From: www.kernel.org
Subject: Update to 2.6.23
Patch-mainline: 2.6.23
Automatically created from "patches.kernel.org/patch-2.6.23" by xen-port-patches.py
Acked-by: jbeulich@novell.com
--- head.orig/arch/x86/Kbuild 2012-06-12 15:06:23.000000000 +0200
+++ head/arch/x86/Kbuild 2011-01-31 17:49:31.000000000 +0100
@@ -2,7 +2,7 @@
obj-$(CONFIG_KVM) += kvm/
# Xen paravirtualization support
-obj-$(CONFIG_XEN) += xen/
+obj-$(CONFIG_PARAVIRT_XEN) += xen/
# lguest paravirtualization support
obj-$(CONFIG_LGUEST_GUEST) += lguest/
--- head.orig/arch/x86/kernel/asm-offsets.c 2012-06-12 15:06:23.000000000 +0200
+++ head/arch/x86/kernel/asm-offsets.c 2011-04-13 12:42:43.000000000 +0200
@@ -17,7 +17,7 @@
#include <asm/bootparam.h>
#include <asm/suspend.h>
-#ifdef CONFIG_XEN
+#if defined(CONFIG_XEN) || defined(CONFIG_PARAVIRT_XEN)
#include <xen/interface/xen.h>
#endif
@@ -55,7 +55,7 @@ void common(void) {
OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2);
#endif
-#ifdef CONFIG_XEN
+#ifdef CONFIG_PARAVIRT_XEN
BLANK();
OFFSET(XEN_vcpu_info_mask, vcpu_info, evtchn_upcall_mask);
OFFSET(XEN_vcpu_info_pending, vcpu_info, evtchn_upcall_pending);
--- head.orig/arch/x86/kernel/cpu/common-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/cpu/common-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -362,6 +362,8 @@ static void __cpuinit generic_identify(s
if ( xlvl >= 0x80000004 )
get_model_name(c); /* Default name */
}
+
+ init_scattered_cpuid_features(c);
}
early_intel_workaround(c);
@@ -613,7 +615,6 @@ extern int nsc_init_cpu(void);
extern int amd_init_cpu(void);
extern int centaur_init_cpu(void);
extern int transmeta_init_cpu(void);
-extern int rise_init_cpu(void);
extern int nexgen_init_cpu(void);
extern int umc_init_cpu(void);
@@ -625,7 +626,6 @@ void __init early_cpu_init(void)
amd_init_cpu();
centaur_init_cpu();
transmeta_init_cpu();
- rise_init_cpu();
nexgen_init_cpu();
umc_init_cpu();
early_cpu_detect();
--- head.orig/arch/x86/kernel/cpu/mtrr/main-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/cpu/mtrr/main-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -167,7 +167,7 @@ mtrr_del(int reg, unsigned long base, un
EXPORT_SYMBOL(mtrr_add);
EXPORT_SYMBOL(mtrr_del);
-__init void mtrr_bp_init(void)
+void __init mtrr_bp_init(void)
{
}
--- head.orig/arch/x86/kernel/e820_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/e820_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -10,6 +10,7 @@
#include <linux/efi.h>
#include <linux/pfn.h>
#include <linux/uaccess.h>
+#include <linux/suspend.h>
#include <asm/pgtable.h>
#include <asm/page.h>
@@ -343,6 +344,37 @@ static int __init request_standard_resou
subsys_initcall(request_standard_resources);
+#if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION)
+/**
+ * e820_mark_nosave_regions - Find the ranges of physical addresses that do not
+ * correspond to e820 RAM areas and mark the corresponding pages as nosave for
+ * hibernation.
+ *
+ * This function requires the e820 map to be sorted and without any
+ * overlapping entries and assumes the first e820 area to be RAM.
+ */
+void __init e820_mark_nosave_regions(void)
+{
+ int i;
+ unsigned long pfn;
+
+ pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size);
+ for (i = 1; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+
+ if (pfn < PFN_UP(ei->addr))
+ register_nosave_region(pfn, PFN_UP(ei->addr));
+
+ pfn = PFN_DOWN(ei->addr + ei->size);
+ if (ei->type != E820_RAM)
+ register_nosave_region(PFN_UP(ei->addr), pfn);
+
+ if (pfn >= max_low_pfn)
+ break;
+ }
+}
+#endif
+
void __init add_memory_region(unsigned long long start,
unsigned long long size, int type)
{
@@ -791,7 +823,7 @@ void __init print_memory_map(char *who)
case E820_NVS:
printk("(ACPI NVS)\n");
break;
- default: printk("type %lu\n", e820.map[i].type);
+ default: printk("type %u\n", e820.map[i].type);
break;
}
}
--- head.orig/arch/x86/kernel/entry_32.S 2012-04-04 09:17:16.000000000 +0200
+++ head/arch/x86/kernel/entry_32.S 2012-04-04 10:14:44.000000000 +0200
@@ -1048,7 +1048,7 @@ ENTRY(kernel_thread_helper)
CFI_ENDPROC
ENDPROC(kernel_thread_helper)
-#ifdef CONFIG_XEN
+#ifdef CONFIG_PARAVIRT_XEN
/* Xen doesn't set %esp to be precisely what the normal sysenter
entrypoint expects, so fix it up before using the normal path. */
ENTRY(xen_sysenter_target)
@@ -1140,7 +1140,7 @@ ENDPROC(xen_failsafe_callback)
BUILD_INTERRUPT3(xen_hvm_callback_vector, XEN_HVM_EVTCHN_CALLBACK,
xen_evtchn_do_upcall)
-#endif /* CONFIG_XEN */
+#endif /* CONFIG_PARAVIRT_XEN */
#ifdef CONFIG_FUNCTION_TRACER
#ifdef CONFIG_DYNAMIC_FTRACE
--- head.orig/arch/x86/kernel/entry_32-xen.S 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/entry_32-xen.S 2011-01-31 17:49:31.000000000 +0100
@@ -452,9 +452,6 @@ restore_nocheck_notrace:
1: INTERRUPT_RETURN
.section .fixup,"ax"
iret_exc:
-#ifndef CONFIG_XEN
- ENABLE_INTERRUPTS(CLBR_NONE)
-#endif
pushl $0 # no error code
pushl $do_iret_error
jmp error_code
--- head.orig/arch/x86/kernel/head_32-xen.S 2011-08-09 10:37:32.000000000 +0200
+++ head/arch/x86/kernel/head_32-xen.S 2011-08-09 10:41:46.000000000 +0200
@@ -87,7 +87,10 @@ ENTRY(_stext)
/*
* BSS section
*/
-.section ".bss.page_aligned","w"
+.section ".bss.page_aligned","wa"
+ .align PAGE_SIZE_asm
+ENTRY(swapper_pg_pmd)
+ .fill 1024,4,0
ENTRY(empty_zero_page)
.fill 4096,1,0
@@ -149,36 +152,36 @@ ENTRY(empty_zero_page)
#endif /* CONFIG_XEN_COMPAT <= 0x030002 */
- ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz, "linux")
- ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz, "2.6")
- ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz, "xen-3.0")
- ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, .long, __PAGE_OFFSET)
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz "linux")
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz "2.6")
+ ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz "xen-3.0")
+ ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, .long __PAGE_OFFSET)
#if CONFIG_XEN_COMPAT <= 0x030002
- ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .long, __PAGE_OFFSET)
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .long __PAGE_OFFSET)
#else
- ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .long, 0)
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .long 0)
#endif
- ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .long, startup_32)
- ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .long, hypercall_page)
- ELFNOTE(Xen, XEN_ELFNOTE_HV_START_LOW, .long, HYPERVISOR_VIRT_START)
- ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .ascii, "writable_page_tables";
- .ascii "|writable_descriptor_tables";
- .ascii "|auto_translated_physmap";
- .ascii "|pae_pgdir_above_4gb";
- .ascii "|supervisor_mode_kernel";
- .asciz XEN_DOM0_CAP_STR)
- ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES, .long, XEN_DOM0_CAP |
+ ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .long startup_32)
+ ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .long hypercall_page)
+ ELFNOTE(Xen, XEN_ELFNOTE_HV_START_LOW, .long HYPERVISOR_VIRT_START)
+ ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .ascii "writable_page_tables";
+ .ascii "|writable_descriptor_tables";
+ .ascii "|auto_translated_physmap";
+ .ascii "|pae_pgdir_above_4gb";
+ .ascii "|supervisor_mode_kernel";
+ .asciz XEN_DOM0_CAP_STR)
+ ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES, .long XEN_DOM0_CAP |
(1 << XENFEAT_writable_page_tables) |
(1 << XENFEAT_writable_descriptor_tables) |
(1 << XENFEAT_auto_translated_physmap) |
(1 << XENFEAT_pae_pgdir_above_4gb) |
(1 << XENFEAT_supervisor_mode_kernel))
#ifdef CONFIG_X86_PAE
- ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz, "yes")
- ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .quad, _PAGE_PRESENT,_PAGE_PRESENT)
+ ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz "yes")
+ ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .quad _PAGE_PRESENT, _PAGE_PRESENT)
#else
- ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz, "no")
- ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .long, _PAGE_PRESENT,_PAGE_PRESENT)
+ ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz "no")
+ ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .long _PAGE_PRESENT, _PAGE_PRESENT)
#endif
- ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz, "generic")
- ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long, 1)
+ ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz "generic")
+ ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long 1)
--- head.orig/arch/x86/kernel/io_apic_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/io_apic_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -402,14 +402,6 @@ static void set_ioapic_affinity_irq(unsi
# include <linux/slab.h> /* kmalloc() */
# include <linux/timer.h> /* time_after() */
-#ifdef CONFIG_BALANCED_IRQ_DEBUG
-# define TDprintk(x...) do { printk("<%ld:%s:%d>: ", jiffies, __FILE__, __LINE__); printk(x); } while (0)
-# define Dprintk(x...) do { TDprintk(x); } while (0)
-# else
-# define TDprintk(x...)
-# define Dprintk(x...)
-# endif
-
#define IRQBALANCE_CHECK_ARCH -999
#define MAX_BALANCED_IRQ_INTERVAL (5*HZ)
#define MIN_BALANCED_IRQ_INTERVAL (HZ/2)
@@ -492,7 +484,7 @@ static inline void balance_irq(int cpu,
static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold)
{
int i, j;
- Dprintk("Rotating IRQs among CPUs.\n");
+
for_each_online_cpu(i) {
for (j = 0; j < NR_IRQS; j++) {
if (!irq_desc[j].action)
@@ -609,19 +601,11 @@ tryanothercpu:
max_loaded = tmp_loaded; /* processor */
imbalance = (max_cpu_irq - min_cpu_irq) / 2;
- Dprintk("max_loaded cpu = %d\n", max_loaded);
- Dprintk("min_loaded cpu = %d\n", min_loaded);
- Dprintk("max_cpu_irq load = %ld\n", max_cpu_irq);
- Dprintk("min_cpu_irq load = %ld\n", min_cpu_irq);
- Dprintk("load imbalance = %lu\n", imbalance);
-
/* if imbalance is less than approx 10% of max load, then
* observe diminishing returns action. - quit
*/
- if (imbalance < (max_cpu_irq >> 3)) {
- Dprintk("Imbalance too trivial\n");
+ if (imbalance < (max_cpu_irq >> 3))
goto not_worth_the_effort;
- }
tryanotherirq:
/* if we select an IRQ to move that can't go where we want, then
@@ -678,9 +662,6 @@ tryanotherirq:
cpus_and(tmp, target_cpu_mask, allowed_mask);
if (!cpus_empty(tmp)) {
-
- Dprintk("irq = %d moved to cpu = %d\n",
- selected_irq, min_loaded);
/* mark for change destination */
set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded));
@@ -700,7 +681,6 @@ not_worth_the_effort:
*/
balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL,
balanced_irq_interval + BALANCED_IRQ_MORE_DELTA);
- Dprintk("IRQ worth rotating not found\n");
return;
}
@@ -716,6 +696,7 @@ static int balanced_irq(void *unused)
set_pending_irq(i, cpumask_of_cpu(0));
}
+ set_freezable();
for ( ; ; ) {
time_remaining = schedule_timeout_interruptible(time_remaining);
try_to_freeze();
@@ -825,14 +806,6 @@ static int pirq_entries [MAX_PIRQS];
static int pirqs_enabled;
int skip_ioapic_setup;
-static int __init ioapic_setup(char *str)
-{
- skip_ioapic_setup = 1;
- return 1;
-}
-
-__setup("noapic", ioapic_setup);
-
static int __init ioapic_pirq_setup(char *str)
{
int i, max;
@@ -1323,12 +1296,15 @@ static struct irq_chip ioapic_chip;
static void ioapic_register_intr(int irq, int vector, unsigned long trigger)
{
if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
- trigger == IOAPIC_LEVEL)
+ trigger == IOAPIC_LEVEL) {
+ irq_desc[irq].status |= IRQ_LEVEL;
set_irq_chip_and_handler_name(irq, &ioapic_chip,
handle_fasteoi_irq, "fasteoi");
- else
+ } else {
+ irq_desc[irq].status &= ~IRQ_LEVEL;
set_irq_chip_and_handler_name(irq, &ioapic_chip,
handle_edge_irq, "edge");
+ }
set_intr_gate(vector, interrupt[irq]);
}
#else
@@ -1957,7 +1933,7 @@ __setup("no_timer_check", notimercheck);
* - if this function detects that timer IRQs are defunct, then we fall
* back to ISA timer IRQs
*/
-int __init timer_irq_works(void)
+static int __init timer_irq_works(void)
{
unsigned long t1 = jiffies;
--- head.orig/arch/x86/kernel/microcode_core-xen.c 2011-12-01 15:01:09.000000000 +0100
+++ head/arch/x86/kernel/microcode_core-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -33,6 +33,7 @@
#include <linux/miscdevice.h>
#include <linux/spinlock.h>
#include <linux/mm.h>
+#include <linux/fs.h>
#include <linux/mutex.h>
#include <linux/cpu.h>
#include <linux/firmware.h>
--- head.orig/arch/x86/kernel/pci-dma-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/pci-dma-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -24,7 +24,7 @@
#include <asm/bug.h>
#ifdef __x86_64__
-#include <asm/proto.h>
+#include <asm/iommu.h>
int iommu_merge __read_mostly = 0;
EXPORT_SYMBOL(iommu_merge);
--- head.orig/arch/x86/kernel/process_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/process_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -241,6 +241,7 @@ early_param("idle", idle_setup);
void show_regs(struct pt_regs * regs)
{
unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
+ unsigned long d0, d1, d2, d3, d6, d7;
printk("\n");
printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
@@ -265,6 +266,17 @@ void show_regs(struct pt_regs * regs)
cr3 = read_cr3();
cr4 = read_cr4_safe();
printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n", cr0, cr2, cr3, cr4);
+
+ get_debugreg(d0, 0);
+ get_debugreg(d1, 1);
+ get_debugreg(d2, 2);
+ get_debugreg(d3, 3);
+ printk("DR0: %08lx DR1: %08lx DR2: %08lx DR3: %08lx\n",
+ d0, d1, d2, d3);
+ get_debugreg(d6, 6);
+ get_debugreg(d7, 7);
+ printk("DR6: %08lx DR7: %08lx\n", d6, d7);
+
show_trace(NULL, regs, ®s->esp);
}
@@ -473,7 +485,30 @@ int dump_task_regs(struct task_struct *t
return 1;
}
-static noinline void __switch_to_xtra(struct task_struct *next_p)
+#ifdef CONFIG_SECCOMP
+void hard_disable_TSC(void)
+{
+ write_cr4(read_cr4() | X86_CR4_TSD);
+}
+void disable_TSC(void)
+{
+ preempt_disable();
+ if (!test_and_set_thread_flag(TIF_NOTSC))
+ /*
+ * Must flip the CPU state synchronously with
+ * TIF_NOTSC in the current running context.
+ */
+ hard_disable_TSC();
+ preempt_enable();
+}
+void hard_enable_TSC(void)
+{
+ write_cr4(read_cr4() & ~X86_CR4_TSD);
+}
+#endif /* CONFIG_SECCOMP */
+
+static noinline void
+__switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p)
{
struct thread_struct *next;
@@ -488,33 +523,17 @@ static noinline void __switch_to_xtra(st
set_debugreg(next->debugreg[6], 6);
set_debugreg(next->debugreg[7], 7);
}
-}
-/*
- * This function selects if the context switch from prev to next
- * has to tweak the TSC disable bit in the cr4.
- */
-static inline void disable_tsc(struct task_struct *prev_p,
- struct task_struct *next_p)
-{
- struct thread_info *prev, *next;
-
- /*
- * gcc should eliminate the ->thread_info dereference if
- * has_secure_computing returns 0 at compile time (SECCOMP=n).
- */
- prev = task_thread_info(prev_p);
- next = task_thread_info(next_p);
-
- if (has_secure_computing(prev) || has_secure_computing(next)) {
- /* slow path here */
- if (has_secure_computing(prev) &&
- !has_secure_computing(next)) {
- write_cr4(read_cr4() & ~X86_CR4_TSD);
- } else if (!has_secure_computing(prev) &&
- has_secure_computing(next))
- write_cr4(read_cr4() | X86_CR4_TSD);
+#ifdef CONFIG_SECCOMP
+ if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
+ test_tsk_thread_flag(next_p, TIF_NOTSC)) {
+ /* prev and next are different */
+ if (test_tsk_thread_flag(next_p, TIF_NOTSC))
+ hard_disable_TSC();
+ else
+ hard_enable_TSC();
}
+#endif
}
/*
@@ -649,10 +668,9 @@ struct task_struct fastcall * __switch_t
/*
* Now maybe handle debug registers
*/
- if (unlikely(task_thread_info(next_p)->flags & _TIF_WORK_CTXSW))
- __switch_to_xtra(next_p);
-
- disable_tsc(prev_p, next_p);
+ if (unlikely(task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV ||
+ task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
+ __switch_to_xtra(prev_p, next_p);
/*
* Leave lazy mode, flushing any hypercalls made here.
