lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

i386: move kernel/cpu/mtrr

Browse Source 
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Thomas Gleixner
2007-10-11 11:16:28 +02:00
parent ee580dc91e
commit 2ec1df4130
11 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
arch/x86/kernel/cpu/mtrr/Makefile Normal file
View File

@ -0,0 +1,3 @@
obj-y := main.o if.o generic.o state.o
obj-$(CONFIG_X86_32) += amd.o cyrix.o centaur.o

121
arch/x86/kernel/cpu/mtrr/amd.c Normal file
View File

@ -0,0 +1,121 @@
#include <linux/init.h>
#include <linux/mm.h>
#include <asm/mtrr.h>
#include <asm/msr.h>
#include "mtrr.h"
static void
amd_get_mtrr(unsigned int reg, unsigned long *base,
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 */
if (reg == 1)
low = high;
/* The base masks off on the right alignment */
*base = (low & 0xFFFE0000) >> PAGE_SHIFT;
*type = 0;
if (low & 1)
*type = MTRR_TYPE_UNCACHABLE;
if (low & 2)
*type = MTRR_TYPE_WRCOMB;
if (!(low & 3)) {
*size = 0;
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
*
* 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
*
* 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.
<do_safe> If TRUE, do the change safely. If FALSE, safety measures should
be done externally.
[RETURNS] Nothing.
*/
{
u32 regs[2];
/*
* Low is MTRR0 , High MTRR 1
*/
rdmsr(MSR_K6_UWCCR, regs[0], regs[1]);
/*
* Blank to disable
*/
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! */
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
*/
wbinvd();
wrmsr(MSR_K6_UWCCR, regs[0], regs[1]);
}
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
*/
if (type > MTRR_TYPE_WRCOMB || size < (1 << (17 - PAGE_SHIFT))
|| (size & ~(size - 1)) - size || (base & (size - 1)))
return -EINVAL;
return 0;
}
static struct mtrr_ops amd_mtrr_ops = {
.vendor = X86_VENDOR_AMD,
.set = amd_set_mtrr,
.get = amd_get_mtrr,
.get_free_region = generic_get_free_region,
.validate_add_page = amd_validate_add_page,
.have_wrcomb = positive_have_wrcomb,
};
int __init amd_init_mtrr(void)
{
set_mtrr_ops(&amd_mtrr_ops);
return 0;
}
//arch_initcall(amd_mtrr_init);

224
arch/x86/kernel/cpu/mtrr/centaur.c Normal file
View File

@ -0,0 +1,224 @@
#include <linux/init.h>
#include <linux/mm.h>
#include <asm/mtrr.h>
#include <asm/msr.h>
#include "mtrr.h"
static struct {
unsigned long high;
unsigned long low;
} centaur_mcr[8];
static u8 centaur_mcr_reserved;
static u8 centaur_mcr_type; /* 0 for winchip, 1 for winchip2 */
/*
* Report boot time MCR setups
*/
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;
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;
mtrr_if->get(i, &lbase, &lsize, &ltype);
if (lsize == 0)
return i;
}
return -ENOSPC;
}
void
mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
{
centaur_mcr[mcr].low = lo;
centaur_mcr[mcr].high = hi;
}
static void
centaur_get_mcr(unsigned int reg, unsigned long *base,
unsigned long *size, mtrr_type * type)
{
*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 */
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)
{
unsigned long low, high;
if (size == 0) {
/* 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 (type == MTRR_TYPE_UNCACHABLE)
low = -size << PAGE_SHIFT | 0x02; /* NC */
else
low = -size << PAGE_SHIFT | 0x09; /* WWO,WC */
}
}
centaur_mcr[reg].high = high;
centaur_mcr[reg].low = low;
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)
{
/*
* FIXME: Winchip2 supports uncached
*/
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");
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,
.validate_add_page = centaur_validate_add_page,
.have_wrcomb = positive_have_wrcomb,
};
int __init centaur_init_mtrr(void)
{
set_mtrr_ops(&centaur_mtrr_ops);
return 0;
}
//arch_initcall(centaur_init_mtrr);

