
-PAGEFLAG(NonWB, non_WB)
+#define PG_WC PG_arch_1
+PAGEFLAG(WC, WC)
+
+#ifdef CONFIG_X86_PAT
+/*
+ * X86 PAT uses page flags WC and Uncached together to keep track of
+ * memory type of pages that have backing page struct. X86 PAT supports 3
+ * different memory types, _PAGE_CACHE_WB, _PAGE_CACHE_WC and
+ * _PAGE_CACHE_UC_MINUS and fourth state where page's memory type has not
+ * been changed from its default (value of -1 used to denote this).
+ * Note we do not support _PAGE_CACHE_UC here.
+ *
+ * Caller must hold memtype_lock for atomicity.
+ */
+static inline unsigned long get_page_memtype(struct page *pg)
+{
+ if (!PageUncached(pg) && !PageWC(pg))
+ return -1;
+ else if (!PageUncached(pg) && PageWC(pg))
+ return _PAGE_CACHE_WC;
+ else if (PageUncached(pg) && !PageWC(pg))
+ return _PAGE_CACHE_UC_MINUS;
+ else
+ return _PAGE_CACHE_WB;
+}
+
+static inline void set_page_memtype(struct page *pg, unsigned long memtype)
+{
+ switch (memtype) {
+ case _PAGE_CACHE_WC:
+ ClearPageUncached(pg);
+ SetPageWC(pg);
+ break;
+ case _PAGE_CACHE_UC_MINUS:
+ SetPageUncached(pg);
+ ClearPageWC(pg);
+ break;
+ case _PAGE_CACHE_WB:
+ SetPageUncached(pg);
+ SetPageWC(pg);
+ break;
+ default:
+ case -1:
+ ClearPageUncached(pg);
+ ClearPageWC(pg);
+ break;
+ }
+}
+#else
+static inline unsigned long get_page_memtype(struct page *pg) { return -1; }
+static inline void set_page_memtype(struct page *pg, unsigned long memtype) { }
+#endif
/*
* The set_memory_* API can be used to change various attributes of a virtual
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
-int
-is_io_mapping_possible(resource_size_t base, unsigned long size);
-
void *
iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot);
void
iounmap_atomic(void *kvaddr, enum km_type type);
+int
+iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot);
+
+void
+iomap_free(resource_size_t base, unsigned long size);
+
#endif /* _ASM_X86_IOMAP_H */
extern void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi);
extern void mtrr_ap_init(void);
extern void mtrr_bp_init(void);
+extern void set_mtrr_aps_delayed_init(void);
+extern void mtrr_aps_init(void);
+extern void mtrr_bp_restore(void);
extern int mtrr_trim_uncached_memory(unsigned long end_pfn);
extern int amd_special_default_mtrr(void);
# else
#define mtrr_ap_init() do {} while (0)
#define mtrr_bp_init() do {} while (0)
+#define set_mtrr_aps_delayed_init() do {} while (0)
+#define mtrr_aps_init() do {} while (0)
+#define mtrr_bp_restore() do {} while (0)
# endif
#ifdef CONFIG_COMPAT
extern int kernel_map_sync_memtype(u64 base, unsigned long size,
unsigned long flag);
+int io_reserve_memtype(resource_size_t start, resource_size_t end,
+ unsigned long *type);
+
+void io_free_memtype(resource_size_t start, resource_size_t end);
+
#endif /* _ASM_X86_PAT_H */
{
int apicid, cpuid;
- if (!boot_cpu_has(X86_FEATURE_APIC))
+ if (!cpu_has_apic)
return 0;
apicid = hard_smp_processor_id();
#include <linux/bitops.h>
#include <linux/mm.h>
-#include <asm/io.h>
+#include <linux/io.h>
#include <asm/processor.h>
#include <asm/apic.h>
#include <asm/cpu.h>
#define CBAR_ENB (0x80000000)
#define CBAR_KEY (0X000000CB)
if (c->x86_model == 9 || c->x86_model == 10) {
- if (inl (CBAR) & CBAR_ENB)
- outl (0 | CBAR_KEY, CBAR);
+ if (inl(CBAR) & CBAR_ENB)
+ outl(0 | CBAR_KEY, CBAR);
}
}
d = d2-d;
if (d > 20*K6_BUG_LOOP)
- printk("system stability may be impaired when more than 32 MB are used.\n");
+ printk(KERN_CONT
+ "system stability may be impaired when more than 32 MB are used.\n");
else
- printk("probably OK (after B9730xxxx).\n");
+ printk(KERN_CONT "probably OK (after B9730xxxx).\n");
printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
}
if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
rdmsr(MSR_K7_CLK_CTL, l, h);
if ((l & 0xfff00000) != 0x20000000) {
- printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
- ((l & 0x000fffff)|0x20000000));
+ printk(KERN_INFO
+ "CPU: CLK_CTL MSR was %x. Reprogramming to %x\n",
+ l, ((l & 0x000fffff)|0x20000000));
wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
}
}
u32 level;
level = cpuid_eax(1);
- if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
+ if ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
set_cpu_cap(c, X86_FEATURE_REP_GOOD);
/*
* benefit in doing so.
*/
if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
- printk(KERN_DEBUG "tseg: %010llx\n", tseg);
- if ((tseg>>PMD_SHIFT) <
+ printk(KERN_DEBUG "tseg: %010llx\n", tseg);
+ if ((tseg>>PMD_SHIFT) <
(max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
- ((tseg>>PMD_SHIFT) <
+ ((tseg>>PMD_SHIFT) <
(max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
- (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
- set_memory_4k((unsigned long)__va(tseg), 1);
+ (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
+ set_memory_4k((unsigned long)__va(tseg), 1);
}
}
#endif
}
#ifdef CONFIG_X86_32
-static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)
+static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c,
+ unsigned int size)
{
/* AMD errata T13 (order #21922) */
if ((c->x86 == 6)) {
- if (c->x86_model == 3 && c->x86_mask == 0) /* Duron Rev A0 */
+ /* Duron Rev A0 */
+ if (c->x86_model == 3 && c->x86_mask == 0)
size = 64;
+ /* Tbird rev A1/A2 */
if (c->x86_model == 4 &&
- (c->x86_mask == 0 || c->x86_mask == 1)) /* Tbird rev A1/A2 */
+ (c->x86_mask == 0 || c->x86_mask == 1))
size = 256;
}
return size;
boot_cpu_data.fdiv_bug = fdiv_bug;
if (boot_cpu_data.fdiv_bug)
- printk("Hmm, FPU with FDIV bug.\n");
+ printk(KERN_WARNING "Hmm, FPU with FDIV bug.\n");
}
static void __init check_hlt(void)
halt();
halt();
halt();
- printk("OK.\n");
+ printk(KERN_CONT "OK.\n");
}
/*
* CPU hard. Too bad.
*/
if (res != 12345678)
- printk("Buggy.\n");
+ printk(KERN_CONT "Buggy.\n");
else
- printk("OK.\n");
+ printk(KERN_CONT "OK.\n");
#endif
}
{
identify_boot_cpu();
#ifndef CONFIG_SMP
- printk("CPU: ");
+ printk(KERN_INFO "CPU: ");
print_cpu_info(&boot_cpu_data);
#endif
check_config();
{
identify_boot_cpu();
#if !defined(CONFIG_SMP)
- printk("CPU: ");
+ printk(KERN_INFO "CPU: ");
print_cpu_info(&boot_cpu_data);
#endif
alternative_instructions();
#include <asm/hypervisor.h>
#include <asm/processor.h>
#include <asm/sections.h>
-#include <asm/topology.h>
-#include <asm/cpumask.h>
+#include <linux/topology.h>
+#include <linux/cpumask.h>
#include <asm/pgtable.h>
#include <asm/atomic.h>
#include <asm/proto.h>
#include <asm/desc.h>
#include <asm/i387.h>
#include <asm/mtrr.h>
-#include <asm/numa.h>
+#include <linux/numa.h>
#include <asm/asm.h>
#include <asm/cpu.h>
#include <asm/mce.h>
#include <asm/msr.h>
#include <asm/pat.h>
-#include <asm/smp.h>
+#include <linux/smp.h>
#ifdef CONFIG_X86_LOCAL_APIC
#include <asm/uv/uv.h>
__setup("clearcpuid=", setup_disablecpuid);
#ifdef CONFIG_X86_64
-struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
+struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table };
DEFINE_PER_CPU_FIRST(union irq_stack_union,
irq_stack_union) __aligned(PAGE_SIZE);
#include <linux/delay.h>
#include <linux/pci.h>
#include <asm/dma.h>
-#include <asm/io.h>
+#include <linux/io.h>
#include <asm/processor-cyrix.h>
#include <asm/processor-flags.h>
-#include <asm/timer.h>
+#include <linux/timer.h>
#include <asm/pci-direct.h>
#include <asm/tsc.h>
* The 5510/5520 companion chips have a funky PIT.
*/
if (vendor == PCI_VENDOR_ID_CYRIX &&
- (device == PCI_DEVICE_ID_CYRIX_5510 || device == PCI_DEVICE_ID_CYRIX_5520))
+ (device == PCI_DEVICE_ID_CYRIX_5510 ||
+ device == PCI_DEVICE_ID_CYRIX_5520))
mark_tsc_unstable("cyrix 5510/5520 detected");
}
#endif
* ? : 0x7x
* GX1 : 0x8x GX1 datasheet 56
*/
- if ((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <= dir1 && dir1 <= 0x8f))
+ if ((0x30 <= dir1 && dir1 <= 0x6f) ||
+ (0x80 <= dir1 && dir1 <= 0x8f))
geode_configure();
return;
} else { /* MediaGX */
printk(KERN_INFO "Enabling CPUID on Cyrix processor.\n");
local_irq_save(flags);
ccr3 = getCx86(CX86_CCR3);
- setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */
- setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80); /* enable cpuid */
- setCx86(CX86_CCR3, ccr3); /* disable MAPEN */
+ /* enable MAPEN */
+ setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);
+ /* enable cpuid */
+ setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80);
+ /* disable MAPEN */
+ setCx86(CX86_CCR3, ccr3);
local_irq_restore(flags);
}
}
static inline void __cpuinit
detect_hypervisor_vendor(struct cpuinfo_x86 *c)
{
- if (vmware_platform()) {
+ if (vmware_platform())
c->x86_hyper_vendor = X86_HYPER_VENDOR_VMWARE;
- } else {
+ else
c->x86_hyper_vendor = X86_HYPER_VENDOR_NONE;
- }
}
unsigned long get_hypervisor_tsc_freq(void)
#include <linux/sched.h>
#include <linux/thread_info.h>
#include <linux/module.h>
+#include <linux/uaccess.h>
#include <asm/processor.h>
#include <asm/pgtable.h>
#include <asm/msr.h>
-#include <asm/uaccess.h>
#include <asm/ds.h>
#include <asm/bugs.h>
#include <asm/cpu.h>
#ifdef CONFIG_X86_64
-#include <asm/topology.h>
+#include <linux/topology.h>
#include <asm/numa_64.h>
#endif
#ifdef CONFIG_X86_F00F_BUG
/*
* All current models of Pentium and Pentium with MMX technology CPUs
- * have the F0 0F bug, which lets nonprivileged users lock up the system.
+ * have the F0 0F bug, which lets nonprivileged users lock up the
+ * system.
* Note that the workaround only should be initialized once...
*/
c->f00f_bug = 0;
printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n");
printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n");
lo |= MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE;
- wrmsr (MSR_IA32_MISC_ENABLE, lo, hi);
+ wrmsr(MSR_IA32_MISC_ENABLE, lo, hi);
}
}
/* Intel has a non-standard dependency on %ecx for this CPUID level. */
cpuid_count(4, 0, &eax, &ebx, &ecx, &edx);
if (eax & 0x1f)
- return ((eax >> 26) + 1);
+ return (eax >> 26) + 1;
else
return 1;
}
*
* Changes:
* Venkatesh Pallipadi : Adding cache identification through cpuid(4)
- *
Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
+ * Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
* Andi Kleen / Andreas Herrmann : CPUID4 emulation on AMD.
*/
#include <linux/pci.h>
#include <asm/processor.h>
-#include <asm/smp.h>
+#include <linux/smp.h>
#include <asm/k8.h>
#define LVL_1_INST 1
#define LVL_3 4
#define LVL_TRACE 5
-struct _cache_table
-{
+struct _cache_table {
unsigned char descriptor;
char cache_type;
short size;
};
-/* all the cache descriptor types we care about (no TLB or trace cache entries) */
+/* All the cache descriptor types we care about (no TLB or
+ trace cache entries) */
+
static const struct _cache_table __cpuinitconst cache_table[] =
{
{ 0x06, LVL_1_INST, 8 }, /* 4-way set assoc, 32 byte line size */
};
-enum _cache_type
-{
+enum _cache_type {
CACHE_TYPE_NULL = 0,
CACHE_TYPE_DATA = 1,
CACHE_TYPE_INST = 2,
Maybe later */
union l1_cache {
struct {
- unsigned line_size : 8;
- unsigned lines_per_tag : 8;
- unsigned assoc : 8;
- unsigned size_in_kb : 8;
+ unsigned line_size:8;
+ unsigned lines_per_tag:8;
+ unsigned assoc:8;
+ unsigned size_in_kb:8;
};
unsigned val;
};
union l2_cache {
struct {
- unsigned line_size : 8;
- unsigned lines_per_tag : 4;
- unsigned assoc : 4;
- unsigned size_in_kb : 16;
+ unsigned line_size:8;
+ unsigned lines_per_tag:4;
+ unsigned assoc:4;
+ unsigned size_in_kb:16;
};
unsigned val;
};
union l3_cache {
struct {
- unsigned line_size : 8;
- unsigned lines_per_tag : 4;
- unsigned assoc : 4;
- unsigned res : 2;
- unsigned size_encoded : 14;
+ unsigned line_size:8;
+ unsigned lines_per_tag:4;
+ unsigned assoc:4;
+ unsigned res:2;
+ unsigned size_encoded:14;
};
unsigned val;
};
unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
{
- unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; /* Cache sizes */
+ /* Cache sizes */
+ unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0;
unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */
unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */
unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb;
retval = cpuid4_cache_lookup_regs(i, &this_leaf);
if (retval >= 0) {
- switch(this_leaf.eax.split.level) {
- case 1:
+ switch (this_leaf.eax.split.level) {
+ case 1:
if (this_leaf.eax.split.type ==
CACHE_TYPE_DATA)
new_l1d = this_leaf.size/1024;
CACHE_TYPE_INST)
new_l1i = this_leaf.size/1024;
break;
- case 2:
+ case 2:
new_l2 = this_leaf.size/1024;
num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing;
index_msb = get_count_order(num_threads_sharing);
l2_id = c->apicid >> index_msb;
break;
- case 3:
+ case 3:
new_l3 = this_leaf.size/1024;
num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing;
- index_msb = get_count_order(num_threads_sharing);
+ index_msb = get_count_order(
+ num_threads_sharing);
l3_id = c->apicid >> index_msb;
break;
- default:
+ default:
break;
}
}
/* Number of times to iterate */
n = cpuid_eax(2) & 0xFF;
- for ( i = 0 ; i < n ; i++ ) {
+ for (i = 0 ; i < n ; i++) {
cpuid(2, ®s[0], ®s[1], ®s[2], ®s[3]);
/* If bit 31 is set, this is an unknown format */
- for ( j = 0 ; j < 3 ; j++ ) {
- if (regs[j] & (1 << 31)) regs[j] = 0;
- }
+ for (j = 0 ; j < 3 ; j++)
+ if (regs[j] & (1 << 31))
+ regs[j] = 0;
/* Byte 0 is level count, not a descriptor */
- for ( j = 1 ; j < 16 ; j++ ) {
+ for (j = 1 ; j < 16 ; j++) {
unsigned char des = dp[j];
unsigned char k = 0;
/* look up this descriptor in the table */
- while (cache_table[k].descriptor != 0)
- {
+ while (cache_table[k].descriptor != 0) {
if (cache_table[k].descriptor == des) {
if (only_trace && cache_table[k].cache_type != LVL_TRACE)
break;
}
if (trace)
- printk (KERN_INFO "CPU: Trace cache: %dK uops", trace);
- else if ( l1i )
- printk (KERN_INFO "CPU: L1 I cache: %dK", l1i);
+ printk(KERN_INFO "CPU: Trace cache: %dK uops", trace);
+ else if (l1i)
+ printk(KERN_INFO "CPU: L1 I cache: %dK", l1i);
if (l1d)
- printk(", L1 D cache: %dK\n", l1d);
+ printk(KERN_CONT ", L1 D cache: %dK\n", l1d);
else
- printk("\n");
+ printk(KERN_CONT "\n");
if (l2)
printk(KERN_INFO "CPU: L2 cache: %dK\n", l2);
}
}
#else
-static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) {}
-static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) {}
+static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
+{
+}
+
+static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
+{
+}
#endif
static void __cpuinit free_cache_attributes(unsigned int cpu)
static ssize_t show_##file_name \
(struct _cpuid4_info *this_leaf, char *buf) \
{ \
- return sprintf (buf, "%lu\n", (unsigned long)this_leaf->object + val); \
+ return sprintf(buf, "%lu\n", (unsigned long)this_leaf->object + val); \
}
show_one_plus(level, eax.split.level, 0);
static ssize_t show_size(struct _cpuid4_info *this_leaf, char *buf)
{
- return sprintf (buf, "%luK\n", this_leaf->size / 1024);
+ return sprintf(buf, "%luK\n", this_leaf->size / 1024);
}
static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
const struct cpumask *mask;
mask = to_cpumask(this_leaf->shared_cpu_map);
- n = type?
+ n = type ?
cpulist_scnprintf(buf, len-2, mask) :
cpumask_scnprintf(buf, len-2, mask);
buf[n++] = '\n';
static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644,
show_cache_disable_1, store_cache_disable_1);
-static struct attribute * default_attrs[] = {
+static struct attribute *default_attrs[] = {
&type.attr,
&level.attr,
&coherency_line_size.attr,
NULL
};
-static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf)
+static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf)
{
struct _cache_attr *fattr = to_attr(attr);
struct _index_kobject *this_leaf = to_object(kobj);
return ret;
}
-static ssize_t store(struct kobject * kobj, struct attribute * attr,
- const char * buf, size_t count)
+static ssize_t store(struct kobject *kobj, struct attribute *attr,
+ const char *buf, size_t count)
{
struct _cache_attr *fattr = to_attr(attr);
struct _index_kobject *this_leaf = to_object(kobj);
goto err_out;
per_cpu(index_kobject, cpu) = kzalloc(
- sizeof(struct _index_kobject ) * num_cache_leaves, GFP_KERNEL);
+ sizeof(struct _index_kobject) * num_cache_leaves, GFP_KERNEL);
if (unlikely(per_cpu(index_kobject, cpu) == NULL))
goto err_out;
}
for (i = 0; i < num_cache_leaves; i++) {
- this_object = INDEX_KOBJECT_PTR(cpu,i);
+ this_object = INDEX_KOBJECT_PTR(cpu, i);
this_object->cpu = cpu;
this_object->index = i;
retval = kobject_init_and_add(&(this_object->kobj),
per_cpu(cache_kobject, cpu),
"index%1lu", i);
if (unlikely(retval)) {
- for (j = 0; j < i; j++) {
- kobject_put(&(INDEX_KOBJECT_PTR(cpu,j)->kobj));
- }
+ for (j = 0; j < i; j++)
+ kobject_put(&(INDEX_KOBJECT_PTR(cpu, j)->kobj));
kobject_put(per_cpu(cache_kobject, cpu));
cpuid4_cache_sysfs_exit(cpu);
return retval;
cpumask_clear_cpu(cpu, to_cpumask(cache_dev_map));
for (i = 0; i < num_cache_leaves; i++)
- kobject_put(&(INDEX_KOBJECT_PTR(cpu,i)->kobj));
+ kobject_put(&(INDEX_KOBJECT_PTR(cpu, i)->kobj));
kobject_put(per_cpu(cache_kobject, cpu));
cpuid4_cache_sysfs_exit(cpu);
}
return NOTIFY_OK;
}
-static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier =
-{
+static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier = {
.notifier_call = cacheinfo_cpu_callback,
};
static void
amd_get_mtrr(unsigned int reg, unsigned long *base,
- unsigned long *size, mtrr_type * type)
+ unsigned long *size, mtrr_type *type)
{
unsigned long low, high;
rdmsr(MSR_K6_UWCCR, low, high);
- /* Upper dword is region 1, lower is region 0 */
+ /* Upper dword is region 1, lower is region 0 */
if (reg == 1)
low = high;
- /* The base masks off on the right alignment */
+ /* The base masks off on the right alignment */
*base = (low & 0xFFFE0000) >> PAGE_SHIFT;
*type = 0;
if (low & 1)
return;
}
/*
- * This needs a little explaining. The size is stored as an
- * inverted mask of bits of 128K granularity 15 bits long offset
- * 2 bits
+ * This needs a little explaining. The size is stored as an
+ * inverted mask of bits of 128K granularity 15 bits long offset
+ * 2 bits.