--- head.orig/arch/x86/kernel/setup_32-xen.c 2012-06-06 13:49:07.000000000 +0200
+++ head/arch/x86/kernel/setup_32-xen.c 2012-06-06 13:52:42.000000000 +0200
@@ -114,19 +114,10 @@ static unsigned int highmem_pages = -1;
/*
* Setup options
*/
-struct drive_info_struct { char dummy[32]; } drive_info;
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || \
- defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
-EXPORT_SYMBOL(drive_info);
-#endif
struct screen_info screen_info;
EXPORT_SYMBOL(screen_info);
struct apm_info apm_info;
EXPORT_SYMBOL(apm_info);
-struct sys_desc_table_struct {
- unsigned short length;
- unsigned char table[0];
-};
struct edid_info edid_info;
EXPORT_SYMBOL_GPL(edid_info);
#ifndef CONFIG_XEN
@@ -149,7 +140,7 @@ unsigned long saved_videomode;
static char __initdata command_line[COMMAND_LINE_SIZE];
-unsigned char __initdata boot_params[PARAM_SIZE];
+struct boot_params __initdata boot_params;
/*
* Point at the empty zero page to start with. We map the real shared_info
@@ -316,18 +307,18 @@ unsigned long __init find_max_low_pfn(vo
printk(KERN_WARNING "Warning only %ldMB will be used.\n",
MAXMEM>>20);
if (max_pfn > MAX_NONPAE_PFN)
- printk(KERN_WARNING "Use a PAE enabled kernel.\n");
+ printk(KERN_WARNING "Use a HIGHMEM64G enabled kernel.\n");
else
printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
max_pfn = MAXMEM_PFN;
#else /* !CONFIG_HIGHMEM */
-#ifndef CONFIG_X86_PAE
+#ifndef CONFIG_HIGHMEM64G
if (max_pfn > MAX_NONPAE_PFN) {
max_pfn = MAX_NONPAE_PFN;
printk(KERN_WARNING "Warning only 4GB will be used.\n");
- printk(KERN_WARNING "Use a PAE enabled kernel.\n");
+ printk(KERN_WARNING "Use a HIGHMEM64G enabled kernel.\n");
}
-#endif /* !CONFIG_X86_PAE */
+#endif /* !CONFIG_HIGHMEM64G */
#endif /* !CONFIG_HIGHMEM */
} else {
if (highmem_pages == -1)
@@ -516,7 +507,7 @@ void __init setup_bootmem_allocator(void
*
* This should all compile down to nothing when NUMA is off.
*/
-void __init remapped_pgdat_init(void)
+static void __init remapped_pgdat_init(void)
{
int nid;
@@ -591,7 +582,6 @@ void __init setup_arch(char **cmdline_p)
properly. Setting ROOT_DEV to default to /dev/ram0 breaks initrd.
*/
ROOT_DEV = MKDEV(UNNAMED_MAJOR,0);
- drive_info = DRIVE_INFO;
screen_info = SCREEN_INFO;
copy_edid();
apm_info.bios = APM_BIOS_INFO;
@@ -767,6 +757,8 @@ void __init setup_arch(char **cmdline_p)
* NOTE: at this point the bootmem allocator is fully available.
*/
+ paravirt_post_allocator_init();
+
if (is_initial_xendomain())
dmi_scan_machine();
@@ -814,6 +806,7 @@ void __init setup_arch(char **cmdline_p)
#endif
e820_register_memory();
+ e820_mark_nosave_regions();
if (is_initial_xendomain()) {
#ifdef CONFIG_VT
--- head.orig/arch/x86/kernel/smp_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/smp_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -22,6 +22,7 @@
#include <asm/mtrr.h>
#include <asm/tlbflush.h>
+#include <asm/mmu_context.h>
#if 0
#include <mach_apic.h>
#endif
@@ -217,13 +218,13 @@ static unsigned long flush_va;
static DEFINE_SPINLOCK(tlbstate_lock);
/*
- * We cannot call mmdrop() because we are in interrupt context,
+ * We cannot call mmdrop() because we are in interrupt context,
* instead update mm->cpu_vm_mask.
*
* We need to reload %cr3 since the page tables may be going
* away from under us..
*/
-static inline void leave_mm (unsigned long cpu)
+void leave_mm(unsigned long cpu)
{
if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
BUG();
--- head.orig/arch/x86/kernel/time-xen.c 2012-02-10 13:26:34.000000000 +0100
+++ head/arch/x86/kernel/time-xen.c 2012-02-10 13:26:46.000000000 +0100
@@ -74,11 +74,12 @@
#include <xen/evtchn.h>
#include <xen/interface/vcpu.h>
-#ifdef CONFIG_X86_32
#include <asm/i8253.h>
DEFINE_SPINLOCK(i8253_lock);
EXPORT_SYMBOL(i8253_lock);
-#else
+
+#ifdef CONFIG_X86_64
+#include <asm/vsyscall.h>
volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES;
#endif
@@ -772,56 +773,10 @@ unsigned long read_persistent_clock(void
return retval;
}
-static void sync_cmos_clock(unsigned long dummy);
-
-static DEFINE_TIMER(sync_cmos_timer, sync_cmos_clock, 0, 0);
-int no_sync_cmos_clock;
-
-static void sync_cmos_clock(unsigned long dummy)
-{
- struct timeval now, next;
- int fail = 1;
-
- /*
- * If we have an externally synchronized Linux clock, then update
- * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
- * called as close as possible to 500 ms before the new second starts.
- * This code is run on a timer. If the clock is set, that timer
- * may not expire at the correct time. Thus, we adjust...
- */
- if (!ntp_synced())
- /*
- * Not synced, exit, do not restart a timer (if one is
- * running, let it run out).
- */
- return;
-
- do_gettimeofday(&now);
- if (now.tv_usec >= USEC_AFTER - ((unsigned) TICK_SIZE) / 2 &&
- now.tv_usec <= USEC_BEFORE + ((unsigned) TICK_SIZE) / 2)
- fail = set_rtc_mmss(now.tv_sec);
-
- next.tv_usec = USEC_AFTER - now.tv_usec;
- if (next.tv_usec <= 0)
- next.tv_usec += USEC_PER_SEC;
-
- if (!fail)
- next.tv_sec = 659;
- else
- next.tv_sec = 0;
-
- if (next.tv_usec >= USEC_PER_SEC) {
- next.tv_sec++;
- next.tv_usec -= USEC_PER_SEC;
- }
- mod_timer(&sync_cmos_timer, jiffies + timeval_to_jiffies(&next));
-}
-
-void notify_arch_cmos_timer(void)
+int update_persistent_clock(struct timespec now)
{
- if (!no_sync_cmos_clock)
- mod_timer(&sync_cmos_timer, jiffies + 1);
mod_timer(&sync_xen_wallclock_timer, jiffies + 1);
+ return set_rtc_mmss(now.tv_sec);
}
extern void (*late_time_init)(void);
--- head.orig/arch/x86/kernel/traps_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/traps_32-xen.c 2012-06-18 09:23:11.000000000 +0200
@@ -41,6 +41,10 @@
#include <linux/mca.h>
#endif
+#if defined(CONFIG_EDAC)
+#include <linux/edac.h>
+#endif
+
#include <asm/processor.h>
#include <asm/system.h>
#include <asm/io.h>
@@ -102,36 +106,45 @@ asmlinkage void machine_check(void);
int kstack_depth_to_print = 24;
static unsigned int code_bytes = 64;
-static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
+static inline int valid_stack_ptr(struct thread_info *tinfo, void *p, unsigned size)
{
return p > (void *)tinfo &&
- p < (void *)tinfo + THREAD_SIZE - 3;
+ p <= (void *)tinfo + THREAD_SIZE - size;
}
+/* The form of the top of the frame on the stack */
+struct stack_frame {
+ struct stack_frame *next_frame;
+ unsigned long return_address;
+};
+
static inline unsigned long print_context_stack(struct thread_info *tinfo,
unsigned long *stack, unsigned long ebp,
struct stacktrace_ops *ops, void *data)
{
- unsigned long addr;
-
#ifdef CONFIG_FRAME_POINTER
- while (valid_stack_ptr(tinfo, (void *)ebp)) {
- unsigned long new_ebp;
- addr = *(unsigned long *)(ebp + 4);
+ struct stack_frame *frame = (struct stack_frame *)ebp;
+ while (valid_stack_ptr(tinfo, frame, sizeof(*frame))) {
+ struct stack_frame *next;
+ unsigned long addr;
+
+ addr = frame->return_address;
ops->address(data, addr);
/*
* break out of recursive entries (such as
* end_of_stack_stop_unwind_function). Also,
* we can never allow a frame pointer to
* move downwards!
- */
- new_ebp = *(unsigned long *)ebp;
- if (new_ebp <= ebp)
+ */
+ next = frame->next_frame;
+ if (next <= frame)
break;
- ebp = new_ebp;
+ frame = next;
}
#else
- while (valid_stack_ptr(tinfo, stack)) {
+ while (valid_stack_ptr(tinfo, stack, sizeof(*stack))) {
+ unsigned long addr;
+
addr = *stack++;
if (__kernel_text_address(addr))
ops->address(data, addr);
@@ -154,7 +167,7 @@ void dump_trace(struct task_struct *task
if (!stack) {
unsigned long dummy;
stack = &dummy;
- if (task && task != current)
+ if (task != current)
stack = (unsigned long *)task->thread.esp;
}
@@ -213,6 +226,7 @@ static void print_trace_address(void *da
{
printk("%s [<%08lx>] ", (char *)data, addr);
print_symbol("%s\n", addr);
+ touch_nmi_watchdog();
}
static struct stacktrace_ops print_trace_ops = {
@@ -396,7 +410,7 @@ void die(const char * str, struct pt_reg
unsigned long esp;
unsigned short ss;
- report_bug(regs->eip);
+ report_bug(regs->eip, regs);
printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter);
#ifdef CONFIG_PREEMPT
@@ -439,6 +453,7 @@ void die(const char * str, struct pt_reg
bust_spinlocks(0);
die.lock_owner = -1;
+ add_taint(TAINT_DIE);
spin_unlock_irqrestore(&die.lock, flags);
if (!regs)
@@ -523,10 +538,12 @@ fastcall void do_##name(struct pt_regs *
do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \
}
-#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
+#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr, irq) \
fastcall void do_##name(struct pt_regs * regs, long error_code) \
{ \
siginfo_t info; \
+ if (irq) \
+ local_irq_enable(); \
info.si_signo = signr; \
info.si_errno = 0; \
info.si_code = sicode; \
@@ -566,13 +583,13 @@ DO_VM86_ERROR( 3, SIGTRAP, "int3", int3)
#endif
DO_VM86_ERROR( 4, SIGSEGV, "overflow", overflow)
DO_VM86_ERROR( 5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->eip)
+DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->eip, 0)
DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
-DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
-DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0)
+DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0, 0)
+DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0, 1)
fastcall void __kprobes do_general_protection(struct pt_regs * regs,
long error_code)
@@ -585,6 +602,13 @@ fastcall void __kprobes do_general_prote
current->thread.error_code = error_code;
current->thread.trap_no = 13;
+ if (show_unhandled_signals && unhandled_signal(current, SIGSEGV) &&
+ printk_ratelimit())
+ printk(KERN_INFO
+ "%s[%d] general protection eip:%lx esp:%lx error:%lx\n",
+ current->comm, current->pid,
+ regs->eip, regs->esp, error_code);
+
force_sig(SIGSEGV, current);
return;
@@ -610,6 +634,14 @@ mem_parity_error(unsigned char reason, s
printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on "
"CPU %d.\n", reason, smp_processor_id());
printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n");
+
+#if defined(CONFIG_EDAC)
+ if(edac_handler_set()) {
+ edac_atomic_assert_error();
+ return;
+ }
+#endif
+
if (panic_on_unrecovered_nmi)
panic("NMI: Not continuing");
@@ -720,6 +752,8 @@ static __kprobes void default_do_nmi(str
reassert_nmi();
}
+static int ignore_nmis;
+
fastcall __kprobes void do_nmi(struct pt_regs * regs, long error_code)
{
int cpu;
@@ -730,11 +764,24 @@ fastcall __kprobes void do_nmi(struct pt
++nmi_count(cpu);
- default_do_nmi(regs);
+ if (!ignore_nmis)
+ default_do_nmi(regs);
nmi_exit();
}
+void stop_nmi(void)
+{
+ acpi_nmi_disable();
+ ignore_nmis++;
+}
+
+void restart_nmi(void)
+{
+ ignore_nmis--;
+ acpi_nmi_enable();
+}
+
#ifdef CONFIG_KPROBES
fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code)
{
--- head.orig/arch/x86/mach-xen/setup.c 2011-08-09 10:35:06.000000000 +0200
+++ head/arch/x86/mach-xen/setup.c 2011-09-07 15:28:26.000000000 +0200
@@ -12,6 +12,7 @@
#include <asm/e820.h>
#include <asm/setup.h>
#include <asm/fixmap.h>
+#include <asm/pgtable.h>
#include <xen/interface/callback.h>
#include <xen/interface/memory.h>
@@ -100,7 +101,7 @@ void __init pre_setup_arch_hook(void)
init_mm.pgd = swapper_pg_dir = (pgd_t *)xen_start_info->pt_base;
- setup_xen_features();
+ xen_setup_features();
if (HYPERVISOR_xen_version(XENVER_platform_parameters, &pp) == 0) {
hypervisor_virt_start = pp.virt_start;
@@ -157,4 +158,18 @@ void __init machine_specific_arch_setup(
HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
}
#endif
+
+ /* Do an early initialization of the fixmap area */
+ {
+ extern pte_t swapper_pg_pmd[PTRS_PER_PTE];
+ unsigned long addr = __fix_to_virt(FIX_EARLYCON_MEM_BASE);
+ pgd_t *pgd = (pgd_t *)xen_start_info->pt_base;
+ pud_t *pud = pud_offset(pgd + pgd_index(addr), addr);
+ pmd_t *pmd = pmd_offset(pud, addr);
+
+ swapper_pg_dir = pgd;
+ init_mm.pgd = pgd;
+ make_lowmem_page_readonly(swapper_pg_pmd, XENFEAT_writable_page_tables);
+ set_pmd(pmd, __pmd(__pa_symbol(swapper_pg_pmd) | _PAGE_TABLE));
+ }
}
--- head.orig/arch/x86/mm/fault_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/mm/fault_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -346,7 +346,10 @@ static inline pmd_t *vmalloc_sync_one(pg
pmd_k = pmd_offset(pud_k, address);
if (!pmd_present(*pmd_k))
return NULL;
- if (!pmd_present(*pmd))
+ if (!pmd_present(*pmd)) {
+ bool lazy = x86_read_percpu(xen_lazy_mmu);
+
+ x86_write_percpu(xen_lazy_mmu, false);
#if CONFIG_XEN_COMPAT > 0x030002
set_pmd(pmd, *pmd_k);
#else
@@ -356,7 +359,8 @@ static inline pmd_t *vmalloc_sync_one(pg
*/
set_pmd(pmd, __pmd(pmd_val(*pmd_k)));
#endif
- else
+ x86_write_percpu(xen_lazy_mmu, lazy);
+ } else
BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
return pmd_k;
}
@@ -388,6 +392,8 @@ static inline int vmalloc_fault(unsigned
return 0;
}
+int show_unhandled_signals = 1;
+
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
@@ -408,6 +414,7 @@ fastcall void __kprobes do_page_fault(st
struct vm_area_struct * vma;
unsigned long address;
int write, si_code;
+ int fault;
/* get the address */
address = read_cr2();
@@ -541,20 +548,18 @@ good_area:
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
- switch (handle_mm_fault(mm, vma, address, write)) {
- case VM_FAULT_MINOR:
- tsk->min_flt++;
- break;
- case VM_FAULT_MAJOR:
- tsk->maj_flt++;
- break;
- case VM_FAULT_SIGBUS:
- goto do_sigbus;
- case VM_FAULT_OOM:
+ fault = handle_mm_fault(mm, vma, address, write);
+ if (unlikely(fault & VM_FAULT_ERROR)) {
+ if (fault & VM_FAULT_OOM)
goto out_of_memory;
- default:
- BUG();
+ else if (fault & VM_FAULT_SIGBUS)
+ goto do_sigbus;
+ BUG();
}
+ if (fault & VM_FAULT_MAJOR)
+ tsk->maj_flt++;
+ else
+ tsk->min_flt++;
/*
* Did it hit the DOS screen memory VA from vm86 mode?
@@ -589,6 +594,14 @@ bad_area_nosemaphore:
if (is_prefetch(regs, address, error_code))
return;
+ if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
+ printk_ratelimit()) {
+ printk("%s%s[%d]: segfault at %08lx eip %08lx "
+ "esp %08lx error %lx\n",
+ tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
+ tsk->comm, tsk->pid, address, regs->eip,
+ regs->esp, error_code);
+ }
tsk->thread.cr2 = address;
/* Kernel addresses are always protection faults */
tsk->thread.error_code = error_code | (address >= TASK_SIZE);
--- head.orig/arch/x86/mm/highmem_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/mm/highmem_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -34,17 +34,16 @@ void *kmap_atomic_prot(struct page *page
/* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
pagefault_disable();
- idx = type + KM_TYPE_NR*smp_processor_id();
- BUG_ON(!pte_none(*(kmap_pte-idx)));
-
if (!PageHighMem(page))
return page_address(page);
+ idx = type + KM_TYPE_NR*smp_processor_id();
vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+ BUG_ON(!pte_none(*(kmap_pte-idx)));
set_pte_at(&init_mm, vaddr, kmap_pte-idx, mk_pte(page, prot));
/*arch_flush_lazy_mmu_mode();*/
- return (void*) vaddr;
+ return (void *)vaddr;
}
void *kmap_atomic(struct page *page, enum km_type type)
--- head.orig/arch/x86/mm/init_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/mm/init_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -96,7 +96,7 @@ static pte_t * __init one_page_table_ini
#endif
pte_t *page_table = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
- paravirt_alloc_pt(__pa(page_table) >> PAGE_SHIFT);
+ paravirt_alloc_pt(&init_mm, __pa(page_table) >> PAGE_SHIFT);
make_lowmem_page_readonly(page_table,
XENFEAT_writable_page_tables);
set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
@@ -453,7 +453,7 @@ static void __init pagetable_init (void)
xen_pagetable_setup_done(pgd_base);
}
-#if defined(CONFIG_SOFTWARE_SUSPEND) || defined(CONFIG_ACPI_SLEEP)
+#if defined(CONFIG_HIBERNATION) || defined(CONFIG_ACPI)
/*
* Swap suspend & friends need this for resume because things like the intel-agp
* driver might have split up a kernel 4MB mapping.