380
arch/x86/kernel/cpu/mtrr/cyrix.c Normal file
View File

@ -0,0 +1,380 @@
#include <linux/init.h>
#include <linux/mm.h>
#include <asm/mtrr.h>
#include <asm/msr.h>
#include <asm/io.h>
#include <asm/processor-cyrix.h>
#include "mtrr.h"
int arr3_protected;
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;
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);
rcr = getCx86(CX86_RCR_BASE + reg);
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
* Note: shift==0xf means 4G, this is unsupported.
*/
if (shift)
*size = (reg < 7 ? 0x1UL : 0x40UL) << (shift - 1);
else
*size = 0;
/* Bit 0 is Cache Enable on ARR7, Cache Disable on ARR0-ARR6 */
if (reg < 7) {
switch (rcr) {
case 1:
*type = MTRR_TYPE_UNCACHABLE;
break;
case 8:
*type = MTRR_TYPE_WRBACK;
break;
case 9:
*type = MTRR_TYPE_WRCOMB;
break;
case 24:
default:
*type = MTRR_TYPE_WRTHROUGH;
break;
}
} else {
switch (rcr) {
case 0:
*type = MTRR_TYPE_UNCACHABLE;
break;
case 8:
*type = MTRR_TYPE_WRCOMB;
break;
case 9:
*type = MTRR_TYPE_WRBACK;
break;
case 25:
default:
*type = MTRR_TYPE_WRTHROUGH;
break;
}
}
}
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;
switch (replace_reg) {
case 7:
if (size < 0x40)
break;
case 6:
case 5:
case 4:
return replace_reg;
case 3:
if (arr3_protected)
break;
case 2:
case 1:
case 0:
return replace_reg;
}
/* If we are to set up a region >32M then look at ARR7 immediately */
if (size > 0x2000) {
cyrix_get_arr(7, &lbase, &lsize, &ltype);
if (lsize == 0)
return 7;
/* Else try ARR0-ARR6 first */
} else {
for (i = 0; i < 7; i++) {
cyrix_get_arr(i, &lbase, &lsize, &ltype);
if ((i == 3) && arr3_protected)
continue;
if (lsize == 0)
return i;
}
/* ARR0-ARR6 isn't free, try ARR7 but its size must be at least 256K */
cyrix_get_arr(i, &lbase, &lsize, &ltype);
if ((lsize == 0) && (size >= 0x40))
return i;
}
return -ENOSPC;
}
static u32 cr4 = 0;
static u32 ccr3;
static void prepare_set(void)
{
u32 cr0;
/* Save value of CR4 and clear Page Global Enable (bit 7) */
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 */
cr0 = read_cr0() | 0x40000000;
wbinvd();
write_cr0(cr0);
wbinvd();
/* Cyrix ARRs - everything else were excluded at the top */
ccr3 = getCx86(CX86_CCR3);
/* Cyrix ARRs - everything else were excluded at the top */
setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);
}
static void post_set(void)
{
/* Flush caches and TLBs */
wbinvd();
/* Cyrix ARRs - everything else was excluded at the top */
setCx86(CX86_CCR3, ccr3);
/* Enable caches */
write_cr0(read_cr0() & 0xbfffffff);
/* Restore value of CR4 */
if ( cpu_has_pge )
write_cr4(cr4);
}
static void cyrix_set_arr(unsigned int reg, unsigned long base,
unsigned long size, mtrr_type type)
{
unsigned char arr, arr_type, arr_size;
arr = CX86_ARR_BASE + (reg << 1) + reg; /* avoid multiplication by 3 */
/* count down from 32M (ARR0-ARR6) or from 2G (ARR7) */
if (reg >= 7)
size >>= 6;
size &= 0x7fff; /* make sure arr_size <= 14 */
for (arr_size = 0; size; arr_size++, size >>= 1) ;
if (reg < 7) {
switch (type) {
case MTRR_TYPE_UNCACHABLE:
arr_type = 1;
break;
case MTRR_TYPE_WRCOMB:
arr_type = 9;
break;
case MTRR_TYPE_WRTHROUGH:
arr_type = 24;
break;
default:
arr_type = 8;
break;
}
} else {
switch (type) {
case MTRR_TYPE_UNCACHABLE:
arr_type = 0;
break;
case MTRR_TYPE_WRCOMB:
arr_type = 8;
break;
case MTRR_TYPE_WRTHROUGH:
arr_type = 25;
break;
default:
arr_type = 9;
break;
}
}
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(CX86_RCR_BASE + reg, arr_type);
post_set();
}
typedef struct {
unsigned long base;
unsigned long size;
mtrr_type type;
} arr_state_t;
static arr_state_t arr_state[8] = {
{0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL},
{0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}, {0UL, 0UL, 0UL}
};
static unsigned char ccr_state[7] = { 0, 0, 0, 0, 0, 0, 0 };
static void cyrix_set_all(void)
{
int i;
prepare_set();
/* the CCRs are not contiguous */
for (i = 0; i < 4; i++)
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,
arr_state[i].size, arr_state[i].type);
post_set();
}
#if 0
/*
* On Cyrix 6x86(MX) and M II the ARR3 is special: it has connection
* with the SMM (System Management Mode) mode. So we need the following:
* Check whether SMI_LOCK (CCR3 bit 0) is set
* if it is set, write a warning message: ARR3 cannot be changed!
* (it cannot be changed until the next processor reset)
* if it is reset, then we can change it, set all the needed bits:
* - disable access to SMM memory through ARR3 range (CCR1 bit 7 reset)
* - disable access to SMM memory (CCR1 bit 2 reset)
* - disable SMM mode (CCR1 bit 1 reset)
* - disable write protection of ARR3 (CCR6 bit 1 reset)
* - (maybe) disable ARR3
* Just to be sure, we enable ARR usage by the processor (CCR5 bit 5 set)
*/
static void __init
cyrix_arr_init(void)
{
struct set_mtrr_context ctxt;
unsigned char ccr[7];
int ccrc[7] = { 0, 0, 0, 0, 0, 0, 0 };
#ifdef CONFIG_SMP
int i;
#endif
/* flush cache and enable MAPEN */
set_mtrr_prepare_save(&ctxt);
set_mtrr_cache_disable(&ctxt);
/* Save all CCRs locally */
ccr[0] = getCx86(CX86_CCR0);
ccr[1] = getCx86(CX86_CCR1);
ccr[2] = getCx86(CX86_CCR2);
ccr[3] = ctxt.ccr3;
ccr[4] = getCx86(CX86_CCR4);
ccr[5] = getCx86(CX86_CCR5);
ccr[6] = getCx86(CX86_CCR6);
if (ccr[3] & 1) {
ccrc[3] = 1;
arr3_protected = 1;
} else {
/* Disable SMM mode (bit 1), access to SMM memory (bit 2) and
* access to SMM memory through ARR3 (bit 7).
*/
if (ccr[1] & 0x80) {
ccr[1] &= 0x7f;
ccrc[1] |= 0x80;
}
if (ccr[1] & 0x04) {
ccr[1] &= 0xfb;
ccrc[1] |= 0x04;
}
if (ccr[1] & 0x02) {
ccr[1] &= 0xfd;
ccrc[1] |= 0x02;
}
arr3_protected = 0;
if (ccr[6] & 0x02) {
ccr[6] &= 0xfd;
ccrc[6] = 1; /* Disable write protection of ARR3 */
setCx86(CX86_CCR6, ccr[6]);
}
/* Disable ARR3. This is safe now that we disabled SMM. */
/* cyrix_set_arr_up (3, 0, 0, 0, FALSE); */
}
/* If we changed CCR1 in memory, change it in the processor, too. */
if (ccrc[1])
setCx86(CX86_CCR1, ccr[1]);
/* Enable ARR usage by the processor */
if (!(ccr[5] & 0x20)) {
ccr[5] |= 0x20;
ccrc[5] = 1;
setCx86(CX86_CCR5, ccr[5]);
}
#ifdef CONFIG_SMP
for (i = 0; i < 7; i++)
ccr_state[i] = ccr[i];
for (i = 0; i < 8; i++)
cyrix_get_arr(i,
&arr_state[i].base, &arr_state[i].size,
&arr_state[i].type);
#endif
set_mtrr_done(&ctxt); /* flush cache and disable MAPEN */
if (ccrc[5])
printk(KERN_INFO "mtrr: ARR usage was not enabled, enabled manually\n");
if (ccrc[3])
printk(KERN_INFO "mtrr: ARR3 cannot be changed\n");
/*
if ( ccrc[1] & 0x80) printk ("mtrr: SMM memory access through ARR3 disabled\n");
if ( ccrc[1] & 0x04) printk ("mtrr: SMM memory access disabled\n");
if ( ccrc[1] & 0x02) printk ("mtrr: SMM mode disabled\n");
*/
if (ccrc[6])
printk(KERN_INFO "mtrr: ARR3 was write protected, unprotected\n");
}
#endif
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,
.get_free_region = cyrix_get_free_region,
.validate_add_page = generic_validate_add_page,
.have_wrcomb = positive_have_wrcomb,
};
int __init cyrix_init_mtrr(void)
{
set_mtrr_ops(&cyrix_mtrr_ops);
return 0;
}
//arch_initcall(cyrix_init_mtrr);