*
- * So to get a size we do invert the mask and add 1 to the lowest
- * mask bit (4 as its 2 bits in). This gives us a size we then shift
- * to turn into 128K blocks
+ * So to get a size we do invert the mask and add 1 to the lowest
+ * mask bit (4 as its 2 bits in). This gives us a size we then shift
+ * to turn into 128K blocks.
*
- * eg 111 1111 1111 1100 is 512K
+ * eg 111 1111 1111 1100 is 512K
*
- * invert 000 0000 0000 0011
- * +1 000 0000 0000 0100
- * *128K ...
+ * invert 000 0000 0000 0011
+ * +1 000 0000 0000 0100
+ * *128K ...
*/
low = (~low) & 0x1FFFC;
*size = (low + 4) << (15 - PAGE_SHIFT);
- return;
}
-static void amd_set_mtrr(unsigned int reg, unsigned long base,
- unsigned long size, mtrr_type type)
-/* [SUMMARY] Set variable MTRR register on the local CPU.
- <reg> The register to set.
- <base> The base address of the region.
- <size> The size of the region. If this is 0 the region is disabled.
- <type> The type of the region.
- [RETURNS] Nothing.
-*/
+/**
+ * amd_set_mtrr - Set variable MTRR register on the local CPU.
+ *
+ * @reg The register to set.
+ * @base The base address of the region.
+ * @size The size of the region. If this is 0 the region is disabled.
+ * @type The type of the region.
+ *
+ * Returns nothing.
+ */
+static void
+amd_set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type)
{
u32 regs[2];
/*
- * Low is MTRR0 , High MTRR 1
+ * Low is MTRR0, High MTRR 1
*/
rdmsr(MSR_K6_UWCCR, regs[0], regs[1]);
/*
- * Blank to disable
+ * Blank to disable
*/
- if (size == 0)
+ if (size == 0) {
regs[reg] = 0;
- else
- /* Set the register to the base, the type (off by one) and an
- inverted bitmask of the size The size is the only odd
- bit. We are fed say 512K We invert this and we get 111 1111
- 1111 1011 but if you subtract one and invert you get the
- desired 111 1111 1111 1100 mask
-
- But ~(x - 1) == ~x + 1 == -x. Two's complement rocks! */
+ } else {
+ /*
+ * Set the register to the base, the type (off by one) and an
+ * inverted bitmask of the size The size is the only odd
+ * bit. We are fed say 512K We invert this and we get 111 1111
+ * 1111 1011 but if you subtract one and invert you get the
+ * desired 111 1111 1111 1100 mask
+ *
+ * But ~(x - 1) == ~x + 1 == -x. Two's complement rocks!
+ */
regs[reg] = (-size >> (15 - PAGE_SHIFT) & 0x0001FFFC)
| (base << PAGE_SHIFT) | (type + 1);
+ }
/*
- * The writeback rule is quite specific. See the manual. Its
- * disable local interrupts, write back the cache, set the mtrr
+ * The writeback rule is quite specific. See the manual. Its
+ * disable local interrupts, write back the cache, set the mtrr
*/
wbinvd();
wrmsr(MSR_K6_UWCCR, regs[0], regs[1]);
}
-static int amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
+static int
+amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
{
- /* Apply the K6 block alignment and size rules
- In order
- o Uncached or gathering only
- o 128K or bigger block
- o Power of 2 block
- o base suitably aligned to the power
- */
+ /*
+ * Apply the K6 block alignment and size rules
+ * In order
+ * o Uncached or gathering only
+ * o 128K or bigger block
+ * o Power of 2 block
+ * o base suitably aligned to the power
+ */
if (type > MTRR_TYPE_WRCOMB || size < (1 << (17 - PAGE_SHIFT))
|| (size & ~(size - 1)) - size || (base & (size - 1)))
return -EINVAL;
set_mtrr_ops(&amd_mtrr_ops);
return 0;
}
-
-//arch_initcall(amd_mtrr_init);
#include <linux/init.h>
#include <linux/mm.h>
+
#include <asm/mtrr.h>
#include <asm/msr.h>
+
#include "mtrr.h"
static struct {
static u8 centaur_mcr_reserved;
static u8 centaur_mcr_type; /* 0 for winchip, 1 for winchip2 */
-/*
- * Report boot time MCR setups
+/**
+ * centaur_get_free_region - Get a free MTRR.
+ *
+ * @base: The starting (base) address of the region.
+ * @size: The size (in bytes) of the region.
+ *
+ * Returns: the index of the region on success, else -1 on error.
*/
-
static int
centaur_get_free_region(unsigned long base, unsigned long size, int replace_reg)
-/* [SUMMARY] Get a free MTRR.
- <base> The starting (base) address of the region.
- <size> The size (in bytes) of the region.
- [RETURNS] The index of the region on success, else -1 on error.
-*/
{
- int i, max;
- mtrr_type ltype;
unsigned long lbase, lsize;
+ mtrr_type ltype;
+ int i, max;
max = num_var_ranges;
if (replace_reg >= 0 && replace_reg < max)
return replace_reg;
+
for (i = 0; i < max; ++i) {
if (centaur_mcr_reserved & (1 << i))
continue;
if (lsize == 0)
return i;
}
+
return -ENOSPC;
}
-void
-mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
+/*
+ * Report boot time MCR setups
+ */
+void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
{
centaur_mcr[mcr].low = lo;
centaur_mcr[mcr].high = hi;
{
*base = centaur_mcr[reg].high >> PAGE_SHIFT;
*size = -(centaur_mcr[reg].low & 0xfffff000) >> PAGE_SHIFT;
- *type = MTRR_TYPE_WRCOMB; /* If it is there, it is write-combining */
+ *type = MTRR_TYPE_WRCOMB; /* write-combining */
+
if (centaur_mcr_type == 1 && ((centaur_mcr[reg].low & 31) & 2))
*type = MTRR_TYPE_UNCACHABLE;
if (centaur_mcr_type == 1 && (centaur_mcr[reg].low & 31) == 25)
*type = MTRR_TYPE_WRBACK;
if (centaur_mcr_type == 0 && (centaur_mcr[reg].low & 31) == 31)
*type = MTRR_TYPE_WRBACK;
-
}
-static void centaur_set_mcr(unsigned int reg, unsigned long base,
- unsigned long size, mtrr_type type)
+static void
+centaur_set_mcr(unsigned int reg, unsigned long base,
+ unsigned long size, mtrr_type type)
{
unsigned long low, high;
if (size == 0) {
- /* Disable */
+ /* Disable */
high = low = 0;
} else {
high = base << PAGE_SHIFT;
- if (centaur_mcr_type == 0)
- low = -size << PAGE_SHIFT | 0x1f; /* only support write-combining... */
- else {
+ if (centaur_mcr_type == 0) {
+ /* Only support write-combining... */
+ low = -size << PAGE_SHIFT | 0x1f;
+ } else {
if (type == MTRR_TYPE_UNCACHABLE)
- low = -size << PAGE_SHIFT | 0x02; /* NC */
+ low = -size << PAGE_SHIFT | 0x02; /* NC */
else
- low = -size << PAGE_SHIFT | 0x09; /* WWO,WC */
+ low = -size << PAGE_SHIFT | 0x09; /* WWO, WC */
}
}
centaur_mcr[reg].high = high;
wrmsr(MSR_IDT_MCR0 + reg, low, high);
}
-#if 0
-/*
- * Initialise the later (saner) Winchip MCR variant. In this version
- * the BIOS can pass us the registers it has used (but not their values)
- * and the control register is read/write
- */
-
-static void __init
-centaur_mcr1_init(void)
-{
- unsigned i;
- u32 lo, hi;
-
- /* Unfortunately, MCR's are read-only, so there is no way to
- * find out what the bios might have done.
- */
-
- rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
- if (((lo >> 17) & 7) == 1) { /* Type 1 Winchip2 MCR */
- lo &= ~0x1C0; /* clear key */
- lo |= 0x040; /* set key to 1 */
- wrmsr(MSR_IDT_MCR_CTRL, lo, hi); /* unlock MCR */
- }
-
- centaur_mcr_type = 1;
-
- /*
- * Clear any unconfigured MCR's.
- */
-
- for (i = 0; i < 8; ++i) {
- if (centaur_mcr[i].high == 0 && centaur_mcr[i].low == 0) {
- if (!(lo & (1 << (9 + i))))
- wrmsr(MSR_IDT_MCR0 + i, 0, 0);
- else
- /*
- * If the BIOS set up an MCR we cannot see it
- * but we don't wish to obliterate it
- */
- centaur_mcr_reserved |= (1 << i);
- }
- }
- /*
- * Throw the main write-combining switch...
- * However if OOSTORE is enabled then people have already done far
- * cleverer things and we should behave.
- */
-
- lo |= 15; /* Write combine enables */
- wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
-}
-
-/*
- * Initialise the original winchip with read only MCR registers
- * no used bitmask for the BIOS to pass on and write only control
- */
-
-static void __init
-centaur_mcr0_init(void)
-{
- unsigned i;
-
- /* Unfortunately, MCR's are read-only, so there is no way to
- * find out what the bios might have done.
- */
-
- /* Clear any unconfigured MCR's.
- * This way we are sure that the centaur_mcr array contains the actual
- * values. The disadvantage is that any BIOS tweaks are thus undone.
- *
- */
- for (i = 0; i < 8; ++i) {
- if (centaur_mcr[i].high == 0 && centaur_mcr[i].low == 0)
- wrmsr(MSR_IDT_MCR0 + i, 0, 0);
- }
-
- wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0); /* Write only */
-}
-
-/*
- * Initialise Winchip series MCR registers
- */
-
-static void __init
-centaur_mcr_init(void)
-{
- struct set_mtrr_context ctxt;
-
- set_mtrr_prepare_save(&ctxt);
- set_mtrr_cache_disable(&ctxt);
-
- if (boot_cpu_data.x86_model == 4)
- centaur_mcr0_init();
- else if (boot_cpu_data.x86_model == 8 || boot_cpu_data.x86_model == 9)
- centaur_mcr1_init();
-
- set_mtrr_done(&ctxt);
-}
-#endif
-
-static int centaur_validate_add_page(unsigned long base,
- unsigned long size, unsigned int type)
+static int
+centaur_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
{
/*
- * FIXME: Winchip2 supports uncached
+ * FIXME: Winchip2 supports uncached
*/
- if (type != MTRR_TYPE_WRCOMB &&
+ if (type != MTRR_TYPE_WRCOMB &&
(centaur_mcr_type == 0 || type != MTRR_TYPE_UNCACHABLE)) {
- printk(KERN_WARNING
- "mtrr: only write-combining%s supported\n",
- centaur_mcr_type ? " and uncacheable are"
- : " is");
+ pr_warning("mtrr: only write-combining%s supported\n",
+ centaur_mcr_type ? " and uncacheable are" : " is");
return -EINVAL;
}
return 0;
static struct mtrr_ops centaur_mtrr_ops = {
.vendor = X86_VENDOR_CENTAUR,
-// .init = centaur_mcr_init,
.set = centaur_set_mcr,
.get = centaur_get_mcr,
.get_free_region = centaur_get_free_region,
set_mtrr_ops(¢aur_mtrr_ops);
return 0;
}
-
-//arch_initcall(centaur_init_mtrr);
-/* MTRR (Memory Type Range Register) cleanup
-
- Copyright (C) 2009 Yinghai Lu
-
- This library is free software; you can redistribute it and/or
- modify it under the terms of the GNU Library General Public
- License as published by the Free Software Foundation; either
- version 2 of the License, or (at your option) any later version.
-
- This library is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- Library General Public License for more details.
-
- You should have received a copy of the GNU Library General Public
- License along with this library; if not, write to the Free
- Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-*/
-
+/*
+ * MTRR (Memory Type Range Register) cleanup
+ *
+ * Copyright (C) 2009 Yinghai Lu
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Library General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Library General Public License for more details.
+ *
+ * You should have received a copy of the GNU Library General Public
+ * License along with this library; if not, write to the Free
+ * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
#include <linux/module.h>
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/smp.h>
#include <linux/cpu.h>
-#include <linux/mutex.h>
#include <linux/sort.h>
+#include <linux/mutex.h>
+#include <linux/uaccess.h>
+#include <linux/kvm_para.h>
+#include <asm/processor.h>
#include <asm/e820.h>
#include <asm/mtrr.h>
-#include <asm/uaccess.h>
-#include <asm/processor.h>
#include <asm/msr.h>
-#include <asm/kvm_para.h>
-#include "mtrr.h"
-/* should be related to MTRR_VAR_RANGES nums */
-#define RANGE_NUM 256
+#include "mtrr.h"
struct res_range {
- unsigned long start;
- unsigned long end;
+ unsigned long start;
+ unsigned long end;
+};
+
+struct var_mtrr_range_state {
+ unsigned long base_pfn;
+ unsigned long size_pfn;
+ mtrr_type type;
+};
+
+struct var_mtrr_state {
+ unsigned long range_startk;
+ unsigned long range_sizek;
+ unsigned long chunk_sizek;
+ unsigned long gran_sizek;
+ unsigned int reg;
};
+/* Should be related to MTRR_VAR_RANGES nums */
+#define RANGE_NUM 256
+
+static struct res_range __initdata range[RANGE_NUM];
+static int __initdata nr_range;
+
+static struct var_mtrr_range_state __initdata range_state[RANGE_NUM];
+
+static int __initdata debug_print;
+#define Dprintk(x...) do { if (debug_print) printk(KERN_DEBUG x); } while (0)
+
+
static int __init
-add_range(struct res_range *range, int nr_range, unsigned long start,
- unsigned long end)
+add_range(struct res_range *range, int nr_range,
+ unsigned long start, unsigned long end)
{
- /* out of slots */
+ /* Out of slots: */
if (nr_range >= RANGE_NUM)
return nr_range;
}
static int __init
-add_range_with_merge(struct res_range *range, int nr_range, unsigned long start,
- unsigned long end)
+add_range_with_merge(struct res_range *range, int nr_range,
+ unsigned long start, unsigned long end)
{
int i;
- /* try to merge it with old one */
+ /* Try to merge it with old one: */
for (i = 0; i < nr_range; i++) {
unsigned long final_start, final_end;
unsigned long common_start, common_end;
return nr_range;
}
- /* need to add that */
+ /* Need to add it: */
return add_range(range, nr_range, start, end);
}
}
if (start > range[j].start && end < range[j].end) {
- /* find the new spare */
+ /* Find the new spare: */
for (i = 0; i < RANGE_NUM; i++) {
if (range[i].end == 0)
break;
return start1 - start2;
}
-struct var_mtrr_range_state {
- unsigned long base_pfn;
- unsigned long size_pfn;
- mtrr_type type;
-};
-
-static struct var_mtrr_range_state __initdata range_state[RANGE_NUM];
-static int __initdata debug_print;
+#define BIOS_BUG_MSG KERN_WARNING \
+ "WARNING: BIOS bug: VAR MTRR %d contains strange UC entry under 1M, check with your system vendor!\n"
static int __init
x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
range[i].start, range[i].end + 1);
}
- /* take out UC ranges */
+ /* Take out UC ranges: */
for (i = 0; i < num_var_ranges; i++) {
type = range_state[i].type;
if (type != MTRR_TYPE_UNCACHABLE &&
if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed &&
(mtrr_state.enabled & 1)) {
/* Var MTRR contains UC entry below 1M? Skip it: */
- printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d "
- "contains strange UC entry under 1M, check "
- "with your system vendor!\n", i);
+ printk(BIOS_BUG_MSG, i);
if (base + size <= (1<<(20-PAGE_SHIFT)))
continue;
size -= (1<<(20-PAGE_SHIFT)) - base;
return nr_range;
}
-static struct res_range __initdata range[RANGE_NUM];
-static int __initdata nr_range;
-
#ifdef CONFIG_MTRR_SANITIZER
static unsigned long __init sum_ranges(struct res_range *range, int nr_range)
{
- unsigned long sum;
+ unsigned long sum = 0;
int i;
- sum = 0;
for (i = 0; i < nr_range; i++)
sum += range[i].end + 1 - range[i].start;
}
early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup);
-struct var_mtrr_state {
- unsigned long range_startk;
- unsigned long range_sizek;
- unsigned long chunk_sizek;
- unsigned long gran_sizek;
- unsigned int reg;
-};
-
static void __init
set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
- unsigned char type, unsigned int address_bits)
+ unsigned char type, unsigned int address_bits)
{
u32 base_lo, base_hi, mask_lo, mask_hi;
u64 base, mask;
mask = (1ULL << address_bits) - 1;
mask &= ~((((u64)sizek) << 10) - 1);
- base = ((u64)basek) << 10;
+ base = ((u64)basek) << 10;
base |= type;
mask |= 0x800;
static void __init
save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
- unsigned char type)
+ unsigned char type)
{
range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10);
range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10);
range_state[reg].type = type;
}
-static void __init
-set_var_mtrr_all(unsigned int address_bits)
+static void __init set_var_mtrr_all(unsigned int address_bits)
{
unsigned long basek, sizek;
unsigned char type;
static unsigned long to_size_factor(unsigned long sizek, char *factorp)
{
- char factor;
unsigned long base = sizek;
+ char factor;
if (base & ((1<<10) - 1)) {
- /* not MB alignment */
+ /* Not MB-aligned: */
factor = 'K';
} else if (base & ((1<<20) - 1)) {
factor = 'M';
unsigned long max_align, align;
unsigned long sizek;
- /* Compute the maximum size I can make a range */
+ /* Compute the maximum size with which we can make a range: */
if (range_startk)
max_align = ffs(range_startk) - 1;
else
max_align = 32;
+
align = fls(range_sizek) - 1;
if (align > max_align)
align = max_align;
char start_factor = 'K', size_factor = 'K';
unsigned long start_base, size_base;
- start_base = to_size_factor(range_startk,
- &start_factor),
- size_base = to_size_factor(sizek, &size_factor),
+ start_base = to_size_factor(range_startk, &start_factor);
+ size_base = to_size_factor(sizek, &size_factor);
- printk(KERN_DEBUG "Setting variable MTRR %d, "
+ Dprintk("Setting variable MTRR %d, "
"base: %ld%cB, range: %ld%cB, type %s\n",
reg, start_base, start_factor,
size_base, size_factor,
chunk_sizek = state->chunk_sizek;
gran_sizek = state->gran_sizek;
- /* align with gran size, prevent small block used up MTRRs */
+ /* Align with gran size, prevent small block used up MTRRs: */
range_basek = ALIGN(state->range_startk, gran_sizek);
if ((range_basek > basek) && basek)
return second_sizek;
+
state->range_sizek -= (range_basek - state->range_startk);
range_sizek = ALIGN(state->range_sizek, gran_sizek);
}
state->range_sizek = range_sizek;
- /* try to append some small hole */
+ /* Try to append some small hole: */
range0_basek = state->range_startk;
range0_sizek = ALIGN(state->range_sizek, chunk_sizek);
- /* no increase */
+ /* No increase: */
if (range0_sizek == state->range_sizek) {
- if (debug_print)
- printk(KERN_DEBUG "rangeX: %016lx - %016lx\n",
- range0_basek<<10,
- (range0_basek + state->range_sizek)<<10);
