aboutsummaryrefslogtreecommitdiff
path: root/arch/x86
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86')
-rw-r--r--arch/x86/Kconfig.debug2
-rw-r--r--arch/x86/Makefile2
-rw-r--r--arch/x86/kernel/Makefile10
-rw-r--r--arch/x86/kernel/asm-offsets_32.c6
-rw-r--r--arch/x86/kernel/cpu/bugs.c8
-rw-r--r--arch/x86/kernel/efi.c2
-rw-r--r--arch/x86/kernel/efi_32.c1
-rw-r--r--arch/x86/kernel/entry_32.S2
-rw-r--r--arch/x86/kernel/entry_64.S2
-rw-r--r--arch/x86/kernel/head64.c3
-rw-r--r--arch/x86/kernel/head_32.S2
-rw-r--r--arch/x86/kernel/head_64.S9
-rw-r--r--arch/x86/kernel/i387.c2
-rw-r--r--arch/x86/kernel/i8259_32.c25
-rw-r--r--arch/x86/kernel/io_delay.c1
-rw-r--r--arch/x86/kernel/kprobes.c4
-rw-r--r--arch/x86/kernel/nmi_32.c21
-rw-r--r--arch/x86/kernel/nmi_64.c21
-rw-r--r--arch/x86/kernel/setup_32.c1
-rw-r--r--arch/x86/kernel/setup_64.c6
-rw-r--r--arch/x86/kernel/topology.c2
-rw-r--r--arch/x86/kernel/vmlinux_32.lds.S26
-rw-r--r--arch/x86/kernel/vmlinux_64.lds.S30
-rw-r--r--arch/x86/lib/csum-wrappers_64.c147
-rw-r--r--arch/x86/lib/io_64.c18
-rw-r--r--arch/x86/mm/init_64.c27
-rw-r--r--arch/x86/mm/ioremap.c26
-rw-r--r--arch/x86/mm/numa_64.c5
-rw-r--r--arch/x86/mm/pageattr.c232
-rw-r--r--arch/x86/mm/srat_64.c3
-rw-r--r--arch/x86/pci/common.c2
-rw-r--r--arch/x86/power/hibernate_asm_64.S5
32 files changed, 362 insertions, 291 deletions
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 864affc9a7b..702eb39901c 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -156,7 +156,7 @@ config IO_DELAY_TYPE_NONE
choice
prompt "IO delay type"
- default IO_DELAY_0XED
+ default IO_DELAY_0X80
config IO_DELAY_0X80
bool "port 0x80 based port-IO delay [recommended]"
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 204af43535c..f1e739a43d4 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -229,7 +229,7 @@ zdisk bzdisk: vmlinux
fdimage fdimage144 fdimage288 isoimage: vmlinux
$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@
-install: vdso_install
+install:
$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) install
PHONY += vdso_install
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 76ec0f8f138..4eb5ce84110 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -6,7 +6,15 @@ extra-y := head_$(BITS).o init_task.o vmlinux.lds
extra-$(CONFIG_X86_64) += head64.o
CPPFLAGS_vmlinux.lds += -U$(UTS_MACHINE)
-CFLAGS_vsyscall_64.o := $(PROFILING) -g0
+
+#
+# vsyscalls (which work on the user stack) should have
+# no stack-protector checks:
+#
+nostackp := $(call cc-option, -fno-stack-protector)
+CFLAGS_vsyscall_64.o := $(PROFILING) -g0 $(nostackp)
+CFLAGS_hpet.o := $(nostackp)
+CFLAGS_tsc_64.o := $(nostackp)
obj-y := process_$(BITS).o signal_$(BITS).o entry_$(BITS).o
obj-y += traps_$(BITS).o irq_$(BITS).o
diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index afd84463b71..a33d5301799 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -20,10 +20,8 @@
#include <xen/interface/xen.h>
-#ifdef CONFIG_LGUEST_GUEST
#include <linux/lguest.h>
#include "../../../drivers/lguest/lg.h"
-#endif
#define DEFINE(sym, val) \
asm volatile("\n->" #sym " %0 " #val : : "i" (val))
@@ -134,6 +132,10 @@ void foo(void)
BLANK();
OFFSET(LGUEST_DATA_irq_enabled, lguest_data, irq_enabled);
OFFSET(LGUEST_DATA_pgdir, lguest_data, pgdir);
+#endif
+
+#ifdef CONFIG_LGUEST
+ BLANK();
OFFSET(LGUEST_PAGES_host_gdt_desc, lguest_pages, state.host_gdt_desc);
OFFSET(LGUEST_PAGES_host_idt_desc, lguest_pages, state.host_idt_desc);
OFFSET(LGUEST_PAGES_host_cr3, lguest_pages, state.host_cr3);
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index 9b95edcfc6a..027e5c003b1 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -25,14 +25,6 @@ static int __init no_halt(char *s)
__setup("no-hlt", no_halt);
-static int __init mca_pentium(char *s)
-{
- mca_pentium_flag = 1;
- return 1;
-}
-
-__setup("mca-pentium", mca_pentium);
-
static int __init no_387(char *s)
{
boot_cpu_data.hard_math = 0;
diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 0c0eeb163d9..759e02bec07 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -54,7 +54,7 @@ EXPORT_SYMBOL(efi);
struct efi_memory_map memmap;
-struct efi efi_phys __initdata;
+static struct efi efi_phys __initdata;
static efi_system_table_t efi_systab __initdata;
static int __init setup_noefi(char *arg)
diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c
index cb91f985b4a..5d23d85624d 100644
--- a/arch/x86/kernel/efi_32.c
+++ b/arch/x86/kernel/efi_32.c
@@ -28,6 +28,7 @@
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
+#include <asm/efi.h>
/*
* To make EFI call EFI runtime service in physical addressing mode we need
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 824e21b80aa..4b87c32b639 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -409,7 +409,7 @@ restore_nocheck_notrace:
RESTORE_REGS
addl $4, %esp # skip orig_eax/error_code
CFI_ADJUST_CFA_OFFSET -4
-ENTRY(irq_return)
+irq_return:
INTERRUPT_RETURN
.section .fixup,"ax"
iret_exc:
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 6be39a387c5..2ad9a1bc6a7 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -583,7 +583,7 @@ retint_restore_args: /* return to kernel space */
restore_args:
RESTORE_ARGS 0,8,0
-ENTRY(irq_return)
+irq_return:
INTERRUPT_RETURN
.section __ex_table, "a"
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 24dbf56928d..ad2440832de 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -88,6 +88,9 @@ void __init x86_64_start_kernel(char * real_mode_data)
/* Make NULL pointers segfault */
zap_identity_mappings();
+ /* Cleanup the over mapped high alias */
+ cleanup_highmap();
+
for (i = 0; i < IDT_ENTRIES; i++) {
#ifdef CONFIG_EARLY_PRINTK
set_intr_gate(i, &early_idt_handlers[i]);
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 74ef4a41f22..25eb98540a4 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -612,7 +612,7 @@ ENTRY(swapper_pg_pmd)
ENTRY(swapper_pg_dir)
.fill 1024,4,0
#endif
-ENTRY(swapper_pg_fixmap)
+swapper_pg_fixmap:
.fill 1024,4,0
ENTRY(empty_zero_page)
.fill 4096,1,0
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 09b38d539b0..eb415043a92 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -107,8 +107,13 @@ startup_64:
movq %rdx, 0(%rbx, %rax, 8)
ident_complete:
- /* Fixup the kernel text+data virtual addresses
+ /*
+ * Fixup the kernel text+data virtual addresses. Note that
+ * we might write invalid pmds, when the kernel is relocated
+ * cleanup_highmap() fixes this up along with the mappings
+ * beyond _end.