509
arch/x86/kernel/cpu/mtrr/generic.c Normal file
View File

@ -0,0 +1,509 @@
/* This only handles 32bit MTRR on 32bit hosts. This is strictly wrong
because MTRRs can span upto 40 bits (36bits on most modern x86) */
#include <linux/init.h>
#include <linux/slab.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/tlbflush.h>
#include "mtrr.h"
struct mtrr_state {
struct mtrr_var_range *var_ranges;
mtrr_type fixed_ranges[NUM_FIXED_RANGES];
unsigned char enabled;
unsigned char have_fixed;
mtrr_type def_type;
};
struct fixed_range_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[] = {
{ MTRRfix64K_00000_MSR, 1 }, /* one 64k MTRR */
{ MTRRfix16K_80000_MSR, 2 }, /* two 16k MTRRs */
{ MTRRfix4K_C0000_MSR, 8 }, /* eight 4k MTRRs */
{}
};
static unsigned long smp_changes_mask;
static struct mtrr_state mtrr_state = {};
#undef MODULE_PARAM_PREFIX
#define MODULE_PARAM_PREFIX "mtrr."
static int mtrr_show;
module_param_named(show, mtrr_show, bool, 0);
/* Get the MSR pair relating to a var range */
static void
get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr)
{
rdmsr(MTRRphysBase_MSR(index), vr->base_lo, vr->base_hi);
rdmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi);
}
static void
get_fixed_ranges(mtrr_type * frs)
{
unsigned int *p = (unsigned int *) frs;
int i;
rdmsr(MTRRfix64K_00000_MSR, p[0], p[1]);
for (i = 0; i < 2; i++)
rdmsr(MTRRfix16K_80000_MSR + i, p[2 + i * 2], p[3 + i * 2]);
for (i = 0; i < 8; i++)
rdmsr(MTRRfix4K_C0000_MSR + i, p[6 + i * 2], p[7 + i * 2]);
}
void mtrr_save_fixed_ranges(void *info)
{
if (cpu_has_mtrr)
get_fixed_ranges(mtrr_state.fixed_ranges);
}
static void print_fixed(unsigned base, unsigned step, const mtrr_type*types)
{
unsigned i;
for (i = 0; i < 8; ++i, ++types, base += step)
printk(KERN_INFO "MTRR %05X-%05X %s\n",
base, base + step - 1, mtrr_attrib_to_str(*types));
}
/* 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;
if (!mtrr_state.var_ranges) {
mtrr_state.var_ranges = kmalloc(num_var_ranges * sizeof (struct mtrr_var_range),
GFP_KERNEL);
if (!mtrr_state.var_ranges)
return;
}
vrs = mtrr_state.var_ranges;
rdmsr(MTRRcap_MSR, lo, dummy);
mtrr_state.have_fixed = (lo >> 8) & 1;
for (i = 0; i < num_var_ranges; i++)
get_mtrr_var_range(i, &vrs[i]);
if (mtrr_state.have_fixed)
get_fixed_ranges(mtrr_state.fixed_ranges);
rdmsr(MTRRdefType_MSR, lo, dummy);
mtrr_state.def_type = (lo & 0xff);
mtrr_state.enabled = (lo & 0xc00) >> 10;
if (mtrr_show) {
int high_width;
printk(KERN_INFO "MTRR default type: %s\n", mtrr_attrib_to_str(mtrr_state.def_type));
if (mtrr_state.have_fixed) {
printk(KERN_INFO "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);
for (i = 0; i < 8; ++i)
print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8);
}
printk(KERN_INFO "MTRR variable ranges %sabled:\n",
mtrr_state.enabled & 2 ? "en" : "dis");
high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (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_INFO "MTRR %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_INFO "MTRR %u disabled\n", i);
}
}
}
/* Some BIOS's are fucked 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");
if (mask & MTRR_CHANGE_MASK_VARIABLE)
printk(KERN_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");
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. */
void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b)
{
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);
}
/**
* Enable and allow read/write of extended fixed-range MTRR bits on K8 CPUs
* see AMD publication no. 24593, chapter 3.2.1 for more information
*/
static inline void k8_enable_fixed_iorrs(void)
{
unsigned lo, hi;
rdmsr(MSR_K8_SYSCFG, lo, hi);
mtrr_wrmsr(MSR_K8_SYSCFG, lo
| K8_MTRRFIXRANGE_DRAM_ENABLE
| K8_MTRRFIXRANGE_DRAM_MODIFY, hi);
}
/**
* Checks and updates an fixed-range MTRR if it differs from the value it
* should have. If K8 extenstions are wanted, update the K8 SYSCFG MSR also.
* see AMD publication no. 24593, chapter 7.8.1, page 233 for more information
* \param msr MSR address of the MTTR which should be checked and updated
* \param changed pointer which indicates whether the MTRR needed to be changed
* \param msrwords pointer to the MSR values which the MSR should have
*/
static void set_fixed_range(int msr, int * changed, unsigned int * msrwords)
{
unsigned lo, hi;
rdmsr(msr, lo, hi);
if (lo != msrwords[0] || hi != msrwords[1]) {
if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
boot_cpu_data.x86 == 15 &&
((msrwords[0] | msrwords[1]) & K8_MTRR_RDMEM_WRMEM_MASK))
k8_enable_fixed_iorrs();
mtrr_wrmsr(msr, msrwords[0], msrwords[1]);
*changed = TRUE;
}
}
int generic_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;