+ Dprintk("rangeX: %016lx - %016lx\n",
+ range0_basek<<10,
+ (range0_basek + state->range_sizek)<<10);
state->reg = range_to_mtrr(state->reg, range0_basek,
state->range_sizek, MTRR_TYPE_WRBACK);
return 0;
}
- /* only cut back, when it is not the last */
+ /* Only cut back when it is not the last: */
if (sizek) {
while (range0_basek + range0_sizek > (basek + sizek)) {
if (range0_sizek >= chunk_sizek)
second_try:
range_basek = range0_basek + range0_sizek;
- /* one hole in the middle */
+ /* One hole in the middle: */
if (range_basek > basek && range_basek <= (basek + sizek))
second_sizek = range_basek - basek;
if (range0_sizek > state->range_sizek) {
- /* one hole in middle or at end */
+ /* One hole in middle or at the end: */
hole_sizek = range0_sizek - state->range_sizek - second_sizek;
- /* hole size should be less than half of range0 size */
+ /* Hole size should be less than half of range0 size: */
if (hole_sizek >= (range0_sizek >> 1) &&
range0_sizek >= chunk_sizek) {
range0_sizek -= chunk_sizek;
}
if (range0_sizek) {
- if (debug_print)
- printk(KERN_DEBUG "range0: %016lx - %016lx\n",
- range0_basek<<10,
- (range0_basek + range0_sizek)<<10);
+ Dprintk("range0: %016lx - %016lx\n",
+ range0_basek<<10,
+ (range0_basek + range0_sizek)<<10);
state->reg = range_to_mtrr(state->reg, range0_basek,
range0_sizek, MTRR_TYPE_WRBACK);
}
if (range0_sizek < state->range_sizek) {
- /* need to handle left over */
+ /* Need to handle left over range: */
range_sizek = state->range_sizek - range0_sizek;
- if (debug_print)
- printk(KERN_DEBUG "range: %016lx - %016lx\n",
- range_basek<<10,
- (range_basek + range_sizek)<<10);
+ Dprintk("range: %016lx - %016lx\n",
+ range_basek<<10,
+ (range_basek + range_sizek)<<10);
+
state->reg = range_to_mtrr(state->reg, range_basek,
range_sizek, MTRR_TYPE_WRBACK);
}
if (hole_sizek) {
hole_basek = range_basek - hole_sizek - second_sizek;
- if (debug_print)
- printk(KERN_DEBUG "hole: %016lx - %016lx\n",
- hole_basek<<10,
- (hole_basek + hole_sizek)<<10);
+ Dprintk("hole: %016lx - %016lx\n",
+ hole_basek<<10,
+ (hole_basek + hole_sizek)<<10);
state->reg = range_to_mtrr(state->reg, hole_basek,
hole_sizek, MTRR_TYPE_UNCACHABLE);
}
basek = base_pfn << (PAGE_SHIFT - 10);
sizek = size_pfn << (PAGE_SHIFT - 10);
- /* See if I can merge with the last range */
+ /* See if I can merge with the last range: */
if ((basek <= 1024) ||
(state->range_startk + state->range_sizek == basek)) {
unsigned long endk = basek + sizek;
state->range_sizek = endk - state->range_startk;
return;
}
- /* Write the range mtrrs */
+ /* Write the range mtrrs: */
if (state->range_sizek != 0)
second_sizek = range_to_mtrr_with_hole(state, basek, sizek);
- /* Allocate an msr */
+ /* Allocate an msr: */
state->range_startk = basek + second_sizek;
state->range_sizek = sizek - second_sizek;
}
-/* mininum size of mtrr block that can take hole */
+/* Mininum size of mtrr block that can take hole: */
static u64 mtrr_chunk_size __initdata = (256ULL<<20);
static int __init parse_mtrr_chunk_size_opt(char *p)
}
early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt);
-/* granity of mtrr of block */
+/* Granularity of mtrr of block: */
static u64 mtrr_gran_size __initdata;
static int __init parse_mtrr_gran_size_opt(char *p)
}
early_param("mtrr_gran_size", parse_mtrr_gran_size_opt);
-static int nr_mtrr_spare_reg __initdata =
+static unsigned long nr_mtrr_spare_reg __initdata =
CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT;
static int __init parse_mtrr_spare_reg(char *arg)
nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0);
return 0;
}
-
early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg);
static int __init
u64 chunk_size, u64 gran_size)
{
struct var_mtrr_state var_state;
- int i;
int num_reg;
+ int i;
var_state.range_startk = 0;
var_state.range_sizek = 0;
memset(range_state, 0, sizeof(range_state));
- /* Write the range etc */
- for (i = 0; i < nr_range; i++)
+ /* Write the range: */
+ for (i = 0; i < nr_range; i++) {
set_var_mtrr_range(&var_state, range[i].start,
range[i].end - range[i].start + 1);
+ }
- /* Write the last range */
+ /* Write the last range: */
if (var_state.range_sizek != 0)
range_to_mtrr_with_hole(&var_state, 0, 0);
num_reg = var_state.reg;
- /* Clear out the extra MTRR's */
+ /* Clear out the extra MTRR's: */
while (var_state.reg < num_var_ranges) {
save_var_mtrr(var_state.reg, 0, 0, 0);
var_state.reg++;
}
struct mtrr_cleanup_result {
- unsigned long gran_sizek;
- unsigned long chunk_sizek;
- unsigned long lose_cover_sizek;
- unsigned int num_reg;
- int bad;
+ unsigned long gran_sizek;
+ unsigned long chunk_sizek;
+ unsigned long lose_cover_sizek;
+ unsigned int num_reg;
+ int bad;
};
/*
static void __init print_out_mtrr_range_state(void)
{
- int i;
char start_factor = 'K', size_factor = 'K';
unsigned long start_base, size_base;
mtrr_type type;
+ int i;
for (i = 0; i < num_var_ranges; i++) {
int i;
mtrr_type type;
unsigned long size;
- /* extra one for all 0 */
+ /* Extra one for all 0: */
int num[MTRR_NUM_TYPES + 1];
- /* check entries number */
+ /* Check entries number: */
memset(num, 0, sizeof(num));
for (i = 0; i < num_var_ranges; i++) {
type = range_state[i].type;
num[type]++;
}
- /* check if we got UC entries */
+ /* Check if we got UC entries: */
if (!num[MTRR_TYPE_UNCACHABLE])
return 0;
- /* check if we only had WB and UC */
+ /* Check if we only had WB and UC */
if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] !=
- num_var_ranges - num[MTRR_NUM_TYPES])
+ num_var_ranges - num[MTRR_NUM_TYPES])
return 0;
return 1;
}
static unsigned long __initdata range_sums;
-static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size,
- unsigned long extra_remove_base,
- unsigned long extra_remove_size,
- int i)
+
+static void __init
+mtrr_calc_range_state(u64 chunk_size, u64 gran_size,
+ unsigned long x_remove_base,
+ unsigned long x_remove_size, int i)
{
- int num_reg;
static struct res_range range_new[RANGE_NUM];
- static int nr_range_new;
unsigned long range_sums_new;
+ static int nr_range_new;
+ int num_reg;
- /* convert ranges to var ranges state */
- num_reg = x86_setup_var_mtrrs(range, nr_range,
- chunk_size, gran_size);
+ /* Convert ranges to var ranges state: */
+ num_reg = x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size);
- /* we got new setting in range_state, check it */
+ /* We got new setting in range_state, check it: */
memset(range_new, 0, sizeof(range_new));
nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
- extra_remove_base, extra_remove_size);
+ x_remove_base, x_remove_size);
range_sums_new = sum_ranges(range_new, nr_range_new);
result[i].chunk_sizek = chunk_size >> 10;
result[i].gran_sizek = gran_size >> 10;
result[i].num_reg = num_reg;
+
if (range_sums < range_sums_new) {
- result[i].lose_cover_sizek =
- (range_sums_new - range_sums) << PSHIFT;
+ result[i].lose_cover_sizek = (range_sums_new - range_sums) << PSHIFT;
result[i].bad = 1;
- } else
- result[i].lose_cover_sizek =
- (range_sums - range_sums_new) << PSHIFT;
+ } else {
+ result[i].lose_cover_sizek = (range_sums - range_sums_new) << PSHIFT;
+ }
- /* double check it */
+ /* Double check it: */
if (!result[i].bad && !result[i].lose_cover_sizek) {
- if (nr_range_new != nr_range ||
- memcmp(range, range_new, sizeof(range)))
- result[i].bad = 1;
+ if (nr_range_new != nr_range || memcmp(range, range_new, sizeof(range)))
+ result[i].bad = 1;
}
- if (!result[i].bad && (range_sums - range_sums_new <
- min_loss_pfn[num_reg])) {
- min_loss_pfn[num_reg] =
- range_sums - range_sums_new;
- }
+ if (!result[i].bad && (range_sums - range_sums_new < min_loss_pfn[num_reg]))
+ min_loss_pfn[num_reg] = range_sums - range_sums_new;
}
static void __init mtrr_print_out_one_result(int i)
{
- char gran_factor, chunk_factor, lose_factor;
unsigned long gran_base, chunk_base, lose_base;
+ char gran_factor, chunk_factor, lose_factor;
gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
- printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
- result[i].bad ? "*BAD*" : " ",
- gran_base, gran_factor, chunk_base, chunk_factor);
- printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n",
- result[i].num_reg, result[i].bad ? "-" : "",
- lose_base, lose_factor);
+
+ pr_info("%sgran_size: %ld%c \tchunk_size: %ld%c \t",
+ result[i].bad ? "*BAD*" : " ",
+ gran_base, gran_factor, chunk_base, chunk_factor);
+ pr_cont("num_reg: %d \tlose cover RAM: %s%ld%c\n",
+ result[i].num_reg, result[i].bad ? "-" : "",
+ lose_base, lose_factor);
}
static int __init mtrr_search_optimal_index(void)
{
- int i;
int num_reg_good;
int index_good;
+ int i;
if (nr_mtrr_spare_reg >= num_var_ranges)
nr_mtrr_spare_reg = num_var_ranges - 1;
+
num_reg_good = -1;
for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
if (!min_loss_pfn[i])
return index_good;
}
-
int __init mtrr_cleanup(unsigned address_bits)
{
- unsigned long extra_remove_base, extra_remove_size;
+ unsigned long x_remove_base, x_remove_size;
unsigned long base, size, def, dummy;
- mtrr_type type;
u64 chunk_size, gran_size;
+ mtrr_type type;
int index_good;
int i;
if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
return 0;
+
rdmsr(MSR_MTRRdefType, def, dummy);
def &= 0xff;
if (def != MTRR_TYPE_UNCACHABLE)
return 0;
- /* get it and store it aside */
+ /* Get it and store it aside: */
memset(range_state, 0, sizeof(range_state));
for (i = 0; i < num_var_ranges; i++) {
mtrr_if->get(i, &base, &size, &type);
range_state[i].type = type;
}
- /* check if we need handle it and can handle it */
+ /* Check if we need handle it and can handle it: */
if (!mtrr_need_cleanup())
return 0;
- /* print original var MTRRs at first, for debugging: */
+ /* Print original var MTRRs at first, for debugging: */
printk(KERN_DEBUG "original variable MTRRs\n");
print_out_mtrr_range_state();
memset(range, 0, sizeof(range));
- extra_remove_size = 0;
- extra_remove_base = 1 << (32 - PAGE_SHIFT);
+ x_remove_size = 0;
+ x_remove_base = 1 << (32 - PAGE_SHIFT);
if (mtrr_tom2)
- extra_remove_size =
- (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base;
- nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base,
- extra_remove_size);
+ x_remove_size = (mtrr_tom2 >> PAGE_SHIFT) - x_remove_base;
+
+ nr_range = x86_get_mtrr_mem_range(range, 0, x_remove_base, x_remove_size);
/*
- * [0, 1M) should always be coverred by var mtrr with WB
- * and fixed mtrrs should take effective before var mtrr for it
+ * [0, 1M) should always be covered by var mtrr with WB
+ * and fixed mtrrs should take effect before var mtrr for it:
*/
nr_range = add_range_with_merge(range, nr_range, 0,
(1ULL<<(20 - PAGE_SHIFT)) - 1);
- /* sort the ranges */
+ /* Sort the ranges: */
sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL);
range_sums = sum_ranges(range, nr_range);
if (mtrr_chunk_size && mtrr_gran_size) {
i = 0;
mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size,
- extra_remove_base, extra_remove_size, i);
+ x_remove_base, x_remove_size, i);
mtrr_print_out_one_result(i);
continue;
mtrr_calc_range_state(chunk_size, gran_size,
- extra_remove_base, extra_remove_size, i);
+ x_remove_base, x_remove_size, i);
if (debug_print) {
mtrr_print_out_one_result(i);
printk(KERN_INFO "\n");
}
}
- /* try to find the optimal index */
+ /* Try to find the optimal index: */
index_good = mtrr_search_optimal_index();
if (index_good != -1) {
i = index_good;
mtrr_print_out_one_result(i);
- /* convert ranges to var ranges state */
+ /* Convert ranges to var ranges state: */
chunk_size = result[i].chunk_sizek;
chunk_size <<= 10;
gran_size = result[i].gran_sizek;
* Note this won't check if the MTRRs < 4GB where the magic bit doesn't
* apply to are wrong, but so far we don't know of any such case in the wild.
*/
-#define Tom2Enabled (1U << 21)
-#define Tom2ForceMemTypeWB (1U << 22)
+#define Tom2Enabled (1U << 21)
+#define Tom2ForceMemTypeWB (1U << 22)
int __init amd_special_default_mtrr(void)
{
return 0;
if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11)
return 0;
- /* In case some hypervisor doesn't pass SYSCFG through */
+ /* In case some hypervisor doesn't pass SYSCFG through: */
if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0)
return 0;
/*
return 0;
}
-static u64 __init real_trim_memory(unsigned long start_pfn,
- unsigned long limit_pfn)
+static u64 __init
+real_trim_memory(unsigned long start_pfn, unsigned long limit_pfn)
{
u64 trim_start, trim_size;
+
trim_start = start_pfn;
trim_start <<= PAGE_SHIFT;
+
trim_size = limit_pfn;
trim_size <<= PAGE_SHIFT;
trim_size -= trim_start;
- return e820_update_range(trim_start, trim_size, E820_RAM,
- E820_RESERVED);
+ return e820_update_range(trim_start, trim_size, E820_RAM, E820_RESERVED);
}
+
/**
* mtrr_trim_uncached_memory - trim RAM not covered by MTRRs
* @end_pfn: ending page frame number
* Some buggy BIOSes don't setup the MTRRs properly for systems with certain
* memory configurations. This routine checks that the highest MTRR matches
* the end of memory, to make sure the MTRRs having a write back type cover
- * all of the memory the kernel is intending to use. If not, it'll trim any
+ * all of the memory the kernel is intending to use. If not, it'll trim any
* memory off the end by adjusting end_pfn, removing it from the kernel's
* allocation pools, warning the user with an obnoxious message.
*/
unsigned long i, base, size, highest_pfn = 0, def, dummy;
mtrr_type type;
u64 total_trim_size;
-
/* extra one for all 0 */
int num[MTRR_NUM_TYPES + 1];
+
/*
* Make sure we only trim uncachable memory on machines that
* support the Intel MTRR architecture:
*/
if (!is_cpu(INTEL) || disable_mtrr_trim)
return 0;
+
rdmsr(MSR_MTRRdefType, def, dummy);
def &= 0xff;
if (def != MTRR_TYPE_UNCACHABLE)
return 0;
- /* get it and store it aside */
+ /* Get it and store it aside: */
memset(range_state, 0, sizeof(range_state));
for (i = 0; i < num_var_ranges; i++) {
mtrr_if->get(i, &base, &size, &type);
range_state[i].type = type;
}
- /* Find highest cached pfn */
+ /* Find highest cached pfn: */
for (i = 0; i < num_var_ranges; i++) {
type = range_state[i].type;
if (type != MTRR_TYPE_WRBACK)
highest_pfn = base + size;
}
- /* kvm/qemu doesn't have mtrr set right, don't trim them all */
+ /* kvm/qemu doesn't have mtrr set right, don't trim them all: */
if (!highest_pfn) {
printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n");
return 0;
}
- /* check entries number */
+ /* Check entries number: */
memset(num, 0, sizeof(num));
for (i = 0; i < num_var_ranges; i++) {
type = range_state[i].type;
num[type]++;
}
- /* no entry for WB? */
+ /* No entry for WB? */
if (!num[MTRR_TYPE_WRBACK])
return 0;
- /* check if we only had WB and UC */
+ /* Check if we only had WB and UC: */
if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] !=
num_var_ranges - num[MTRR_NUM_TYPES])
return 0;
}
nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0);
+ /* Check the head: */
total_trim_size = 0;
- /* check the head */
if (range[0].start)
total_trim_size += real_trim_memory(0, range[0].start);
- /* check the holes */
+
+ /* Check the holes: */
for (i = 0; i < nr_range - 1; i++) {
if (range[i].end + 1 < range[i+1].start)
total_trim_size += real_trim_memory(range[i].end + 1,
range[i+1].start);
}
- /* check the top */
+
+ /* Check the top: */
i = nr_range - 1;
if (range[i].end + 1 < end_pfn)
total_trim_size += real_trim_memory(range[i].end + 1,
end_pfn);
if (total_trim_size) {
- printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover"
- " all of memory, losing %lluMB of RAM.\n",
- total_trim_size >> 20);
+ pr_warning("WARNING: BIOS bug: CPU MTRRs don't cover all of memory, losing %lluMB of RAM.\n", total_trim_size >> 20);
if (!changed_by_mtrr_cleanup)
WARN_ON(1);
- printk(KERN_INFO "update e820 for mtrr\n");
+ pr_info("update e820 for mtrr\n");
update_e820();
return 1;
return 0;
}
-
#include <linux/init.h>
+#include <linux/io.h>
#include <linux/mm.h>
-#include <asm/mtrr.h>
-#include <asm/msr.h>
-#include <asm/io.h>
+
#include <asm/processor-cyrix.h>
#include <asm/processor-flags.h>
+#include <asm/mtrr.h>
+#include <asm/msr.h>
+
#include "mtrr.h"
static void
cyrix_get_arr(unsigned int reg, unsigned long *base,
unsigned long *size, mtrr_type * type)
{
- unsigned long flags;
unsigned char arr, ccr3, rcr, shift;
+ unsigned long flags;
arr = CX86_ARR_BASE + (reg << 1) + reg; /* avoid multiplication by 3 */
- /* Save flags and disable interrupts */
local_irq_save(flags);
ccr3 = getCx86(CX86_CCR3);
setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */
- ((unsigned char *) base)[3] = getCx86(arr);
- ((unsigned char *) base)[2] = getCx86(arr + 1);
- ((unsigned char *) base)[1] = getCx86(arr + 2);
+ ((unsigned char *)base)[3] = getCx86(arr);
+ ((unsigned char *)base)[2] = getCx86(arr + 1);
+ ((unsigned char *)base)[1] = getCx86(arr + 2);
rcr = getCx86(CX86_RCR_BASE + reg);
- setCx86(CX86_CCR3, ccr3); /* disable MAPEN */
+ setCx86(CX86_CCR3, ccr3); /* disable MAPEN */
- /* Enable interrupts if it was enabled previously */
local_irq_restore(flags);
+
shift = ((unsigned char *) base)[1] & 0x0f;
*base >>= PAGE_SHIFT;
- /* Power of two, at least 4K on ARR0-ARR6, 256K on ARR7
+ /*
+ * Power of two, at least 4K on ARR0-ARR6, 256K on ARR7
* Note: shift==0xf means 4G, this is unsupported.