*/
+
leaq level2_kernel_pgt(%rip), %rdi
leaq 4096(%rdi), %r8
/* See if it is a valid page table entry */
@@ -250,7 +255,7 @@ ENTRY(secondary_startup_64)
lretq
/* SMP bootup changes these two */
- __CPUINITDATA
+ __REFDATA
.align 8
ENTRY(initial_code)
.quad x86_64_start_kernel
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index 26719bd2c77..763dfc40723 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -39,7 +39,7 @@
#define HAVE_HWFP 1
#endif
-unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu;
+static unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu;
void mxcsr_feature_mask_init(void)
{
diff --git a/arch/x86/kernel/i8259_32.c b/arch/x86/kernel/i8259_32.c
index 2d25b77102f..fe631967d62 100644
--- a/arch/x86/kernel/i8259_32.c
+++ b/arch/x86/kernel/i8259_32.c
@@ -26,8 +26,6 @@
* present in the majority of PC/AT boxes.
* plus some generic x86 specific things if generic specifics makes
* any sense at all.
- * this file should become arch/i386/kernel/irq.c when the old irq.c
- * moves to arch independent land
*/
static int i8259A_auto_eoi;
@@ -362,23 +360,12 @@ void __init init_ISA_irqs (void)
#endif
init_8259A(0);
- for (i = 0; i < NR_IRQS; i++) {
- irq_desc[i].status = IRQ_DISABLED;
- irq_desc[i].action = NULL;
- irq_desc[i].depth = 1;
-
- if (i < 16) {
- /*
- * 16 old-style INTA-cycle interrupts:
- */
- set_irq_chip_and_handler_name(i, &i8259A_chip,
- handle_level_irq, "XT");
- } else {
- /*
- * 'high' PCI IRQs filled in on demand
- */
- irq_desc[i].chip = &no_irq_chip;
- }
+ /*
+ * 16 old-style INTA-cycle interrupts:
+ */
+ for (i = 0; i < 16; i++) {
+ set_irq_chip_and_handler_name(i, &i8259A_chip,
+ handle_level_irq, "XT");
}
}
diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c
index bd49321034d..c706a306155 100644
--- a/arch/x86/kernel/io_delay.c
+++ b/arch/x86/kernel/io_delay.c
@@ -13,7 +13,6 @@
#include <asm/io.h>
int io_delay_type __read_mostly = CONFIG_DEFAULT_IO_DELAY_TYPE;
-EXPORT_SYMBOL_GPL(io_delay_type);
static int __initdata io_delay_override;
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index a99e764fd66..34a591283f5 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -581,7 +581,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
* When a retprobed function returns, this code saves registers and
* calls trampoline_handler() runs, which calls the kretprobe's handler.
*/
-void __kprobes kretprobe_trampoline_holder(void)
+static void __used __kprobes kretprobe_trampoline_holder(void)
{
asm volatile (
".global kretprobe_trampoline\n"
@@ -673,7 +673,7 @@ void __kprobes kretprobe_trampoline_holder(void)
/*
* Called from kretprobe_trampoline
*/
-void * __kprobes trampoline_handler(struct pt_regs *regs)
+static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
{
struct kretprobe_instance *ri = NULL;
struct hlist_head *head, empty_rp;
diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c
index edd413650b3..6a0aa703868 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -46,9 +46,6 @@ static unsigned int nmi_hz = HZ;
static DEFINE_PER_CPU(short, wd_enabled);
-/* local prototypes */
-static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
-
static int endflag __initdata = 0;
#ifdef CONFIG_SMP
@@ -391,15 +388,6 @@ __kprobes int nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
return rc;
}
-int do_nmi_callback(struct pt_regs * regs, int cpu)
-{
-#ifdef CONFIG_SYSCTL
- if (unknown_nmi_panic)
- return unknown_nmi_panic_callback(regs, cpu);
-#endif
- return 0;
-}
-
#ifdef CONFIG_SYSCTL
static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
@@ -453,6 +441,15 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
#endif
+int do_nmi_callback(struct pt_regs *regs, int cpu)
+{
+#ifdef CONFIG_SYSCTL
+ if (unknown_nmi_panic)
+ return unknown_nmi_panic_callback(regs, cpu);
+#endif
+ return 0;
+}
+
void __trigger_all_cpu_backtrace(void)
{
int i;
diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index fb99484d21c..9a4fde74bee 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -46,9 +46,6 @@ static unsigned int nmi_hz = HZ;
static DEFINE_PER_CPU(short, wd_enabled);
-/* local prototypes */
-static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
-
/* Run after command line and cpu_init init, but before all other checks */
void nmi_watchdog_default(void)
{
@@ -394,15 +391,6 @@ asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
nmi_exit();
}
-int do_nmi_callback(struct pt_regs * regs, int cpu)
-{
-#ifdef CONFIG_SYSCTL
- if (unknown_nmi_panic)
- return unknown_nmi_panic_callback(regs, cpu);
-#endif
- return 0;
-}
-
void stop_nmi(void)
{
acpi_nmi_disable();
@@ -464,6 +452,15 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
#endif
+int do_nmi_callback(struct pt_regs *regs, int cpu)
+{
+#ifdef CONFIG_SYSCTL
+ if (unknown_nmi_panic)
+ return unknown_nmi_panic_callback(regs, cpu);
+#endif
+ return 0;
+}
+
void __trigger_all_cpu_backtrace(void)
{
int i;
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 691ab4cb167..a1d7071a51c 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -164,7 +164,6 @@ unsigned long mmu_cr4_features = X86_CR4_PAE;
unsigned int machine_id;
unsigned int machine_submodel_id;
unsigned int BIOS_revision;
-unsigned int mca_pentium_flag;
/* Boot loader ID as an integer, for the benefit of proc_dointvec */
int bootloader_type;
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index c0d8208af12..6fd804f0782 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -518,7 +518,7 @@ static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
}
#ifdef CONFIG_NUMA
-static int nearby_node(int apicid)
+static int __cpuinit nearby_node(int apicid)
{
int i, node;
@@ -791,7 +791,7 @@ static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
return 1;
}
-static void srat_detect_node(void)
+static void __cpuinit srat_detect_node(void)
{
#ifdef CONFIG_NUMA