@@ -492,9 +492,13 @@ void zap_low_mappings (void)
flush_tlb_all();
}
+int nx_enabled = 0;
+
+#ifdef CONFIG_X86_PAE
+
static int disable_nx __initdata = 0;
u64 __supported_pte_mask __read_mostly = ~_PAGE_NX;
-EXPORT_SYMBOL(__supported_pte_mask);
+EXPORT_SYMBOL_GPL(__supported_pte_mask);
/*
* noexec = on|off
@@ -521,9 +525,6 @@ static int __init noexec_setup(char *str
}
early_param("noexec", noexec_setup);
-int nx_enabled = 0;
-#ifdef CONFIG_X86_PAE
-
static void __init set_nx(void)
{
unsigned int v[4], l, h;
@@ -770,7 +771,7 @@ void __init mem_init(void)
zap_low_mappings();
#endif
- set_bit(PG_pinned, &virt_to_page(init_mm.pgd)->flags);
+ SetPagePinned(virt_to_page(init_mm.pgd));
}
#ifdef CONFIG_MEMORY_HOTPLUG
@@ -802,8 +803,7 @@ void __init pgtable_cache_init(void)
PTRS_PER_PMD*sizeof(pmd_t),
PTRS_PER_PMD*sizeof(pmd_t),
SLAB_PANIC,
- pmd_ctor,
- NULL);
+ pmd_ctor);
if (!SHARED_KERNEL_PMD) {
/* If we're in PAE mode and have a non-shared
kernel pmd, then the pgd size must be a
--- head.orig/arch/x86/mm/ioremap-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/mm/ioremap-xen.c 2011-02-07 15:38:30.000000000 +0100
@@ -344,9 +344,8 @@ void iounmap(volatile void __iomem *addr
/* Reset the direct mapping. Can block */
if ((p->flags >> 20) && is_local_lowmem(p->phys_addr)) {
- /* p->size includes the guard page, but cpa doesn't like that */
change_page_attr(virt_to_page(bus_to_virt(p->phys_addr)),
- (p->size - PAGE_SIZE) >> PAGE_SHIFT,
+ get_vm_area_size(p) >> PAGE_SHIFT,
PAGE_KERNEL);
global_flush_tlb();
}
--- head.orig/arch/x86/mm/pgtable_32-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/mm/pgtable_32-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -198,7 +198,7 @@ void pte_free(struct page *pte)
va, pfn_pte(pfn, PAGE_KERNEL), 0))
BUG();
} else
- clear_bit(PG_pinned, &pte->flags);
+ ClearPagePinned(pte);
ClearPageForeign(pte);
init_page_count(pte);
@@ -249,7 +249,7 @@ static inline void pgd_list_del(pgd_t *p
#if (PTRS_PER_PMD == 1)
/* Non-PAE pgd constructor */
-void pgd_ctor(void *pgd)
+static void pgd_ctor(void *pgd)
{
unsigned long flags;
@@ -272,7 +272,7 @@ void pgd_ctor(void *pgd)
}
#else /* PTRS_PER_PMD > 1 */
/* PAE pgd constructor */
-void pgd_ctor(void *pgd)
+static void pgd_ctor(void *pgd)
{
/* PAE, kernel PMD may be shared */
@@ -286,7 +286,7 @@ void pgd_ctor(void *pgd)
}
#endif /* PTRS_PER_PMD */
-void pgd_dtor(void *pgd)
+static void pgd_dtor(void *pgd)
{
unsigned long flags; /* can be called from interrupt context */
@@ -644,9 +644,9 @@ static inline unsigned int pgd_walk_set_
if (PageHighMem(page)) {
if (pgprot_val(flags) & _PAGE_RW)
- clear_bit(PG_pinned, &page->flags);
+ ClearPagePinned(page);
else
- set_bit(PG_pinned, &page->flags);
+ SetPagePinned(page);
} else {
MULTI_update_va_mapping(per_cpu(pb_mcl, cpu) + seq,
(unsigned long)__va(pfn << PAGE_SHIFT),
@@ -716,19 +716,19 @@ static void __pgd_pin(pgd_t *pgd)
pgd_walk(pgd, PAGE_KERNEL_RO);
kmap_flush_unused();
xen_pgd_pin(__pa(pgd));
- set_bit(PG_pinned, &virt_to_page(pgd)->flags);
+ SetPagePinned(virt_to_page(pgd));
}
static void __pgd_unpin(pgd_t *pgd)
{
xen_pgd_unpin(__pa(pgd));
pgd_walk(pgd, PAGE_KERNEL);
- clear_bit(PG_pinned, &virt_to_page(pgd)->flags);
+ ClearPagePinned(virt_to_page(pgd));
}
static void pgd_test_and_unpin(pgd_t *pgd)
{
- if (test_bit(PG_pinned, &virt_to_page(pgd)->flags))
+ if (PagePinned(virt_to_page(pgd)))
__pgd_unpin(pgd);
}
@@ -766,7 +766,7 @@ void mm_pin_all(void)
*/
spin_lock_irqsave(&pgd_lock, flags);
for (page = pgd_list; page; page = (struct page *)page->index) {
- if (!test_bit(PG_pinned, &page->flags))
+ if (!PagePinned(page))
__pgd_pin((pgd_t *)page_address(page));
}
spin_unlock_irqrestore(&pgd_lock, flags);
@@ -774,7 +774,7 @@ void mm_pin_all(void)
void arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
{
- if (!test_bit(PG_pinned, &virt_to_page(mm->pgd)->flags))
+ if (!PagePinned(virt_to_page(mm->pgd)))
mm_pin(mm);
}
@@ -800,7 +800,7 @@ void arch_exit_mmap(struct mm_struct *mm
task_unlock(tsk);
- if (test_bit(PG_pinned, &virt_to_page(mm->pgd)->flags) &&
+ if (PagePinned(virt_to_page(mm->pgd)) &&
(atomic_read(&mm->mm_count) == 1) &&
!mm->context.has_foreign_mappings)
mm_unpin(mm);
--- head.orig/arch/x86/xen/Kconfig 2012-06-12 15:06:23.000000000 +0200
+++ head/arch/x86/xen/Kconfig 2011-01-31 17:49:31.000000000 +0100
@@ -2,7 +2,7 @@
# This Kconfig describes xen options
#
-config XEN
+config PARAVIRT_XEN
bool "Xen guest support"
select PARAVIRT
select PARAVIRT_CLOCK
--- head.orig/arch/x86/xen/xen-head.S 2012-06-12 15:06:23.000000000 +0200
+++ head/arch/x86/xen/xen-head.S 2011-01-31 17:49:31.000000000 +0100
@@ -1,7 +1,7 @@
/* Xen-specific pieces of head.S, intended to be included in the right
place in head.S */
-#ifdef CONFIG_XEN
+#ifdef CONFIG_PARAVIRT_XEN
#include <linux/elfnote.h>
#include <linux/init.h>
@@ -52,4 +52,4 @@ ENTRY(hypercall_page)
ELFNOTE(Xen, XEN_ELFNOTE_HV_START_LOW, _ASM_PTR __HYPERVISOR_VIRT_START)
ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, _ASM_PTR 0)
-#endif /*CONFIG_XEN */
+#endif /* CONFIG_PARAVIRT_XEN */
--- head.orig/arch/x86/ia32/ia32entry-xen.S 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/ia32/ia32entry-xen.S 2011-01-31 17:49:31.000000000 +0100
@@ -105,7 +105,7 @@ ENTRY(ia32_sysenter_target)
movl $VSYSCALL32_SYSEXIT,8(%rsp)
movq %rax,(%rsp)
cld
- SAVE_ARGS 0,0,0
+ SAVE_ARGS 0,0,1
/* no need to do an access_ok check here because rbp has been
32bit zero extended */
1: movl (%rbp),%r9d
@@ -244,7 +244,7 @@ ia32_badarg:
*/
ENTRY(ia32_syscall)
- CFI_STARTPROC simple
+ CFI_STARTPROC32 simple
CFI_SIGNAL_FRAME
CFI_DEF_CFA rsp,SS+8-RIP+16
/*CFI_REL_OFFSET ss,SS-RIP+16*/
@@ -280,6 +280,7 @@ ia32_sysret:
ia32_tracesys:
SAVE_REST
+ CLEAR_RREGS
movq $-ENOSYS,RAX(%rsp) /* really needed? */
movq %rsp,%rdi /* &pt_regs -> arg1 */
call syscall_trace_enter
@@ -476,7 +477,7 @@ ia32_sys_call_table:
.quad sys_init_module
.quad sys_delete_module
.quad quiet_ni_syscall /* 130 get_kernel_syms */
- .quad sys_quotactl
+ .quad sys32_quotactl
.quad sys_getpgid
.quad sys_fchdir
.quad quiet_ni_syscall /* bdflush */
@@ -669,4 +670,5 @@ ia32_sys_call_table:
.quad compat_sys_signalfd
.quad compat_sys_timerfd
.quad sys_eventfd
+ .quad sys32_fallocate
ia32_syscall_end:
--- head.orig/arch/x86/kernel/e820_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/e820_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -224,37 +224,6 @@ unsigned long __init e820_end_of_ram(voi
}
/*
- * Find the hole size in the range.
- */
-unsigned long __init e820_hole_size(unsigned long start, unsigned long end)
-{
- unsigned long ram = 0;
- int i;
-
- for (i = 0; i < e820.nr_map; i++) {
- struct e820entry *ei = &e820.map[i];
- unsigned long last, addr;
-
- if (ei->type != E820_RAM ||
- ei->addr+ei->size <= start ||
- ei->addr >= end)
- continue;
-
- addr = round_up(ei->addr, PAGE_SIZE);
- if (addr < start)
- addr = start;
-
- last = round_down(ei->addr + ei->size, PAGE_SIZE);
- if (last >= end)
- last = end;
-
- if (last > addr)
- ram += last - addr;
- }
- return ((end - start) - ram);
-}
-
-/*
* Mark e820 reserved areas as busy for the resource manager.
*/
void __init e820_reserve_resources(struct e820entry *e820, int nr_map)
@@ -327,54 +296,66 @@ void __init e820_mark_nosave_regions(voi
}
#endif
+/*
+ * Finds an active region in the address range from start_pfn to end_pfn and
+ * returns its range in ei_startpfn and ei_endpfn for the e820 entry.
+ */
+static int __init e820_find_active_region(const struct e820entry *ei,
+ unsigned long start_pfn,
+ unsigned long end_pfn,
+ unsigned long *ei_startpfn,
+ unsigned long *ei_endpfn)
+{
+#ifdef CONFIG_XEN
+ if (end_pfn > xen_start_info->nr_pages)
+ end_pfn = xen_start_info->nr_pages;
+#endif
+
+ *ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT;
+ *ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE) >> PAGE_SHIFT;
+
+ /* Skip map entries smaller than a page */
+ if (*ei_startpfn >= *ei_endpfn)
+ return 0;
+
+ /* Check if end_pfn_map should be updated */
+ if (ei->type != E820_RAM && *ei_endpfn > end_pfn_map)
+ end_pfn_map = *ei_endpfn;
+
+ /* Skip if map is outside the node */
+ if (ei->type != E820_RAM || *ei_endpfn <= start_pfn ||
+ *ei_startpfn >= end_pfn)
+ return 0;
+
+ /* Check for overlaps */
+ if (*ei_startpfn < start_pfn)
+ *ei_startpfn = start_pfn;
+ if (*ei_endpfn > end_pfn)
+ *ei_endpfn = end_pfn;
+
+ /* Obey end_user_pfn to save on memmap */
+ if (*ei_startpfn >= end_user_pfn)
+ return 0;
+ if (*ei_endpfn > end_user_pfn)
+ *ei_endpfn = end_user_pfn;
+
+ return 1;
+}
+
/* Walk the e820 map and register active regions within a node */
void __init
e820_register_active_regions(int nid, unsigned long start_pfn,
unsigned long end_pfn)
{
+ unsigned long ei_startpfn;
+ unsigned long ei_endpfn;
int i;
- unsigned long ei_startpfn, ei_endpfn;
- for (i = 0; i < e820.nr_map; i++) {
- struct e820entry *ei = &e820.map[i];
- ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT;
- ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE)
- >> PAGE_SHIFT;
-
- /* Skip map entries smaller than a page */
- if (ei_startpfn >= ei_endpfn)
- continue;
-
- /* Check if end_pfn_map should be updated */
- if (ei->type != E820_RAM && ei_endpfn > end_pfn_map)
- end_pfn_map = ei_endpfn;
-
- /* Skip if map is outside the node */
- if (ei->type != E820_RAM ||
- ei_endpfn <= start_pfn ||
- ei_startpfn >= end_pfn)
- continue;
-
- /* Check for overlaps */
- if (ei_startpfn < start_pfn)
- ei_startpfn = start_pfn;
- if (ei_endpfn > end_pfn)
- ei_endpfn = end_pfn;
-
- /* Obey end_user_pfn to save on memmap */
- if (ei_startpfn >= end_user_pfn)
- continue;
- if (ei_endpfn > end_user_pfn)
- ei_endpfn = end_user_pfn;
-
-#ifdef CONFIG_XEN
- if (ei_startpfn >= xen_start_info->nr_pages)
- continue;
- if (ei_endpfn > xen_start_info->nr_pages)
- ei_endpfn = xen_start_info->nr_pages;
-#endif
- add_active_range(nid, ei_startpfn, ei_endpfn);
- }
+ for (i = 0; i < e820.nr_map; i++)
+ if (e820_find_active_region(&e820.map[i],
+ start_pfn, end_pfn,
+ &ei_startpfn, &ei_endpfn))
+ add_active_range(nid, ei_startpfn, ei_endpfn);
#ifdef CONFIG_XEN
BUG_ON(nid);
add_active_range(nid, end_pfn, end_pfn);
@@ -399,12 +380,35 @@ void __init add_memory_region(unsigned l
e820.nr_map++;
}
+/*
+ * Find the hole size (in bytes) in the memory range.
+ * @start: starting address of the memory range to scan
+ * @end: ending address of the memory range to scan
+ */
+unsigned long __init e820_hole_size(unsigned long start, unsigned long end)
+{
+ unsigned long start_pfn = start >> PAGE_SHIFT;
+ unsigned long end_pfn = end >> PAGE_SHIFT;
+ unsigned long ei_startpfn;
+ unsigned long ei_endpfn;
+ unsigned long ram = 0;
+ int i;
+
+ for (i = 0; i < e820.nr_map; i++) {
+ if (e820_find_active_region(&e820.map[i],
+ start_pfn, end_pfn,
+ &ei_startpfn, &ei_endpfn))
+ ram += ei_endpfn - ei_startpfn;
+ }
+ return end - start - (ram << PAGE_SHIFT);
+}
+
void __init e820_print_map(char *who)
{
int i;
for (i = 0; i < e820.nr_map; i++) {
- printk(" %s: %016Lx - %016Lx ", who,
+ printk(KERN_INFO " %s: %016Lx - %016Lx ", who,
(unsigned long long) e820.map[i].addr,
(unsigned long long) (e820.map[i].addr + e820.map[i].size));
switch (e820.map[i].type) {
--- head.orig/arch/x86/kernel/early_printk-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/early_printk-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -174,6 +174,7 @@ static __init void early_serial_init(cha
* mappings. Someone should fix this for domain 0. For now, use fake serial.