*/
if (shift)
}
}
+/*
+ * cyrix_get_free_region - get a free ARR.
+ *
+ * @base: the starting (base) address of the region.
+ * @size: the size (in bytes) of the region.
+ *
+ * Returns: the index of the region on success, else -1 on error.
+*/
static int
cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg)
-/* [SUMMARY] Get a free ARR.
- <base> The starting (base) address of the region.
- <size> The size (in bytes) of the region.
- [RETURNS] The index of the region on success, else -1 on error.
-*/
{
- int i;
- mtrr_type ltype;
unsigned long lbase, lsize;
+ mtrr_type ltype;
+ int i;
switch (replace_reg) {
case 7:
cyrix_get_arr(7, &lbase, &lsize, <ype);
if (lsize == 0)
return 7;
- /* Else try ARR0-ARR6 first */
+ /* Else try ARR0-ARR6 first */
} else {
for (i = 0; i < 7; i++) {
cyrix_get_arr(i, &lbase, &lsize, <ype);
if (lsize == 0)
return i;
}
- /* ARR0-ARR6 isn't free, try ARR7 but its size must be at least 256K */
+ /*
+ * ARR0-ARR6 isn't free
+ * try ARR7 but its size must be at least 256K
+ */
cyrix_get_arr(i, &lbase, &lsize, <ype);
if ((lsize == 0) && (size >= 0x40))
return i;
return -ENOSPC;
}
-static u32 cr4 = 0;
-static u32 ccr3;
+static u32 cr4, ccr3;
static void prepare_set(void)
{
u32 cr0;
/* Save value of CR4 and clear Page Global Enable (bit 7) */
- if ( cpu_has_pge ) {
+ if (cpu_has_pge) {
cr4 = read_cr4();
write_cr4(cr4 & ~X86_CR4_PGE);
}
- /* Disable and flush caches. Note that wbinvd flushes the TLBs as
- a side-effect */
+ /*
+ * Disable and flush caches.
+ * Note that wbinvd flushes the TLBs as a side-effect
+ */
cr0 = read_cr0() | X86_CR0_CD;
wbinvd();
write_cr0(cr0);
/* Cyrix ARRs - everything else was excluded at the top */
setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);
-
}
static void post_set(void)
{
- /* Flush caches and TLBs */
+ /* Flush caches and TLBs */
wbinvd();
/* Cyrix ARRs - everything else was excluded at the top */
setCx86(CX86_CCR3, ccr3);
-
- /* Enable caches */
+
+ /* Enable caches */
write_cr0(read_cr0() & 0xbfffffff);
- /* Restore value of CR4 */
- if ( cpu_has_pge )
+ /* Restore value of CR4 */
+ if (cpu_has_pge)
write_cr4(cr4);
}
size >>= 6;
size &= 0x7fff; /* make sure arr_size <= 14 */
- for (arr_size = 0; size; arr_size++, size >>= 1) ;
+ for (arr_size = 0; size; arr_size++, size >>= 1)
+ ;
if (reg < 7) {
switch (type) {
prepare_set();
base <<= PAGE_SHIFT;
- setCx86(arr, ((unsigned char *) &base)[3]);
- setCx86(arr + 1, ((unsigned char *) &base)[2]);
- setCx86(arr + 2, (((unsigned char *) &base)[1]) | arr_size);
+ setCx86(arr + 0, ((unsigned char *)&base)[3]);
+ setCx86(arr + 1, ((unsigned char *)&base)[2]);
+ setCx86(arr + 2, (((unsigned char *)&base)[1]) | arr_size);
setCx86(CX86_RCR_BASE + reg, arr_type);
post_set();
}
typedef struct {
- unsigned long base;
- unsigned long size;
- mtrr_type type;
+ unsigned long base;
+ unsigned long size;
+ mtrr_type type;
} arr_state_t;
static arr_state_t arr_state[8] = {
setCx86(CX86_CCR0 + i, ccr_state[i]);
for (; i < 7; i++)
setCx86(CX86_CCR4 + i, ccr_state[i]);
- for (i = 0; i < 8; i++)
- cyrix_set_arr(i, arr_state[i].base,
+
+ for (i = 0; i < 8; i++) {
+ cyrix_set_arr(i, arr_state[i].base,
arr_state[i].size, arr_state[i].type);
+ }
post_set();
}
static struct mtrr_ops cyrix_mtrr_ops = {
.vendor = X86_VENDOR_CYRIX,
-// .init = cyrix_arr_init,
.set_all = cyrix_set_all,
.set = cyrix_set_arr,
.get = cyrix_get_arr,
set_mtrr_ops(&cyrix_mtrr_ops);
return 0;
}
-
-//arch_initcall(cyrix_init_mtrr);
-/* This only handles 32bit MTRR on 32bit hosts. This is strictly wrong
- because MTRRs can span upto 40 bits (36bits on most modern x86) */
+/*
+ * This only handles 32bit MTRR on 32bit hosts. This is strictly wrong
+ * because MTRRs can span upto 40 bits (36bits on most modern x86)
+ */
+#define DEBUG
+
+#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
+#include <linux/io.h>
#include <linux/mm.h>
-#include <linux/module.h>
-#include <asm/io.h>
-#include <asm/mtrr.h>
-#include <asm/msr.h>
-#include <asm/system.h>
-#include <asm/cpufeature.h>
+
#include <asm/processor-flags.h>
+#include <asm/cpufeature.h>
#include <asm/tlbflush.h>
+#include <asm/system.h>
+#include <asm/mtrr.h>
+#include <asm/msr.h>
#include <asm/pat.h>
+
#include "mtrr.h"
struct fixed_range_block {
- int base_msr; /* start address of an MTRR block */
- int ranges; /* number of MTRRs in this block */
+ int base_msr; /* start address of an MTRR block */
+ int ranges; /* number of MTRRs in this block */
};
static struct fixed_range_block fixed_range_blocks[] = {
- { MSR_MTRRfix64K_00000, 1 }, /* one 64k MTRR */
- { MSR_MTRRfix16K_80000, 2 }, /* two 16k MTRRs */
- { MSR_MTRRfix4K_C0000, 8 }, /* eight 4k MTRRs */
+ { MSR_MTRRfix64K_00000, 1 }, /* one 64k MTRR */
+ { MSR_MTRRfix16K_80000, 2 }, /* two 16k MTRRs */
+ { MSR_MTRRfix4K_C0000, 8 }, /* eight 4k MTRRs */
{}
};
static int mtrr_state_set;
u64 mtrr_tom2;
-struct mtrr_state_type mtrr_state = {};
+struct mtrr_state_type mtrr_state;
EXPORT_SYMBOL_GPL(mtrr_state);
-/**
+/*
* BIOS is expected to clear MtrrFixDramModEn bit, see for example
* "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD
* Opteron Processors" (26094 Rev. 3.30 February 2006), section
* Look of multiple ranges matching this address and pick type
* as per MTRR precedence
*/
- if (!(mtrr_state.enabled & 2)) {
+ if (!(mtrr_state.enabled & 2))
return mtrr_state.def_type;
- }
prev_match = 0xFF;
for (i = 0; i < num_var_ranges; ++i) {
if (start_state != end_state)
return 0xFE;
- if ((start & mask) != (base & mask)) {
+ if ((start & mask) != (base & mask))
continue;
- }
curr_match = mtrr_state.var_ranges[i].base_lo & 0xff;
if (prev_match == 0xFF) {
curr_match = MTRR_TYPE_WRTHROUGH;
}
- if (prev_match != curr_match) {
+ if (prev_match != curr_match)
return MTRR_TYPE_UNCACHABLE;
- }
}
if (mtrr_tom2) {
return mtrr_state.def_type;
}
-/* Get the MSR pair relating to a var range */
+/* Get the MSR pair relating to a var range */
static void
get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr)
{
rdmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi);
}
-/* fill the MSR pair relating to a var range */
+/* Fill the MSR pair relating to a var range */
void fill_mtrr_var_range(unsigned int index,
u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi)
{
vr[index].mask_hi = mask_hi;
}
-static void
-get_fixed_ranges(mtrr_type * frs)
+static void get_fixed_ranges(mtrr_type *frs)
{
- unsigned int *p = (unsigned int *) frs;
+ unsigned int *p = (unsigned int *)frs;
int i;
k8_check_syscfg_dram_mod_en();
if (!last_fixed_end)
return;
- printk(KERN_DEBUG " %05X-%05X %s\n", last_fixed_start,
- last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type));
+ pr_debug(" %05X-%05X %s\n", last_fixed_start,
+ last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type));
last_fixed_end = 0;
}
static void __init update_fixed_last(unsigned base, unsigned end,
- mtrr_type type)
+ mtrr_type type)
{
last_fixed_start = base;
last_fixed_end = end;
last_fixed_type = type;
}
-static void __init print_fixed(unsigned base, unsigned step,
- const mtrr_type *types)
+static void __init
+print_fixed(unsigned base, unsigned step, const mtrr_type *types)
{
unsigned i;
unsigned int i;
int high_width;
- printk(KERN_DEBUG "MTRR default type: %s\n",
- mtrr_attrib_to_str(mtrr_state.def_type));
+ pr_debug("MTRR default type: %s\n",
+ mtrr_attrib_to_str(mtrr_state.def_type));
if (mtrr_state.have_fixed) {
- printk(KERN_DEBUG "MTRR fixed ranges %sabled:\n",
- mtrr_state.enabled & 1 ? "en" : "dis");
+ pr_debug("MTRR fixed ranges %sabled:\n",
+ mtrr_state.enabled & 1 ? "en" : "dis");
print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0);
for (i = 0; i < 2; ++i)
- print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8);
+ print_fixed(0x80000 + i * 0x20000, 0x04000,
+ mtrr_state.fixed_ranges + (i + 1) * 8);
for (i = 0; i < 8; ++i)
- print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8);
+ print_fixed(0xC0000 + i * 0x08000, 0x01000,
+ mtrr_state.fixed_ranges + (i + 3) * 8);
/* tail */
print_fixed_last();
}
- printk(KERN_DEBUG "MTRR variable ranges %sabled:\n",
- mtrr_state.enabled & 2 ? "en" : "dis");
+ pr_debug("MTRR variable ranges %sabled:\n",
+ mtrr_state.enabled & 2 ? "en" : "dis");
if (size_or_mask & 0xffffffffUL)
high_width = ffs(size_or_mask & 0xffffffffUL) - 1;
else
high_width = ffs(size_or_mask>>32) + 32 - 1;
high_width = (high_width - (32 - PAGE_SHIFT) + 3) / 4;
+
for (i = 0; i < num_var_ranges; ++i) {
if (mtrr_state.var_ranges[i].mask_lo & (1 << 11))
- printk(KERN_DEBUG " %u base %0*X%05X000 mask %0*X%05X000 %s\n",
- i,
- high_width,
- mtrr_state.var_ranges[i].base_hi,
- mtrr_state.var_ranges[i].base_lo >> 12,
- high_width,
- mtrr_state.var_ranges[i].mask_hi,
- mtrr_state.var_ranges[i].mask_lo >> 12,
- mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff));
+ pr_debug(" %u base %0*X%05X000 mask %0*X%05X000 %s\n",
+ i,
+ high_width,
+ mtrr_state.var_ranges[i].base_hi,
+ mtrr_state.var_ranges[i].base_lo >> 12,
+ high_width,
+ mtrr_state.var_ranges[i].mask_hi,
+ mtrr_state.var_ranges[i].mask_lo >> 12,
+ mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff));
else
- printk(KERN_DEBUG " %u disabled\n", i);
- }
- if (mtrr_tom2) {
- printk(KERN_DEBUG "TOM2: %016llx aka %lldM\n",
- mtrr_tom2, mtrr_tom2>>20);
+ pr_debug(" %u disabled\n", i);
}
+ if (mtrr_tom2)
+ pr_debug("TOM2: %016llx aka %lldM\n", mtrr_tom2, mtrr_tom2>>20);
}
-/* Grab all of the MTRR state for this CPU into *state */
+/* Grab all of the MTRR state for this CPU into *state */
void __init get_mtrr_state(void)
{
- unsigned int i;
struct mtrr_var_range *vrs;
- unsigned lo, dummy;
unsigned long flags;
+ unsigned lo, dummy;
+ unsigned int i;
vrs = mtrr_state.var_ranges;
if (amd_special_default_mtrr()) {
unsigned low, high;
+
/* TOP_MEM2 */
rdmsr(MSR_K8_TOP_MEM2, low, high);
mtrr_tom2 = high;
post_set();
local_irq_restore(flags);
-
}
-/* Some BIOS's are fucked and don't set all MTRRs the same! */
+/* Some BIOS's are messed up and don't set all MTRRs the same! */
void __init mtrr_state_warn(void)
{
unsigned long mask = smp_changes_mask;
if (!mask)
return;
if (mask & MTRR_CHANGE_MASK_FIXED)
- printk(KERN_WARNING "mtrr: your CPUs had inconsistent fixed MTRR settings\n");
+ pr_warning("mtrr: your CPUs had inconsistent fixed MTRR settings\n");
if (mask & MTRR_CHANGE_MASK_VARIABLE)
- printk(KERN_WARNING "mtrr: your CPUs had inconsistent variable MTRR settings\n");
+ pr_warning("mtrr: your CPUs had inconsistent variable MTRR settings\n");
if (mask & MTRR_CHANGE_MASK_DEFTYPE)
- printk(KERN_WARNING "mtrr: your CPUs had inconsistent MTRRdefType settings\n");
+ pr_warning("mtrr: your CPUs had inconsistent MTRRdefType settings\n");
+
printk(KERN_INFO "mtrr: probably your BIOS does not setup all CPUs.\n");
printk(KERN_INFO "mtrr: corrected configuration.\n");
}
-/* Doesn't attempt to pass an error out to MTRR users
- because it's quite complicated in some cases and probably not
- worth it because the best error handling is to ignore it. */
+/*
+ * Doesn't attempt to pass an error out to MTRR users
+ * because it's quite complicated in some cases and probably not
+ * worth it because the best error handling is to ignore it.
+ */
void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b)
{
- if (wrmsr_safe(msr, a, b) < 0)
+ if (wrmsr_safe(msr, a, b) < 0) {
printk(KERN_ERR
"MTRR: CPU %u: Writing MSR %x to %x:%x failed\n",
smp_processor_id(), msr, a, b);
+ }
}
/**
- * set_fixed_range - checks & updates a fixed-range MTRR if it differs from the value it should have
+ * set_fixed_range - checks & updates a fixed-range MTRR if it
+ * differs from the value it should have
* @msr: MSR address of the MTTR which should be checked and updated
* @changed: pointer which indicates whether the MTRR needed to be changed
* @msrwords: pointer to the MSR values which the MSR should have
*
* Returns: The index of the region on success, else negative on error.
*/
-int generic_get_free_region(unsigned long base, unsigned long size, int replace_reg)
+int
+generic_get_free_region(unsigned long base, unsigned long size, int replace_reg)
{
- int i, max;
- mtrr_type ltype;
unsigned long lbase, lsize;
+ mtrr_type ltype;
+ int i, max;
max = num_var_ranges;
if (replace_reg >= 0 && replace_reg < max)
return replace_reg;
+
for (i = 0; i < max; ++i) {
mtrr_if->get(i, &lbase, &lsize, <ype);
if (lsize == 0)
return i;
}
+
return -ENOSPC;
}
rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi);
if ((mask_lo & 0x800) == 0) {
- /* Invalid (i.e. free) range */
+ /* Invalid (i.e. free) range */
*base = 0;
*size = 0;
*type = 0;
}
/**
- * set_fixed_ranges - checks & updates the fixed-range MTRRs if they differ from the saved set
+ * set_fixed_ranges - checks & updates the fixed-range MTRRs if they
+ * differ from the saved set
* @frs: pointer to fixed-range MTRR values, saved by get_fixed_ranges()
*/
-static int set_fixed_ranges(mtrr_type * frs)
+static int set_fixed_ranges(mtrr_type *frs)
{
- unsigned long long *saved = (unsigned long long *) frs;
+ unsigned long long *saved = (unsigned long long *)frs;
bool changed = false;
- int block=-1, range;
+ int block = -1, range;
k8_check_syscfg_dram_mod_en();
- while (fixed_range_blocks[++block].ranges)
- for (range=0; range < fixed_range_blocks[block].ranges; range++)
- set_fixed_range(fixed_range_blocks[block].base_msr + range,
- &changed, (unsigned int *) saved++);
+ while (fixed_range_blocks[++block].ranges) {
+ for (range = 0; range < fixed_range_blocks[block].ranges; range++)
+ set_fixed_range(fixed_range_blocks[block].base_msr + range,
+ &changed, (unsigned int *)saved++);
+ }
return changed;
}
-/* Set the MSR pair relating to a var range. Returns TRUE if
- changes are made */
+/*
+ * Set the MSR pair relating to a var range.
+ * Returns true if changes are made.
+ */
static bool set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
{
unsigned int lo, hi;
if ((vr->base_lo & 0xfffff0ffUL) != (lo & 0xfffff0ffUL)
|| (vr->base_hi & (size_and_mask >> (32 - PAGE_SHIFT))) !=
(hi & (size_and_mask >> (32 - PAGE_SHIFT)))) {
+
mtrr_wrmsr(MTRRphysBase_MSR(index), vr->base_lo, vr->base_hi);
changed = true;
}
*/
static unsigned long set_mtrr_state(void)
{
- unsigned int i;
unsigned long change_mask = 0;
+ unsigned int i;
- for (i = 0; i < num_var_ranges; i++)
+ for (i = 0; i < num_var_ranges; i++) {
if (set_mtrr_var_ranges(i, &mtrr_state.var_ranges[i]))
change_mask |= MTRR_CHANGE_MASK_VARIABLE;
+ }
if (mtrr_state.have_fixed && set_fixed_ranges(mtrr_state.fixed_ranges))
change_mask |= MTRR_CHANGE_MASK_FIXED;
- /* Set_mtrr_restore restores the old value of MTRRdefType,
- so to set it we fiddle with the saved value */
+ /*
+ * Set_mtrr_restore restores the old value of MTRRdefType,
+ * so to set it we fiddle with the saved value:
+ */
if ((deftype_lo & 0xff) != mtrr_state.def_type
|| ((deftype_lo & 0xc00) >> 10) != mtrr_state.enabled) {
- deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type | (mtrr_state.enabled << 10);
+
+ deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type |
+ (mtrr_state.enabled << 10);
change_mask |= MTRR_CHANGE_MASK_DEFTYPE;
}
}
-static unsigned long cr4 = 0;
+static unsigned long cr4;
static DEFINE_SPINLOCK(set_atomicity_lock);
/*
- * Since we are disabling the cache don't allow any interrupts - they
- * would run extremely slow and would only increase the pain. The caller must
- * ensure that local interrupts are disabled and are reenabled after post_set()
- * has been called.
+ * Since we are disabling the cache don't allow any interrupts,
+ * they would run extremely slow and would only increase the pain.
+ *
+ * The caller must ensure that local interrupts are disabled and
+ * are reenabled after post_set() has been called.