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, &ltype);
if (lsize == 0)
return i;
}
return -ENOSPC;
}
static void generic_get_mtrr(unsigned int reg, unsigned long *base,
unsigned long *size, mtrr_type *type)
{
unsigned int mask_lo, mask_hi, base_lo, base_hi;
rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi);
if ((mask_lo & 0x800) == 0) {
/* Invalid (i.e. free) range */
*base = 0;
*size = 0;
*type = 0;
return;
}
rdmsr(MTRRphysBase_MSR(reg), base_lo, base_hi);
/* Work out the shifted address mask. */
mask_lo = size_or_mask | mask_hi << (32 - PAGE_SHIFT)
| mask_lo >> PAGE_SHIFT;
/* This works correctly if size is a power of two, i.e. a
contiguous range. */
*size = -mask_lo;
*base = base_hi << (32 - PAGE_SHIFT) | base_lo >> PAGE_SHIFT;
*type = base_lo & 0xff;
}
/**
* Checks and updates the fixed-range MTRRs if they differ from the saved set
* \param frs pointer to fixed-range MTRR values, saved by get_fixed_ranges()
*/
static int set_fixed_ranges(mtrr_type * frs)
{
unsigned long long *saved = (unsigned long long *) frs;
int changed = FALSE;
int block=-1, range;
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 */
static int set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
{
unsigned int lo, hi;
int changed = FALSE;
rdmsr(MTRRphysBase_MSR(index), 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;
}
rdmsr(MTRRphysMask_MSR(index), lo, hi);
if ((vr->mask_lo & 0xfffff800UL) != (lo & 0xfffff800UL)
|| (vr->mask_hi & (size_and_mask >> (32 - PAGE_SHIFT))) !=
(hi & (size_and_mask >> (32 - PAGE_SHIFT)))) {
mtrr_wrmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi);
changed = TRUE;
}
return changed;
}
static u32 deftype_lo, deftype_hi;
static unsigned long set_mtrr_state(void)
/* [SUMMARY] Set the MTRR state for this CPU.
<state> The MTRR state information to read.
<ctxt> Some relevant CPU context.
[NOTE] The CPU must already be in a safe state for MTRR changes.
[RETURNS] 0 if no changes made, else a mask indication what was changed.
*/
{
unsigned int i;
unsigned long change_mask = 0;
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 */
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);
change_mask |= MTRR_CHANGE_MASK_DEFTYPE;
}
return change_mask;
}
static unsigned long cr4 = 0;
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.
*/
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 */
spin_lock(&set_atomicity_lock);
/* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */
cr0 = read_cr0() | 0x40000000; /* set CD flag */
write_cr0(cr0);
wbinvd();
/* 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 */
rdmsr(MTRRdefType_MSR, deftype_lo, deftype_hi);
/* Disable MTRRs, and set the default type to uncached */
mtrr_wrmsr(MTRRdefType_MSR, 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_tlb();
/* Intel (P6) standard MTRRs */
mtrr_wrmsr(MTRRdefType_MSR, deftype_lo, deftype_hi);
/* Enable caches */
write_cr0(read_cr0() & 0xbfffffff);
/* Restore value of CR4 */
if ( cpu_has_pge )
write_cr4(cr4);
spin_unlock(&set_atomicity_lock);
}
static void generic_set_all(void)
{
unsigned long mask, count;
unsigned long flags;
local_irq_save(flags);
prepare_set();
/* Actually set the state */
mask = set_mtrr_state();
post_set();
local_irq_restore(flags);
/* 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;
}
}
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.
<do_safe> If TRUE, do the change safely. If FALSE, safety measures should
be done externally.
[RETURNS] Nothing.
*/
{
unsigned long flags;
struct mtrr_var_range *vr;
vr = &mtrr_state.var_ranges[reg];
local_irq_save(flags);
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. */
mtrr_wrmsr(MTRRphysMask_MSR(reg), 0, 0);
memset(vr, 0, sizeof(struct mtrr_var_range));
} else {
vr->base_lo = base << PAGE_SHIFT | type;
vr->base_hi = (base & size_and_mask) >> (32 - PAGE_SHIFT);
vr->mask_lo = -size << PAGE_SHIFT | 0x800;
vr->mask_hi = (-size & size_and_mask) >> (32 - PAGE_SHIFT);
mtrr_wrmsr(MTRRphysBase_MSR(reg), vr->base_lo, vr->base_hi);
mtrr_wrmsr(MTRRphysMask_MSR(reg), vr->mask_lo, vr->mask_hi);
}
post_set();
local_irq_restore(flags);
}
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 */
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);
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");
return -EINVAL;
}
}
/* 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) ;
if (lbase != last) {
printk(KERN_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(MTRRcap_MSR, config, dummy);
return (config & (1 << 10));
}
int positive_have_wrcomb(void)
{
return 1;
}
/* 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,
};