unsigned node;
@@ -1046,7 +1046,7 @@ __setup("noclflush", setup_noclflush);
void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
{
if (c->x86_model_id[0])
- printk(KERN_INFO "%s", c->x86_model_id);
+ printk(KERN_CONT "%s", c->x86_model_id);
if (c->x86_mask || c->cpuid_level >= 0)
printk(KERN_CONT " stepping %02x\n", c->x86_mask);
diff --git a/arch/x86/kernel/topology.c b/arch/x86/kernel/topology.c
index a40051b71d9..0fcc95a354f 100644
--- a/arch/x86/kernel/topology.c
+++ b/arch/x86/kernel/topology.c
@@ -34,7 +34,7 @@
static DEFINE_PER_CPU(struct x86_cpu, cpu_devices);
#ifdef CONFIG_HOTPLUG_CPU
-int arch_register_cpu(int num)
+int __ref arch_register_cpu(int num)
{
/*
* CPU0 cannot be offlined due to several
diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S
index f1148ac8abe..2ffa9656fe7 100644
--- a/arch/x86/kernel/vmlinux_32.lds.S
+++ b/arch/x86/kernel/vmlinux_32.lds.S
@@ -38,7 +38,7 @@ SECTIONS
/* read-only */
.text : AT(ADDR(.text) - LOAD_OFFSET) {
- . = ALIGN(4096); /* not really needed, already page aligned */
+ . = ALIGN(PAGE_SIZE); /* not really needed, already page aligned */
*(.text.page_aligned)
TEXT_TEXT
SCHED_TEXT
@@ -70,21 +70,21 @@ SECTIONS
RODATA
/* writeable */
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.data : AT(ADDR(.data) - LOAD_OFFSET) { /* Data */
DATA_DATA
CONSTRUCTORS
} :data
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) {
__nosave_begin = .;
*(.data.nosave)
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__nosave_end = .;
}
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) {
*(.data.page_aligned)
*(.data.idt)
@@ -108,7 +108,7 @@ SECTIONS
}
/* might get freed after init */
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) {
__smp_locks = .;
*(.smp_locks)
@@ -120,10 +120,10 @@ SECTIONS
* after boot. Always make sure that ALIGN() directive is present after
* the section which contains __smp_alt_end.
*/
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
/* will be freed after init */
- . = ALIGN(4096); /* Init code and data */
+ . = ALIGN(PAGE_SIZE); /* Init code and data */
.init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
__init_begin = .;
_sinittext = .;
@@ -174,23 +174,23 @@ SECTIONS
EXIT_DATA
}
#if defined(CONFIG_BLK_DEV_INITRD)
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) {
__initramfs_start = .;
*(.init.ramfs)
__initramfs_end = .;
}
#endif
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) {
__per_cpu_start = .;
*(.data.percpu)
*(.data.percpu.shared_aligned)
__per_cpu_end = .;
}
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
/* freed after init ends here */
-
+
.bss : AT(ADDR(.bss) - LOAD_OFFSET) {
__init_end = .;
__bss_start = .; /* BSS */
@@ -200,7 +200,7 @@ SECTIONS
__bss_stop = .;
_end = . ;
/* This is where the kernel creates the early boot page tables */
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
pg0 = . ;
}
diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S
index 0992b9946c6..fab13229973 100644
--- a/arch/x86/kernel/vmlinux_64.lds.S
+++ b/arch/x86/kernel/vmlinux_64.lds.S
@@ -37,7 +37,7 @@ SECTIONS
KPROBES_TEXT
*(.fixup)
*(.gnu.warning)
- _etext = .; /* End of text section */
+ _etext = .; /* End of text section */
} :text = 0x9090
. = ALIGN(16); /* Exception table */
@@ -60,7 +60,7 @@ SECTIONS
__tracedata_end = .;
}
- . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */
+ . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */
/* Data */
.data : AT(ADDR(.data) - LOAD_OFFSET) {
DATA_DATA
@@ -119,7 +119,7 @@ SECTIONS
.vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3))
{ *(.vsyscall_3) }
- . = VSYSCALL_VIRT_ADDR + 4096;
+ . = VSYSCALL_VIRT_ADDR + PAGE_SIZE;
#undef VSYSCALL_ADDR
#undef VSYSCALL_PHYS_ADDR
@@ -129,28 +129,28 @@ SECTIONS
#undef VVIRT_OFFSET
#undef VVIRT
- . = ALIGN(8192); /* init_task */
+ . = ALIGN(THREAD_SIZE); /* init_task */
.data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) {
*(.data.init_task)
}:data.init
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
.data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) {
*(.data.page_aligned)
}
/* might get freed after init */
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__smp_alt_begin = .;
__smp_locks = .;
.smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) {
*(.smp_locks)
}
__smp_locks_end = .;
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__smp_alt_end = .;
- . = ALIGN(4096); /* Init code and data */
+ . = ALIGN(PAGE_SIZE); /* Init code and data */
__init_begin = .;
.init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
_sinittext = .;
@@ -191,7 +191,7 @@ SECTIONS
.altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) {
*(.altinstructions)
}
- __alt_instructions_end = .;
+ __alt_instructions_end = .;
.altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) {
*(.altinstr_replacement)
}
@@ -207,25 +207,25 @@ SECTIONS
/* vdso blob that is mapped into user space */
vdso_start = . ;
.vdso : AT(ADDR(.vdso) - LOAD_OFFSET) { *(.vdso) }
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
vdso_end = .;
#ifdef CONFIG_BLK_DEV_INITRD
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__initramfs_start = .;
.init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) }
__initramfs_end = .;
#endif
- PERCPU(4096)
+ PERCPU(PAGE_SIZE)
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__init_end = .;
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__nosave_begin = .;
.data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) }
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE);
__nosave_end = .;
__bss_start = .; /* BSS */
diff --git a/arch/x86/lib/csum-wrappers_64.c b/arch/x86/lib/csum-wrappers_64.c
index fd42a4a095f..459b58a8a15 100644
--- a/arch/x86/lib/csum-wrappers_64.c
+++ b/arch/x86/lib/csum-wrappers_64.c
@@ -1,117 +1,129 @@
-/* Copyright 2002,2003 Andi Kleen, SuSE Labs.