*/
-
static void prepare_set(void) __acquires(set_atomicity_lock)
{
unsigned long cr0;
- /* Note that this is not ideal, since the cache is only flushed/disabled
- for this CPU while the MTRRs are changed, but changing this requires
- more invasive changes to the way the kernel boots */
+ /*
+ * Note that this is not ideal
+ * since the cache is only flushed/disabled for this CPU while the
+ * MTRRs are changed, but changing this requires more invasive
+ * changes to the way the kernel boots
+ */
spin_lock(&set_atomicity_lock);
- /* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */
+ /* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */
cr0 = read_cr0() | X86_CR0_CD;
write_cr0(cr0);
wbinvd();
- /* Save value of CR4 and clear Page Global Enable (bit 7) */
- if ( cpu_has_pge ) {
+ /* Save value of CR4 and clear Page Global Enable (bit 7) */
+ if (cpu_has_pge) {
cr4 = read_cr4();
write_cr4(cr4 & ~X86_CR4_PGE);
}
/* Flush all TLBs via a mov %cr3, %reg; mov %reg, %cr3 */
__flush_tlb();
- /* Save MTRR state */
+ /* Save MTRR state */
rdmsr(MSR_MTRRdefType, deftype_lo, deftype_hi);
- /* Disable MTRRs, and set the default type to uncached */
+ /* Disable MTRRs, and set the default type to uncached */
mtrr_wrmsr(MSR_MTRRdefType, deftype_lo & ~0xcff, deftype_hi);
}
static void post_set(void) __releases(set_atomicity_lock)
{
- /* Flush TLBs (no need to flush caches - they are disabled) */
+ /* Flush TLBs (no need to flush caches - they are disabled) */
__flush_tlb();
/* Intel (P6) standard MTRRs */
mtrr_wrmsr(MSR_MTRRdefType, deftype_lo, deftype_hi);
-
- /* Enable caches */
+
+ /* Enable caches */
write_cr0(read_cr0() & 0xbfffffff);
- /* Restore value of CR4 */
- if ( cpu_has_pge )
+ /* Restore value of CR4 */
+ if (cpu_has_pge)
write_cr4(cr4);
spin_unlock(&set_atomicity_lock);
}
post_set();
local_irq_restore(flags);
- /* Use the atomic bitops to update the global mask */
+ /* Use the atomic bitops to update the global mask */
for (count = 0; count < sizeof mask * 8; ++count) {
if (mask & 0x01)
set_bit(count, &smp_changes_mask);
mask >>= 1;
}
-
+
}
+/**
+ * generic_set_mtrr - set variable MTRR register on the local CPU.
+ *
+ * @reg: The register to set.
+ * @base: The base address of the region.
+ * @size: The size of the region. If this is 0 the region is disabled.
+ * @type: The type of the region.
+ *
+ * Returns nothing.
+ */
static void generic_set_mtrr(unsigned int reg, unsigned long base,
unsigned long size, mtrr_type type)
-/* [SUMMARY] Set variable MTRR register on the local CPU.
- <reg> The register to set.
- <base> The base address of the region.
- <size> The size of the region. If this is 0 the region is disabled.
- <type> The type of the region.
- [RETURNS] Nothing.
-*/
{
unsigned long flags;
struct mtrr_var_range *vr;
prepare_set();
if (size == 0) {
- /* The invalid bit is kept in the mask, so we simply clear the
- relevant mask register to disable a range. */
+ /*
+ * The invalid bit is kept in the mask, so we simply
+ * clear the relevant mask register to disable a range.
+ */
mtrr_wrmsr(MTRRphysMask_MSR(reg), 0, 0);
memset(vr, 0, sizeof(struct mtrr_var_range));
} else {
local_irq_restore(flags);
}
-int generic_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
+int generic_validate_add_page(unsigned long base, unsigned long size,
+ unsigned int type)
{
unsigned long lbase, last;
- /* For Intel PPro stepping <= 7, must be 4 MiB aligned
- and not touch 0x70000000->0x7003FFFF */
+ /*
+ * For Intel PPro stepping <= 7
+ * must be 4 MiB aligned and not touch 0x70000000 -> 0x7003FFFF
+ */
if (is_cpu(INTEL) && boot_cpu_data.x86 == 6 &&
boot_cpu_data.x86_model == 1 &&
boot_cpu_data.x86_mask <= 7) {
if (base & ((1 << (22 - PAGE_SHIFT)) - 1)) {
- printk(KERN_WARNING "mtrr: base(0x%lx000) is not 4 MiB aligned\n", base);
+ pr_warning("mtrr: base(0x%lx000) is not 4 MiB aligned\n", base);
return -EINVAL;
}
if (!(base + size < 0x70000 || base > 0x7003F) &&
(type == MTRR_TYPE_WRCOMB
|| type == MTRR_TYPE_WRBACK)) {
- printk(KERN_WARNING "mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n");
+ pr_warning("mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n");
return -EINVAL;
}
}
- /* Check upper bits of base and last are equal and lower bits are 0
- for base and 1 for last */
+ /*
+ * Check upper bits of base and last are equal and lower bits are 0
+ * for base and 1 for last
+ */
last = base + size - 1;
for (lbase = base; !(lbase & 1) && (last & 1);
- lbase = lbase >> 1, last = last >> 1) ;
+ lbase = lbase >> 1, last = last >> 1)
+ ;
if (lbase != last) {
- printk(KERN_WARNING "mtrr: base(0x%lx000) is not aligned on a size(0x%lx000) boundary\n",
- base, size);
+ pr_warning("mtrr: base(0x%lx000) is not aligned on a size(0x%lx000) boundary\n", base, size);
return -EINVAL;
}
return 0;
}
-
static int generic_have_wrcomb(void)
{
unsigned long config, dummy;
rdmsr(MSR_MTRRcap, config, dummy);
- return (config & (1 << 10));
+ return config & (1 << 10);
}
int positive_have_wrcomb(void)
return 1;
}
-/* generic structure...
+/*
+ * Generic structure...
*/
struct mtrr_ops generic_mtrr_ops = {
- .use_intel_if = 1,
- .set_all = generic_set_all,
- .get = generic_get_mtrr,
- .get_free_region = generic_get_free_region,
- .set = generic_set_mtrr,
- .validate_add_page = generic_validate_add_page,
- .have_wrcomb = generic_have_wrcomb,
+ .use_intel_if = 1,
+ .set_all = generic_set_all,
+ .get = generic_get_mtrr,
+ .get_free_region = generic_get_free_region,
+ .set = generic_set_mtrr,
+ .validate_add_page = generic_validate_add_page,
+ .have_wrcomb = generic_have_wrcomb,
};
-#include <linux/init.h>
-#include <linux/proc_fs.h>
#include <linux/capability.h>
-#include <linux/ctype.h>
-#include <linux/module.h>
#include <linux/seq_file.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
+#include <linux/proc_fs.h>
+#include <linux/module.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
#define LINE_SIZE 80
#include <asm/mtrr.h>
+
#include "mtrr.h"
#define FILE_FCOUNT(f) (((struct seq_file *)((f)->private_data))->private)
static const char *const mtrr_strings[MTRR_NUM_TYPES] =
{
- "uncachable", /* 0 */
- "write-combining", /* 1 */
- "?", /* 2 */
- "?", /* 3 */
- "write-through", /* 4 */
- "write-protect", /* 5 */
- "write-back", /* 6 */
+ "uncachable", /* 0 */
+ "write-combining", /* 1 */
+ "?", /* 2 */
+ "?", /* 3 */
+ "write-through", /* 4 */
+ "write-protect", /* 5 */
+ "write-back", /* 6 */
};
const char *mtrr_attrib_to_str(int x)
mtrr_file_add(unsigned long base, unsigned long size,
unsigned int type, bool increment, struct file *file, int page)
{
+ unsigned int *fcount = FILE_FCOUNT(file);
int reg, max;
- unsigned int *fcount = FILE_FCOUNT(file);
max = num_var_ranges;
if (fcount == NULL) {
mtrr_file_del(unsigned long base, unsigned long size,
struct file *file, int page)
{
- int reg;
unsigned int *fcount = FILE_FCOUNT(file);
+ int reg;
if (!page) {
if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1)))
return reg;
}
-/* RED-PEN: seq_file can seek now. this is ignored. */
+/*
+ * seq_file can seek but we ignore it.
+ *
+ * Format of control line:
+ * "base=%Lx size=%Lx type=%s" or "disable=%d"
+ */
static ssize_t
mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos)
-/* Format of control line:
- "base=%Lx size=%Lx type=%s" OR:
- "disable=%d"
-*/
{
int i, err;
unsigned long reg;
return -EPERM;
if (!len)
return -EINVAL;
+
memset(line, 0, LINE_SIZE);
if (len > LINE_SIZE)
len = LINE_SIZE;
if (copy_from_user(line, buf, len - 1))
return -EFAULT;
+
linelen = strlen(line);
ptr = line + linelen - 1;
if (linelen && *ptr == '\n')
*ptr = '\0';
+
if (!strncmp(line, "disable=", 8)) {
reg = simple_strtoul(line + 8, &ptr, 0);
err = mtrr_del_page(reg, 0, 0);
return err;
return len;
}
+
if (strncmp(line, "base=", 5))
return -EINVAL;
+
base = simple_strtoull(line + 5, &ptr, 0);
- for (; isspace(*ptr); ++ptr) ;
+ for (; isspace(*ptr); ++ptr)
+ ;
+
if (strncmp(ptr, "size=", 5))
return -EINVAL;
+
size = simple_strtoull(ptr + 5, &ptr, 0);
if ((base & 0xfff) || (size & 0xfff))
return -EINVAL;
- for (; isspace(*ptr); ++ptr) ;
+ for (; isspace(*ptr); ++ptr)
+ ;
+
if (strncmp(ptr, "type=", 5))
return -EINVAL;
ptr += 5;
- for (; isspace(*ptr); ++ptr) ;
+ for (; isspace(*ptr); ++ptr)
+ ;
+
for (i = 0; i < MTRR_NUM_TYPES; ++i) {
if (strcmp(ptr, mtrr_strings[i]))
continue;
base >>= PAGE_SHIFT;
size >>= PAGE_SHIFT;
- err =
- mtrr_add_page((unsigned long) base, (unsigned long) size, i,
- true);
+ err = mtrr_add_page((unsigned long)base, (unsigned long)size, i, true);
if (err < 0)
return err;
return len;
case MTRRIOC32_SET_PAGE_ENTRY:
case MTRRIOC32_DEL_PAGE_ENTRY:
case MTRRIOC32_KILL_PAGE_ENTRY: {
- struct mtrr_sentry32 __user *s32 = (struct mtrr_sentry32 __user *)__arg;
+ struct mtrr_sentry32 __user *s32;
+
+ s32 = (struct mtrr_sentry32 __user *)__arg;
err = get_user(sentry.base, &s32->base);
err |= get_user(sentry.size, &s32->size);
err |= get_user(sentry.type, &s32->type);
}
case MTRRIOC32_GET_ENTRY:
case MTRRIOC32_GET_PAGE_ENTRY: {
- struct mtrr_gentry32 __user *g32 = (struct mtrr_gentry32 __user *)__arg;
+ struct mtrr_gentry32 __user *g32;
+
+ g32 = (struct mtrr_gentry32 __user *)__arg;
err = get_user(gentry.regnum, &g32->regnum);
err |= get_user(gentry.base, &g32->base);
err |= get_user(gentry.size, &g32->size);
if (err)
return err;
- switch(cmd) {
+ switch (cmd) {
case MTRRIOC_GET_ENTRY:
case MTRRIOC_GET_PAGE_ENTRY:
if (copy_to_user(arg, &gentry, sizeof gentry))
#ifdef CONFIG_COMPAT
case MTRRIOC32_GET_ENTRY:
case MTRRIOC32_GET_PAGE_ENTRY: {
- struct mtrr_gentry32 __user *g32 = (struct mtrr_gentry32 __user *)__arg;
+ struct mtrr_gentry32 __user *g32;
+
+ g32 = (struct mtrr_gentry32 __user *)__arg;
err = put_user(gentry.base, &g32->base);
err |= put_user(gentry.size, &g32->size);
err |= put_user(gentry.regnum, &g32->regnum);
return err;
}
-static int
-mtrr_close(struct inode *ino, struct file *file)
+static int mtrr_close(struct inode *ino, struct file *file)
{
- int i, max;
unsigned int *fcount = FILE_FCOUNT(file);
+ int i, max;
if (fcount != NULL) {
max = num_var_ranges;
static int mtrr_open(struct inode *inode, struct file *file)
{
- if (!mtrr_if)
+ if (!mtrr_if)
return -EIO;
- if (!mtrr_if->get)
- return -ENXIO;
+ if (!mtrr_if->get)
+ return -ENXIO;
return single_open(file, mtrr_seq_show, NULL);
}
static const struct file_operations mtrr_fops = {
- .owner = THIS_MODULE,
- .open = mtrr_open,
- .read = seq_read,
- .llseek = seq_lseek,
- .write = mtrr_write,
- .unlocked_ioctl = mtrr_ioctl,
- .compat_ioctl = mtrr_ioctl,
- .release = mtrr_close,
+ .owner = THIS_MODULE,
+ .open = mtrr_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .write = mtrr_write,
+ .unlocked_ioctl = mtrr_ioctl,
+ .compat_ioctl = mtrr_ioctl,
+ .release = mtrr_close,
};
static int mtrr_seq_show(struct seq_file *seq, void *offset)
max = num_var_ranges;
for (i = 0; i < max; i++) {
mtrr_if->get(i, &base, &size, &type);
- if (size == 0)
+ if (size == 0) {
mtrr_usage_table[i] = 0;
- else {
- if (size < (0x100000 >> PAGE_SHIFT)) {
- /* less than 1MB */
- factor = 'K';
- size <<= PAGE_SHIFT - 10;
- } else {
- factor = 'M';
- size >>= 20 - PAGE_SHIFT;
- }
- /* RED-PEN: base can be > 32bit */
- len += seq_printf(seq,
- "reg%02i: base=0x%06lx000 (%5luMB), size=%5lu%cB, count=%d: %s\n",
- i, base, base >> (20 - PAGE_SHIFT), size, factor,
- mtrr_usage_table[i], mtrr_attrib_to_str(type));
+ continue;
}
+ if (size < (0x100000 >> PAGE_SHIFT)) {
+ /* less than 1MB */
+ factor = 'K';
+ size <<= PAGE_SHIFT - 10;
+ } else {
+ factor = 'M';
+ size >>= 20 - PAGE_SHIFT;
+ }
+ /* Base can be > 32bit */
+ len += seq_printf(seq, "reg%02i: base=0x%06lx000 "
+ "(%5luMB), size=%5lu%cB, count=%d: %s\n",
+ i, base, base >> (20 - PAGE_SHIFT), size,
+ factor, mtrr_usage_table[i],
+ mtrr_attrib_to_str(type));
}
return 0;
}
proc_create("mtrr", S_IWUSR | S_IRUGO, NULL, &mtrr_fops);
return 0;
}
-
arch_initcall(mtrr_if_init);
#endif /* CONFIG_PROC_FS */
Operating System Writer's Guide" (Intel document number 242692),
section 11.11.7
- This was cleaned and made readable by Patrick Mochel <mochel@osdl.org>
- on 6-7 March 2002.
- Source: Intel Architecture Software Developers Manual, Volume 3:
+ This was cleaned and made readable by Patrick Mochel <mochel@osdl.org>
+ on 6-7 March 2002.
+ Source: Intel Architecture Software Developers Manual, Volume 3:
System Programming Guide; Section 9.11. (1997 edition - PPro).
*/
+#define DEBUG
+
+#include <linux/types.h> /* FIXME: kvm_para.h needs this */
+
+#include <linux/kvm_para.h>
+#include <linux/uaccess.h>
#include <linux/module.h>
+#include <linux/mutex.h>
#include <linux/init.h>
+#include <linux/sort.h>
+#include <linux/cpu.h>
#include <linux/pci.h>
#include <linux/smp.h>
-#include <linux/cpu.h>
-#include <linux/mutex.h>
-#include <linux/sort.h>
+#include <asm/processor.h>
#include <asm/e820.h>
#include <asm/mtrr.h>
-#include <asm/uaccess.h>
-#include <asm/processor.h>
#include <asm/msr.h>
-#include <asm/kvm_para.h>
+
#include "mtrr.h"
-u32 num_var_ranges = 0;
+u32 num_var_ranges;
unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
static DEFINE_MUTEX(mtrr_mutex);
u64 size_or_mask, size_and_mask;
+static bool mtrr_aps_delayed_init;
-static struct mtrr_ops * mtrr_ops[X86_VENDOR_NUM] = {};
+static struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM];
-struct mtrr_ops * mtrr_if = NULL;
+struct mtrr_ops *mtrr_if;
static void set_mtrr(unsigned int reg, unsigned long base,
unsigned long size, mtrr_type type);
-void set_mtrr_ops(struct mtrr_ops * ops)
+void set_mtrr_ops(struct mtrr_ops *ops)
{
if (ops->vendor && ops->vendor < X86_VENDOR_NUM)
mtrr_ops[ops->vendor] = ops;
{
struct pci_dev *dev;
u8 rev;
-
- if ((dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL)) != NULL) {
- /* ServerWorks LE chipsets < rev 6 have problems with write-combining
- Don't allow it and leave room for other chipsets to be tagged */
+
+ dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL);
+ if (dev != NULL) {
+ /*
+ * ServerWorks LE chipsets < rev 6 have problems with
+ * write-combining. Don't allow it and leave room for other
+ * chipsets to be tagged
+ */
if (dev->vendor == PCI_VENDOR_ID_SERVERWORKS &&
dev->device == PCI_DEVICE_ID_SERVERWORKS_LE) {
pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev);
if (rev <= 5) {
- printk(KERN_INFO "mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n");
+ pr_info("mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n");
pci_dev_put(dev);
return 0;
}
}
- /* Intel 450NX errata # 23. Non ascending cacheline evictions to
- write combining memory may resulting in data corruption */
+ /*
+ * Intel 450NX errata # 23. Non ascending cacheline evictions to
+ * write combining memory may resulting in data corruption
+ */
if (dev->vendor == PCI_VENDOR_ID_INTEL &&
dev->device == PCI_DEVICE_ID_INTEL_82451NX) {
- printk(KERN_INFO "mtrr: Intel 450NX MMC detected. Write-combining disabled.\n");
+ pr_info("mtrr: Intel 450NX MMC detected. Write-combining disabled.\n");
pci_dev_put(dev);
return 0;
}
pci_dev_put(dev);
- }
- return (mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0);
+ }
+ return mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0;
}
/* This function returns the number of variable MTRRs */
{
unsigned long config = 0, dummy;
- if (use_intel()) {
+ if (use_intel())
rdmsr(MSR_MTRRcap, config, dummy);
- } else if (is_cpu(AMD))
+ else if (is_cpu(AMD))
config = 2;
else if (is_cpu(CYRIX) || is_cpu(CENTAUR))
config = 8;
+
num_var_ranges = config & 0xff;
}
mtrr_type smp_type;
};
+/**
+ * ipi_handler - Synchronisation handler. Executed by "other" CPUs.
+ *
+ * Returns nothing.
+ */
static void ipi_handler(void *info)
-/* [SUMMARY] Synchronisation handler. Executed by "other" CPUs.
- [RETURNS] Nothing.
-*/
{
#ifdef CONFIG_SMP
struct set_mtrr_data *data = info;
local_irq_save(flags);
atomic_dec(&data->count);
- while(!atomic_read(&data->gate))
+ while (!atomic_read(&data->gate))
cpu_relax();
/* The master has cleared me to execute */
- if (data->smp_reg != ~0U)
- mtrr_if->set(data->smp_reg, data->smp_base,
+ if (data->smp_reg != ~0U) {
+ mtrr_if->set(data->smp_reg, data->smp_base,
data->smp_size, data->smp_type);
- else
+ } else if (mtrr_aps_delayed_init) {
+ /*
+ * Initialize the MTRRs inaddition to the synchronisation.