439
arch/x86/kernel/cpu/mtrr/if.c Normal file
View File

@ -0,0 +1,439 @@
#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>
#define LINE_SIZE 80
#include <asm/mtrr.h>
#include "mtrr.h"
/* RED-PEN: this is accessed without any locking */
extern unsigned int *usage_table;
#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 */
};
const char *mtrr_attrib_to_str(int x)
{
return (x <= 6) ? mtrr_strings[x] : "?";
}
#ifdef CONFIG_PROC_FS
static int
mtrr_file_add(unsigned long base, unsigned long size,
unsigned int type, char increment, struct file *file, int page)
{
int reg, max;
unsigned int *fcount = FILE_FCOUNT(file);
max = num_var_ranges;
if (fcount == NULL) {
fcount = kzalloc(max * sizeof *fcount, GFP_KERNEL);
if (!fcount)
return -ENOMEM;
FILE_FCOUNT(file) = fcount;
}
if (!page) {
if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1)))
return -EINVAL;
base >>= PAGE_SHIFT;
size >>= PAGE_SHIFT;
}
reg = mtrr_add_page(base, size, type, 1);
if (reg >= 0)
++fcount[reg];
return reg;
}
static int
mtrr_file_del(unsigned long base, unsigned long size,
struct file *file, int page)
{
int reg;
unsigned int *fcount = FILE_FCOUNT(file);
if (!page) {
if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1)))
return -EINVAL;
base >>= PAGE_SHIFT;
size >>= PAGE_SHIFT;
}
reg = mtrr_del_page(-1, base, size);
if (reg < 0)
return reg;
if (fcount == NULL)
return reg;
if (fcount[reg] < 1)
return -EINVAL;
--fcount[reg];
return reg;
}
/* RED-PEN: seq_file can seek now. this is ignored. */
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;
unsigned long long base, size;
char *ptr;
char line[LINE_SIZE];
size_t linelen;
if (!capable(CAP_SYS_ADMIN))
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);
if (err < 0)
return err;
return len;
}
if (strncmp(line, "base=", 5))
return -EINVAL;
base = simple_strtoull(line + 5, &ptr, 0);
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) ;
if (strncmp(ptr, "type=", 5))
return -EINVAL;
ptr += 5;
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,
1);
if (err < 0)
return err;
return len;
}
return -EINVAL;
}
static long
mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
{
int err = 0;
mtrr_type type;
unsigned long size;
struct mtrr_sentry sentry;
struct mtrr_gentry gentry;
void __user *arg = (void __user *) __arg;
switch (cmd) {
case MTRRIOC_ADD_ENTRY:
case MTRRIOC_SET_ENTRY:
case MTRRIOC_DEL_ENTRY:
case MTRRIOC_KILL_ENTRY:
case MTRRIOC_ADD_PAGE_ENTRY:
case MTRRIOC_SET_PAGE_ENTRY:
case MTRRIOC_DEL_PAGE_ENTRY:
case MTRRIOC_KILL_PAGE_ENTRY:
if (copy_from_user(&sentry, arg, sizeof sentry))
return -EFAULT;
break;
case MTRRIOC_GET_ENTRY:
case MTRRIOC_GET_PAGE_ENTRY:
if (copy_from_user(&gentry, arg, sizeof gentry))
return -EFAULT;
break;
#ifdef CONFIG_COMPAT
case MTRRIOC32_ADD_ENTRY:
case MTRRIOC32_SET_ENTRY:
case MTRRIOC32_DEL_ENTRY:
case MTRRIOC32_KILL_ENTRY:
case MTRRIOC32_ADD_PAGE_ENTRY:
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;
err = get_user(sentry.base, &s32->base);
err |= get_user(sentry.size, &s32->size);
err |= get_user(sentry.type, &s32->type);
if (err)
return err;
break;
}
case MTRRIOC32_GET_ENTRY:
case MTRRIOC32_GET_PAGE_ENTRY: {
struct mtrr_gentry32 __user *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);
err |= get_user(gentry.type, &g32->type);
if (err)
return err;
break;
}
#endif
}
switch (cmd) {
default:
return -ENOTTY;
case MTRRIOC_ADD_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_ADD_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err =
mtrr_file_add(sentry.base, sentry.size, sentry.type, 1,
file, 0);
break;
case MTRRIOC_SET_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_SET_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err = mtrr_add(sentry.base, sentry.size, sentry.type, 0);
break;
case MTRRIOC_DEL_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_DEL_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err = mtrr_file_del(sentry.base, sentry.size, file, 0);
break;
case MTRRIOC_KILL_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_KILL_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err = mtrr_del(-1, sentry.base, sentry.size);
break;
case MTRRIOC_GET_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_GET_ENTRY:
#endif
if (gentry.regnum >= num_var_ranges)
return -EINVAL;
mtrr_if->get(gentry.regnum, &gentry.base, &size, &type);
/* Hide entries that go above 4GB */
if (gentry.base + size - 1 >= (1UL << (8 * sizeof(gentry.size) - PAGE_SHIFT))
|| size >= (1UL << (8 * sizeof(gentry.size) - PAGE_SHIFT)))
gentry.base = gentry.size = gentry.type = 0;
else {
gentry.base <<= PAGE_SHIFT;
gentry.size = size << PAGE_SHIFT;
gentry.type = type;
}
break;
case MTRRIOC_ADD_PAGE_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_ADD_PAGE_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err =
mtrr_file_add(sentry.base, sentry.size, sentry.type, 1,
file, 1);
break;
case MTRRIOC_SET_PAGE_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_SET_PAGE_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err = mtrr_add_page(sentry.base, sentry.size, sentry.type, 0);
break;
case MTRRIOC_DEL_PAGE_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_DEL_PAGE_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err = mtrr_file_del(sentry.base, sentry.size, file, 1);
break;
case MTRRIOC_KILL_PAGE_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_KILL_PAGE_ENTRY:
#endif
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
err = mtrr_del_page(-1, sentry.base, sentry.size);
break;
case MTRRIOC_GET_PAGE_ENTRY:
#ifdef CONFIG_COMPAT
case MTRRIOC32_GET_PAGE_ENTRY:
#endif
if (gentry.regnum >= num_var_ranges)
return -EINVAL;
mtrr_if->get(gentry.regnum, &gentry.base, &size, &type);
/* Hide entries that would overflow */
if (size != (__typeof__(gentry.size))size)
gentry.base = gentry.size = gentry.type = 0;
else {
gentry.size = size;
gentry.type = type;
}
break;
}
if (err)
return err;
switch(cmd) {
case MTRRIOC_GET_ENTRY:
case MTRRIOC_GET_PAGE_ENTRY:
if (copy_to_user(arg, &gentry, sizeof gentry))
err = -EFAULT;
break;
#ifdef CONFIG_COMPAT
case MTRRIOC32_GET_ENTRY:
case MTRRIOC32_GET_PAGE_ENTRY: {
struct mtrr_gentry32 __user *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);
err |= put_user(gentry.type, &g32->type);
break;
}
#endif
}
return err;
}
static int
mtrr_close(struct inode *ino, struct file *file)
{
int i, max;
unsigned int *fcount = FILE_FCOUNT(file);
if (fcount != NULL) {
max = num_var_ranges;
for (i = 0; i < max; ++i) {
while (fcount[i] > 0) {
mtrr_del(i, 0, 0);
--fcount[i];
}
}
kfree(fcount);
FILE_FCOUNT(file) = NULL;
}
return single_release(ino, file);
}
static int mtrr_seq_show(struct seq_file *seq, void *offset);
static int mtrr_open(struct inode *inode, struct file *file)
{
if (!mtrr_if)
return -EIO;
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,
};
static struct proc_dir_entry *proc_root_mtrr;
static int mtrr_seq_show(struct seq_file *seq, void *offset)
{
char factor;
int i, max, len;
mtrr_type type;
unsigned long base, size;
len = 0;
max = num_var_ranges;
for (i = 0; i < max; i++) {
mtrr_if->get(i, &base, &size, &type);
if (size == 0)
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%05lx000 (%4luMB), size=%4lu%cB: %s, count=%d\n",
i, base, base >> (20 - PAGE_SHIFT), size, factor,
mtrr_attrib_to_str(type), usage_table[i]);
}
}
return 0;
}
static int __init mtrr_if_init(void)
{
struct cpuinfo_x86 *c = &boot_cpu_data;
if ((!cpu_has(c, X86_FEATURE_MTRR)) &&
(!cpu_has(c, X86_FEATURE_K6_MTRR)) &&
(!cpu_has(c, X86_FEATURE_CYRIX_ARR)) &&
(!cpu_has(c, X86_FEATURE_CENTAUR_MCR)))
return -ENODEV;
proc_root_mtrr =
create_proc_entry("mtrr", S_IWUSR | S_IRUGO, &proc_root);
if (proc_root_mtrr) {
proc_root_mtrr->owner = THIS_MODULE;
proc_root_mtrr->proc_fops = &mtrr_fops;
}
return 0;
}
arch_initcall(mtrr_if_init);
#endif /* CONFIG_PROC_FS */