*/
#define early_vga_console early_serial_console
+#define xenboot_console early_serial_console
#endif
@@ -261,20 +262,22 @@ static int __init setup_early_printk(cha
} else if (!strncmp(buf, "ttyS", 4)) {
early_serial_init(buf);
early_console = &early_serial_console;
- } else if (!strncmp(buf, "vga", 3)
+ } else if (!strncmp(buf, "vga", 3)) {
#ifndef CONFIG_XEN
&& SCREEN_INFO.orig_video_isVGA == 1) {
max_xpos = SCREEN_INFO.orig_video_cols;
max_ypos = SCREEN_INFO.orig_video_lines;
current_ypos = SCREEN_INFO.orig_y;
-#else
- || !strncmp(buf, "xen", 3)) {
#endif
early_console = &early_vga_console;
} else if (!strncmp(buf, "simnow", 6)) {
simnow_init(buf + 6);
early_console = &simnow_console;
keep_early = 1;
+#ifdef CONFIG_XEN
+ } else if (!strncmp(buf, "xen", 3)) {
+ early_console = &xenboot_console;
+#endif
}
if (keep_early)
--- head.orig/arch/x86/kernel/entry_64-xen.S 2011-10-06 11:55:00.000000000 +0200
+++ head/arch/x86/kernel/entry_64-xen.S 2011-10-07 11:21:43.000000000 +0200
@@ -313,7 +313,7 @@ sysret_signal:
TRACE_IRQS_ON
/* sti */
XEN_UNBLOCK_EVENTS(%rsi)
- testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
+ testl $(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY),%edx
jz 1f
/* Really a signal */
@@ -412,7 +412,7 @@ int_very_careful:
jmp int_restore_rest
int_signal:
- testl $(_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_SINGLESTEP),%edx
+ testl $(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY),%edx
jz 1f
movq %rsp,%rdi # &ptregs -> arg1
xorl %esi,%esi # oldset -> arg2
@@ -555,7 +555,7 @@ retint_careful:
jmp retint_check
retint_signal:
- testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
+ testl $(_TIF_SIGPENDING|_TIF_SINGLESTEP|_TIF_MCE_NOTIFY),%edx
jz retint_restore_args
TRACE_IRQS_ON
XEN_UNBLOCK_EVENTS(%rsi)
--- head.orig/arch/x86/kernel/head_64-xen.S 2011-08-09 10:37:34.000000000 +0200
+++ head/arch/x86/kernel/head_64-xen.S 2011-08-09 10:41:44.000000000 +0200
@@ -24,7 +24,7 @@
#include <xen/interface/elfnote.h>
#include <xen/interface/features.h>
- .section .bootstrap.text, "ax", @progbits
+ .section .text.head, "ax", @progbits
.code64
.globl startup_64
startup_64:
@@ -40,7 +40,7 @@ startup_64:
#define NEXT_PAGE(name) \
.balign PAGE_SIZE; \
- phys_##name = . - .bootstrap.text; \
+ phys_##name = . - .text.head; \
ENTRY(name)
NEXT_PAGE(init_level4_pgt)
@@ -64,6 +64,12 @@ NEXT_PAGE(level3_kernel_pgt)
NEXT_PAGE(level3_user_pgt)
.fill 512,8,0
+NEXT_PAGE(level2_fixmap_pgt)
+ .fill 512,8,0
+
+NEXT_PAGE(level1_fixmap_pgt)
+ .fill 512,8,0
+
NEXT_PAGE(hypercall_page)
CFI_STARTPROC
.rept 0x1000 / 0x20
@@ -182,27 +188,27 @@ ENTRY(empty_zero_page)
.byte 0
#endif /* CONFIG_XEN_COMPAT <= 0x030002 */
- ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz, "linux")
- ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz, "2.6")
- ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz, "xen-3.0")
- ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, .quad, __START_KERNEL_map)
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz "linux")
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz "2.6")
+ ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz "xen-3.0")
+ ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, .quad __START_KERNEL_map)
#if CONFIG_XEN_COMPAT <= 0x030002
- ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .quad, __START_KERNEL_map)
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .quad __START_KERNEL_map)
#else
- ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .quad, 0)
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .quad 0)
#endif
- ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .quad, startup_64)
- ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .quad, hypercall_page)
- ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .quad, _PAGE_PRESENT,_PAGE_PRESENT)
- ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .ascii, "writable_page_tables";
- .ascii "|writable_descriptor_tables";
- .ascii "|auto_translated_physmap";
- .ascii "|supervisor_mode_kernel";
- .asciz XEN_DOM0_CAP_STR)
- ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES, .long, XEN_DOM0_CAP |
+ ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .quad startup_64)
+ ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .quad hypercall_page)
+ ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .quad _PAGE_PRESENT, _PAGE_PRESENT)
+ ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .ascii "writable_page_tables";
+ .ascii "|writable_descriptor_tables";
+ .ascii "|auto_translated_physmap";
+ .ascii "|supervisor_mode_kernel";
+ .asciz XEN_DOM0_CAP_STR)
+ ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES, .long XEN_DOM0_CAP |
(1 << XENFEAT_writable_page_tables) |
(1 << XENFEAT_writable_descriptor_tables) |
(1 << XENFEAT_auto_translated_physmap) |
(1 << XENFEAT_supervisor_mode_kernel))
- ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz, "generic")
- ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long, 1)
+ ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz "generic")
+ ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long 1)
--- head.orig/arch/x86/kernel/head64-xen.c 2011-08-09 10:37:35.000000000 +0200
+++ head/arch/x86/kernel/head64-xen.c 2011-08-09 10:41:42.000000000 +0200
@@ -89,7 +89,7 @@ void __init x86_64_start_kernel(char * r
struct xen_machphys_mapping mapping;
int i;
- setup_xen_features();
+ xen_setup_features();
xen_start_info = (struct start_info *)real_mode_data;
if (!xen_feature(XENFEAT_auto_translated_physmap))
--- head.orig/arch/x86/kernel/io_apic_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/io_apic_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -165,7 +165,9 @@ static inline void io_apic_write(unsigne
#endif
}
-#ifndef CONFIG_XEN
+#ifdef CONFIG_XEN
+#define io_apic_modify io_apic_write
+#else
/*
* Re-write a value: to be used for read-modify-write
* cycles where the read already set up the index register.
@@ -175,8 +177,32 @@ static inline void io_apic_modify(unsign
struct io_apic __iomem *io_apic = io_apic_base(apic);
writel(value, &io_apic->data);
}
-#else
-#define io_apic_modify io_apic_write
+
+static int io_apic_level_ack_pending(unsigned int irq)
+{
+ struct irq_pin_list *entry;
+ unsigned long flags;
+ int pending = 0;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ entry = irq_2_pin + irq;
+ for (;;) {
+ unsigned int reg;
+ int pin;
+
+ pin = entry->pin;
+ if (pin == -1)
+ break;
+ reg = io_apic_read(entry->apic, 0x10 + pin*2);
+ /* Is the remote IRR bit set? */
+ pending |= (reg >> 14) & 1;
+ if (!entry->next)
+ break;
+ entry = irq_2_pin + entry->next;
+ }
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ return pending;
+}
#endif
/*
@@ -406,14 +432,12 @@ static void clear_IO_APIC (void)
int skip_ioapic_setup;
int ioapic_force;
-/* dummy parsing: see setup.c */
-
-static int __init disable_ioapic_setup(char *str)
+static int __init parse_noapic(char *str)
{
- skip_ioapic_setup = 1;
+ disable_ioapic_setup();
return 0;
}
-early_param("noapic", disable_ioapic_setup);
+early_param("noapic", parse_noapic);
/* Actually the next is obsolete, but keep it for paranoid reasons -AK */
static int __init disable_timer_pin_setup(char *arg)
@@ -765,12 +789,15 @@ static struct irq_chip ioapic_chip;
static void ioapic_register_intr(int irq, unsigned long trigger)
{
- if (trigger)
+ if (trigger) {
+ irq_desc[irq].status |= IRQ_LEVEL;
set_irq_chip_and_handler_name(irq, &ioapic_chip,
handle_fasteoi_irq, "fasteoi");
- else
+ } else {
+ irq_desc[irq].status &= ~IRQ_LEVEL;
set_irq_chip_and_handler_name(irq, &ioapic_chip,
handle_edge_irq, "edge");
+ }
}
#else
#define ioapic_register_intr(irq, trigger) evtchn_register_pirq(irq)
@@ -1419,9 +1446,37 @@ static void ack_apic_level(unsigned int
ack_APIC_irq();
/* Now we can move and renable the irq */
- move_masked_irq(irq);
- if (unlikely(do_unmask_irq))
+ if (unlikely(do_unmask_irq)) {
+ /* Only migrate the irq if the ack has been received.
+ *
+ * On rare occasions the broadcast level triggered ack gets
+ * delayed going to ioapics, and if we reprogram the
+ * vector while Remote IRR is still set the irq will never
+ * fire again.
+ *
+ * To prevent this scenario we read the Remote IRR bit
+ * of the ioapic. This has two effects.
+ * - On any sane system the read of the ioapic will
+ * flush writes (and acks) going to the ioapic from
+ * this cpu.
+ * - We get to see if the ACK has actually been delivered.
+ *
+ * Based on failed experiments of reprogramming the
+ * ioapic entry from outside of irq context starting
+ * with masking the ioapic entry and then polling until
+ * Remote IRR was clear before reprogramming the
+ * ioapic I don't trust the Remote IRR bit to be
+ * completey accurate.
+ *
+ * However there appears to be no other way to plug
+ * this race, so if the Remote IRR bit is not
+ * accurate and is causing problems then it is a hardware bug
+ * and you can go talk to the chipset vendor about it.
+ */
+ if (!io_apic_level_ack_pending(irq))
+ move_masked_irq(irq);
unmask_IO_APIC_irq(irq);
+ }
}
static struct irq_chip ioapic_chip __read_mostly = {
--- head.orig/arch/x86/kernel/ldt_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/ldt_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -114,6 +114,8 @@ int init_new_context(struct task_struct
memset(&mm->context, 0, sizeof(mm->context));
init_MUTEX(&mm->context.sem);
old_mm = current->mm;
+ if (old_mm)
+ mm->context.vdso = old_mm->context.vdso;
if (old_mm && old_mm->context.size > 0) {
down(&old_mm->context.sem);
retval = copy_ldt(&mm->context, &old_mm->context);
@@ -146,7 +148,7 @@ void destroy_context(struct mm_struct *m
kfree(mm->context.ldt);
mm->context.size = 0;
}
- if (!mm->context.pinned) {
+ if (!PagePinned(virt_to_page(mm->pgd))) {
spin_lock(&mm_unpinned_lock);
list_del(&mm->context.unpinned);
spin_unlock(&mm_unpinned_lock);
--- head.orig/arch/x86/kernel/mpparse_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/mpparse_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -32,7 +32,6 @@
/* Have we found an MP table */
int smp_found_config;
-unsigned int __initdata maxcpus = NR_CPUS;
/*
* Various Linux-internal data structures created from the
@@ -657,6 +656,20 @@ static int mp_find_ioapic(int gsi)
return -1;
}
+static u8 uniq_ioapic_id(u8 id)
+{
+ int i;
+ DECLARE_BITMAP(used, 256);
+ bitmap_zero(used, 256);
+ for (i = 0; i < nr_ioapics; i++) {
+ struct mpc_config_ioapic *ia = &mp_ioapics[i];
+ __set_bit(ia->mpc_apicid, used);
+ }
+ if (!test_bit(id, used))
+ return id;
+ return find_first_zero_bit(used, 256);
+}
+
void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
{
int idx = 0;
@@ -664,7 +677,7 @@ void __init mp_register_ioapic(u8 id, u3
if (bad_ioapic(address))
return;
- idx = nr_ioapics++;
+ idx = nr_ioapics;
mp_ioapics[idx].mpc_type = MP_IOAPIC;
mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
@@ -673,7 +686,7 @@ void __init mp_register_ioapic(u8 id, u3
#ifndef CONFIG_XEN
set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
#endif
- mp_ioapics[idx].mpc_apicid = id;
+ mp_ioapics[idx].mpc_apicid = uniq_ioapic_id(id);
mp_ioapics[idx].mpc_apicver = 0;
/*
@@ -690,6 +703,8 @@ void __init mp_register_ioapic(u8 id, u3
mp_ioapics[idx].mpc_apicaddr,
mp_ioapic_routing[idx].gsi_start,
mp_ioapic_routing[idx].gsi_end);
+
+ nr_ioapics++;
}
void __init
--- head.orig/arch/x86/kernel/process_64-xen.c 2011-02-02 08:31:39.000000000 +0100
+++ head/arch/x86/kernel/process_64-xen.c 2011-02-02 08:31:50.000000000 +0100
@@ -26,6 +26,7 @@
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/mm.h>
+#include <linux/fs.h>
#include <linux/elfcore.h>
#include <linux/smp.h>
#include <linux/slab.h>
@@ -248,6 +249,7 @@ early_param("idle", idle_setup);
void __show_regs(struct pt_regs * regs)
{
unsigned long fs, gs, shadowgs;
+ unsigned long d0, d1, d2, d3, d6, d7;
unsigned int fsindex,gsindex;
unsigned int ds,cs,es;
@@ -287,6 +289,14 @@ void __show_regs(struct pt_regs * regs)
fs,fsindex,gs,gsindex,shadowgs);
printk("CS: %04x DS: %04x ES: %04x\n", cs, ds, es);
+ get_debugreg(d0, 0);
+ get_debugreg(d1, 1);
+ get_debugreg(d2, 2);
+ printk("DR0: %016lx DR1: %016lx DR2: %016lx\n", d0, d1, d2);
+ get_debugreg(d3, 3);
+ get_debugreg(d6, 6);
+ get_debugreg(d7, 7);
+ printk("DR3: %016lx DR6: %016lx DR7: %016lx\n", d3, d6, d7);
}
void show_regs(struct pt_regs *regs)
--- head.orig/arch/x86/kernel/setup_64-xen.c 2012-06-06 13:49:31.000000000 +0200
+++ head/arch/x86/kernel/setup_64-xen.c 2012-06-06 13:52:43.000000000 +0200
@@ -826,6 +826,8 @@ static void __cpuinit init_amd(struct cp
level = cpuid_eax(1);
if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58))
set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
+ if (c->x86 == 0x10)
+ set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
/* Enable workaround for FXSAVE leak */
if (c->x86 >= 6)
@@ -851,8 +853,14 @@ static void __cpuinit init_amd(struct cp
if (c->extended_cpuid_level >= 0x80000008)
amd_detect_cmp(c);
- /* Fix cpuid4 emulation for more */
- num_cache_leaves = 3;
+ if (c->extended_cpuid_level >= 0x80000006 &&
+ (cpuid_edx(0x80000006) & 0xf000))
+ num_cache_leaves = 4;
+ else
+ num_cache_leaves = 3;
+
+ if (c->x86 == 0xf || c->x86 == 0x10 || c->x86 == 0x11)
+ set_bit(X86_FEATURE_K8, &c->x86_capability);
/* RDTSC can be speculated around */
clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
@@ -1097,6 +1105,8 @@ void __cpuinit identify_cpu(struct cpuin
c->x86_capability[2] = cpuid_edx(0x80860001);
}
+ init_scattered_cpuid_features(c);
+
#ifndef CONFIG_XEN
c->apicid = phys_pkg_id(0);
#endif
@@ -1184,7 +1194,7 @@ static int show_cpuinfo(struct seq_file
"fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce",
"cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov",
"pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx",
- "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL,
+ "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe",
/* AMD-defined */
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
@@ -1200,10 +1210,11 @@ static int show_cpuinfo(struct seq_file
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* Other (Linux-defined) */
- "cxmmx", NULL, "cyrix_arr", "centaur_mcr", NULL,
- "constant_tsc", NULL, NULL,
- "up", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
- NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ "cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr",
+ NULL, NULL, NULL, NULL,
+ "constant_tsc", "up", NULL, "arch_perfmon",
+ "pebs", "bts", NULL, "sync_rdtsc",
+ "rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* Intel-defined (#2) */
@@ -1214,7 +1225,7 @@ static int show_cpuinfo(struct seq_file
/* VIA/Cyrix/Centaur-defined */
NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en",
- NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ "ace2", "ace2_en", "phe", "phe_en", "pmm", "pmm_en", NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
@@ -1225,6 +1236,12 @@ static int show_cpuinfo(struct seq_file
"osvw", "ibs", NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+ /* Auxiliary (Linux-defined) */
+ "ida", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
};
static char *x86_power_flags[] = {
"ts", /* temperature sensor */
--- head.orig/arch/x86/kernel/setup64-xen.c 2012-04-20 15:04:40.000000000 +0200
+++ head/arch/x86/kernel/setup64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -125,11 +125,14 @@ void __init setup_per_cpu_areas(void)
}
#ifdef CONFIG_XEN
-static void switch_pt(void)
+static void __init_refok switch_pt(int cpu)
{
+ if (cpu == 0)
+ xen_init_pt();
xen_pt_switch(__pa_symbol(init_level4_pgt));
xen_new_user_pt(__pa_symbol(__user_pgd(init_level4_pgt)));
}
+#define switch_pt() switch_pt(cpu)
static void __cpuinit cpu_gdt_init(const struct desc_ptr *gdt_descr)
{
@@ -185,9 +188,6 @@ void pda_init(int cpu)
pda->mmu_state = 0;
if (cpu == 0) {
-#ifdef CONFIG_XEN
- xen_init_pt();
-#endif
/* others are initialized in smpboot.c */
pda->pcurrent = &init_task;
pda->irqstackptr = boot_cpu_stack;
--- head.orig/arch/x86/kernel/smp_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/smp_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -362,7 +362,7 @@ __smp_call_function_single(int cpu, void
}
/*
- * smp_call_function_single - Run a function on another CPU
+ * smp_call_function_single - Run a function on a specific CPU
* @func: The function to run. This must be fast and non-blocking.
* @info: An arbitrary pointer to pass to the function.
* @nonatomic: Currently unused.