+ */
mtrr_if->set_all();
+ }
atomic_dec(&data->count);
- while(atomic_read(&data->gate))
+ while (atomic_read(&data->gate))
cpu_relax();
atomic_dec(&data->count);
#endif
}
-static inline int types_compatible(mtrr_type type1, mtrr_type type2) {
+static inline int types_compatible(mtrr_type type1, mtrr_type type2)
+{
return type1 == MTRR_TYPE_UNCACHABLE ||
type2 == MTRR_TYPE_UNCACHABLE ||
(type1 == MTRR_TYPE_WRTHROUGH && type2 == MTRR_TYPE_WRBACK) ||
* @type: mtrr type
*
* This is kinda tricky, but fortunately, Intel spelled it out for us cleanly:
- *
+ *
* 1. Send IPI to do the following:
* 2. Disable Interrupts
- * 3. Wait for all procs to do so
+ * 3. Wait for all procs to do so
* 4. Enter no-fill cache mode
* 5. Flush caches
* 6. Clear PGE bit
* 10. Enable all range registers
* 11. Flush all TLBs and caches again
* 12. Enter normal cache mode and reenable caching
- * 13. Set PGE
+ * 13. Set PGE
* 14. Wait for buddies to catch up
* 15. Enable interrupts.
- *
+ *
* What does that mean for us? Well, first we set data.count to the number
* of CPUs. As each CPU disables interrupts, it'll decrement it once. We wait
* until it hits 0 and proceed. We set the data.gate flag and reset data.count.
- * Meanwhile, they are waiting for that flag to be set. Once it's set, each
- * CPU goes through the transition of updating MTRRs. The CPU vendors may each do it
- * differently, so we call mtrr_if->set() callback and let them take care of it.
- * When they're done, they again decrement data->count and wait for data.gate to
- * be reset.
- * When we finish, we wait for data.count to hit 0 and toggle the data.gate flag.
+ * Meanwhile, they are waiting for that flag to be set. Once it's set, each
+ * CPU goes through the transition of updating MTRRs.
+ * The CPU vendors may each do it differently,
+ * so we call mtrr_if->set() callback and let them take care of it.
+ * When they're done, they again decrement data->count and wait for data.gate
+ * to be reset.
+ * When we finish, we wait for data.count to hit 0 and toggle the data.gate flag
* Everyone then enables interrupts and we all continue on.
*
* Note that the mechanism is the same for UP systems, too; all the SMP stuff
* becomes nops.
*/
-static void set_mtrr(unsigned int reg, unsigned long base,
- unsigned long size, mtrr_type type)
+static void
+set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type)
{
struct set_mtrr_data data;
unsigned long flags;
data.smp_size = size;
data.smp_type = type;
atomic_set(&data.count, num_booting_cpus() - 1);
- /* make sure data.count is visible before unleashing other CPUs */
+
+ /* Make sure data.count is visible before unleashing other CPUs */
smp_wmb();
- atomic_set(&data.gate,0);
+ atomic_set(&data.gate, 0);
- /* Start the ball rolling on other CPUs */
+ /* Start the ball rolling on other CPUs */
if (smp_call_function(ipi_handler, &data, 0) != 0)
panic("mtrr: timed out waiting for other CPUs\n");
local_irq_save(flags);
- while(atomic_read(&data.count))
+ while (atomic_read(&data.count))
cpu_relax();
- /* ok, reset count and toggle gate */
+ /* Ok, reset count and toggle gate */
atomic_set(&data.count, num_booting_cpus() - 1);
smp_wmb();
- atomic_set(&data.gate,1);
+ atomic_set(&data.gate, 1);
- /* do our MTRR business */
+ /* Do our MTRR business */
- /* HACK!
+ /*
+ * HACK!
* We use this same function to initialize the mtrrs on boot.
* The state of the boot cpu's mtrrs has been saved, and we want
- * to replicate across all the APs.
+ * to replicate across all the APs.
* If we're doing that @reg is set to something special...
*/
- if (reg != ~0U)
- mtrr_if->set(reg,base,size,type);
+ if (reg != ~0U)
+ mtrr_if->set(reg, base, size, type);
+ else if (!mtrr_aps_delayed_init)
+ mtrr_if->set_all();
- /* wait for the others */
- while(atomic_read(&data.count))
+ /* Wait for the others */
+ while (atomic_read(&data.count))
cpu_relax();
atomic_set(&data.count, num_booting_cpus() - 1);
smp_wmb();
- atomic_set(&data.gate,0);
+ atomic_set(&data.gate, 0);
/*
* Wait here for everyone to have seen the gate change
* So we're the last ones to touch 'data'
*/
- while(atomic_read(&data.count))
+ while (atomic_read(&data.count))
cpu_relax();
local_irq_restore(flags);
}
/**
- * mtrr_add_page - Add a memory type region
- * @base: Physical base address of region in pages (in units of 4 kB!)
- * @size: Physical size of region in pages (4 kB)
- * @type: Type of MTRR desired
- * @increment: If this is true do usage counting on the region
+ * mtrr_add_page - Add a memory type region
+ * @base: Physical base address of region in pages (in units of 4 kB!)
+ * @size: Physical size of region in pages (4 kB)
+ * @type: Type of MTRR desired
+ * @increment: If this is true do usage counting on the region
*
- * Memory type region registers control the caching on newer Intel and
- * non Intel processors. This function allows drivers to request an
- * MTRR is added. The details and hardware specifics of each processor's
- * implementation are hidden from the caller, but nevertheless the
- * caller should expect to need to provide a power of two size on an
- * equivalent power of two boundary.
+ * Memory type region registers control the caching on newer Intel and
+ * non Intel processors. This function allows drivers to request an
+ * MTRR is added. The details and hardware specifics of each processor's
+ * implementation are hidden from the caller, but nevertheless the
+ * caller should expect to need to provide a power of two size on an
+ * equivalent power of two boundary.
*
- * If the region cannot be added either because all regions are in use
- * or the CPU cannot support it a negative value is returned. On success
- * the register number for this entry is returned, but should be treated
- * as a cookie only.
+ * If the region cannot be added either because all regions are in use
+ * or the CPU cannot support it a negative value is returned. On success
+ * the register number for this entry is returned, but should be treated
+ * as a cookie only.
*
- * On a multiprocessor machine the changes are made to all processors.
- * This is required on x86 by the Intel processors.
+ * On a multiprocessor machine the changes are made to all processors.
+ * This is required on x86 by the Intel processors.
*
- * The available types are
+ * The available types are
*
- * %MTRR_TYPE_UNCACHABLE - No caching
+ * %MTRR_TYPE_UNCACHABLE - No caching
*
- * %MTRR_TYPE_WRBACK - Write data back in bursts whenever
+ * %MTRR_TYPE_WRBACK - Write data back in bursts whenever
*
- * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
+ * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
*
- * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes
+ * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes
*
- * BUGS: Needs a quiet flag for the cases where drivers do not mind
- * failures and do not wish system log messages to be sent.
+ * BUGS: Needs a quiet flag for the cases where drivers do not mind
+ * failures and do not wish system log messages to be sent.
*/
-
-int mtrr_add_page(unsigned long base, unsigned long size,
+int mtrr_add_page(unsigned long base, unsigned long size,
unsigned int type, bool increment)
{
+ unsigned long lbase, lsize;
int i, replace, error;
mtrr_type ltype;
- unsigned long lbase, lsize;
if (!mtrr_if)
return -ENXIO;
-
- if ((error = mtrr_if->validate_add_page(base,size,type)))
+
+ error = mtrr_if->validate_add_page(base, size, type);
+ if (error)
return error;
if (type >= MTRR_NUM_TYPES) {
- printk(KERN_WARNING "mtrr: type: %u invalid\n", type);
+ pr_warning("mtrr: type: %u invalid\n", type);
return -EINVAL;
}
- /* If the type is WC, check that this processor supports it */
+ /* If the type is WC, check that this processor supports it */
if ((type == MTRR_TYPE_WRCOMB) && !have_wrcomb()) {
- printk(KERN_WARNING
- "mtrr: your processor doesn't support write-combining\n");
+ pr_warning("mtrr: your processor doesn't support write-combining\n");
return -ENOSYS;
}
if (!size) {
- printk(KERN_WARNING "mtrr: zero sized request\n");
+ pr_warning("mtrr: zero sized request\n");
return -EINVAL;
}
if (base & size_or_mask || size & size_or_mask) {
- printk(KERN_WARNING "mtrr: base or size exceeds the MTRR width\n");
+ pr_warning("mtrr: base or size exceeds the MTRR width\n");
return -EINVAL;
}
/* No CPU hotplug when we change MTRR entries */
get_online_cpus();
- /* Search for existing MTRR */
+
+ /* Search for existing MTRR */
mutex_lock(&mtrr_mutex);
for (i = 0; i < num_var_ranges; ++i) {
mtrr_if->get(i, &lbase, &lsize, <ype);
- if (!lsize || base > lbase + lsize - 1 || base + size - 1 < lbase)
+ if (!lsize || base > lbase + lsize - 1 ||
+ base + size - 1 < lbase)
continue;
- /* At this point we know there is some kind of overlap/enclosure */
+ /*
+ * At this point we know there is some kind of
+ * overlap/enclosure
+ */
if (base < lbase || base + size - 1 > lbase + lsize - 1) {
- if (base <= lbase && base + size - 1 >= lbase + lsize - 1) {
+ if (base <= lbase &&
+ base + size - 1 >= lbase + lsize - 1) {
/* New region encloses an existing region */
if (type == ltype) {
replace = replace == -1 ? i : -2;
continue;
- }
- else if (types_compatible(type, ltype))
+ } else if (types_compatible(type, ltype))
continue;
}
- printk(KERN_WARNING
- "mtrr: 0x%lx000,0x%lx000 overlaps existing"
- " 0x%lx000,0x%lx000\n", base, size, lbase,
- lsize);
+ pr_warning("mtrr: 0x%lx000,0x%lx000 overlaps existing"
+ " 0x%lx000,0x%lx000\n", base, size, lbase,
+ lsize);
goto out;
}
- /* New region is enclosed by an existing region */
+ /* New region is enclosed by an existing region */
if (ltype != type) {
if (types_compatible(type, ltype))
continue;
- printk (KERN_WARNING "mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n",
- base, size, mtrr_attrib_to_str(ltype),
- mtrr_attrib_to_str(type));
+ pr_warning("mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n",
+ base, size, mtrr_attrib_to_str(ltype),
+ mtrr_attrib_to_str(type));
goto out;
}
if (increment)
error = i;
goto out;
}
- /* Search for an empty MTRR */
+ /* Search for an empty MTRR */
i = mtrr_if->get_free_region(base, size, replace);
if (i >= 0) {
set_mtrr(i, base, size, type);
mtrr_usage_table[replace] = 0;
}
}
- } else
- printk(KERN_INFO "mtrr: no more MTRRs available\n");
+ } else {
+ pr_info("mtrr: no more MTRRs available\n");
+ }
error = i;
out:
mutex_unlock(&mtrr_mutex);
static int mtrr_check(unsigned long base, unsigned long size)
{
if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
- printk(KERN_WARNING
- "mtrr: size and base must be multiples of 4 kiB\n");
- printk(KERN_DEBUG
- "mtrr: size: 0x%lx base: 0x%lx\n", size, base);
+ pr_warning("mtrr: size and base must be multiples of 4 kiB\n");
+ pr_debug("mtrr: size: 0x%lx base: 0x%lx\n", size, base);
dump_stack();
return -1;
}
}
/**
- * mtrr_add - Add a memory type region
- * @base: Physical base address of region
- * @size: Physical size of region
- * @type: Type of MTRR desired
- * @increment: If this is true do usage counting on the region
+ * mtrr_add - Add a memory type region
+ * @base: Physical base address of region
+ * @size: Physical size of region
+ * @type: Type of MTRR desired
+ * @increment: If this is true do usage counting on the region
*
- * Memory type region registers control the caching on newer Intel and
- * non Intel processors. This function allows drivers to request an
- * MTRR is added. The details and hardware specifics of each processor's
- * implementation are hidden from the caller, but nevertheless the
- * caller should expect to need to provide a power of two size on an
- * equivalent power of two boundary.
+ * Memory type region registers control the caching on newer Intel and
+ * non Intel processors. This function allows drivers to request an
+ * MTRR is added. The details and hardware specifics of each processor's
+ * implementation are hidden from the caller, but nevertheless the
+ * caller should expect to need to provide a power of two size on an
+ * equivalent power of two boundary.
*
- * If the region cannot be added either because all regions are in use
- * or the CPU cannot support it a negative value is returned. On success
- * the register number for this entry is returned, but should be treated
- * as a cookie only.
+ * If the region cannot be added either because all regions are in use
+ * or the CPU cannot support it a negative value is returned. On success
+ * the register number for this entry is returned, but should be treated
+ * as a cookie only.
*
- * On a multiprocessor machine the changes are made to all processors.
- * This is required on x86 by the Intel processors.
+ * On a multiprocessor machine the changes are made to all processors.
+ * This is required on x86 by the Intel processors.
*
- * The available types are
+ * The available types are
*
- * %MTRR_TYPE_UNCACHABLE - No caching
+ * %MTRR_TYPE_UNCACHABLE - No caching
*
- * %MTRR_TYPE_WRBACK - Write data back in bursts whenever
+ * %MTRR_TYPE_WRBACK - Write data back in bursts whenever
*
- * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
+ * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
*
- * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes
+ * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes
*
- * BUGS: Needs a quiet flag for the cases where drivers do not mind
- * failures and do not wish system log messages to be sent.
+ * BUGS: Needs a quiet flag for the cases where drivers do not mind
+ * failures and do not wish system log messages to be sent.
*/
-
-int
-mtrr_add(unsigned long base, unsigned long size, unsigned int type,
- bool increment)
+int mtrr_add(unsigned long base, unsigned long size, unsigned int type,
+ bool increment)
{
if (mtrr_check(base, size))
return -EINVAL;
return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
increment);
}
+EXPORT_SYMBOL(mtrr_add);
/**
- * mtrr_del_page - delete a memory type region
- * @reg: Register returned by mtrr_add
- * @base: Physical base address
- * @size: Size of region
+ * mtrr_del_page - delete a memory type region
+ * @reg: Register returned by mtrr_add
+ * @base: Physical base address
+ * @size: Size of region
*
- * If register is supplied then base and size are ignored. This is
- * how drivers should call it.
+ * If register is supplied then base and size are ignored. This is
+ * how drivers should call it.
*
- * Releases an MTRR region. If the usage count drops to zero the
- * register is freed and the region returns to default state.
- * On success the register is returned, on failure a negative error
- * code.
+ * Releases an MTRR region. If the usage count drops to zero the
+ * register is freed and the region returns to default state.
+ * On success the register is returned, on failure a negative error
+ * code.
*/
-
int mtrr_del_page(int reg, unsigned long base, unsigned long size)
{
int i, max;
}
}
if (reg < 0) {
- printk(KERN_DEBUG "mtrr: no MTRR for %lx000,%lx000 found\n", base,
- size);
+ pr_debug("mtrr: no MTRR for %lx000,%lx000 found\n",
+ base, size);
goto out;
}
}
if (reg >= max) {
- printk(KERN_WARNING "mtrr: register: %d too big\n", reg);
+ pr_warning("mtrr: register: %d too big\n", reg);
goto out;
}
mtrr_if->get(reg, &lbase, &lsize, <ype);
if (lsize < 1) {
- printk(KERN_WARNING "mtrr: MTRR %d not used\n", reg);
+ pr_warning("mtrr: MTRR %d not used\n", reg);
goto out;
}
if (mtrr_usage_table[reg] < 1) {
- printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg);
+ pr_warning("mtrr: reg: %d has count=0\n", reg);
goto out;
}
if (--mtrr_usage_table[reg] < 1)
put_online_cpus();
return error;
}
+
/**
- * mtrr_del - delete a memory type region
- * @reg: Register returned by mtrr_add
- * @base: Physical base address
- * @size: Size of region
+ * mtrr_del - delete a memory type region
+ * @reg: Register returned by mtrr_add
+ * @base: Physical base address
+ * @size: Size of region
*
- * If register is supplied then base and size are ignored. This is
- * how drivers should call it.
+ * If register is supplied then base and size are ignored. This is
+ * how drivers should call it.
*
- * Releases an MTRR region. If the usage count drops to zero the
- * register is freed and the region returns to default state.
- * On success the register is returned, on failure a negative error
- * code.
+ * Releases an MTRR region. If the usage count drops to zero the
+ * register is freed and the region returns to default state.
+ * On success the register is returned, on failure a negative error
+ * code.
*/
-
-int
-mtrr_del(int reg, unsigned long base, unsigned long size)
+int mtrr_del(int reg, unsigned long base, unsigned long size)
{
if (mtrr_check(base, size))
return -EINVAL;
return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT);
}
-
-EXPORT_SYMBOL(mtrr_add);
EXPORT_SYMBOL(mtrr_del);
-/* HACK ALERT!
+/*
+ * HACK ALERT!
* These should be called implicitly, but we can't yet until all the initcall
* stuff is done...
*/
static struct mtrr_value mtrr_value[MTRR_MAX_VAR_RANGES];
-static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
+static int mtrr_save(struct sys_device *sysdev, pm_message_t state)
{
int i;
for (i = 0; i < num_var_ranges; i++) {
- mtrr_if->get(i,
- &mtrr_value[i].lbase,
- &mtrr_value[i].lsize,
- &mtrr_value[i].ltype);
+ mtrr_if->get(i, &mtrr_value[i].lbase,
+ &mtrr_value[i].lsize,
+ &mtrr_value[i].ltype);
}
return 0;
}
-static int mtrr_restore(struct sys_device * sysdev)
+static int mtrr_restore(struct sys_device *sysdev)
{
int i;
for (i = 0; i < num_var_ranges; i++) {
- if (mtrr_value[i].lsize)
- set_mtrr(i,
- mtrr_value[i].lbase,
- mtrr_value[i].lsize,
- mtrr_value[i].ltype);
+ if (mtrr_value[i].lsize) {
+ set_mtrr(i, mtrr_value[i].lbase,
+ mtrr_value[i].lsize,
+ mtrr_value[i].ltype);
+ }
}
return 0;
}
/**
* mtrr_bp_init - initialize mtrrs on the boot CPU
*
- * This needs to be called early; before any of the other CPUs are
+ * This needs to be called early; before any of the other CPUs are
* initialized (i.e. before smp_init()).
- *
+ *
*/
void __init mtrr_bp_init(void)
{
u32 phys_addr;
+
init_ifs();
phys_addr = 32;
if (cpu_has_mtrr) {
mtrr_if = &generic_mtrr_ops;
- size_or_mask = 0xff000000; /* 36 bits */
+ size_or_mask = 0xff000000; /* 36 bits */
size_and_mask = 0x00f00000;
phys_addr = 36;
- /* This is an AMD specific MSR, but we assume(hope?) that
- Intel will implement it to when they extend the address
- bus of the Xeon. */
+ /*
+ * This is an AMD specific MSR, but we assume(hope?) that
+ * Intel will implement it to when they extend the address
+ * bus of the Xeon.
+ */
if (cpuid_eax(0x80000000) >= 0x80000008) {
phys_addr = cpuid_eax(0x80000008) & 0xff;
/* CPUID workaround for Intel 0F33/0F34 CPU */
size_and_mask = ~size_or_mask & 0xfffff00000ULL;
} else if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR &&
boot_cpu_data.x86 == 6) {
- /* VIA C* family have Intel style MTRRs, but
- don't support PAE */
- size_or_mask = 0xfff00000; /* 32 bits */
+ /*
+ * VIA C* family have Intel style MTRRs,
+ * but don't support PAE
+ */
+ size_or_mask = 0xfff00000; /* 32 bits */
size_and_mask = 0;
phys_addr = 32;
}
changed_by_mtrr_cleanup = 1;
mtrr_if->set_all();
}
-
}
}
}
void mtrr_ap_init(void)
{
- unsigned long flags;
-
- if (!mtrr_if || !use_intel())
+ if (!use_intel() || mtrr_aps_delayed_init)
return;
/*
- * Ideally we should hold mtrr_mutex here to avoid mtrr entries changed,
- * but this routine will be called in cpu boot time, holding the lock
- * breaks it. This routine is called in two cases: 1.very earily time
- * of software resume, when there absolutely isn't mtrr entry changes;
- * 2.cpu hotadd time. We let mtrr_add/del_page hold cpuhotplug lock to
- * prevent mtrr entry changes
+ * Ideally we should hold mtrr_mutex here to avoid mtrr entries
+ * changed, but this routine will be called in cpu boot time,
+ * holding the lock breaks it.