768
arch/x86/kernel/cpu/mtrr/main.c Normal file
View File

@ -0,0 +1,768 @@
/* Generic MTRR (Memory Type Range Register) driver.
Copyright (C) 1997-2000 Richard Gooch
Copyright (c) 2002 Patrick Mochel
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.
Richard Gooch may be reached by email at rgooch@atnf.csiro.au
The postal address is:
Richard Gooch, c/o ATNF, P. O. Box 76, Epping, N.S.W., 2121, Australia.
Source: "Pentium Pro Family Developer's Manual, Volume 3:
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:
System Programming Guide; Section 9.11. (1997 edition - PPro).
*/
#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 <asm/mtrr.h>
#include <asm/uaccess.h>
#include <asm/processor.h>
#include <asm/msr.h>
#include "mtrr.h"
u32 num_var_ranges = 0;
unsigned int *usage_table;
static DEFINE_MUTEX(mtrr_mutex);
u64 size_or_mask, size_and_mask;
static struct mtrr_ops * mtrr_ops[X86_VENDOR_NUM] = {};
struct mtrr_ops * mtrr_if = NULL;
static void set_mtrr(unsigned int reg, unsigned long base,
unsigned long size, mtrr_type type);
#ifndef CONFIG_X86_64
extern int arr3_protected;
#else
#define arr3_protected 0
#endif
void set_mtrr_ops(struct mtrr_ops * ops)
{
if (ops->vendor && ops->vendor < X86_VENDOR_NUM)
mtrr_ops[ops->vendor] = ops;
}
/* Returns non-zero if we have the write-combining memory type */
static int have_wrcomb(void)
{
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 */
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");
pci_dev_put(dev);
return 0;
}
}
/* 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");
pci_dev_put(dev);
return 0;
}
pci_dev_put(dev);
}
return (mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0);
}
/* This function returns the number of variable MTRRs */
static void __init set_num_var_ranges(void)
{
unsigned long config = 0, dummy;
if (use_intel()) {
rdmsr(MTRRcap_MSR, config, dummy);
} else if (is_cpu(AMD))
config = 2;
else if (is_cpu(CYRIX) || is_cpu(CENTAUR))
config = 8;
num_var_ranges = config & 0xff;
}
static void __init init_table(void)
{
int i, max;
max = num_var_ranges;
if ((usage_table = kmalloc(max * sizeof *usage_table, GFP_KERNEL))
== NULL) {
printk(KERN_ERR "mtrr: could not allocate\n");
return;
}
for (i = 0; i < max; i++)
usage_table[i] = 1;
}
struct set_mtrr_data {
atomic_t count;
atomic_t gate;
unsigned long smp_base;
unsigned long smp_size;
unsigned int smp_reg;
mtrr_type smp_type;
};
#ifdef CONFIG_SMP
static void ipi_handler(void *info)
/* [SUMMARY] Synchronisation handler. Executed by "other" CPUs.
[RETURNS] Nothing.
*/
{
struct set_mtrr_data *data = info;
unsigned long flags;
local_irq_save(flags);
atomic_dec(&data->count);
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,
data->smp_size, data->smp_type);
else
mtrr_if->set_all();
atomic_dec(&data->count);
while(atomic_read(&data->gate))
cpu_relax();
atomic_dec(&data->count);
local_irq_restore(flags);
}
#endif
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) ||
(type1 == MTRR_TYPE_WRBACK && type2 == MTRR_TYPE_WRTHROUGH);
}
/**
* set_mtrr - update mtrrs on all processors
* @reg: mtrr in question
* @base: mtrr base
* @size: mtrr size
* @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
* 4. Enter no-fill cache mode
* 5. Flush caches
* 6. Clear PGE bit
* 7. Flush all TLBs
* 8. Disable all range registers
* 9. Update the MTRRs
* 10. Enable all range registers
* 11. Flush all TLBs and caches again
* 12. Enter normal cache mode and reenable caching
* 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.
* 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)
{
struct set_mtrr_data data;
unsigned long flags;
data.smp_reg = reg;
data.smp_base = base;
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 */
smp_wmb();
atomic_set(&data.gate,0);
/* Start the ball rolling on other CPUs */
if (smp_call_function(ipi_handler, &data, 1, 0) != 0)
panic("mtrr: timed out waiting for other CPUs\n");
local_irq_save(flags);
while(atomic_read(&data.count))
cpu_relax();
/* ok, reset count and toggle gate */
atomic_set(&data.count, num_booting_cpus() - 1);
smp_wmb();
atomic_set(&data.gate,1);
/* do our MTRR business */
/* 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.
* If we're doing that @reg is set to something special...
*/
if (reg != ~0U)
mtrr_if->set(reg,base,size,type);
/* 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);
/*
* Wait here for everyone to have seen the gate change
* So we're the last ones to touch 'data'
*/
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
*
* 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.
*
* On a multiprocessor machine the changes are made to all processors.
* This is required on x86 by the Intel processors.
*
* The available types are
*
* %MTRR_TYPE_UNCACHABLE - No caching
*
* %MTRR_TYPE_WRBACK - Write data back in bursts whenever
*
* %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
*
* %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.
*/
int mtrr_add_page(unsigned long base, unsigned long size,
unsigned int type, char increment)
{
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)))
return error;
if (type >= MTRR_NUM_TYPES) {
printk(KERN_WARNING "mtrr: type: %u invalid\n", type);
return -EINVAL;
}
/* 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");
return -ENOSYS;
}
if (!size) {
printk(KERN_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");
return -EINVAL;
}
error = -EINVAL;
replace = -1;
/* No CPU hotplug when we change MTRR entries */
lock_cpu_hotplug();
/* Search for existing MTRR */
mutex_lock(&mtrr_mutex);
for (i = 0; i < num_var_ranges; ++i) {
mtrr_if->get(i, &lbase, &lsize, &ltype);
if (!lsize || base > lbase + lsize - 1 || base + size - 1 < lbase)
continue;
/* 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) {
/* New region encloses an existing region */
if (type == ltype) {
replace = replace == -1 ? i : -2;
continue;
}