@@ -379,17 +379,21 @@ int smp_call_function_single (int cpu, v
{
/* prevent preemption and reschedule on another processor */
int me = get_cpu();
+
+ /* Can deadlock when called with interrupts disabled */
+ WARN_ON(irqs_disabled());
+
if (cpu == me) {
+ local_irq_disable();
+ func(info);
+ local_irq_enable();
put_cpu();
return 0;
}
- /* Can deadlock when called with interrupts disabled */
- WARN_ON(irqs_disabled());
-
- spin_lock_bh(&call_lock);
+ spin_lock(&call_lock);
__smp_call_function_single(cpu, func, info, nonatomic, wait);
- spin_unlock_bh(&call_lock);
+ spin_unlock(&call_lock);
put_cpu();
return 0;
}
--- head.orig/arch/x86/kernel/traps_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/traps_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -34,6 +34,10 @@
#include <linux/bug.h>
#include <linux/kdebug.h>
+#if defined(CONFIG_EDAC)
+#include <linux/edac.h>
+#endif
+
#include <asm/system.h>
#include <asm/io.h>
#include <asm/atomic.h>
@@ -332,6 +336,7 @@ static int print_trace_stack(void *data,
static void print_trace_address(void *data, unsigned long addr)
{
+ touch_nmi_watchdog();
printk_address(addr);
}
@@ -520,6 +525,7 @@ void __kprobes __die(const char * str, s
printk("\n");
notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV);
show_registers(regs);
+ add_taint(TAINT_DIE);
/* Executive summary in case the oops scrolled away */
printk(KERN_ALERT "RIP ");
printk_address(regs->rip);
@@ -533,7 +539,7 @@ void die(const char * str, struct pt_reg
unsigned long flags = oops_begin();
if (!user_mode(regs))
- report_bug(regs->rip);
+ report_bug(regs->rip, regs);
__die(str, regs, err);
oops_end(flags);
@@ -582,7 +588,8 @@ static void __kprobes do_trap(int trapnr
tsk->thread.error_code = error_code;
tsk->thread.trap_no = trapnr;
- if (exception_trace && unhandled_signal(tsk, signr))
+ if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
+ printk_ratelimit())
printk(KERN_INFO
"%s[%d] trap %s rip:%lx rsp:%lx error:%lx\n",
tsk->comm, tsk->pid, str,
@@ -686,7 +693,8 @@ asmlinkage void __kprobes do_general_pro
tsk->thread.error_code = error_code;
tsk->thread.trap_no = 13;
- if (exception_trace && unhandled_signal(tsk, SIGSEGV))
+ if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
+ printk_ratelimit())
printk(KERN_INFO
"%s[%d] general protection rip:%lx rsp:%lx error:%lx\n",
tsk->comm, tsk->pid,
@@ -721,6 +729,13 @@ mem_parity_error(unsigned char reason, s
reason);
printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n");
+#if defined(CONFIG_EDAC)
+ if(edac_handler_set()) {
+ edac_atomic_assert_error();
+ return;
+ }
+#endif
+
if (panic_on_unrecovered_nmi)
panic("NMI: Not continuing");
--- head.orig/arch/x86/kernel/vsyscall_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/kernel/vsyscall_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -42,6 +42,7 @@
#include <asm/segment.h>
#include <asm/desc.h>
#include <asm/topology.h>
+#include <asm/vgtod.h>
#define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
#define __syscall_clobber "r11","rcx","memory"
@@ -57,26 +58,9 @@
* - writen by timer interrupt or systcl (/proc/sys/kernel/vsyscall64)
* Try to keep this structure as small as possible to avoid cache line ping pongs
*/
-struct vsyscall_gtod_data_t {
- seqlock_t lock;
-
- /* open coded 'struct timespec' */
- time_t wall_time_sec;
- u32 wall_time_nsec;
-
- int sysctl_enabled;
- struct timezone sys_tz;
- struct { /* extract of a clocksource struct */
- cycle_t (*vread)(void);
- cycle_t cycle_last;
- cycle_t mask;
- u32 mult;
- u32 shift;
- } clock;
-};
int __vgetcpu_mode __section_vgetcpu_mode;
-struct vsyscall_gtod_data_t __vsyscall_gtod_data __section_vsyscall_gtod_data =
+struct vsyscall_gtod_data __vsyscall_gtod_data __section_vsyscall_gtod_data =
{
.lock = SEQLOCK_UNLOCKED,
.sysctl_enabled = 1,
@@ -96,6 +80,8 @@ void update_vsyscall(struct timespec *wa
vsyscall_gtod_data.wall_time_sec = wall_time->tv_sec;
vsyscall_gtod_data.wall_time_nsec = wall_time->tv_nsec;
vsyscall_gtod_data.sys_tz = sys_tz;
+ vsyscall_gtod_data.wall_time_nsec = wall_time->tv_nsec;
+ vsyscall_gtod_data.wall_to_monotonic = wall_to_monotonic;
write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags);
}
--- head.orig/arch/x86/mm/fault_64-xen.c 2011-07-26 09:27:41.000000000 +0200
+++ head/arch/x86/mm/fault_64-xen.c 2011-08-15 10:44:41.000000000 +0200
@@ -160,7 +160,9 @@ void dump_pagetable(unsigned long addres
pmd_t *pmd;
pte_t *pte;
- pgd = __va(read_cr3() & PHYSICAL_PAGE_MASK);
+ pgd = (pgd_t *)read_cr3();
+
+ pgd = __va((unsigned long)pgd & PHYSICAL_PAGE_MASK);
pgd += pgd_index(address);
if (bad_address(pgd)) goto bad;
printk("PGD %lx ", pgd_val(*pgd));
@@ -220,16 +222,6 @@ static int is_errata93(struct pt_regs *r
return 0;
}
-int unhandled_signal(struct task_struct *tsk, int sig)
-{
- if (is_init(tsk))
- return 1;
- if (tsk->ptrace & PT_PTRACED)
- return 0;
- return (tsk->sighand->action[sig-1].sa.sa_handler == SIG_IGN) ||
- (tsk->sighand->action[sig-1].sa.sa_handler == SIG_DFL);
-}
-
static noinline void pgtable_bad(unsigned long address, struct pt_regs *regs,
unsigned long error_code)
{
@@ -303,8 +295,8 @@ static int vmalloc_fault(unsigned long a
return 0;
}
-int page_fault_trace = 0;
-int exception_trace = 1;
+static int page_fault_trace;
+int show_unhandled_signals = 1;
#define MEM_VERBOSE 1
@@ -373,7 +365,7 @@ asmlinkage void __kprobes do_page_fault(
struct vm_area_struct * vma;
unsigned long address;
const struct exception_table_entry *fixup;
- int write;
+ int write, fault;
unsigned long flags;
siginfo_t info;
@@ -385,7 +377,7 @@ asmlinkage void __kprobes do_page_fault(
prefetchw(&mm->mmap_sem);
/* get the address */
- address = current_vcpu_info()->arch.cr2;
+ address = read_cr2();
info.si_code = SEGV_MAPERR;
@@ -446,6 +438,13 @@ asmlinkage void __kprobes do_page_fault(
if (unlikely(in_atomic() || !mm))
goto bad_area_nosemaphore;
+ /*
+ * User-mode registers count as a user access even for any
+ * potential system fault or CPU buglet.
+ */
+ if (user_mode_vm(regs))
+ error_code |= PF_USER;
+
again:
/* When running in the kernel we expect faults to occur only to
* addresses in user space. All other faults represent errors in the
@@ -512,19 +511,18 @@ good_area:
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
- switch (handle_mm_fault(mm, vma, address, write)) {
- case VM_FAULT_MINOR:
- tsk->min_flt++;
- break;
- case VM_FAULT_MAJOR:
- tsk->maj_flt++;
- break;
- case VM_FAULT_SIGBUS:
- goto do_sigbus;
- default:
- goto out_of_memory;
+ fault = handle_mm_fault(mm, vma, address, write);
+ if (unlikely(fault & VM_FAULT_ERROR)) {
+ if (fault & VM_FAULT_OOM)
+ goto out_of_memory;
+ else if (fault & VM_FAULT_SIGBUS)
+ goto do_sigbus;
+ BUG();
}
-
+ if (fault & VM_FAULT_MAJOR)
+ tsk->maj_flt++;
+ else
+ tsk->min_flt++;
up_read(&mm->mmap_sem);
return;
@@ -557,7 +555,8 @@ bad_area_nosemaphore:
(regs->cs & (1<<2))) && (address >> 32))
return;
- if (exception_trace && unhandled_signal(tsk, SIGSEGV)) {
+ if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
+ printk_ratelimit()) {
printk(
"%s%s[%d]: segfault at %016lx rip %016lx rsp %016lx error %lx\n",
tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
@@ -631,7 +630,7 @@ out_of_memory:
}
printk("VM: killing process %s\n", tsk->comm);
if (error_code & 4)
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
goto no_context;
do_sigbus:
--- head.orig/arch/x86/mm/init_64-xen.c 2011-11-03 12:08:43.000000000 +0100
+++ head/arch/x86/mm/init_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -66,6 +66,9 @@ int after_bootmem;
DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
extern unsigned long start_pfn;
+extern pmd_t level2_fixmap_pgt[PTRS_PER_PMD];
+extern pte_t level1_fixmap_pgt[PTRS_PER_PTE];
+
/*
* Use this until direct mapping is established, i.e. before __va() is
* available in init_memory_mapping().
@@ -362,6 +365,10 @@ __set_fixmap (enum fixed_addresses idx,
set_pte_phys(address, phys, prot, 0);
set_pte_phys(address, phys, prot, 1);
break;
+ case FIX_EARLYCON_MEM_BASE:
+ xen_l1_entry_update(level1_fixmap_pgt + pte_index(address),
+ pfn_pte_ma(phys >> PAGE_SHIFT, prot));
+ break;
default:
set_pte_phys_ma(address, phys, prot);
break;
@@ -595,6 +602,19 @@ void __init xen_init_pt(void)
__user_pgd(init_level4_pgt)[pgd_index(VSYSCALL_START)] =
__pgd(__pa_symbol(level3_user_pgt) | _PAGE_TABLE);
+ /* Do an early initialization of the fixmap area. */
+ addr = __fix_to_virt(FIX_EARLYCON_MEM_BASE);
+ if (pud_present(level3_kernel_pgt[pud_index(addr)])) {
+ unsigned long adr = page[pud_index(addr)];
+
+ addr_to_page(adr, page);
+ copy_page(level2_fixmap_pgt, page);
+ }
+ level3_kernel_pgt[pud_index(addr)] =
+ __pud(__pa_symbol(level2_fixmap_pgt) | _PAGE_TABLE);
+ level2_fixmap_pgt[pmd_index(addr)] =
+ __pmd(__pa_symbol(level1_fixmap_pgt) | _PAGE_TABLE);
+
early_make_page_readonly(init_level4_pgt,
XENFEAT_writable_page_tables);
early_make_page_readonly(__user_pgd(init_level4_pgt),
@@ -603,6 +623,10 @@ void __init xen_init_pt(void)
XENFEAT_writable_page_tables);
early_make_page_readonly(level3_user_pgt,
XENFEAT_writable_page_tables);
+ early_make_page_readonly(level2_fixmap_pgt,
+ XENFEAT_writable_page_tables);
+ early_make_page_readonly(level1_fixmap_pgt,
+ XENFEAT_writable_page_tables);
if (!xen_feature(XENFEAT_writable_page_tables)) {
xen_pgd_pin(__pa_symbol(init_level4_pgt));
@@ -832,7 +856,7 @@ void __init paging_init(void)
sparse_init();
free_area_init_nodes(max_zone_pfns);
- init_mm.context.pinned = 1;
+ SetPagePinned(virt_to_page(init_mm.pgd));
}
#endif
@@ -1155,41 +1179,6 @@ int kern_addr_valid(unsigned long addr)
return pfn_valid(pte_pfn(*pte));
}
-#ifdef CONFIG_SYSCTL
-#include <linux/sysctl.h>
-
-extern int exception_trace, page_fault_trace;
-
-static ctl_table debug_table2[] = {
- {
- .ctl_name = 99,
- .procname = "exception-trace",
- .data = &exception_trace,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = proc_dointvec
- },
- {}
-};
-
-static ctl_table debug_root_table2[] = {
- {
- .ctl_name = CTL_DEBUG,
- .procname = "debug",
- .mode = 0555,
- .child = debug_table2
- },
- {}
-};
-
-static __init int x8664_sysctl_init(void)
-{
- register_sysctl_table(debug_root_table2);
- return 0;
-}
-__initcall(x8664_sysctl_init);
-#endif
-
/* A pseudo VMA to allow ptrace access for the vsyscall page. This only
covers the 64bit vsyscall page now. 32bit has a real VMA now and does
not need special handling anymore. */
@@ -1228,9 +1217,18 @@ int in_gate_area_no_task(unsigned long a
}
#ifndef CONFIG_XEN
-void *alloc_bootmem_high_node(pg_data_t *pgdat, unsigned long size)
+void * __init alloc_bootmem_high_node(pg_data_t *pgdat, unsigned long size)
{
return __alloc_bootmem_core(pgdat->bdata, size,
SMP_CACHE_BYTES, (4UL*1024*1024*1024), 0);
}
#endif
+
+const char *arch_vma_name(struct vm_area_struct *vma)
+{
+ if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
+ return "[vdso]";
+ if (vma == &gate_vma)
+ return "[vsyscall]";
+ return NULL;
+}
--- head.orig/arch/x86/mm/pageattr_64-xen.c 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/mm/pageattr_64-xen.c 2011-01-31 17:49:31.000000000 +0100
@@ -171,7 +171,7 @@ void mm_pin(struct mm_struct *mm)
mm_walk(mm, PAGE_KERNEL_RO);
xen_pgd_pin(__pa(mm->pgd)); /* kernel */
xen_pgd_pin(__pa(__user_pgd(mm->pgd))); /* user */
- mm->context.pinned = 1;
+ SetPagePinned(virt_to_page(mm->pgd));
spin_lock(&mm_unpinned_lock);
list_del(&mm->context.unpinned);
spin_unlock(&mm_unpinned_lock);
@@ -189,7 +189,7 @@ void mm_unpin(struct mm_struct *mm)
xen_pgd_unpin(__pa(mm->pgd));
xen_pgd_unpin(__pa(__user_pgd(mm->pgd)));
mm_walk(mm, PAGE_KERNEL);
- mm->context.pinned = 0;
+ ClearPagePinned(virt_to_page(mm->pgd));
spin_lock(&mm_unpinned_lock);
list_add(&mm->context.unpinned, &mm_unpinned);
spin_unlock(&mm_unpinned_lock);
@@ -217,7 +217,7 @@ void mm_pin_all(void)
void arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
{
- if (!mm->context.pinned)
+ if (!PagePinned(virt_to_page(mm->pgd)))
mm_pin(mm);
}
@@ -243,8 +243,9 @@ void arch_exit_mmap(struct mm_struct *mm
task_unlock(tsk);
- if ( mm->context.pinned && (atomic_read(&mm->mm_count) == 1) &&
- !mm->context.has_foreign_mappings )
+ if (PagePinned(virt_to_page(mm->pgd))
+ && (atomic_read(&mm->mm_count) == 1)
+ && !mm->context.has_foreign_mappings)
mm_unpin(mm);
}
@@ -343,14 +344,13 @@ static void flush_kernel_map(void *arg)
struct page *pg;
/* When clflush is available always use it because it is
- much cheaper than WBINVD. Disable clflush for now because
- the high level code is not ready yet */
+ much cheaper than WBINVD. */
+ /* clflush is still broken. Disable for now. */
if (1 || !cpu_has_clflush)
asm volatile("wbinvd" ::: "memory");
else list_for_each_entry(pg, l, lru) {
void *adr = page_address(pg);
- if (cpu_has_clflush)
- cache_flush_page(adr);
+ cache_flush_page(adr);
}
__flush_tlb_all();
}
@@ -364,7 +364,8 @@ static LIST_HEAD(deferred_pages); /* pro
static inline void save_page(struct page *fpage)
{
- list_add(&fpage->lru, &deferred_pages);
+ if (!test_and_set_bit(PG_arch_1, &fpage->flags))
+ list_add(&fpage->lru, &deferred_pages);
}
/*
@@ -398,9 +399,12 @@ __change_page_attr(unsigned long address
pte_t *kpte;
struct page *kpte_page;
pgprot_t ref_prot2;
+
kpte = lookup_address(address);
if (!kpte) return 0;
kpte_page = virt_to_page(((unsigned long)kpte) & PAGE_MASK);
+ BUG_ON(PageLRU(kpte_page));
+ BUG_ON(PageCompound(kpte_page));
if (pgprot_val(prot) != pgprot_val(ref_prot)) {
if (!pte_huge(*kpte)) {
set_pte(kpte, pfn_pte(pfn, prot));
@@ -439,10 +443,9 @@ __change_page_attr(unsigned long address
return 0;
#endif
- if (page_private(kpte_page) == 0) {
- save_page(kpte_page);
+ save_page(kpte_page);
+ if (page_private(kpte_page) == 0)
revert_page(address, ref_prot);
- }
return 0;
}
@@ -514,6 +517,10 @@ void global_flush_tlb(void)
flush_map(&l);
list_for_each_entry_safe(pg, next, &l, lru) {
+ list_del(&pg->lru);
+ clear_bit(PG_arch_1, &pg->flags);
+ if (page_private(pg) != 0)
+ continue;
ClearPagePrivate(pg);
__free_page(pg);
}
--- head.orig/arch/x86/oprofile/xenoprof.c 2008-01-28 12:24:19.000000000 +0100
+++ head/arch/x86/oprofile/xenoprof.c 2011-01-31 17:49:31.000000000 +0100
@@ -18,9 +18,9 @@
#include <linux/init.h>
#include <linux/oprofile.h>
#include <linux/sched.h>
+#include <linux/vmalloc.h>