+/*
+ * Copyright 2002, 2003 Andi Kleen, SuSE Labs.
* Subject to the GNU Public License v.2
- *
+ *
* Wrappers of assembly checksum functions for x86-64.
*/
-
#include <asm/checksum.h>
#include <linux/module.h>
-/**
- * csum_partial_copy_from_user - Copy and checksum from user space.
- * @src: source address (user space)
+/**
+ * csum_partial_copy_from_user - Copy and checksum from user space.
+ * @src: source address (user space)
* @dst: destination address
* @len: number of bytes to be copied.
* @isum: initial sum that is added into the result (32bit unfolded)
* @errp: set to -EFAULT for an bad source address.
- *
+ *
* Returns an 32bit unfolded checksum of the buffer.
- * src and dst are best aligned to 64bits.
- */
+ * src and dst are best aligned to 64bits.
+ */
__wsum
csum_partial_copy_from_user(const void __user *src, void *dst,
int len, __wsum isum, int *errp)
-{
+{
might_sleep();
*errp = 0;
- if (likely(access_ok(VERIFY_READ,src, len))) {
- /* Why 6, not 7? To handle odd addresses aligned we
- would need to do considerable complications to fix the
- checksum which is defined as an 16bit accumulator. The
- fix alignment code is primarily for performance
- compatibility with 32bit and that will handle odd
- addresses slowly too. */
- if (unlikely((unsigned long)src & 6)) {
- while (((unsigned long)src & 6) && len >= 2) {
- __u16 val16;
- *errp = __get_user(val16, (const __u16 __user *)src);
- if (*errp)
- return isum;
- *(__u16 *)dst = val16;
- isum = (__force __wsum)add32_with_carry(
- (__force unsigned)isum, val16);
- src += 2;
- dst += 2;
- len -= 2;
- }
+
+ if (!likely(access_ok(VERIFY_READ, src, len)))
+ goto out_err;
+
+ /*
+ * Why 6, not 7? To handle odd addresses aligned we
+ * would need to do considerable complications to fix the
+ * checksum which is defined as an 16bit accumulator. The
+ * fix alignment code is primarily for performance
+ * compatibility with 32bit and that will handle odd
+ * addresses slowly too.
+ */
+ if (unlikely((unsigned long)src & 6)) {
+ while (((unsigned long)src & 6) && len >= 2) {
+ __u16 val16;
+
+ *errp = __get_user(val16, (const __u16 __user *)src);
+ if (*errp)
+ return isum;
+
+ *(__u16 *)dst = val16;
+ isum = (__force __wsum)add32_with_carry(
+ (__force unsigned)isum, val16);
+ src += 2;
+ dst += 2;
+ len -= 2;
}
- isum = csum_partial_copy_generic((__force const void *)src,
- dst, len, isum, errp, NULL);
- if (likely(*errp == 0))
- return isum;
- }
+ }
+ isum = csum_partial_copy_generic((__force const void *)src,
+ dst, len, isum, errp, NULL);
+ if (unlikely(*errp))
+ goto out_err;
+
+ return isum;
+
+out_err:
*errp = -EFAULT;
- memset(dst,0,len);
- return isum;
-}
+ memset(dst, 0, len);
+ return isum;
+}
EXPORT_SYMBOL(csum_partial_copy_from_user);
-/**
- * csum_partial_copy_to_user - Copy and checksum to user space.
+/**
+ * csum_partial_copy_to_user - Copy and checksum to user space.
* @src: source address
* @dst: destination address (user space)
* @len: number of bytes to be copied.
* @isum: initial sum that is added into the result (32bit unfolded)
* @errp: set to -EFAULT for an bad destination address.
- *
+ *
* Returns an 32bit unfolded checksum of the buffer.
* src and dst are best aligned to 64bits.
- */
+ */
__wsum
csum_partial_copy_to_user(const void *src, void __user *dst,
int len, __wsum isum, int *errp)
-{
+{
might_sleep();
+
if (unlikely(!access_ok(VERIFY_WRITE, dst, len))) {
*errp = -EFAULT;
- return 0;
+ return 0;
}
if (unlikely((unsigned long)dst & 6)) {
- while (((unsigned long)dst & 6) && len >= 2) {
+ while (((unsigned long)dst & 6) && len >= 2) {
__u16 val16 = *(__u16 *)src;
+
isum = (__force __wsum)add32_with_carry(
(__force unsigned)isum, val16);
*errp = __put_user(val16, (__u16 __user *)dst);
if (*errp)
return isum;
- src += 2;
- dst += 2;
+ src += 2;
+ dst += 2;
len -= 2;
}
}
*errp = 0;
- return csum_partial_copy_generic(src, (void __force *)dst,len,isum,NULL,errp);
-}
-
+ return csum_partial_copy_generic(src, (void __force *)dst,
+ len, isum, NULL, errp);
+}
EXPORT_SYMBOL(csum_partial_copy_to_user);
-/**
+/**
* csum_partial_copy_nocheck - Copy and checksum.