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);
goto out;
}
/* 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));
goto out;
}
if (increment)
++usage_table[i];
error = i;
goto out;
}
/* Search for an empty MTRR */
i = mtrr_if->get_free_region(base, size, replace);
if (i >= 0) {
set_mtrr(i, base, size, type);
if (likely(replace < 0))
usage_table[i] = 1;
else {
usage_table[i] = usage_table[replace] + !!increment;
if (unlikely(replace != i)) {
set_mtrr(replace, 0, 0, 0);
usage_table[replace] = 0;
}
}
} else
printk(KERN_INFO "mtrr: no more MTRRs available\n");
error = i;
out:
mutex_unlock(&mtrr_mutex);
unlock_cpu_hotplug();
return error;
}
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);
dump_stack();
return -1;
}
return 0;
}
/**
* 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.
*
* 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.
*
* The available types are
*
* %MTRR_TYPE_UNCACHABLE - No caching
*
* %MTRR_TYPE_WRBACK - Write data back in bursts whenever
*
* %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
*
* %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.
*/
int
mtrr_add(unsigned long base, unsigned long size, unsigned int type,
char increment)
{
if (mtrr_check(base, size))
return -EINVAL;
return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
increment);
}
/**
* 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.
*
* 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;
mtrr_type ltype;
unsigned long lbase, lsize;
int error = -EINVAL;
if (!mtrr_if)
return -ENXIO;
max = num_var_ranges;
/* No CPU hotplug when we change MTRR entries */
lock_cpu_hotplug();
mutex_lock(&mtrr_mutex);
if (reg < 0) {
/* Search for existing MTRR */
for (i = 0; i < max; ++i) {
mtrr_if->get(i, &lbase, &lsize, &ltype);
if (lbase == base && lsize == size) {
reg = i;
break;
}
}
if (reg < 0) {
printk(KERN_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);
goto out;
}
if (is_cpu(CYRIX) && !use_intel()) {
if ((reg == 3) && arr3_protected) {
printk(KERN_WARNING "mtrr: ARR3 cannot be changed\n");
goto out;
}
}
mtrr_if->get(reg, &lbase, &lsize, &ltype);
if (lsize < 1) {
printk(KERN_WARNING "mtrr: MTRR %d not used\n", reg);
goto out;
}
if (usage_table[reg] < 1) {
printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg);
goto out;
}
if (--usage_table[reg] < 1)
set_mtrr(reg, 0, 0, 0);
error = reg;
out:
mutex_unlock(&mtrr_mutex);
unlock_cpu_hotplug();
return error;
}
/**
* 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.
*
* 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)
{
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!
* These should be called implicitly, but we can't yet until all the initcall
* stuff is done...
*/
extern void amd_init_mtrr(void);
extern void cyrix_init_mtrr(void);
extern void centaur_init_mtrr(void);
static void __init init_ifs(void)
{
#ifndef CONFIG_X86_64
amd_init_mtrr();
cyrix_init_mtrr();
centaur_init_mtrr();
#endif
}
/* The suspend/resume methods are only for CPU without MTRR. CPU using generic
* MTRR driver doesn't require this
*/
struct mtrr_value {
mtrr_type ltype;
unsigned long lbase;
unsigned long lsize;
};
static struct mtrr_value * mtrr_state;
static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
{
int i;
int size = num_var_ranges * sizeof(struct mtrr_value);
mtrr_state = kzalloc(size,GFP_ATOMIC);
if (!mtrr_state)
return -ENOMEM;
for (i = 0; i < num_var_ranges; i++) {
mtrr_if->get(i,
&mtrr_state[i].lbase,
&mtrr_state[i].lsize,
&mtrr_state[i].ltype);
}
return 0;
}
static int mtrr_restore(struct sys_device * sysdev)
{
int i;
for (i = 0; i < num_var_ranges; i++) {
if (mtrr_state[i].lsize)
set_mtrr(i,
mtrr_state[i].lbase,
mtrr_state[i].lsize,
mtrr_state[i].ltype);
}
kfree(mtrr_state);
return 0;
}
static struct sysdev_driver mtrr_sysdev_driver = {
.suspend = mtrr_save,
.resume = mtrr_restore,
};
/**
* mtrr_bp_init - initialize mtrrs on the boot CPU
*
* 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)
{
init_ifs();
if (cpu_has_mtrr) {
mtrr_if = &generic_mtrr_ops;
size_or_mask = 0xff000000; /* 36 bits */
size_and_mask = 0x00f00000;
/* 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) {
u32 phys_addr;
phys_addr = cpuid_eax(0x80000008) & 0xff;
/* CPUID workaround for Intel 0F33/0F34 CPU */
if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
boot_cpu_data.x86 == 0xF &&
boot_cpu_data.x86_model == 0x3 &&
(boot_cpu_data.x86_mask == 0x3 ||
boot_cpu_data.x86_mask == 0x4))
phys_addr = 36;
size_or_mask = ~((1ULL << (phys_addr - PAGE_SHIFT)) - 1);
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 */
size_and_mask = 0;
}
} else {
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_AMD:
if (cpu_has_k6_mtrr) {
/* Pre-Athlon (K6) AMD CPU MTRRs */
mtrr_if = mtrr_ops[X86_VENDOR_AMD];
size_or_mask = 0xfff00000; /* 32 bits */
size_and_mask = 0;
}
break;
case X86_VENDOR_CENTAUR:
if (cpu_has_centaur_mcr) {
mtrr_if = mtrr_ops[X86_VENDOR_CENTAUR];
size_or_mask = 0xfff00000; /* 32 bits */
size_and_mask = 0;
}
break;
case X86_VENDOR_CYRIX:
if (cpu_has_cyrix_arr) {
mtrr_if = mtrr_ops[X86_VENDOR_CYRIX];
size_or_mask = 0xfff00000; /* 32 bits */
size_and_mask = 0;
}
break;
default:
break;
}
}
if (mtrr_if) {
set_num_var_ranges();
init_table();
if (use_intel())
get_mtrr_state();
}
}
void mtrr_ap_init(void)
{
unsigned long flags;
if (!mtrr_if || !use_intel())
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
*/
local_irq_save(flags);
mtrr_if->set_all();
local_irq_restore(flags);
}
/**
* Save current fixed-range MTRR state of the BSP
*/
void mtrr_save_state(void)
{
int cpu = get_cpu();
if (cpu == 0)
mtrr_save_fixed_ranges(NULL);
else
smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1, 1);
put_cpu();
}
static int __init mtrr_init_finialize(void)
{
if (!mtrr_if)
return 0;
if (use_intel())
mtrr_state_warn();
else {
/* The CPUs haven't MTRR and seemes 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);