#include <asm/pgtable.h>
-#include <xen/driver_util.h>
#include <xen/interface/xen.h>
#include <xen/interface/xenoprof.h>
#include <xen/xenoprof.h>
--- head.orig/arch/x86/vdso/vdso32/note.S 2012-06-12 15:06:23.000000000 +0200
+++ head/arch/x86/vdso/vdso32/note.S 2011-01-31 17:49:31.000000000 +0100
@@ -13,7 +13,7 @@ ELFNOTE_START(Linux, 0, "a")
.long LINUX_VERSION_CODE
ELFNOTE_END
-#ifdef CONFIG_XEN
+#if defined(CONFIG_X86_XEN) || defined(CONFIG_PARAVIRT_XEN)
/*
* Add a special note telling glibc's dynamic linker a fake hardware
* flavor that it will use to choose the search path for libraries in the
@@ -37,8 +37,12 @@ ELFNOTE_END
ELFNOTE_START(GNU, 2, "a")
.long 1 /* ncaps */
+#ifdef CONFIG_PARAVIRT_XEN
VDSO32_NOTE_MASK: /* Symbol used by arch/x86/xen/setup.c */
.long 0 /* mask */
+#else
+ .long 1 << VDSO_NOTE_NONEGSEG_BIT /* mask */
+#endif
.byte VDSO_NOTE_NONEGSEG_BIT; .asciz "nosegneg" /* bit, name */
ELFNOTE_END
#endif
--- head.orig/drivers/Makefile 2011-11-16 17:01:23.000000000 +0100
+++ head/drivers/Makefile 2011-11-17 10:57:00.000000000 +0100
@@ -24,7 +24,7 @@ obj-$(CONFIG_ARM_AMBA) += amba/
obj-$(CONFIG_DMA_ENGINE) += dma/
obj-$(CONFIG_VIRTIO) += virtio/
-obj-$(CONFIG_XEN) += xen/
+obj-$(CONFIG_PARAVIRT_XEN) += xen/
# regulators early, since some subsystems rely on them to initialize
obj-$(CONFIG_REGULATOR) += regulator/
--- head.orig/drivers/block/Kconfig 2012-06-12 15:06:23.000000000 +0200
+++ head/drivers/block/Kconfig 2012-03-12 13:39:16.000000000 +0100
@@ -488,9 +488,9 @@ config XILINX_SYSACE
help
Include support for the Xilinx SystemACE CompactFlash interface
-config XEN_BLKDEV_FRONTEND
+config PARAVIRT_XEN_BLKDEV_FRONTEND
tristate "Xen virtual block device support"
- depends on XEN
+ depends on PARAVIRT_XEN
default y
select XEN_XENBUS_FRONTEND
help
--- head.orig/drivers/block/Makefile 2012-06-12 15:06:23.000000000 +0200
+++ head/drivers/block/Makefile 2012-02-08 11:35:25.000000000 +0100
@@ -36,7 +36,7 @@ obj-$(CONFIG_BLK_DEV_SX8) += sx8.o
obj-$(CONFIG_BLK_DEV_UB) += ub.o
obj-$(CONFIG_BLK_DEV_HD) += hd.o
-obj-$(CONFIG_XEN_BLKDEV_FRONTEND) += xen-blkfront.o
+obj-$(CONFIG_PARAVIRT_XEN_BLKDEV_FRONTEND) += xen-blkfront.o
obj-$(CONFIG_XEN_BLKDEV_BACKEND) += xen-blkback/
obj-$(CONFIG_BLK_DEV_DRBD) += drbd/
obj-$(CONFIG_BLK_DEV_RBD) += rbd.o
--- head.orig/drivers/tty/hvc/Kconfig 2012-06-12 15:06:23.000000000 +0200
+++ head/drivers/tty/hvc/Kconfig 2012-04-10 16:36:21.000000000 +0200
@@ -59,7 +59,7 @@ config HVC_IUCV
config HVC_XEN
bool "Xen Hypervisor Console support"
- depends on XEN
+ depends on PARAVIRT_XEN
select HVC_DRIVER
select HVC_IRQ
default y
--- head.orig/drivers/net/Kconfig 2012-06-12 15:06:23.000000000 +0200
+++ head/drivers/net/Kconfig 2012-02-08 11:35:09.000000000 +0100
@@ -295,9 +295,9 @@ source "drivers/net/wimax/Kconfig"
source "drivers/net/wan/Kconfig"
-config XEN_NETDEV_FRONTEND
+config PARAVIRT_XEN_NETDEV_FRONTEND
tristate "Xen network device frontend driver"
- depends on XEN
+ depends on PARAVIRT_XEN
select XEN_XENBUS_FRONTEND
default y
help
--- head.orig/drivers/net/Makefile 2012-06-12 15:06:23.000000000 +0200
+++ head/drivers/net/Makefile 2012-02-08 11:35:14.000000000 +0100
@@ -56,7 +56,7 @@ obj-$(CONFIG_WLAN) += wireless/
obj-$(CONFIG_WIMAX) += wimax/
obj-$(CONFIG_VMXNET3) += vmxnet3/
-obj-$(CONFIG_XEN_NETDEV_FRONTEND) += xen-netfront.o
+obj-$(CONFIG_PARAVIRT_XEN_NETDEV_FRONTEND) += xen-netfront.o
obj-$(CONFIG_XEN_NETDEV_BACKEND) += xen-netback/
obj-$(CONFIG_USB_CATC) += usb/
--- head.orig/drivers/xen/Kconfig 2011-11-03 12:01:19.000000000 +0100
+++ head/drivers/xen/Kconfig 2011-01-31 17:49:31.000000000 +0100
@@ -23,6 +23,9 @@ config XEN_PRIVILEGED_GUEST
config XEN_UNPRIVILEGED_GUEST
def_bool !XEN_PRIVILEGED_GUEST
select PM
+ select PM_SLEEP
+ select PM_SLEEP_SMP if SMP
+ select SUSPEND
config XEN_PRIVCMD
def_bool y
--- head.orig/drivers/xen/Makefile 2011-10-19 10:32:29.000000000 +0200
+++ head/drivers/xen/Makefile 2011-01-31 17:49:31.000000000 +0100
@@ -1,10 +1,14 @@
-obj-y += core/
-obj-y += console/
-obj-y += evtchn/
-obj-y += xenbus/
-obj-y += char/
+obj-$(CONFIG_PARAVIRT_XEN) += grant-table.o
-obj-y += util.o
+obj-$(CONFIG_XEN) += core/
+obj-$(CONFIG_XEN) += console/
+obj-$(CONFIG_XEN) += evtchn/
+obj-y += xenbus/
+obj-$(CONFIG_XEN) += char/
+
+xen-backend-$(CONFIG_XEN_BACKEND) := util.o
+
+obj-$(CONFIG_XEN) += $(xen-backend-y) $(xen-backend-m)
obj-$(CONFIG_XEN_BALLOON) += balloon/
obj-$(CONFIG_XEN_BLKDEV_BACKEND) += blkback/
obj-$(CONFIG_XEN_BLKDEV_TAP) += blktap/
--- head.orig/drivers/xen/balloon/balloon.c 2012-06-06 13:48:56.000000000 +0200
+++ head/drivers/xen/balloon/balloon.c 2012-06-06 13:52:53.000000000 +0200
@@ -317,13 +317,9 @@ static int increase_reservation(unsigned
#ifndef MODULE
setup_per_zone_pages_min();
-# if defined(CONFIG_MEMORY_HOTPLUG) || defined(CONFIG_ACPI_HOTPLUG_MEMORY) \
- || defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
- /* build_all_zonelists() is __meminit */
if (need_zonelists_rebuild)
build_all_zonelists();
else
-# endif
vm_total_pages = nr_free_pagecache_pages();
#endif
--- head.orig/drivers/xen/blkback/blkback.c 2012-03-26 12:21:29.000000000 +0200
+++ head/drivers/xen/blkback/blkback.c 2012-03-26 12:21:55.000000000 +0200
@@ -157,7 +157,7 @@ static void unplug_queue(blkif_t *blkif)
static void plug_queue(blkif_t *blkif, struct block_device *bdev)
{
- request_queue_t *q = bdev_get_queue(bdev);
+ struct request_queue *q = bdev_get_queue(bdev);
if (q == blkif->plug)
return;
--- head.orig/drivers/xen/blkback/common.h 2012-06-06 13:46:49.000000000 +0200
+++ head/drivers/xen/blkback/common.h 2012-06-06 13:52:57.000000000 +0200
@@ -35,7 +35,6 @@
#include <asm/hypervisor.h>
#include <xen/barrier.h>
#include <xen/blkif.h>
-#include <xen/driver_util.h>
#include <xen/xenbus.h>
#include <xen/interface/event_channel.h>
#include "blkback-pagemap.h"
@@ -77,7 +76,7 @@ typedef struct blkif_st {
wait_queue_head_t wq;
struct task_struct *xenblkd;
unsigned int waiting_reqs;
- request_queue_t *plug;
+ struct request_queue *plug;
/* statistics */
unsigned long st_print;
--- head.orig/drivers/xen/blkback/interface.c 2011-04-11 14:29:25.000000000 +0200
+++ head/drivers/xen/blkback/interface.c 2012-04-04 10:14:31.000000000 +0200
@@ -133,5 +133,5 @@ void blkif_free(blkif_t *blkif)
void __init blkif_interface_init(void)
{
blkif_cachep = kmem_cache_create("blkif_cache", sizeof(blkif_t),
- 0, 0, NULL, NULL);
+ 0, 0, NULL);
}
--- head.orig/drivers/xen/blkback/xenbus.c 2012-04-04 08:57:09.000000000 +0200
+++ head/drivers/xen/blkback/xenbus.c 2011-01-31 17:49:31.000000000 +0100
@@ -27,8 +27,6 @@
pr_debug("blkback/xenbus (%s:%d) " fmt ".\n", \
__FUNCTION__, __LINE__, ##args)
-static DEFINE_RWLOCK(sysfs_read_lock);
-
static void connect(struct backend_info *);
static int connect_ring(struct backend_info *);
static void backend_changed(struct xenbus_watch *, const char **,
@@ -110,10 +108,8 @@ static void update_blkif_status(blkif_t
if (!get_device(_dev)) \
return ret; \
dev = to_xenbus_device(_dev); \
- read_lock(&sysfs_read_lock); \
if ((be = dev->dev.driver_data) != NULL) \
ret = sprintf(buf, format, ##args); \
- read_unlock(&sysfs_read_lock); \
put_device(_dev); \
return ret; \
} \
@@ -181,7 +177,6 @@ static int blkback_remove(struct xenbus_
DPRINTK("");
- write_lock(&sysfs_read_lock);
if (be->major || be->minor)
xenvbd_sysfs_delif(dev);
@@ -200,7 +195,6 @@ static int blkback_remove(struct xenbus_
kfree(be);
dev->dev.driver_data = NULL;
- write_unlock(&sysfs_read_lock);
return 0;
}
--- head.orig/drivers/xen/blkfront/blkfront.c 2012-03-12 13:33:33.000000000 +0100
+++ head/drivers/xen/blkfront/blkfront.c 2012-06-12 15:12:49.000000000 +0200
@@ -593,14 +593,20 @@ int blkif_ioctl(struct inode *inode, str
return -EINVAL;
default:
- if (info->mi && info->gd) {
+ if (info->mi && info->gd && info->rq) {
switch (info->mi->major) {
case SCSI_DISK0_MAJOR:
case SCSI_DISK1_MAJOR ... SCSI_DISK7_MAJOR:
case SCSI_DISK8_MAJOR ... SCSI_DISK15_MAJOR:
case SCSI_CDROM_MAJOR:
+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,23)
return scsi_cmd_ioctl(filep, info->gd, command,
(void __user *)argument);
+#else
+ return scsi_cmd_ioctl(filep, info->rq,
+ info->gd, command,
+ (void __user *)argument);
+#endif
}
}
@@ -717,7 +723,7 @@ static int blkif_queue_request(struct re
* do_blkif_request
* read a block; request is in a request queue
*/
-void do_blkif_request(request_queue_t *rq)
+void do_blkif_request(struct request_queue *rq)
{
struct blkfront_info *info = NULL;
struct request *req;
--- head.orig/drivers/xen/blkfront/block.h 2012-06-06 13:16:59.000000000 +0200
+++ head/drivers/xen/blkfront/block.h 2012-06-06 13:53:02.000000000 +0200
@@ -101,7 +101,7 @@ struct blkfront_info
struct scatterlist sg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
unsigned int irq;
struct xlbd_major_info *mi;
- request_queue_t *rq;
+ struct request_queue *rq;
struct work_struct work;
struct gnttab_free_callback callback;
struct blk_shadow shadow[BLK_RING_SIZE];
@@ -123,7 +123,7 @@ extern int blkif_ioctl(struct inode *ino
extern int blkif_getgeo(struct block_device *, struct hd_geometry *);
extern int blkif_check(dev_t dev);
extern int blkif_revalidate(dev_t dev);
-extern void do_blkif_request (request_queue_t *rq);
+extern void do_blkif_request (struct request_queue *rq);
/* Virtual block-device subsystem. */
/* Note that xlvbd_add doesn't call add_disk for you: you're expected
--- head.orig/drivers/xen/blkfront/vbd.c 2012-03-12 12:51:05.000000000 +0100
+++ head/drivers/xen/blkfront/vbd.c 2012-03-12 13:38:18.000000000 +0100
@@ -357,7 +357,7 @@ static int
xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size,
struct blkfront_info *info)
{
- request_queue_t *rq;
+ struct request_queue *rq;
rq = blk_init_queue(do_blkif_request, &info->io_lock);
if (rq == NULL)
--- head.orig/drivers/xen/blktap/common.h 2012-06-06 13:46:52.000000000 +0200
+++ head/drivers/xen/blktap/common.h 2012-06-06 13:53:10.000000000 +0200
@@ -34,7 +34,6 @@
#include <asm/hypervisor.h>
#include <xen/barrier.h>
#include <xen/blkif.h>
-#include <xen/driver_util.h>
#include <xen/xenbus.h>
#include <xen/interface/event_channel.h>
@@ -64,7 +63,7 @@ typedef struct blkif_st {
wait_queue_head_t wq;
struct task_struct *xenblkd;
unsigned int waiting_reqs;
- request_queue_t *plug;
+ struct request_queue *plug;
/* statistics */
unsigned long st_print;
--- head.orig/drivers/xen/blktap/interface.c 2011-04-11 14:29:39.000000000 +0200
+++ head/drivers/xen/blktap/interface.c 2012-04-04 10:14:34.000000000 +0200
@@ -129,5 +129,5 @@ void tap_blkif_kmem_cache_free(blkif_t *
void __init tap_blkif_interface_init(void)
{
blkif_cachep = kmem_cache_create("blktapif_cache", sizeof(blkif_t),
- 0, 0, NULL, NULL);
+ 0, 0, NULL);
}
--- head.orig/drivers/xen/blktap/xenbus.c 2012-04-04 08:57:09.000000000 +0200
+++ head/drivers/xen/blktap/xenbus.c 2012-02-16 12:30:42.000000000 +0100
@@ -50,8 +50,6 @@ struct backend_info
int group_added;
};
-static DEFINE_RWLOCK(sysfs_read_lock);
-
static void connect(struct backend_info *);
static int connect_ring(struct backend_info *);
static int blktap_remove(struct xenbus_device *dev);
@@ -128,10 +126,8 @@ static int blktap_name(blkif_t *blkif, c
if (!get_device(_dev)) \
return ret; \
dev = to_xenbus_device(_dev); \
- read_lock(&sysfs_read_lock); \
if ((be = dev->dev.driver_data) != NULL) \
ret = sprintf(buf, format, ##args); \
- read_unlock(&sysfs_read_lock); \
put_device(_dev); \
return ret; \
} \
@@ -178,7 +174,6 @@ static int blktap_remove(struct xenbus_d
{
struct backend_info *be = dev->dev.driver_data;
- write_lock(&sysfs_read_lock);
if (be->group_added)
xentap_sysfs_delif(be->dev);
if (be->backend_watch.node) {
@@ -196,7 +191,6 @@ static int blktap_remove(struct xenbus_d
}
kfree(be);
dev->dev.driver_data = NULL;
- write_unlock(&sysfs_read_lock);
return 0;
}
--- head.orig/drivers/xen/blktap2/blktap.h 2012-06-06 13:16:59.000000000 +0200
+++ head/drivers/xen/blktap2/blktap.h 2012-06-06 13:53:14.000000000 +0200
@@ -204,6 +204,7 @@ blktap_validate_params(struct blktap *ta
}
int blktap_control_destroy_device(struct blktap *);
+int blktap_control_finish_destroy(struct blktap *);
int blktap_ring_init(int *);
int blktap_ring_free(void);
--- head.orig/drivers/xen/blktap2/control.c 2011-04-11 13:43:15.000000000 +0200
+++ head/drivers/xen/blktap2/control.c 2011-01-31 17:49:31.000000000 +0100
@@ -194,14 +194,20 @@ blktap_control_destroy_device(struct blk
clear_bit(BLKTAP_SHUTDOWN_REQUESTED, &tap->dev_inuse);
- if (tap->dev_inuse == (1UL << BLKTAP_CONTROL)) {
+ if (blktap_control_finish_destroy(tap))
err = 0;
- clear_bit(BLKTAP_CONTROL, &tap->dev_inuse);
- }
return err;
}
+int
+blktap_control_finish_destroy(struct blktap *tap)
+{
+ if (tap->dev_inuse == (1UL << BLKTAP_CONTROL))
+ clear_bit(BLKTAP_CONTROL, &tap->dev_inuse);
+ return !tap->dev_inuse;
+}
+
static int __init
blktap_control_init(void)
{
--- head.orig/drivers/xen/blktap2/device.c 2012-02-16 11:09:01.000000000 +0100
+++ head/drivers/xen/blktap2/device.c 2012-02-16 12:30:32.000000000 +0100
@@ -811,7 +811,7 @@ static void
blktap_device_run_queue(struct blktap *tap)
{
int queued, err;
- request_queue_t *rq;
+ struct request_queue *rq;
struct request *req;
struct blktap_ring *ring;
struct blktap_device *dev;
@@ -894,7 +894,7 @@ blktap_device_run_queue(struct blktap *t
* dev->lock held on entry
*/
static void
-blktap_device_do_request(request_queue_t *rq)
+blktap_device_do_request(struct request_queue *rq)
{
struct request *req;
struct blktap *tap;
@@ -1173,6 +1173,5 @@ void
blktap_device_free(void)
{
if (blktap_device_major)
- if (unregister_blkdev(blktap_device_major, "tapdev"))
- BTERR("blktap device unregister failed\n");
+ unregister_blkdev(blktap_device_major, "tapdev");
}
--- head.orig/drivers/xen/blktap2/sysfs.c 2011-01-31 17:29:16.000000000 +0100
+++ head/drivers/xen/blktap2/sysfs.c 2011-01-31 17:49:31.000000000 +0100
@@ -334,6 +334,24 @@ blktap_sysfs_create(struct blktap *tap)
return err;
}
+static void
+_blktap_sysfs_destroy(struct device *_dev)
+{
+ struct class_device *dev = _dev->???;
+ struct blktap *tap = dev->class_data;
+
+ class_device_remove_file(dev, &class_device_attr_name);
+ class_device_remove_file(dev, &class_device_attr_remove);
+ class_device_remove_file(dev, &class_device_attr_pause);