+ *
+ * This routine is called in two cases:
+ *
+ * 1. very earily time of software resume, when there absolutely
+ * isn't mtrr entry changes;
+ *
+ * 2. cpu hotadd time. We let mtrr_add/del_page hold cpuhotplug
+ * lock to prevent mtrr entry changes
*/
- local_irq_save(flags);
-
- mtrr_if->set_all();
-
- local_irq_restore(flags);
+ set_mtrr(~0U, 0, 0, 0);
}
/**
smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1);
}
+void set_mtrr_aps_delayed_init(void)
+{
+ if (!use_intel())
+ return;
+
+ mtrr_aps_delayed_init = true;
+}
+
+/*
+ * MTRR initialization for all AP's
+ */
+void mtrr_aps_init(void)
+{
+ if (!use_intel())
+ return;
+
+ set_mtrr(~0U, 0, 0, 0);
+ mtrr_aps_delayed_init = false;
+}
+
+void mtrr_bp_restore(void)
+{
+ if (!use_intel())
+ return;
+
+ mtrr_if->set_all();
+}
+
static int __init mtrr_init_finialize(void)
{
if (!mtrr_if)
return 0;
+
if (use_intel()) {
if (!changed_by_mtrr_cleanup)
mtrr_state_warn();
- } else {
- /* The CPUs haven't MTRR and seem to not support SMP. They have
- * specific drivers, we use a tricky method to support
- * suspend/resume for them.
- * TBD: is there any system with such CPU which supports
- * suspend/resume? if no, we should remove the code.
- */
- sysdev_driver_register(&cpu_sysdev_class,
- &mtrr_sysdev_driver);
+ return 0;
}
+
+ /*
+ * The CPU has no MTRR and seems to not support SMP. They have
+ * specific drivers, we use a tricky method to support
+ * suspend/resume for them.
+ *
+ * TBD: is there any system with such CPU which supports
+ * suspend/resume? If no, we should remove the code.
+ */
+ sysdev_driver_register(&cpu_sysdev_class, &mtrr_sysdev_driver);
+
return 0;
}
subsys_initcall(mtrr_init_finialize);
/*
- * local mtrr defines.
+ * local MTRR defines.
*/
#include <linux/types.h>
struct mtrr_ops {
u32 vendor;
u32 use_intel_if;
-// void (*init)(void);
void (*set)(unsigned int reg, unsigned long base,
unsigned long size, mtrr_type type);
void (*set_all)(void);
void (*get)(unsigned int reg, unsigned long *base,
- unsigned long *size, mtrr_type * type);
+ unsigned long *size, mtrr_type *type);
int (*get_free_region)(unsigned long base, unsigned long size,
int replace_reg);
int (*validate_add_page)(unsigned long base, unsigned long size,
/* library functions for processor-specific routines */
struct set_mtrr_context {
- unsigned long flags;
- unsigned long cr4val;
- u32 deftype_lo;
- u32 deftype_hi;
- u32 ccr3;
+ unsigned long flags;
+ unsigned long cr4val;
+ u32 deftype_lo;
+ u32 deftype_hi;
+ u32 ccr3;
};
void set_mtrr_done(struct set_mtrr_context *ctxt);
u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi);
void get_mtrr_state(void);
-extern void set_mtrr_ops(struct mtrr_ops * ops);
+extern void set_mtrr_ops(struct mtrr_ops *ops);
extern u64 size_or_mask, size_and_mask;
-extern struct mtrr_ops * mtrr_if;
+extern struct mtrr_ops *mtrr_if;
#define is_cpu(vnd) (mtrr_if && mtrr_if->vendor == X86_VENDOR_##vnd)
#define use_intel() (mtrr_if && mtrr_if->use_intel_if == 1)
-#include <linux/mm.h>
#include <linux/init.h>
-#include <asm/io.h>
-#include <asm/mtrr.h>
-#include <asm/msr.h>
+#include <linux/io.h>
+#include <linux/mm.h>
+
#include <asm/processor-cyrix.h>
#include <asm/processor-flags.h>
-#include "mtrr.h"
+#include <asm/mtrr.h>
+#include <asm/msr.h>
+#include "mtrr.h"
-/* Put the processor into a state where MTRRs can be safely set */
+/* Put the processor into a state where MTRRs can be safely set */
void set_mtrr_prepare_save(struct set_mtrr_context *ctxt)
{
unsigned int cr0;
- /* Disable interrupts locally */
+ /* Disable interrupts locally */
local_irq_save(ctxt->flags);
if (use_intel() || is_cpu(CYRIX)) {
- /* Save value of CR4 and clear Page Global Enable (bit 7) */
+ /* Save value of CR4 and clear Page Global Enable (bit 7) */
if (cpu_has_pge) {
ctxt->cr4val = read_cr4();
write_cr4(ctxt->cr4val & ~X86_CR4_PGE);
write_cr0(cr0);
wbinvd();
- if (use_intel())
- /* Save MTRR state */
+ if (use_intel()) {
+ /* Save MTRR state */
rdmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi);
- else
- /* Cyrix ARRs - everything else were excluded at the top */
+ } else {
+ /*
+ * Cyrix ARRs -
+ * everything else were excluded at the top
+ */
ctxt->ccr3 = getCx86(CX86_CCR3);
+ }
}
}
void set_mtrr_cache_disable(struct set_mtrr_context *ctxt)
{
- if (use_intel())
- /* Disable MTRRs, and set the default type to uncached */
+ if (use_intel()) {
+ /* Disable MTRRs, and set the default type to uncached */
mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo & 0xf300UL,
ctxt->deftype_hi);
- else if (is_cpu(CYRIX))
- /* Cyrix ARRs - everything else were excluded at the top */
- setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10);
+ } else {
+ if (is_cpu(CYRIX)) {
+ /* Cyrix ARRs - everything else were excluded at the top */
+ setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10);
+ }
+ }
}
-/* Restore the processor after a set_mtrr_prepare */
+/* Restore the processor after a set_mtrr_prepare */
void set_mtrr_done(struct set_mtrr_context *ctxt)
{
if (use_intel() || is_cpu(CYRIX)) {
- /* Flush caches and TLBs */
+ /* Flush caches and TLBs */
wbinvd();
- /* Restore MTRRdefType */
- if (use_intel())
+ /* Restore MTRRdefType */
+ if (use_intel()) {
/* Intel (P6) standard MTRRs */
- mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi);
- else
- /* Cyrix ARRs - everything else was excluded at the top */
+ mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo,
+ ctxt->deftype_hi);
+ } else {
+ /*
+ * Cyrix ARRs -
+ * everything else was excluded at the top
+ */
setCx86(CX86_CCR3, ctxt->ccr3);
+ }
- /* Enable caches */
+ /* Enable caches */
write_cr0(read_cr0() & 0xbfffffff);
- /* Restore value of CR4 */
+ /* Restore value of CR4 */
if (cpu_has_pge)
write_cr4(ctxt->cr4val);
}
- /* Re-enable interrupts locally (if enabled previously) */
+ /* Re-enable interrupts locally (if enabled previously) */
local_irq_restore(ctxt->flags);
}
-
/* returns the bit offset of the performance counter register */
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_AMD:
- return (msr - MSR_K7_PERFCTR0);
+ return msr - MSR_K7_PERFCTR0;
case X86_VENDOR_INTEL:
if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
- return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+ return msr - MSR_ARCH_PERFMON_PERFCTR0;
switch (boot_cpu_data.x86) {
case 6:
- return (msr - MSR_P6_PERFCTR0);
+ return msr - MSR_P6_PERFCTR0;
case 15:
- return (msr - MSR_P4_BPU_PERFCTR0);
+ return msr - MSR_P4_BPU_PERFCTR0;
}
}
return 0;
/* returns the bit offset of the event selection register */
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_AMD:
- return (msr - MSR_K7_EVNTSEL0);
+ return msr - MSR_K7_EVNTSEL0;
case X86_VENDOR_INTEL:
if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
- return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+ return msr - MSR_ARCH_PERFMON_EVENTSEL0;
switch (boot_cpu_data.x86) {
case 6:
- return (msr - MSR_P6_EVNTSEL0);
+ return msr - MSR_P6_EVNTSEL0;
case 15:
- return (msr - MSR_P4_BSU_ESCR0);
+ return msr - MSR_P4_BSU_ESCR0;
}
}
return 0;
{
BUG_ON(counter > NMI_MAX_COUNTER_BITS);
- return (!test_bit(counter, perfctr_nmi_owner));
+ return !test_bit(counter, perfctr_nmi_owner);
}
/* checks the an msr for availability */
counter = nmi_perfctr_msr_to_bit(msr);
BUG_ON(counter > NMI_MAX_COUNTER_BITS);
- return (!test_bit(counter, perfctr_nmi_owner));
+ return !test_bit(counter, perfctr_nmi_owner);
}
EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
*/
counter_val = (u64)cpu_khz * 1000;
do_div(counter_val, retval);
- if (counter_val > 0x7fffffffULL) {
+ if (counter_val > 0x7fffffffULL) {
u64 count = (u64)cpu_khz * 1000;
do_div(count, 0x7fffffffUL);
retval = count + 1;
u64 count = (u64)cpu_khz * 1000;
do_div(count, nmi_hz);
- if(descr)
+ if (descr)
pr_debug("setting %s to -0x%08Lx\n", descr, count);
wrmsrl(perfctr_msr, 0 - count);
}
u64 count = (u64)cpu_khz * 1000;
do_div(count, nmi_hz);
- if(descr)
+ if (descr)
pr_debug("setting %s to -0x%08Lx\n", descr, count);
wrmsr(perfctr_msr, (u32)(-count), 0);
}
/* setup the timer */
wrmsr(evntsel_msr, evntsel, 0);
- write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz);
+ write_watchdog_counter(perfctr_msr, "K7_PERFCTR0", nmi_hz);
/* initialize the wd struct before enabling */
wd->perfctr_msr = perfctr_msr;
/* setup the timer */
wrmsr(evntsel_msr, evntsel, 0);
nmi_hz = adjust_for_32bit_ctr(nmi_hz);
- write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz);
+ write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0", nmi_hz);
/* initialize the wd struct before enabling */
wd->perfctr_msr = perfctr_msr;
apic_write(APIC_LVTPC, APIC_DM_NMI);
/* P6/ARCH_PERFMON has 32 bit counter write */
- write_watchdog_counter32(wd->perfctr_msr, NULL,nmi_hz);
+ write_watchdog_counter32(wd->perfctr_msr, NULL, nmi_hz);
}
static const struct wd_ops p6_wd_ops = {
if (smp_num_siblings == 2) {
unsigned int ebx, apicid;
- ebx = cpuid_ebx(1);
- apicid = (ebx >> 24) & 0xff;
- ht_num = apicid & 1;
+ ebx = cpuid_ebx(1);
+ apicid = (ebx >> 24) & 0xff;
+ ht_num = apicid & 1;
} else
#endif
ht_num = 0;
}
evntsel = P4_ESCR_EVENT_SELECT(0x3F)
- | P4_ESCR_OS
+ | P4_ESCR_OS
| P4_ESCR_USR;
cccr_val |= P4_CCCR_THRESHOLD(15)
{
unsigned dummy;
/*
- * P4 quirks:
+ * P4 quirks:
* - An overflown perfctr will assert its interrupt
* until the OVF flag in its CCCR is cleared.
* - LVTPC is masked on interrupt and must be
* NOTE: Corresponding bit = 0 in ebx indicates event present.
*/
cpuid(10, &(eax.full), &ebx, &unused, &unused);
- if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+ if ((eax.split.mask_length <
+ (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
(ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
return 0;
if (i < ARRAY_SIZE(x86_power_flags) &&
x86_power_flags[i])
seq_printf(m, "%s%s",
- x86_power_flags[i][0]?" ":"",
+ x86_power_flags[i][0] ? " " : "",
x86_power_flags[i]);
else
seq_printf(m, " [%d]", i);
static unsigned long __vmware_get_tsc_khz(void)
{
- uint64_t tsc_hz;
- uint32_t eax, ebx, ecx, edx;
+ uint64_t tsc_hz;
+ uint32_t eax, ebx, ecx, edx;
- VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
+ VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
- if (ebx == UINT_MAX)
- return 0;
- tsc_hz = eax | (((uint64_t)ebx) << 32);
- do_div(tsc_hz, 1000);
- BUG_ON(tsc_hz >> 32);
- return tsc_hz;
+ if (ebx == UINT_MAX)
+ return 0;
+ tsc_hz = eax | (((uint64_t)ebx) << 32);
+ do_div(tsc_hz, 1000);
+ BUG_ON(tsc_hz >> 32);
+ return tsc_hz;
}
/*
bts_escape = ((unsigned long)-1 & ~bts_qual_mask)
};
-static inline unsigned long bts_get(const char *base, enum bts_field field)
+static inline unsigned long bts_get(const char *base, unsigned long field)
{
base += (ds_cfg.sizeof_ptr_field * field);
return *(unsigned long *)base;
}
-static inline void bts_set(char *base, enum bts_field field, unsigned long val)
+static inline void bts_set(char *base, unsigned long field, unsigned long val)
{
- base += (ds_cfg.sizeof_ptr_field * field);;
+ base += (ds_cfg.sizeof_ptr_field * field);
(*(unsigned long *)base) = val;
}
#include <linux/bug.h>
#include <linux/nmi.h>
#include <linux/sysfs.h>
-#include <linux/ftrace.h>
#include <asm/stacktrace.h>
void fixup_irqs(void)
{
unsigned int irq;
- static int warned;
struct irq_desc *desc;
for_each_irq_desc(irq, desc) {
}
if (desc->chip->set_affinity)
desc->chip->set_affinity(irq, affinity);
- else if (desc->action && !(warned++))
- printk("Cannot set affinity for irq %i\n", irq);
+ else if (desc->action)
+ printk_once("Cannot set affinity for irq %i\n", irq);
}
#if 0
printk(KERN_INFO "Command line: %s\n", boot_command_line);
#endif
+ strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
+ *cmdline_p = command_line;
+
+#ifdef CONFIG_X86_64
+ /*
+ * Must call this twice: Once just to detect whether hardware doesn't
+ * support NX (so that the early EHCI debug console setup can safely
+ * call set_fixmap(), and then again after parsing early parameters to
+ * honor the respective command line option.
+ */
+ check_efer();
+#endif
+
+ parse_early_param();
+
/* VMI may relocate the fixmap; do this before touching ioremap area */
vmi_init();
#endif
#endif
- strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
- *cmdline_p = command_line;
-
- parse_early_param();
-
#ifdef CONFIG_X86_64
check_efer();
#endif
if (is_uv_system())
uv_system_init();
+
+ set_mtrr_aps_delayed_init();
out:
preempt_enable();
}
+
+void arch_enable_nonboot_cpus_begin(void)
+{
+ set_mtrr_aps_delayed_init();
+}
+
+void arch_enable_nonboot_cpus_end(void)
+{
+ mtrr_aps_init();
+}
+
/*
* Early setup to make printk work.
*/
setup_ioapic_dest();
#endif
check_nmi_watchdog();
+ mtrr_aps_init();
}
static int __initdata setup_possible_cpus = -1;
* F0 0F bug workaround.. We have a special link segment
* for this.
*/
-gate_desc idt_table[256]
+gate_desc idt_table[NR_VECTORS]
__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
#endif
unsigned int bytes,
struct kvm_vcpu *vcpu)
{
- static int reported;
-
- if (!reported) {
- reported = 1;
- printk(KERN_WARNING "kvm: emulating exchange as write\n");
- }
+ printk_once(KERN_WARNING "kvm: emulating exchange as write\n");
#ifndef CONFIG_X86_64
/* guests cmpxchg8b have to be emulated atomically */
if (bytes == 8) {
#include <linux/module.h>
#include <linux/highmem.h>
-int is_io_mapping_possible(resource_size_t base, unsigned long size)
+static int is_io_mapping_possible(resource_size_t base, unsigned long size)
{
#if !defined(CONFIG_X86_PAE) && defined(CONFIG_PHYS_ADDR_T_64BIT)
/* There is no way to map greater than 1 << 32 address without PAE */
#endif
return 1;
}
-EXPORT_SYMBOL_GPL(is_io_mapping_possible);
+
+int iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot)
+{
+ unsigned long flag = _PAGE_CACHE_WC;
+ int ret;
+
+ if (!is_io_mapping_possible(base, size))
+ return -EINVAL;
+
+ ret = io_reserve_memtype(base, base + size, &flag);
+ if (ret)
+ return ret;
+
+ *prot = __pgprot(__PAGE_KERNEL | flag);
+ return 0;
+}
+EXPORT_SYMBOL_GPL(iomap_create_wc);
+
+void
+iomap_free(resource_size_t base, unsigned long size)
+{
+ io_free_memtype(base, base + size);
+}
+EXPORT_SYMBOL_GPL(iomap_free);
void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
{
retval = reserve_memtype(phys_addr, (u64)phys_addr + size,
prot_val, &new_prot_val);
if (retval) {
- pr_debug("Warning: reserve_memtype returned %d\n", retval);
+ printk(KERN_ERR "ioremap reserve_memtype failed %d\n", retval);
return NULL;
}
if (prot_val != new_prot_val) {
- /*
- * Do not fallback to certain memory types with certain
- * requested type:
- * - request is uc-, return cannot be write-back
- * - request is uc-, return cannot be write-combine
- * - request is write-combine, return cannot be write-back
- */
- if ((prot_val == _PAGE_CACHE_UC_MINUS &&
- (new_prot_val == _PAGE_CACHE_WB ||
- new_prot_val == _PAGE_CACHE_WC)) ||
- (prot_val == _PAGE_CACHE_WC &&
- new_prot_val == _PAGE_CACHE_WB)) {
- pr_debug(
+ if (!is_new_memtype_allowed(phys_addr, size,
+ prot_val, new_prot_val)) {
+ printk(KERN_ERR
"ioremap error for 0x%llx-0x%llx, requested 0x%lx, got 0x%lx\n",
(unsigned long long)phys_addr,
(unsigned long long)(phys_addr + size),
{
struct cpa_data cpa;
int ret, cache, checkalias;
+ unsigned long baddr = 0;
/*
* Check, if we are requested to change a not supported
*/
WARN_ON_ONCE(1);
}
+ /*
+ * Save address for cache flush. *addr is modified in the call
+ * to __change_page_attr_set_clr() below.
+ */
+ baddr = *addr;
}
/* Must avoid aliasing mappings in the highmem code */
cpa_flush_array(addr, numpages, cache,
cpa.flags, pages);
} else
- cpa_flush_range(*addr, numpages, cache);
+ cpa_flush_range(baddr, numpages, cache);
} else
cpa_flush_all(cache);
#include <linux/gfp.h>
#include <linux/mm.h>
#include <linux/fs.h>
+#include <linux/rbtree.h>
#include <asm/cacheflush.h>
#include <asm/processor.h>
* areas). All the aliases have the same cache attributes of course.
* Zero attributes are represented as holes.
*
- * Currently the data structure is a list because the number of mappings
- * are expected to be relatively small. If this should be a problem
- * it could be changed to a rbtree or similar.
+ * The data structure is a list that is also organized as an rbtree
+ * sorted on the start address of memtype range.
*
- * memtype_lock protects the whole list.
+ * memtype_lock protects both the linear list and rbtree.