* @src: source address
* @dst: destination address
* @len: number of bytes to be copied.
* @isum: initial sum that is added into the result (32bit unfolded)
- *
+ *
* Returns an 32bit unfolded checksum of the buffer.
- */
+ */
__wsum
csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
-{
- return csum_partial_copy_generic(src,dst,len,sum,NULL,NULL);
-}
+{
+ return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
+}
EXPORT_SYMBOL(csum_partial_copy_nocheck);
__sum16 csum_ipv6_magic(const struct in6_addr *saddr,
@@ -119,17 +131,20 @@ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
__u32 len, unsigned short proto, __wsum sum)
{
__u64 rest, sum64;
-
+
rest = (__force __u64)htonl(len) + (__force __u64)htons(proto) +
(__force __u64)sum;
- asm(" addq (%[saddr]),%[sum]\n"
- " adcq 8(%[saddr]),%[sum]\n"
- " adcq (%[daddr]),%[sum]\n"
- " adcq 8(%[daddr]),%[sum]\n"
- " adcq $0,%[sum]\n"
- : [sum] "=r" (sum64)
- : "[sum]" (rest),[saddr] "r" (saddr), [daddr] "r" (daddr));
- return csum_fold((__force __wsum)add32_with_carry(sum64 & 0xffffffff, sum64>>32));
-}
+ asm(" addq (%[saddr]),%[sum]\n"
+ " adcq 8(%[saddr]),%[sum]\n"
+ " adcq (%[daddr]),%[sum]\n"
+ " adcq 8(%[daddr]),%[sum]\n"
+ " adcq $0,%[sum]\n"
+
+ : [sum] "=r" (sum64)
+ : "[sum]" (rest), [saddr] "r" (saddr), [daddr] "r" (daddr));
+
+ return csum_fold(
+ (__force __wsum)add32_with_carry(sum64 & 0xffffffff, sum64>>32));
+}
EXPORT_SYMBOL(csum_ipv6_magic);
diff --git a/arch/x86/lib/io_64.c b/arch/x86/lib/io_64.c
index 87b4a4e1803..3f1eb59b5f0 100644
--- a/arch/x86/lib/io_64.c
+++ b/arch/x86/lib/io_64.c
@@ -1,23 +1,25 @@
#include <linux/string.h>
-#include <asm/io.h>
#include <linux/module.h>
+#include <asm/io.h>
-void __memcpy_toio(unsigned long dst,const void*src,unsigned len)
+void __memcpy_toio(unsigned long dst, const void *src, unsigned len)
{
- __inline_memcpy((void *) dst,src,len);
+ __inline_memcpy((void *)dst, src, len);
}
EXPORT_SYMBOL(__memcpy_toio);
-void __memcpy_fromio(void *dst,unsigned long src,unsigned len)
+void __memcpy_fromio(void *dst, unsigned long src, unsigned len)
{
- __inline_memcpy(dst,(const void *) src,len);
+ __inline_memcpy(dst, (const void *)src, len);
}
EXPORT_SYMBOL(__memcpy_fromio);
void memset_io(volatile void __iomem *a, int b, size_t c)
{
- /* XXX: memset can mangle the IO patterns quite a bit.
- perhaps it would be better to use a dumb one */
- memset((void *)a,b,c);
+ /*
+ * TODO: memset can mangle the IO patterns quite a bit.
+ * perhaps it would be better to use a dumb one:
+ */
+ memset((void *)a, b, c);
}
EXPORT_SYMBOL(memset_io);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index a4a9cccdd4f..bb652f5a93f 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -171,6 +171,33 @@ set_pte_phys(unsigned long vaddr, unsigned long phys, pgprot_t prot)
__flush_tlb_one(vaddr);
}
+/*
+ * The head.S code sets up the kernel high mapping from:
+ * __START_KERNEL_map to __START_KERNEL_map + KERNEL_TEXT_SIZE
+ *
+ * phys_addr holds the negative offset to the kernel, which is added
+ * to the compile time generated pmds. This results in invalid pmds up
+ * to the point where we hit the physaddr 0 mapping.
+ *
+ * We limit the mappings to the region from _text to _end. _end is
+ * rounded up to the 2MB boundary. This catches the invalid pmds as
+ * well, as they are located before _text:
+ */
+void __init cleanup_highmap(void)
+{
+ unsigned long vaddr = __START_KERNEL_map;
+ unsigned long end = round_up((unsigned long)_end, PMD_SIZE) - 1;
+ pmd_t *pmd = level2_kernel_pgt;
+ pmd_t *last_pmd = pmd + PTRS_PER_PMD;
+
+ for (; pmd < last_pmd; pmd++, vaddr += PMD_SIZE) {
+ if (!pmd_present(*pmd))
+ continue;
+ if (vaddr < (unsigned long) _text || vaddr > end)
+ set_pmd(pmd, __pmd(0));
+ }
+}
+
/* NOTE: this is meant to be run only at boot */
void __init
__set_fixmap(enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 9f42d7e9c15..882328efc3d 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -42,6 +42,22 @@ int page_is_ram(unsigned long pagenr)
unsigned long addr, end;
int i;
+ /*
+ * A special case is the first 4Kb of memory;
+ * This is a BIOS owned area, not kernel ram, but generally
+ * not listed as such in the E820 table.
+ */
+ if (pagenr == 0)
+ return 0;
+
+ /*
+ * Second special case: Some BIOSen report the PC BIOS
+ * area (640->1Mb) as ram even though it is not.