+ class_device_remove_file(dev, &class_device_attr_resume);
+ class_device_remove_file(dev, &class_device_attr_debug);
+ class_device_unregister(dev);
+
+ clear_bit(BLKTAP_SYSFS, &tap->dev_inuse);
+
+ blktap_control_finish_destroy(tap);
+}
+
int
blktap_sysfs_destroy(struct blktap *tap)
{
@@ -350,17 +368,7 @@ blktap_sysfs_destroy(struct blktap *tap)
!atomic_read(&tap->ring.sysfs_refcnt)))
return -EAGAIN;
- /* XXX: is it safe to remove the class from a sysfs attribute? */
- class_device_remove_file(dev, &class_device_attr_name);
- class_device_remove_file(dev, &class_device_attr_remove);
- class_device_remove_file(dev, &class_device_attr_pause);
- class_device_remove_file(dev, &class_device_attr_resume);
- class_device_remove_file(dev, &class_device_attr_debug);
- class_device_destroy(class, ring->devno);
-
- clear_bit(BLKTAP_SYSFS, &tap->dev_inuse);
-
- return 0;
+ return device_schedule_callback(dev->???, _blktap_sysfs_destroy);
}
static ssize_t
--- head.orig/drivers/xen/core/gnttab.c 2012-03-12 13:31:19.000000000 +0100
+++ head/drivers/xen/core/gnttab.c 2012-03-12 13:38:38.000000000 +0100
@@ -43,7 +43,6 @@
#include <asm/synch_bitops.h>
#include <asm/io.h>
#include <xen/interface/memory.h>
-#include <xen/driver_util.h>
#include <asm/gnttab_dma.h>
#ifdef HAVE_XEN_PLATFORM_COMPAT_H
--- head.orig/drivers/xen/core/reboot.c 2011-01-31 17:32:16.000000000 +0100
+++ head/drivers/xen/core/reboot.c 2011-01-31 17:49:31.000000000 +0100
@@ -3,6 +3,7 @@
#include <linux/unistd.h>
#include <linux/module.h>
#include <linux/reboot.h>
+#include <linux/sched.h>
#include <linux/sysrq.h>
#include <asm/hypervisor.h>
#include <xen/xenbus.h>
--- head.orig/drivers/xen/core/xen_proc.c 2011-04-11 13:43:15.000000000 +0200
+++ head/drivers/xen/core/xen_proc.c 2011-04-01 13:07:57.000000000 +0200
@@ -1,11 +1,15 @@
-
+#include <linux/init.h>
#include <linux/module.h>
#include <linux/proc_fs.h>
#include <xen/xen_proc.h>
static struct proc_dir_entry *xen_base;
-struct proc_dir_entry *create_xen_proc_entry(const char *name, mode_t mode)
+struct proc_dir_entry *
+#ifndef MODULE
+__init
+#endif
+create_xen_proc_entry(const char *name, mode_t mode)
{
if ( xen_base == NULL )
if ( (xen_base = proc_mkdir("xen", &proc_root)) == NULL )
@@ -15,7 +19,7 @@ struct proc_dir_entry *create_xen_proc_e
#ifdef MODULE
EXPORT_SYMBOL_GPL(create_xen_proc_entry);
-#else
+#elif defined(CONFIG_XEN_PRIVILEGED_GUEST)
void remove_xen_proc_entry(const char *name)
{
--- head.orig/drivers/xen/netback/common.h 2012-06-06 13:49:42.000000000 +0200
+++ head/drivers/xen/netback/common.h 2012-06-06 13:53:20.000000000 +0200
@@ -39,7 +39,6 @@
#include <linux/wait.h>
#include <xen/interface/io/netif.h>
#include <xen/barrier.h>
-#include <xen/driver_util.h>
#include <xen/xenbus.h>
#include <xen/interface/event_channel.h>
--- head.orig/drivers/xen/scsiback/common.h 2012-06-06 13:48:05.000000000 +0200
+++ head/drivers/xen/scsiback/common.h 2012-06-06 13:53:26.000000000 +0200
@@ -48,7 +48,6 @@
#include <scsi/scsi_eh.h>
#include <asm/hypervisor.h>
#include <xen/barrier.h>
-#include <xen/driver_util.h>
#include <xen/xenbus.h>
#include <xen/interface/io/ring.h>
#include <xen/interface/io/vscsiif.h>
--- head.orig/drivers/xen/scsiback/interface.c 2011-04-11 14:30:02.000000000 +0200
+++ head/drivers/xen/scsiback/interface.c 2011-04-11 14:52:22.000000000 +0200
@@ -126,7 +126,7 @@ void scsiback_free(struct vscsibk_info *
int __init scsiback_interface_init(void)
{
scsiback_cachep = kmem_cache_create("vscsiif_cache",
- sizeof(struct vscsibk_info), 0, 0, NULL, NULL);
+ sizeof(struct vscsibk_info), 0, 0, NULL);
if (!scsiback_cachep) {
printk(KERN_ERR "scsiback: can't init scsi cache\n");
return -ENOMEM;
--- head.orig/drivers/xen/scsifront/scsifront.c 2011-01-31 17:29:16.000000000 +0100
+++ head/drivers/xen/scsifront/scsifront.c 2011-02-08 10:03:55.000000000 +0100
@@ -147,7 +147,7 @@ static void scsifront_cdb_cmd_done(struc
add_id_to_freelist(info, id);
sc->result = ring_res->rslt;
- sc->resid = ring_res->residual_len;
+ scsi_set_resid(sc, ring_res->residual_len);
if (ring_res->sense_len > VSCSIIF_SENSE_BUFFERSIZE)
sense_len = VSCSIIF_SENSE_BUFFERSIZE;
@@ -351,8 +351,7 @@ big_to_sg:
static int scsifront_queuecommand(struct scsi_cmnd *sc,
void (*done)(struct scsi_cmnd *))
{
- struct vscsifrnt_info *info =
- (struct vscsifrnt_info *) sc->device->host->hostdata;
+ struct vscsifrnt_info *info = shost_priv(sc->device->host);
vscsiif_request_t *ring_req;
int ref_cnt;
uint16_t rqid;
@@ -428,8 +427,7 @@ static int scsifront_eh_abort_handler(st
static int scsifront_dev_reset_handler(struct scsi_cmnd *sc)
{
struct Scsi_Host *host = sc->device->host;
- struct vscsifrnt_info *info =
- (struct vscsifrnt_info *) sc->device->host->hostdata;
+ struct vscsifrnt_info *info = shost_priv(host);
vscsiif_request_t *ring_req;
uint16_t rqid;
--- head.orig/drivers/xen/sfc_netback/accel_solarflare.c 2011-01-31 17:32:29.000000000 +0100
+++ head/drivers/xen/sfc_netback/accel_solarflare.c 2011-01-31 17:49:31.000000000 +0100
@@ -37,7 +37,6 @@
#include "ci/efhw/public.h"
#include <xen/evtchn.h>
-#include <xen/driver_util.h>
#include <linux/list.h>
#include <linux/mutex.h>
--- head.orig/drivers/xen/sfc_netutil/accel_util.c 2010-09-23 15:39:04.000000000 +0200
+++ head/drivers/xen/sfc_netutil/accel_util.c 2011-01-31 17:49:31.000000000 +0100
@@ -28,7 +28,6 @@
#include <asm/pgtable.h>
#include <asm/hypercall.h>
#include <xen/xenbus.h>
-#include <xen/driver_util.h>
#include <xen/gnttab.h>
#include "accel_util.h"
--- head.orig/drivers/xen/tpmback/common.h 2011-11-28 09:46:47.000000000 +0100
+++ head/drivers/xen/tpmback/common.h 2011-11-28 09:47:23.000000000 +0100
@@ -9,7 +9,6 @@
#include <linux/interrupt.h>
#include <linux/mm.h>
#include <linux/slab.h>
-#include <xen/driver_util.h>
#include <xen/xenbus.h>
#include <xen/interface/event_channel.h>
#include <xen/interface/io/tpmif.h>
--- head.orig/drivers/xen/tpmback/interface.c 2011-04-11 14:31:42.000000000 +0200
+++ head/drivers/xen/tpmback/interface.c 2011-04-11 14:52:30.000000000 +0200
@@ -13,6 +13,7 @@
#include "common.h"
#include <linux/delay.h>
+#include <linux/err.h>
#include <linux/vmalloc.h>
#include <xen/balloon.h>
#include <xen/evtchn.h>
@@ -122,7 +123,7 @@ void tpmif_disconnect_complete(tpmif_t *
int __init tpmif_interface_init(void)
{
tpmif_cachep = kmem_cache_create("tpmif_cache", sizeof (tpmif_t),
- 0, 0, NULL, NULL);
+ 0, 0, NULL);
return tpmif_cachep ? 0 : -ENOMEM;
}
--- head.orig/drivers/xen/usbback/usbback.h 2012-06-06 13:48:22.000000000 +0200
+++ head/drivers/xen/usbback/usbback.h 2012-06-06 13:53:31.000000000 +0200
@@ -55,7 +55,6 @@
#include <linux/list.h>
#include <linux/kref.h>
#include <xen/barrier.h>
-#include <xen/driver_util.h>
#include <xen/xenbus.h>
#include <xen/interface/event_channel.h>
#include <xen/interface/io/usbif.h>
--- head.orig/drivers/xen/usbfront/xenbus.c 2012-03-12 13:33:24.000000000 +0100
+++ head/drivers/xen/usbfront/xenbus.c 2012-03-12 13:38:43.000000000 +0100
@@ -389,7 +389,7 @@ static int __init usbfront_init(void)
return -ENODEV;
xenhcd_urbp_cachep = kmem_cache_create("xenhcd_urb_priv",
- sizeof(struct urb_priv), 0, 0, NULL, NULL);
+ sizeof(struct urb_priv), 0, 0, NULL);
if (!xenhcd_urbp_cachep) {
printk(KERN_ERR "usbfront failed to create kmem cache\n");
return -ENOMEM;
--- head.orig/drivers/xen/util.c 2007-07-10 09:42:30.000000000 +0200
+++ head/drivers/xen/util.c 2011-01-31 17:49:31.000000000 +0100
@@ -1,8 +1,5 @@
-#include <linux/mm.h>
+#include <linux/err.h>
#include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <asm/uaccess.h>
#include <xen/driver_util.h>
struct class *get_xen_class(void)
@@ -21,45 +18,3 @@ struct class *get_xen_class(void)
return xen_class;
}
EXPORT_SYMBOL_GPL(get_xen_class);
-
-#ifdef CONFIG_X86
-static int f(pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
-{
- /* apply_to_page_range() does all the hard work. */
- return 0;
-}
-
-struct vm_struct *alloc_vm_area(unsigned long size)
-{
- struct vm_struct *area;
-
- area = get_vm_area(size, VM_IOREMAP);
- if (area == NULL)
- return NULL;
-
- /*
- * This ensures that page tables are constructed for this region
- * of kernel virtual address space and mapped into init_mm.
- */
- if (apply_to_page_range(&init_mm, (unsigned long)area->addr,
- area->size, f, NULL)) {
- free_vm_area(area);
- return NULL;
- }
-
- /* Map page directories into every address space. */
- vmalloc_sync_all();
-
- return area;
-}
-EXPORT_SYMBOL_GPL(alloc_vm_area);
-
-void free_vm_area(struct vm_struct *area)
-{
- struct vm_struct *ret;
- ret = remove_vm_area(area->addr);
- BUG_ON(ret != area);
- kfree(area);
-}
-EXPORT_SYMBOL_GPL(free_vm_area);
-#endif /* CONFIG_X86 */
--- head.orig/drivers/xen/xenbus/xenbus_backend_client.c 2011-04-11 13:43:15.000000000 +0200
+++ head/drivers/xen/xenbus/xenbus_backend_client.c 2011-01-31 17:49:31.000000000 +0100
@@ -32,9 +32,9 @@
#include <linux/err.h>
#include <linux/delay.h>
+#include <linux/vmalloc.h>
#include <xen/gnttab.h>
#include <xen/xenbus.h>
-#include <xen/driver_util.h>
/* Based on Rusty Russell's skeleton driver's map_page */
struct vm_struct *xenbus_map_ring_valloc(struct xenbus_device *dev, grant_ref_t gnt_ref)
--- head.orig/drivers/xen/xenbus/xenbus_client.c 2011-11-28 09:44:25.000000000 +0100
+++ head/drivers/xen/xenbus/xenbus_client.c 2011-06-30 16:10:27.000000000 +0200
@@ -30,19 +30,25 @@
* IN THE SOFTWARE.
*/
+#if defined(CONFIG_XEN) || defined(MODULE)
#include <linux/slab.h>
#include <xen/evtchn.h>
#include <xen/gnttab.h>
+#else
+#include <linux/types.h>
+#include <linux/vmalloc.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/event_channel.h>
+#include <xen/events.h>
+#include <xen/grant_table.h>
+#endif
#include <xen/xenbus.h>
-#include <xen/driver_util.h>
#ifdef HAVE_XEN_PLATFORM_COMPAT_H
#include <xen/platform-compat.h>
#endif
-#define DPRINTK(fmt, args...) \
- pr_debug("xenbus_client (%s:%d) " fmt ".\n", __FUNCTION__, __LINE__, ##args)
-
const char *xenbus_strstate(enum xenbus_state state)
{
static const char *const name[] = {
@@ -60,6 +66,20 @@ const char *xenbus_strstate(enum xenbus_
}
EXPORT_SYMBOL_GPL(xenbus_strstate);
+/**
+ * xenbus_watch_path - register a watch
+ * @dev: xenbus device
+ * @path: path to watch
+ * @watch: watch to register
+ * @callback: callback to register
+ *
+ * Register a @watch on the given path, using the given xenbus_watch structure
+ * for storage, and the given @callback function as the callback. Return 0 on
+ * success, or -errno on error. On success, the given @path will be saved as
+ * @watch->node, and remains the caller's to free. On error, @watch->node will
+ * be NULL, the device will switch to %XenbusStateClosing, and the error will
+ * be saved in the store.
+ */
int xenbus_watch_path(struct xenbus_device *dev, const char *path,
struct xenbus_watch *watch,
void (*callback)(struct xenbus_watch *,
@@ -83,6 +103,7 @@ int xenbus_watch_path(struct xenbus_devi
EXPORT_SYMBOL_GPL(xenbus_watch_path);
+#if defined(CONFIG_XEN) || defined(MODULE)
int xenbus_watch_path2(struct xenbus_device *dev, const char *path,
const char *path2, struct xenbus_watch *watch,
void (*callback)(struct xenbus_watch *,
@@ -101,8 +122,60 @@ int xenbus_watch_path2(struct xenbus_dev
return err;
}
EXPORT_SYMBOL_GPL(xenbus_watch_path2);
+#else
+/**
+ * xenbus_watch_pathfmt - register a watch on a sprintf-formatted path
+ * @dev: xenbus device
+ * @watch: watch to register
+ * @callback: callback to register
+ * @pathfmt: format of path to watch
+ *
+ * Register a watch on the given @path, using the given xenbus_watch
+ * structure for storage, and the given @callback function as the callback.
+ * Return 0 on success, or -errno on error. On success, the watched path
+ * (@path/@path2) will be saved as @watch->node, and becomes the caller's to
+ * kfree(). On error, watch->node will be NULL, so the caller has nothing to
+ * free, the device will switch to %XenbusStateClosing, and the error will be
+ * saved in the store.
+ */
+int xenbus_watch_pathfmt(struct xenbus_device *dev,
+ struct xenbus_watch *watch,
+ void (*callback)(struct xenbus_watch *,
+ const char **, unsigned int),
+ const char *pathfmt, ...)
+{
+ int err;
+ va_list ap;
+ char *path;
+
+ va_start(ap, pathfmt);
+ path = kvasprintf(GFP_KERNEL, pathfmt, ap);
+ va_end(ap);
+
+ if (!path) {
+ xenbus_dev_fatal(dev, -ENOMEM, "allocating path for watch");
+ return -ENOMEM;
+ }
+ err = xenbus_watch_path(dev, path, watch, callback);
+
+ if (err)
+ kfree(path);
+ return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_watch_pathfmt);
+#endif
+/**
+ * xenbus_switch_state
+ * @dev: xenbus device
+ * @xbt: transaction handle
+ * @state: new state
+ *
+ * Advertise in the store a change of the given driver to the given new_state.
+ * Return 0 on success, or -errno on error. On error, the device will switch
+ * to XenbusStateClosing, and the error will be saved in the store.
+ */
int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state state)
{
/* We check whether the state is currently set to the given value, and
@@ -161,8 +234,8 @@ static char *error_path(struct xenbus_de
}
-void _dev_error(struct xenbus_device *dev, int err, const char *fmt,
- va_list ap)
+static void _dev_error(struct xenbus_device *dev, int err,
+ const char *fmt, va_list ap)
{
int ret;
unsigned int len;
@@ -183,14 +256,16 @@ void _dev_error(struct xenbus_device *de
path_buffer = error_path(dev);
if (path_buffer == NULL) {
- printk("xenbus: failed to write error node for %s (%s)\n",
- dev->nodename, printf_buffer);
+ dev_err(&dev->dev,
+ "xenbus: failed to write error node for %s (%s)\n",
+ dev->nodename, printf_buffer);
goto fail;
}
if (xenbus_write(XBT_NIL, path_buffer, "error", printf_buffer) != 0) {
- printk("xenbus: failed to write error node for %s (%s)\n",
- dev->nodename, printf_buffer);
+ dev_err(&dev->dev,
+ "xenbus: failed to write error node for %s (%s)\n",
+ dev->nodename, printf_buffer);
goto fail;
}
@@ -200,6 +275,15 @@ fail:
}
+/**
+ * xenbus_dev_error
+ * @dev: xenbus device
+ * @err: error to report
+ * @fmt: error message format
+ *
+ * Report the given negative errno into the store, along with the given
+ * formatted message.
+ */
void xenbus_dev_error(struct xenbus_device *dev, int err, const char *fmt, ...)