*/
struct memtype {
u64 end;
unsigned long type;
struct list_head nd;
+ struct rb_node rb;
};
+static struct rb_root memtype_rbroot = RB_ROOT;
static LIST_HEAD(memtype_list);
static DEFINE_SPINLOCK(memtype_lock); /* protects memtype list */
+static struct memtype *memtype_rb_search(struct rb_root *root, u64 start)
+{
+ struct rb_node *node = root->rb_node;
+ struct memtype *last_lower = NULL;
+
+ while (node) {
+ struct memtype *data = container_of(node, struct memtype, rb);
+
+ if (data->start < start) {
+ last_lower = data;
+ node = node->rb_right;
+ } else if (data->start > start) {
+ node = node->rb_left;
+ } else
+ return data;
+ }
+
+ /* Will return NULL if there is no entry with its start <= start */
+ return last_lower;
+}
+
+static void memtype_rb_insert(struct rb_root *root, struct memtype *data)
+{
+ struct rb_node **new = &(root->rb_node);
+ struct rb_node *parent = NULL;
+
+ while (*new) {
+ struct memtype *this = container_of(*new, struct memtype, rb);
+
+ parent = *new;
+ if (data->start <= this->start)
+ new = &((*new)->rb_left);
+ else if (data->start > this->start)
+ new = &((*new)->rb_right);
+ }
+
+ rb_link_node(&data->rb, parent, new);
+ rb_insert_color(&data->rb, root);
+}
+
/*
* Does intersection of PAT memory type and MTRR memory type and returns
* the resulting memory type as PAT understands it.
return -EBUSY;
}
-static struct memtype *cached_entry;
-static u64 cached_start;
-
static int pat_pagerange_is_ram(unsigned long start, unsigned long end)
{
int ram_page = 0, not_rampage = 0;
}
/*
- * For RAM pages, mark the pages as non WB memory type using
- * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or
- * set_memory_wc() on a RAM page at a time before marking it as WB again.
- * This is ok, because only one driver will be owning the page and
- * doing set_memory_*() calls.
+ * For RAM pages, we use page flags to mark the pages with appropriate type.
+ * Here we do two pass:
+ * - Find the memtype of all the pages in the range, look for any conflicts
+ * - In case of no conflicts, set the new memtype for pages in the range
*
- * For now, we use PageNonWB to track that the RAM page is being mapped
- * as non WB. In future, we will have to use one more flag
- * (or some other mechanism in page_struct) to distinguish between
- * UC and WC mapping.
+ * Caller must hold memtype_lock for atomicity.
*/
static int reserve_ram_pages_type(u64 start, u64 end, unsigned long req_type,
unsigned long *new_type)
{
struct page *page;
- u64 pfn, end_pfn;
+ u64 pfn;
+
+ if (req_type == _PAGE_CACHE_UC) {
+ /* We do not support strong UC */
+ WARN_ON_ONCE(1);
+ req_type = _PAGE_CACHE_UC_MINUS;
+ }
for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) {
- page = pfn_to_page(pfn);
- if (page_mapped(page) || PageNonWB(page))
- goto out;
+ unsigned long type;
- SetPageNonWB(page);
+ page = pfn_to_page(pfn);
+ type = get_page_memtype(page);
+ if (type != -1) {
+ printk(KERN_INFO "reserve_ram_pages_type failed "
+ "0x%Lx-0x%Lx, track 0x%lx, req 0x%lx\n",
+ start, end, type, req_type);
+ if (new_type)
+ *new_type = type;
+
+ return -EBUSY;
+ }
}
- return 0;
-out:
- end_pfn = pfn;
- for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) {
+ if (new_type)
+ *new_type = req_type;
+
+ for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) {
page = pfn_to_page(pfn);
- ClearPageNonWB(page);
+ set_page_memtype(page, req_type);
}
-
- return -EINVAL;
+ return 0;
}
static int free_ram_pages_type(u64 start, u64 end)
{
struct page *page;
- u64 pfn, end_pfn;
+ u64 pfn;
for (pfn = (start >> PAGE_SHIFT); pfn < (end >> PAGE_SHIFT); ++pfn) {
page = pfn_to_page(pfn);
- if (page_mapped(page) || !PageNonWB(page))
- goto out;
-
- ClearPageNonWB(page);
+ set_page_memtype(page, -1);
}
return 0;
-
-out:
- end_pfn = pfn;
- for (pfn = (start >> PAGE_SHIFT); pfn < end_pfn; ++pfn) {
- page = pfn_to_page(pfn);
- SetPageNonWB(page);
- }
- return -EINVAL;
}
/*
if (new_type) {
if (req_type == -1)
*new_type = _PAGE_CACHE_WB;
+ else if (req_type == _PAGE_CACHE_WC)
+ *new_type = _PAGE_CACHE_UC_MINUS;
else
*new_type = req_type & _PAGE_CACHE_MASK;
}
*new_type = actual_type;
is_range_ram = pat_pagerange_is_ram(start, end);
- if (is_range_ram == 1)
- return reserve_ram_pages_type(start, end, req_type,
- new_type);
- else if (is_range_ram < 0)
+ if (is_range_ram == 1) {
+
+ spin_lock(&memtype_lock);
+ err = reserve_ram_pages_type(start, end, req_type, new_type);
+ spin_unlock(&memtype_lock);
+
+ return err;
+ } else if (is_range_ram < 0) {
return -EINVAL;
+ }
new = kmalloc(sizeof(struct memtype), GFP_KERNEL);
if (!new)
spin_lock(&memtype_lock);
- if (cached_entry && start >= cached_start)
- entry = cached_entry;
- else
- entry = list_entry(&memtype_list, struct memtype, nd);
-
/* Search for existing mapping that overlaps the current range */
where = NULL;
- list_for_each_entry_continue(entry, &memtype_list, nd) {
+ list_for_each_entry(entry, &memtype_list, nd) {
if (end <= entry->start) {
where = entry->nd.prev;
- cached_entry = list_entry(where, struct memtype, nd);
break;
} else if (start <= entry->start) { /* end > entry->start */
err = chk_conflict(new, entry, new_type);
dprintk("Overlap at 0x%Lx-0x%Lx\n",
entry->start, entry->end);
where = entry->nd.prev;
- cached_entry = list_entry(where,
- struct memtype, nd);
}
break;
} else if (start < entry->end) { /* start > entry->start */
if (!err) {
dprintk("Overlap at 0x%Lx-0x%Lx\n",
entry->start, entry->end);
- cached_entry = list_entry(entry->nd.prev,
- struct memtype, nd);
/*
* Move to right position in the linked
return err;
}
- cached_start = start;
-
if (where)
list_add(&new->nd, where);
else
list_add_tail(&new->nd, &memtype_list);
+ memtype_rb_insert(&memtype_rbroot, new);
+
spin_unlock(&memtype_lock);
dprintk("reserve_memtype added 0x%Lx-0x%Lx, track %s, req %s, ret %s\n",
int free_memtype(u64 start, u64 end)
{
- struct memtype *entry;
+ struct memtype *entry, *saved_entry;
int err = -EINVAL;
int is_range_ram;
return 0;
is_range_ram = pat_pagerange_is_ram(start, end);
- if (is_range_ram == 1)
- return free_ram_pages_type(start, end);
- else if (is_range_ram < 0)
+ if (is_range_ram == 1) {
+
+ spin_lock(&memtype_lock);
+ err = free_ram_pages_type(start, end);
+ spin_unlock(&memtype_lock);
+
+ return err;
+ } else if (is_range_ram < 0) {
return -EINVAL;
+ }
spin_lock(&memtype_lock);
- list_for_each_entry(entry, &memtype_list, nd) {
+
+ entry = memtype_rb_search(&memtype_rbroot, start);
+ if (unlikely(entry == NULL))
+ goto unlock_ret;
+
+ /*
+ * Saved entry points to an entry with start same or less than what
+ * we searched for. Now go through the list in both directions to look
+ * for the entry that matches with both start and end, with list stored
+ * in sorted start address
+ */
+ saved_entry = entry;
+ list_for_each_entry_from(entry, &memtype_list, nd) {
if (entry->start == start && entry->end == end) {
- if (cached_entry == entry || cached_start == start)
- cached_entry = NULL;
+ rb_erase(&entry->rb, &memtype_rbroot);
+ list_del(&entry->nd);
+ kfree(entry);
+ err = 0;
+ break;
+ } else if (entry->start > start) {
+ break;
+ }
+ }
+
+ if (!err)
+ goto unlock_ret;
+ entry = saved_entry;
+ list_for_each_entry_reverse(entry, &memtype_list, nd) {
+ if (entry->start == start && entry->end == end) {
+ rb_erase(&entry->rb, &memtype_rbroot);
list_del(&entry->nd);
kfree(entry);
err = 0;
break;
+ } else if (entry->start < start) {
+ break;
}
}
+unlock_ret:
spin_unlock(&memtype_lock);
if (err) {
}
+/**
+ * lookup_memtype - Looksup the memory type for a physical address
+ * @paddr: physical address of which memory type needs to be looked up
+ *
+ * Only to be called when PAT is enabled
+ *
+ * Returns _PAGE_CACHE_WB, _PAGE_CACHE_WC, _PAGE_CACHE_UC_MINUS or
+ * _PAGE_CACHE_UC
+ */
+static unsigned long lookup_memtype(u64 paddr)
+{
+ int rettype = _PAGE_CACHE_WB;
+ struct memtype *entry;
+
+ if (is_ISA_range(paddr, paddr + PAGE_SIZE - 1))
+ return rettype;
+
+ if (pat_pagerange_is_ram(paddr, paddr + PAGE_SIZE)) {
+ struct page *page;
+ spin_lock(&memtype_lock);
+ page = pfn_to_page(paddr >> PAGE_SHIFT);
+ rettype = get_page_memtype(page);
+ spin_unlock(&memtype_lock);
+ /*
+ * -1 from get_page_memtype() implies RAM page is in its
+ * default state and not reserved, and hence of type WB
+ */
+ if (rettype == -1)
+ rettype = _PAGE_CACHE_WB;
+
+ return rettype;
+ }
+
+ spin_lock(&memtype_lock);
+
+ entry = memtype_rb_search(&memtype_rbroot, paddr);
+ if (entry != NULL)
+ rettype = entry->type;
+ else
+ rettype = _PAGE_CACHE_UC_MINUS;
+
+ spin_unlock(&memtype_lock);
+ return rettype;
+}
+
+/**
+ * io_reserve_memtype - Request a memory type mapping for a region of memory
+ * @start: start (physical address) of the region
+ * @end: end (physical address) of the region
+ * @type: A pointer to memtype, with requested type. On success, requested
+ * or any other compatible type that was available for the region is returned
+ *
+ * On success, returns 0
+ * On failure, returns non-zero
+ */
+int io_reserve_memtype(resource_size_t start, resource_size_t end,
+ unsigned long *type)
+{
+ resource_size_t size = end - start;
+ unsigned long req_type = *type;
+ unsigned long new_type;
+ int ret;
+
+ WARN_ON_ONCE(iomem_map_sanity_check(start, size));
+
+ ret = reserve_memtype(start, end, req_type, &new_type);
+ if (ret)
+ goto out_err;
+
+ if (!is_new_memtype_allowed(start, size, req_type, new_type))
+ goto out_free;
+
+ if (kernel_map_sync_memtype(start, size, new_type) < 0)
+ goto out_free;
+
+ *type = new_type;
+ return 0;
+
+out_free:
+ free_memtype(start, end);
+ ret = -EBUSY;
+out_err:
+ return ret;
+}
+
+/**
+ * io_free_memtype - Release a memory type mapping for a region of memory
+ * @start: start (physical address) of the region
+ * @end: end (physical address) of the region
+ */
+void io_free_memtype(resource_size_t start, resource_size_t end)
+{
+ free_memtype(start, end);
+}
+
pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
unsigned long size, pgprot_t vma_prot)
{
{
unsigned long id_sz;
- if (!pat_enabled || base >= __pa(high_memory))
+ if (base >= __pa(high_memory))
return 0;
id_sz = (__pa(high_memory) < base + size) ?
is_ram = pat_pagerange_is_ram(paddr, paddr + size);
/*
- * reserve_pfn_range() doesn't support RAM pages. Maintain the current
- * behavior with RAM pages by returning success.
+ * reserve_pfn_range() for RAM pages. We do not refcount to keep
+ * track of number of mappings of RAM pages. We can assert that
+ * the type requested matches the type of first page in the range.
*/
- if (is_ram != 0)
+ if (is_ram) {
+ if (!pat_enabled)
+ return 0;
+
+ flags = lookup_memtype(paddr);
+ if (want_flags != flags) {
+ printk(KERN_WARNING
+ "%s:%d map pfn RAM range req %s for %Lx-%Lx, got %s\n",
+ current->comm, current->pid,
+ cattr_name(want_flags),
+ (unsigned long long)paddr,
+ (unsigned long long)(paddr + size),
+ cattr_name(flags));
+ *vma_prot = __pgprot((pgprot_val(*vma_prot) &
+ (~_PAGE_CACHE_MASK)) |
+ flags);
+ }
return 0;
+ }
ret = reserve_memtype(paddr, paddr + size, want_flags, &flags);
if (ret)
unsigned long vma_size = vma->vm_end - vma->vm_start;
pgprot_t pgprot;
- if (!pat_enabled)
- return 0;
-
- /*
- * For now, only handle remap_pfn_range() vmas where
- * is_linear_pfn_mapping() == TRUE. Handling of
- * vm_insert_pfn() is TBD.
- */
if (is_linear_pfn_mapping(vma)) {
/*
* reserve the whole chunk covered by vma. We need the
int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t *prot,
unsigned long pfn, unsigned long size)
{
+ unsigned long flags;
resource_size_t paddr;
unsigned long vma_size = vma->vm_end - vma->vm_start;
- if (!pat_enabled)
- return 0;
-
- /*
- * For now, only handle remap_pfn_range() vmas where
- * is_linear_pfn_mapping() == TRUE. Handling of
- * vm_insert_pfn() is TBD.
- */
if (is_linear_pfn_mapping(vma)) {
/* reserve the whole chunk starting from vm_pgoff */
paddr = (resource_size_t)vma->vm_pgoff << PAGE_SHIFT;
return reserve_pfn_range(paddr, vma_size, prot, 0);
}
+ if (!pat_enabled)
+ return 0;
+
+ /* for vm_insert_pfn and friends, we set prot based on lookup */
+ flags = lookup_memtype(pfn << PAGE_SHIFT);
+ *prot = __pgprot((pgprot_val(vma->vm_page_prot) & (~_PAGE_CACHE_MASK)) |
+ flags);
+
return 0;
}
resource_size_t paddr;
unsigned long vma_size = vma->vm_end - vma->vm_start;
- if (!pat_enabled)
- return;
-
- /*
- * For now, only handle remap_pfn_range() vmas where
- * is_linear_pfn_mapping() == TRUE. Handling of
- * vm_insert_pfn() is TBD.
- */
if (is_linear_pfn_mapping(vma)) {
/* free the whole chunk starting from vm_pgoff */
paddr = (resource_size_t)vma->vm_pgoff << PAGE_SHIFT;
fix_processor_context();
do_fpu_end();
- mtrr_ap_init();
+ mtrr_bp_restore();
#ifdef CONFIG_X86_OLD_MCE
mcheck_init(&boot_cpu_data);
io_mapping_create_wc(resource_size_t base, unsigned long size)
{
struct io_mapping *iomap;
-
- if (!is_io_mapping_possible(base, size))
- return NULL;
+ pgprot_t prot;
iomap = kmalloc(sizeof(*iomap), GFP_KERNEL);
if (!iomap)
- return NULL;
+ goto out_err;
+
+ if (iomap_create_wc(base, size, &prot))
+ goto out_free;
iomap->base = base;
iomap->size = size;
- iomap->prot = pgprot_writecombine(__pgprot(__PAGE_KERNEL));
+ iomap->prot = prot;
return iomap;
+
+out_free:
+ kfree(iomap);
+out_err:
+ return NULL;
}
static inline void
io_mapping_free(struct io_mapping *mapping)
{
+ iomap_free(mapping->base, mapping->size);
kfree(mapping);
}
#ifdef CONFIG_HAVE_MLOCKED_PAGE_BIT
PG_mlocked, /* Page is vma mlocked */
#endif
-#ifdef CONFIG_IA64_UNCACHED_ALLOCATOR
+#ifdef CONFIG_ARCH_USES_PG_UNCACHED
PG_uncached, /* Page has been mapped as uncached */
#endif
__NR_PAGEFLAGS,
SETPAGEFLAG_NOOP(Mlocked) TESTCLEARFLAG_FALSE(Mlocked)
#endif
-#ifdef CONFIG_IA64_UNCACHED_ALLOCATOR
+#ifdef CONFIG_ARCH_USES_PG_UNCACHED
PAGEFLAG(Uncached, uncached)
#else
PAGEFLAG_FALSE(Uncached)
return error;
}
+void __weak arch_enable_nonboot_cpus_begin(void)
+{
+}
+
+void __weak arch_enable_nonboot_cpus_end(void)
+{
+}
+
void __ref enable_nonboot_cpus(void)
{
int cpu, error;
goto out;
printk("Enabling non-boot CPUs ...\n");
+
+ arch_enable_nonboot_cpus_begin();
+
for_each_cpu(cpu, frozen_cpus) {
error = _cpu_up(cpu, 1);
if (!error) {
}
printk(KERN_WARNING "Error taking CPU%d up: %d\n", cpu, error);
}
+
+ arch_enable_nonboot_cpus_end();
+
cpumask_clear(frozen_cpus);
out:
cpu_maps_update_done();
struct call_function_data *data;
int cpu = get_cpu();
+ /*
+ * Shouldn't receive this interrupt on a cpu that is not yet online.
+ */
+ WARN_ON_ONCE(!cpu_online(cpu));
+
/*
* Ensure entry is visible on call_function_queue after we have
* entered the IPI. See comment in smp_call_function_many.
unsigned int data_flags;
LIST_HEAD(list);
+ /*
+ * Shouldn't receive this interrupt on a cpu that is not yet online.
+ */
+ WARN_ON_ONCE(!cpu_online(smp_processor_id()));
+
spin_lock(&q->lock);
list_replace_init(&q->list, &list);
spin_unlock(&q->lock);
*/
this_cpu = get_cpu();
- /* Can deadlock when called with interrupts disabled */
- WARN_ON_ONCE(irqs_disabled() && !oops_in_progress);
+ /*
+ * Can deadlock when called with interrupts disabled.
+ * We allow cpu's that are not yet online though, as no one else can
+ * send smp call function interrupt to this cpu and as such deadlocks
+ * can't happen.
+ */
+ WARN_ON_ONCE(cpu_online(this_cpu) && irqs_disabled()
+ && !oops_in_progress);
if (cpu == this_cpu) {
local_irq_save(flags);
{
csd_lock(data);
- /* Can deadlock when called with interrupts disabled */
- WARN_ON_ONCE(wait && irqs_disabled() && !oops_in_progress);
+ /*
+ * Can deadlock when called with interrupts disabled.
+ * We allow cpu's that are not yet online though, as no one else can
+ * send smp call function interrupt to this cpu and as such deadlocks
+ * can't happen.
+ */
+ WARN_ON_ONCE(cpu_online(smp_processor_id()) && wait && irqs_disabled()
+ && !oops_in_progress);
generic_exec_single(cpu, data, wait);
}
unsigned long flags;
int cpu, next_cpu, this_cpu = smp_processor_id();
- /* Can deadlock when called with interrupts disabled */
- WARN_ON_ONCE(irqs_disabled() && !oops_in_progress);
+ /*
+ * Can deadlock when called with interrupts disabled.
+ * We allow cpu's that are not yet online though, as no one else can
+ * send smp call function interrupt to this cpu and as such deadlocks
+ * can't happen.
+ */
+ WARN_ON_ONCE(cpu_online(this_cpu) && irqs_disabled()
+ && !oops_in_progress);
/* So, what's a CPU they want? Ignoring this one. */
cpu = cpumask_first_and(mask, cpu_online_mask);
#
config PAGEFLAGS_EXTENDED
def_bool y
- depends on 64BIT || SPARSEMEM_VMEMMAP || !NUMA || !SPARSEMEM
+ depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM
# Heavily threaded applications may benefit from splitting the mm-wide
# page_table_lock, so that faults on different parts of the user address
git.openpandora.org / pandora-kernel.git / commitdiff