+ */
+ if (pagenr >= (BIOS_BEGIN >> PAGE_SHIFT) &&
+ pagenr < (BIOS_END >> PAGE_SHIFT))
+ return 0;
+
for (i = 0; i < e820.nr_map; i++) {
/*
* Not usable memory:
@@ -51,14 +67,6 @@ int page_is_ram(unsigned long pagenr)
addr = (e820.map[i].addr + PAGE_SIZE-1) >> PAGE_SHIFT;
end = (e820.map[i].addr + e820.map[i].size) >> PAGE_SHIFT;
- /*
- * Sanity check: Some BIOSen report areas as RAM that
- * are not. Notably the 640->1Mb area, which is the
- * PCI BIOS area.
- */
- if (addr >= (BIOS_BEGIN >> PAGE_SHIFT) &&
- end < (BIOS_END >> PAGE_SHIFT))
- continue;
if ((pagenr >= addr) && (pagenr < end))
return 1;
@@ -126,6 +134,8 @@ static void __iomem *__ioremap(unsigned long phys_addr, unsigned long size,
return NULL;
}
+ WARN_ON_ONCE(page_is_ram(pfn));
+
switch (mode) {
case IOR_MODE_UNCACHED:
default:
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 1aecc658cd7..59898fb0a4a 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -494,11 +494,13 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
int i;
nodes_clear(node_possible_map);
+ nodes_clear(node_online_map);
#ifdef CONFIG_NUMA_EMU
if (cmdline && !numa_emulation(start_pfn, end_pfn))
return;
nodes_clear(node_possible_map);
+ nodes_clear(node_online_map);
#endif
#ifdef CONFIG_ACPI_NUMA
@@ -506,6 +508,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
end_pfn << PAGE_SHIFT))
return;
nodes_clear(node_possible_map);
+ nodes_clear(node_online_map);
#endif
#ifdef CONFIG_K8_NUMA
@@ -513,6 +516,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
end_pfn<<PAGE_SHIFT))
return;
nodes_clear(node_possible_map);
+ nodes_clear(node_online_map);
#endif
printk(KERN_INFO "%s\n",
numa_off ? "NUMA turned off" : "No NUMA configuration found");
@@ -524,7 +528,6 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
memnode_shift = 63;
memnodemap = memnode.embedded_map;
memnodemap[0] = 0;
- nodes_clear(node_online_map);
node_set_online(0);
node_set(0, node_possible_map);
for (i = 0; i < NR_CPUS; i++)
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 4119379f80f..464d8fc21ce 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -16,6 +16,7 @@
#include <asm/sections.h>
#include <asm/uaccess.h>
#include <asm/pgalloc.h>
+#include <asm/proto.h>
/*
* The current flushing context - we pass it instead of 5 arguments:
@@ -26,8 +27,23 @@ struct cpa_data {
pgprot_t mask_clr;
int numpages;
int flushtlb;
+ unsigned long pfn;
};
+#ifdef CONFIG_X86_64
+
+static inline unsigned long highmap_start_pfn(void)
+{
+ return __pa(_text) >> PAGE_SHIFT;
+}
+
+static inline unsigned long highmap_end_pfn(void)
+{
+ return __pa(round_up((unsigned long)_end, PMD_SIZE)) >> PAGE_SHIFT;
+}
+
+#endif
+
static inline int
within(unsigned long addr, unsigned long start, unsigned long end)
{
@@ -123,29 +139,14 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache)
}
}
-#define HIGH_MAP_START __START_KERNEL_map
-#define HIGH_MAP_END (__START_KERNEL_map + KERNEL_TEXT_SIZE)
-
-
-/*
- * Converts a virtual address to a X86-64 highmap address
- */
-static unsigned long virt_to_highmap(void *address)
-{
-#ifdef CONFIG_X86_64
- return __pa((unsigned long)address) + HIGH_MAP_START - phys_base;
-#else
- return (unsigned long)address;
-#endif
-}
-
/*
* Certain areas of memory on x86 require very specific protection flags,
* for example the BIOS area or kernel text. Callers don't always get this
* right (again, ioremap() on BIOS memory is not uncommon) so this function
* checks and fixes these known static required protection bits.
*/
-static inline pgprot_t static_protections(pgprot_t prot, unsigned long address)
+static inline pgprot_t static_protections(pgprot_t prot, unsigned long address,
+ unsigned long pfn)
{
pgprot_t forbidden = __pgprot(0);
@@ -153,30 +154,23 @@ static inline pgprot_t static_protections(pgprot_t prot, unsigned long address)
* The BIOS area between 640k and 1Mb needs to be executable for
* PCI BIOS based config access (CONFIG_PCI_GOBIOS) support.
*/
- if (within(__pa(address), BIOS_BEGIN, BIOS_END))
+ if (within(pfn, BIOS_BEGIN >> PAGE_SHIFT, BIOS_END >> PAGE_SHIFT))
pgprot_val(forbidden) |= _PAGE_NX;
/*
* The kernel text needs to be executable for obvious reasons
- * Does not cover __inittext since that is gone later on
+ * Does not cover __inittext since that is gone later on. On
+ * 64bit we do not enforce !NX on the low mapping
*/
if (within(address, (unsigned long)_text, (unsigned long)_etext))
pgprot_val(forbidden) |= _PAGE_NX;
- /*
- * Do the same for the x86-64 high kernel mapping
- */
- if (within(address, virt_to_highmap(_text), virt_to_highmap(_etext)))
- pgprot_val(forbidden) |= _PAGE_NX;
- /* The .rodata section needs to be read-only */
- if (within(address, (unsigned long)__start_rodata,
- (unsigned long)__end_rodata))
- pgprot_val(forbidden) |= _PAGE_RW;
/*
- * Do the same for the x86-64 high kernel mapping
+ * The .rodata section needs to be read-only. Using the pfn
+ * catches all aliases.