98
arch/x86/kernel/cpu/mtrr/mtrr.h Normal file
View File

@ -0,0 +1,98 @@
/*
* local mtrr defines.
*/
#ifndef TRUE
#define TRUE 1
#define FALSE 0
#endif
#define MTRRcap_MSR 0x0fe
#define MTRRdefType_MSR 0x2ff
#define MTRRphysBase_MSR(reg) (0x200 + 2 * (reg))
#define MTRRphysMask_MSR(reg) (0x200 + 2 * (reg) + 1)
#define NUM_FIXED_RANGES 88
#define MTRRfix64K_00000_MSR 0x250
#define MTRRfix16K_80000_MSR 0x258
#define MTRRfix16K_A0000_MSR 0x259
#define MTRRfix4K_C0000_MSR 0x268
#define MTRRfix4K_C8000_MSR 0x269
#define MTRRfix4K_D0000_MSR 0x26a
#define MTRRfix4K_D8000_MSR 0x26b
#define MTRRfix4K_E0000_MSR 0x26c
#define MTRRfix4K_E8000_MSR 0x26d
#define MTRRfix4K_F0000_MSR 0x26e
#define MTRRfix4K_F8000_MSR 0x26f
#define MTRR_CHANGE_MASK_FIXED 0x01
#define MTRR_CHANGE_MASK_VARIABLE 0x02
#define MTRR_CHANGE_MASK_DEFTYPE 0x04
/* In the Intel processor's MTRR interface, the MTRR type is always held in
an 8 bit field: */
typedef u8 mtrr_type;
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);
int (*get_free_region)(unsigned long base, unsigned long size,
int replace_reg);
int (*validate_add_page)(unsigned long base, unsigned long size,
unsigned int type);
int (*have_wrcomb)(void);
};
extern int generic_get_free_region(unsigned long base, unsigned long size,
int replace_reg);
extern int generic_validate_add_page(unsigned long base, unsigned long size,
unsigned int type);
extern struct mtrr_ops generic_mtrr_ops;
extern int positive_have_wrcomb(void);
/* library functions for processor-specific routines */
struct set_mtrr_context {
unsigned long flags;
unsigned long cr4val;
u32 deftype_lo;
u32 deftype_hi;
u32 ccr3;
};
struct mtrr_var_range {
u32 base_lo;
u32 base_hi;
u32 mask_lo;
u32 mask_hi;
};
void set_mtrr_done(struct set_mtrr_context *ctxt);
void set_mtrr_cache_disable(struct set_mtrr_context *ctxt);
void set_mtrr_prepare_save(struct set_mtrr_context *ctxt);
void get_mtrr_state(void);
extern void set_mtrr_ops(struct mtrr_ops * ops);
extern u64 size_or_mask, size_and_mask;
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)
extern unsigned int num_var_ranges;
void mtrr_state_warn(void);
const char *mtrr_attrib_to_str(int x);
void mtrr_wrmsr(unsigned, unsigned, unsigned);

79
arch/x86/kernel/cpu/mtrr/state.c Normal file
View File

@ -0,0 +1,79 @@
#include <linux/mm.h>
#include <linux/init.h>
#include <asm/io.h>
#include <asm/mtrr.h>
#include <asm/msr.h>
#include <asm-i386/processor-cyrix.h>
#include "mtrr.h"
/* 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 */
local_irq_save(ctxt->flags);
if (use_intel() || is_cpu(CYRIX)) {
/* 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);
}
/* Disable and flush caches. Note that wbinvd flushes the TLBs as
a side-effect */
cr0 = read_cr0() | 0x40000000;
wbinvd();
write_cr0(cr0);
wbinvd();
if (use_intel())
/* Save MTRR state */
rdmsr(MTRRdefType_MSR, ctxt->deftype_lo, ctxt->deftype_hi);
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 */
mtrr_wrmsr(MTRRdefType_MSR, 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);
}
/* 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 */
wbinvd();
/* Restore MTRRdefType */
if (use_intel())
/* Intel (P6) standard MTRRs */
mtrr_wrmsr(MTRRdefType_MSR, ctxt->deftype_lo, ctxt->deftype_hi);
else
/* Cyrix ARRs - everything else was excluded at the top */
setCx86(CX86_CCR3, ctxt->ccr3);
/* Enable caches */
write_cr0(read_cr0() & 0xbfffffff);
/* Restore value of CR4 */
if ( cpu_has_pge )
write_cr4(ctxt->cr4val);
}
/* Re-enable interrupts locally (if enabled previously) */
local_irq_restore(ctxt->flags);
}