*/
- if (within(address, virt_to_highmap(__start_rodata),
- virt_to_highmap(__end_rodata)))
+ if (within(pfn, __pa((unsigned long)__start_rodata) >> PAGE_SHIFT,
+ __pa((unsigned long)__end_rodata) >> PAGE_SHIFT))
pgprot_val(forbidden) |= _PAGE_RW;
prot = __pgprot(pgprot_val(prot) & ~pgprot_val(forbidden));
@@ -253,7 +247,7 @@ static int
try_preserve_large_page(pte_t *kpte, unsigned long address,
struct cpa_data *cpa)
{
- unsigned long nextpage_addr, numpages, pmask, psize, flags, addr;
+ unsigned long nextpage_addr, numpages, pmask, psize, flags, addr, pfn;
pte_t new_pte, old_pte, *tmp;
pgprot_t old_prot, new_prot;
int i, do_split = 1;
@@ -301,7 +295,15 @@ try_preserve_large_page(pte_t *kpte, unsigned long address,
pgprot_val(new_prot) &= ~pgprot_val(cpa->mask_clr);
pgprot_val(new_prot) |= pgprot_val(cpa->mask_set);
- new_prot = static_protections(new_prot, address);
+
+ /*
+ * old_pte points to the large page base address. So we need
+ * to add the offset of the virtual address:
+ */
+ pfn = pte_pfn(old_pte) + ((address & (psize - 1)) >> PAGE_SHIFT);
+ cpa->pfn = pfn;
+
+ new_prot = static_protections(new_prot, address, pfn);
/*
* We need to check the full range, whether
@@ -309,8 +311,9 @@ try_preserve_large_page(pte_t *kpte, unsigned long address,
* the pages in the range we try to preserve:
*/
addr = address + PAGE_SIZE;
- for (i = 1; i < cpa->numpages; i++, addr += PAGE_SIZE) {
- pgprot_t chk_prot = static_protections(new_prot, addr);
+ pfn++;
+ for (i = 1; i < cpa->numpages; i++, addr += PAGE_SIZE, pfn++) {
+ pgprot_t chk_prot = static_protections(new_prot, addr, pfn);
if (pgprot_val(chk_prot) != pgprot_val(new_prot))
goto out_unlock;
@@ -505,46 +508,46 @@ out_unlock:
return 0;
}
-static int __change_page_attr(unsigned long address, struct cpa_data *cpa)
+static int __change_page_attr(struct cpa_data *cpa, int primary)
{
+ unsigned long address = cpa->vaddr;
int do_split, err;
unsigned int level;
- struct page *kpte_page;
- pte_t *kpte;
+ pte_t *kpte, old_pte;
repeat:
kpte = lookup_address(address, &level);
if (!kpte)
- return -EINVAL;
+ return primary ? -EINVAL : 0;
- kpte_page = virt_to_page(kpte);
- BUG_ON(PageLRU(kpte_page));
- BUG_ON(PageCompound(kpte_page));
+ old_pte = *kpte;
+ if (!pte_val(old_pte)) {
+ if (!primary)
+ return 0;
+ printk(KERN_WARNING "CPA: called for zero pte. "
+ "vaddr = %lx cpa->vaddr = %lx\n", address,
+ cpa->vaddr);
+ WARN_ON(1);
+ return -EINVAL;
+ }
if (level == PG_LEVEL_4K) {
- pte_t new_pte, old_pte = *kpte;
+ pte_t new_pte;
pgprot_t new_prot = pte_pgprot(old_pte);
-
- if(!pte_val(old_pte)) {
- printk(KERN_WARNING "CPA: called for zero pte. "
- "vaddr = %lx cpa->vaddr = %lx\n", address,
- cpa->vaddr);
- WARN_ON(1);
- return -EINVAL;
- }
+ unsigned long pfn = pte_pfn(old_pte);
pgprot_val(new_prot) &= ~pgprot_val(cpa->mask_clr);
pgprot_val(new_prot) |= pgprot_val(cpa->mask_set);
- new_prot = static_protections(new_prot, address);
+ new_prot = static_protections(new_prot, address, pfn);
/*
* We need to keep the pfn from the existing PTE,
* after all we're only going to change it's attributes
* not the memory it points to
*/
- new_pte = pfn_pte(pte_pfn(old_pte), canon_pgprot(new_prot));
-
+ new_pte = pfn_pte(pfn, canon_pgprot(new_prot));
+ cpa->pfn = pfn;
/*
* Do we really change anything ?
*/
@@ -581,67 +584,59 @@ repeat:
return err;
}
-/**
- * change_page_attr_addr - Change page table attributes in linear mapping
- * @address: Virtual address in linear mapping.
- * @prot: New page table attribute (PAGE_*)
- *
- * Change page attributes of a page in the direct mapping. This is a variant
- * of change_page_attr() that also works on memory holes that do not have
- * mem_map entry (pfn_valid() is false).
- *
- * See change_page_attr() documentation for more details.
- *
- * Modules and drivers should use the set_memory_* APIs instead.
- */
-static int change_page_attr_addr(struct cpa_data *cpa)
+static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias);
+
+static int cpa_process_alias(struct cpa_data *cpa)
{
- int err;
- unsigned long address = cpa->vaddr;
+ struct cpa_data alias_cpa;
+ int ret = 0;
-#ifdef CONFIG_X86_64
- unsigned long phys_addr = __pa(address);
+ if (cpa->pfn > max_pfn_mapped)
+ return 0;
/*
- * If we are inside the high mapped kernel range, then we
- * fixup the low mapping first. __va() returns the virtual
- * address in the linear mapping:
+ * No need to redo, when the primary call touched the direct
+ * mapping already:
*/
- if (within(address, HIGH_MAP_START, HIGH_MAP_END))
- address = (unsigned long) __va(phys_addr);
-#endif
+ if (!within(cpa->vaddr, PAGE_OFFSET,
+ PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))) {
- err = __change_page_attr(address, cpa);
- if (err)
- return err;
+ alias_cpa = *cpa;
+ alias_cpa.vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT);
+
+ ret = __change_page_attr_set_clr(&alias_cpa, 0);
+ }
#ifdef CONFIG_X86_64
+ if (ret)
+ return ret;
+ /*
+ * No need to redo, when the primary call touched the high
+ * mapping already:
+ */
+ if (within(cpa->vaddr, (unsigned long) _text, (unsigned long) _end))
+ return 0;
+
/*
* If the physical address is inside the kernel map, we need
* to touch the high mapped kernel as well:
*/
- if (within(phys_addr, 0, KERNEL_TEXT_SIZE)) {
- /*
- * Calc the high mapping address. See __phys_addr()
- * for the non obvious details.
- *
- * Note that NX and other required permissions are
- * checked in static_protections().
- */
- address = phys_addr + HIGH_MAP_START - phys_base;
+ if (!within(cpa->pfn, highmap_start_pfn(), highmap_end_pfn()))
+ return 0;
- /*
- * Our high aliases are imprecise, because we check
- * everything between 0 and KERNEL_TEXT_SIZE, so do
- * not propagate lookup failures back to users:
- */
- __change_page_attr(address, cpa);
- }
+ alias_cpa = *cpa;
+ alias_cpa.vaddr =
+ (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base;
+
+ /*
+ * The high mapping range is imprecise, so ignore the return value.
+ */
+ __change_page_attr_set_clr(&alias_cpa, 0);
#endif
- return err;
+ return ret;
}
-static int __change_page_attr_set_clr(struct cpa_data *cpa)
+static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias)
{
int ret, numpages = cpa->numpages;
@@ -651,10 +646,17 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa)
* preservation check.
*/
cpa->numpages = numpages;
- ret = change_page_attr_addr(cpa);
+
+ ret = __change_page_attr(cpa, checkalias);
if (ret)
return ret;
+ if (checkalias) {
+ ret = cpa_process_alias(cpa);
+ if (ret)
+ return ret;
+ }
+
/*
* Adjust the number of pages with the result of the
* CPA operation. Either a large page has been
@@ -677,7 +679,7 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages,
pgprot_t mask_set, pgprot_t mask_clr)
{
struct cpa_data cpa;
- int ret, cache;
+ int ret, cache, checkalias;
/*
* Check, if we are requested to change a not supported
@@ -703,7 +705,10 @@ static int change_page_attr_set_clr(unsigned long addr, int numpages,
cpa.mask_clr = mask_clr;
cpa.flushtlb = 0;
- ret = __change_page_attr_set_clr(&cpa);
+ /* No alias checking for _NX bit modifications */
+ checkalias = (pgprot_val(mask_set) | pgprot_val(mask_clr)) != _PAGE_NX;
+
+ ret = __change_page_attr_set_clr(&cpa, checkalias);
/*
* Check whether we really changed something:
@@ -841,7 +846,7 @@ static int __set_pages_p(struct page *page, int numpages)
.mask_set = __pgprot(_PAGE_PRESENT | _PAGE_RW),
.mask_clr = __pgprot(0)};
- return __change_page_attr_set_clr(&cpa);
+ return __change_page_attr_set_clr(&cpa, 1);
}
static int __set_pages_np(struct page *page, int numpages)
@@ -851,7 +856,7 @@ static int __set_pages_np(struct page *page, int numpages)
.mask_set = __pgprot(0),
.mask_clr = __pgprot(_PAGE_PRESENT | _PAGE_RW)};
- return __change_page_attr_set_clr(&cpa);
+ return __change_page_attr_set_clr(&cpa, 1);
}
void kernel_map_pages(struct page *page, int numpages, int enable)
@@ -894,7 +899,24 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
*/
cpa_fill_pool();
}
-#endif
+
+#ifdef CONFIG_HIBERNATION
+
+bool kernel_page_present(struct page *page)
+{
+ unsigned int level;
+ pte_t *pte;
+
+ if (PageHighMem(page))
+ return false;
+
+ pte = lookup_address((unsigned long)page_address(page), &level);
+ return (pte_val(*pte) & _PAGE_PRESENT);
+}
+
+#endif /* CONFIG_HIBERNATION */
+
+#endif /* CONFIG_DEBUG_PAGEALLOC */
/*
* The testcases use internal knowledge of the implementation that shouldn't
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index ecd91ea8a8a..845001c617c 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -166,7 +166,8 @@ static inline int save_add_info(void) {return 0;}
* Both SPARSE and RESERVE need nodes_add information.
* This code supports one contiguous hot add area per node.
*/
-static int reserve_hotadd(int node, unsigned long start, unsigned long end)
+static int __init
+reserve_hotadd(int node, unsigned long start, unsigned long end)
{
unsigned long s_pfn = start >> PAGE_SHIFT;
unsigned long e_pfn = end >> PAGE_SHIFT;
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index b7c67a187b6..7b6e3bb9b28 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -541,7 +541,7 @@ void pcibios_disable_device (struct pci_dev *dev)
pcibios_disable_irq(dev);
}
-struct pci_bus *pci_scan_bus_with_sysdata(int busno)
+struct pci_bus *__devinit pci_scan_bus_with_sysdata(int busno)
{
struct pci_bus *bus = NULL;
struct pci_sysdata *sd;
diff --git a/arch/x86/power/hibernate_asm_64.S b/arch/x86/power/hibernate_asm_64.S
index 1deb3244b99..000415947d9 100644
--- a/arch/x86/power/hibernate_asm_64.S
+++ b/arch/x86/power/hibernate_asm_64.S
@@ -20,6 +20,7 @@
#include <asm/segment.h>
#include <asm/page.h>
#include <asm/asm-offsets.h>
+#include <asm/processor-flags.h>
ENTRY(swsusp_arch_suspend)
movq $saved_context, %rax
@@ -60,7 +61,7 @@ ENTRY(restore_image)
/* Flush TLB */
movq mmu_cr4_features(%rip), %rax
movq %rax, %rdx
- andq $~(1<<7), %rdx # PGE
+ andq $~(X86_CR4_PGE), %rdx
movq %rdx, %cr4; # turn off PGE
movq %cr3, %rcx; # flush TLB
movq %rcx, %cr3;
@@ -112,7 +113,7 @@ ENTRY(restore_registers)
/* Flush TLB, including "global" things (vmalloc) */
movq mmu_cr4_features(%rip), %rax
movq %rax, %rdx
- andq $~(1<<7), %rdx; # PGE
+ andq $~(X86_CR4_PGE), %rdx
movq %rdx, %cr4; # turn off PGE
movq %cr3, %rcx; # flush TLB
movq %rcx, %cr3