{
va_list ap;
@@ -211,6 +295,16 @@ void xenbus_dev_error(struct xenbus_devi
EXPORT_SYMBOL_GPL(xenbus_dev_error);
+/**
+ * xenbus_dev_fatal
+ * @dev: xenbus device
+ * @err: error to report
+ * @fmt: error message format
+ *
+ * Equivalent to xenbus_dev_error(dev, err, fmt, args), followed by
+ * xenbus_switch_state(dev, NULL, XenbusStateClosing) to schedule an orderly
+ * closedown of this driver and its peer.
+ */
void xenbus_dev_fatal(struct xenbus_device *dev, int err, const char *fmt, ...)
{
va_list ap;
@@ -224,6 +318,15 @@ void xenbus_dev_fatal(struct xenbus_devi
EXPORT_SYMBOL_GPL(xenbus_dev_fatal);
+/**
+ * xenbus_grant_ring
+ * @dev: xenbus device
+ * @ring_mfn: mfn of ring to grant
+ *
+ * Grant access to the given @ring_mfn to the peer of the given device. Return
+ * 0 on success, or -errno on error. On error, the device will switch to
+ * XenbusStateClosing, and the error will be saved in the store.
+ */
int xenbus_grant_ring(struct xenbus_device *dev, unsigned long ring_mfn)
{
int err = gnttab_grant_foreign_access(dev->otherend_id, ring_mfn, 0);
@@ -234,6 +337,12 @@ int xenbus_grant_ring(struct xenbus_devi
EXPORT_SYMBOL_GPL(xenbus_grant_ring);
+/**
+ * Allocate an event channel for the given xenbus_device, assigning the newly
+ * created local port to *port. Return 0 on success, or -errno on error. On
+ * error, the device will switch to XenbusStateClosing, and the error will be
+ * saved in the store.
+ */
int xenbus_alloc_evtchn(struct xenbus_device *dev, int *port)
{
struct evtchn_alloc_unbound alloc_unbound;
@@ -254,6 +363,38 @@ int xenbus_alloc_evtchn(struct xenbus_de
EXPORT_SYMBOL_GPL(xenbus_alloc_evtchn);
+#if 0 /* !defined(CONFIG_XEN) && !defined(MODULE) */
+/**
+ * Bind to an existing interdomain event channel in another domain. Returns 0
+ * on success and stores the local port in *port. On error, returns -errno,
+ * switches the device to XenbusStateClosing, and saves the error in XenStore.
+ */
+int xenbus_bind_evtchn(struct xenbus_device *dev, int remote_port, int *port)
+{
+ struct evtchn_bind_interdomain bind_interdomain;
+ int err;
+
+ bind_interdomain.remote_dom = dev->otherend_id;
+ bind_interdomain.remote_port = remote_port;
+
+ err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
+ &bind_interdomain);
+ if (err)
+ xenbus_dev_fatal(dev, err,
+ "binding to event channel %d from domain %d",
+ remote_port, dev->otherend_id);
+ else
+ *port = bind_interdomain.local_port;
+
+ return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_bind_evtchn);
+#endif
+
+
+/**
+ * Free an existing event channel. Returns 0 on success or -errno on error.
+ */
int xenbus_free_evtchn(struct xenbus_device *dev, int port)
{
struct evtchn_close close;
@@ -270,6 +411,191 @@ int xenbus_free_evtchn(struct xenbus_dev
EXPORT_SYMBOL_GPL(xenbus_free_evtchn);
+#if 0 /* !defined(CONFIG_XEN) && !defined(MODULE) */
+/**
+ * xenbus_map_ring_valloc
+ * @dev: xenbus device
+ * @gnt_ref: grant reference
+ * @vaddr: pointer to address to be filled out by mapping
+ *
+ * Based on Rusty Russell's skeleton driver's map_page.
+ * Map a page of memory into this domain from another domain's grant table.
+ * xenbus_map_ring_valloc allocates a page of virtual address space, maps the
+ * page to that address, and sets *vaddr to that address.
+ * Returns 0 on success, and GNTST_* (see xen/include/interface/grant_table.h)
+ * or -ENOMEM on error. If an error is returned, device will switch to
+ * XenbusStateClosing and the error message will be saved in XenStore.
+ */
+int xenbus_map_ring_valloc(struct xenbus_device *dev, grant_ref_t gnt_ref, void **vaddr)
+{
+ struct gnttab_map_grant_ref op = {
+ .flags = GNTMAP_host_map,
+ .ref = gnt_ref,
+ .dom = dev->otherend_id,
+ };
+ struct vm_struct *area;
+
+ *vaddr = NULL;
+
+ area = alloc_vm_area(PAGE_SIZE);
+ if (!area)
+ return -ENOMEM;
+
+ op.host_addr = (unsigned long)area->addr;
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
+ BUG();
+
+ if (op.status != GNTST_okay) {
+ free_vm_area(area);
+ xenbus_dev_fatal(dev, op.status,
+ "mapping in shared page %d from domain %d",
+ gnt_ref, dev->otherend_id);
+ return op.status;
+ }
+
+ /* Stuff the handle in an unused field */
+ area->phys_addr = (unsigned long)op.handle;
+
+ *vaddr = area->addr;
+ return 0;
+}
+EXPORT_SYMBOL_GPL(xenbus_map_ring_valloc);
+
+
+/**
+ * xenbus_map_ring
+ * @dev: xenbus device
+ * @gnt_ref: grant reference
+ * @handle: pointer to grant handle to be filled
+ * @vaddr: address to be mapped to
+ *
+ * Map a page of memory into this domain from another domain's grant table.
+ * xenbus_map_ring does not allocate the virtual address space (you must do
+ * this yourself!). It only maps in the page to the specified address.
+ * Returns 0 on success, and GNTST_* (see xen/include/interface/grant_table.h)
+ * or -ENOMEM on error. If an error is returned, device will switch to
+ * XenbusStateClosing and the error message will be saved in XenStore.
+ */
+int xenbus_map_ring(struct xenbus_device *dev, grant_ref_t gnt_ref,
+ grant_handle_t *handle, void *vaddr)
+{
+ struct gnttab_map_grant_ref op = {
+ .host_addr = (unsigned long)vaddr,
+ .flags = GNTMAP_host_map,
+ .ref = gnt_ref,
+ .dom = dev->otherend_id,
+ };
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
+ BUG();
+
+ if (op.status != GNTST_okay) {
+ xenbus_dev_fatal(dev, op.status,
+ "mapping in shared page %d from domain %d",
+ gnt_ref, dev->otherend_id);
+ } else
+ *handle = op.handle;
+
+ return op.status;
+}
+EXPORT_SYMBOL_GPL(xenbus_map_ring);
+
+
+/**
+ * xenbus_unmap_ring_vfree
+ * @dev: xenbus device
+ * @vaddr: addr to unmap
+ *
+ * Based on Rusty Russell's skeleton driver's unmap_page.
+ * Unmap a page of memory in this domain that was imported from another domain.
+ * Use xenbus_unmap_ring_vfree if you mapped in your memory with
+ * xenbus_map_ring_valloc (it will free the virtual address space).
+ * Returns 0 on success and returns GNTST_* on error
+ * (see xen/include/interface/grant_table.h).
+ */
+int xenbus_unmap_ring_vfree(struct xenbus_device *dev, void *vaddr)
+{
+ struct vm_struct *area;
+ struct gnttab_unmap_grant_ref op = {
+ .host_addr = (unsigned long)vaddr,
+ };
+
+ /* It'd be nice if linux/vmalloc.h provided a find_vm_area(void *addr)
+ * method so that we don't have to muck with vmalloc internals here.
+ * We could force the user to hang on to their struct vm_struct from
+ * xenbus_map_ring_valloc, but these 6 lines considerably simplify
+ * this API.
+ */
+ read_lock(&vmlist_lock);
+ for (area = vmlist; area != NULL; area = area->next) {
+ if (area->addr == vaddr)
+ break;
+ }
+ read_unlock(&vmlist_lock);
+
+ if (!area) {
+ xenbus_dev_error(dev, -ENOENT,
+ "can't find mapped virtual address %p", vaddr);
+ return GNTST_bad_virt_addr;
+ }
+
+ op.handle = (grant_handle_t)area->phys_addr;
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
+ BUG();
+
+ if (op.status == GNTST_okay)
+ free_vm_area(area);
+ else
+ xenbus_dev_error(dev, op.status,
+ "unmapping page at handle %d error %d",
+ (int16_t)area->phys_addr, op.status);
+
+ return op.status;
+}
+EXPORT_SYMBOL_GPL(xenbus_unmap_ring_vfree);
+
+
+/**
+ * xenbus_unmap_ring
+ * @dev: xenbus device
+ * @handle: grant handle
+ * @vaddr: addr to unmap
+ *
+ * Unmap a page of memory in this domain that was imported from another domain.
+ * Returns 0 on success and returns GNTST_* on error
+ * (see xen/include/interface/grant_table.h).
+ */
+int xenbus_unmap_ring(struct xenbus_device *dev,
+ grant_handle_t handle, void *vaddr)
+{
+ struct gnttab_unmap_grant_ref op = {
+ .host_addr = (unsigned long)vaddr,
+ .handle = handle,
+ };
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
+ BUG();
+
+ if (op.status != GNTST_okay)
+ xenbus_dev_error(dev, op.status,
+ "unmapping page at handle %d error %d",
+ handle, op.status);
+
+ return op.status;
+}
+EXPORT_SYMBOL_GPL(xenbus_unmap_ring);
+#endif
+
+
+/**
+ * xenbus_read_driver_state
+ * @path: path for driver
+ *
+ * Return the state of the driver rooted at the given store path, or
+ * XenbusStateUnknown if no state can be read.
+ */
enum xenbus_state xenbus_read_driver_state(const char *path)
{
int result;
--- head.orig/drivers/xen/xenbus/xenbus_comms.c 2011-01-31 17:32:16.000000000 +0100
+++ head/drivers/xen/xenbus/xenbus_comms.c 2011-04-11 14:52:53.000000000 +0200
@@ -34,12 +34,15 @@
#include <linux/interrupt.h>
#include <linux/sched.h>
#include <linux/err.h>
-#include <linux/ptrace.h>
-#include <linux/workqueue.h>
-#include <xen/evtchn.h>
#include <xen/xenbus.h>
-
+#if defined(CONFIG_XEN) || defined(MODULE)
+#include <xen/evtchn.h>
#include <asm/hypervisor.h>
+#else
+#include <asm/xen/hypervisor.h>
+#include <xen/events.h>
+#include <xen/page.h>
+#endif
#include "xenbus_comms.h"
@@ -49,7 +52,6 @@
static int xenbus_irq;
-extern void xenbus_probe(struct work_struct *);
static DECLARE_WORK(probe_work, xenbus_probe);
static DECLARE_WAIT_QUEUE_HEAD(xb_waitq);
@@ -112,6 +114,13 @@ static const void *get_input_chunk(XENST
return buf + MASK_XENSTORE_IDX(cons);
}
+/**
+ * xb_write - low level write
+ * @data: buffer to send
+ * @len: length of buffer
+ *
+ * Returns 0 on success, error otherwise.
+ */
int xb_write(const void *data, unsigned len)
{
struct xenstore_domain_interface *intf = xen_store_interface;
@@ -220,7 +229,9 @@ int xb_read(void *data, unsigned len)
return 0;
}
-/* Set up interrupt handler off store event channel. */
+/**
+ * xb_init_comms - Set up interrupt handler off store event channel.
+ */
int xb_init_comms(void)
{
struct xenstore_domain_interface *intf = xen_store_interface;
@@ -242,7 +253,11 @@ int xb_init_comms(void)
if (xenbus_irq)
unbind_from_irqhandler(xenbus_irq, &xb_waitq);
+#if defined(CONFIG_XEN) || defined(MODULE)
err = bind_caller_port_to_irqhandler(
+#else
+ err = bind_evtchn_to_irqhandler(
+#endif
xen_store_evtchn, wake_waiting,
0, "xenbus", &xb_waitq);
if (err <= 0) {
--- head.orig/drivers/xen/xenbus/xenbus_dev.c 2012-02-16 11:09:01.000000000 +0100
+++ head/drivers/xen/xenbus/xenbus_dev.c 2012-02-16 12:30:22.000000000 +0100
@@ -493,7 +493,13 @@ static const struct file_operations xenb
#endif
};
-int xenbus_dev_init(void)
+int
+#ifndef MODULE
+__init
+#else
+__devinit
+#endif
+xenbus_dev_init(void)
{
xenbus_dev_intf = create_xen_proc_entry("xenbus", 0400);
if (xenbus_dev_intf)
--- head.orig/drivers/xen/xenbus/xenbus_probe.c 2012-03-12 13:31:47.000000000 +0100
+++ head/drivers/xen/xenbus/xenbus_probe.c 2012-03-12 13:39:04.000000000 +0100
@@ -43,21 +43,26 @@
#include <linux/mm.h>
#include <linux/notifier.h>
#include <linux/mutex.h>
-#include <linux/module.h>
-#include <xen/gnttab.h>
+#include <linux/io.h>
-#include <asm/io.h>
#include <asm/page.h>
-#include <asm/maddr.h>
#include <asm/pgtable.h>
+#if defined(CONFIG_XEN) || defined(MODULE)
#include <asm/hypervisor.h>
#include <xen/xenbus.h>
#include <xen/xen_proc.h>
#include <xen/evtchn.h>
#include <xen/features.h>
+#include <xen/gnttab.h>
#ifdef MODULE
#include <xen/hvm.h>
#endif
+#else
+#include <asm/xen/hypervisor.h>
+#include <xen/xenbus.h>
+#include <xen/events.h>
+#include <xen/page.h>
+#endif
#include "xenbus_comms.h"
#include "xenbus_probe.h"
@@ -170,7 +175,7 @@ static int read_backend_details(struct x
return read_otherend_details(xendev, "backend-id", "backend");
}
-#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16)
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16) && (defined(CONFIG_XEN) || defined(MODULE))
static int xenbus_uevent_frontend(struct device *dev, char **envp,
int num_envp, char *buffer, int buffer_size)
{
@@ -209,12 +214,16 @@ static struct xen_bus_type xenbus_fronte
.probe = xenbus_dev_probe,
.remove = xenbus_dev_remove,
.shutdown = xenbus_dev_shutdown,
+#if defined(CONFIG_XEN) || defined(MODULE)
.uevent = xenbus_uevent_frontend,
#endif
+#endif
},
+#if defined(CONFIG_XEN) || defined(MODULE)
.dev = {
.bus_id = "xen",
},
+#endif
};
static void otherend_changed(struct xenbus_watch *watch,
@@ -230,14 +239,15 @@ static void otherend_changed(struct xenb
if (!dev->otherend ||
strncmp(dev->otherend, vec[XS_WATCH_PATH],
strlen(dev->otherend))) {
- DPRINTK("Ignoring watch at %s", vec[XS_WATCH_PATH]);
+ dev_dbg(&dev->dev, "Ignoring watch at %s", vec[XS_WATCH_PATH]);
return;
}
state = xenbus_read_driver_state(dev->otherend);
- DPRINTK("state is %d (%s), %s, %s", state, xenbus_strstate(state),
- dev->otherend_watch.node, vec[XS_WATCH_PATH]);
+ dev_dbg(&dev->dev, "state is %d (%s), %s, %s",
+ state, xenbus_strstate(state), dev->otherend_watch.node,
+ vec[XS_WATCH_PATH]);
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,16)
/*
@@ -274,8 +284,13 @@ static int talk_to_otherend(struct xenbu
static int watch_otherend(struct xenbus_device *dev)
{
+#if defined(CONFIG_XEN) || defined(MODULE)
return xenbus_watch_path2(dev, dev->otherend, "state",
&dev->otherend_watch, otherend_changed);
+#else
+ return xenbus_watch_pathfmt(dev, &dev->otherend_watch, otherend_changed,
+ "%s/%s", dev->otherend, "state");
+#endif
}
@@ -301,9 +316,9 @@ int xenbus_dev_probe(struct device *_dev
err = talk_to_otherend(dev);
if (err) {
- printk(KERN_WARNING
- "xenbus_probe: talk_to_otherend on %s failed.\n",
- dev->nodename);
+ dev_warn(&dev->dev,
+ "xenbus_probe: talk_to_otherend on %s failed.\n",
+ dev->nodename);
return err;
}
@@ -313,9 +328,9 @@ int xenbus_dev_probe(struct device *_dev
err = watch_otherend(dev);
if (err) {
- printk(KERN_WARNING
- "xenbus_probe: watch_otherend on %s failed.\n",
- dev->nodename);
+ dev_warn(&dev->dev,
+ "xenbus_probe: watch_otherend on %s failed.\n",
+ dev->nodename);
return err;
}
@@ -361,8 +376,8 @@ static void xenbus_dev_shutdown(struct d
get_device(&dev->dev);
if (dev->state != XenbusStateConnected) {
- printk("%s: %s: %s != Connected, skipping\n", __FUNCTION__,
- dev->nodename, xenbus_strstate(dev->state));
+ dev_info(&dev->dev, "%s: %s: %s != Connected, skipping\n", __FUNCTION__,
+ dev->nodename, xenbus_strstate(dev->state));
goto out;
}
xenbus_switch_state(dev, XenbusStateClosing);
@@ -372,7 +387,8 @@ static void xenbus_dev_shutdown(struct d
timeout = wait_for_completion_timeout(&dev->down, timeout);
if (!timeout)
- printk("%s: %s timeout closing device\n", __FUNCTION__, dev->nodename);
+ dev_info(&dev->dev, "%s: %s timeout closing device\n",
+ __FUNCTION__, dev->nodename);
out:
put_device(&dev->dev);
}
@@ -548,7 +564,9 @@ int xenbus_probe_node(struct xen_bus_typ
xendev->devicetype = tmpstring;
init_completion(&xendev->down);
+#if defined(CONFIG_XEN) || defined(MODULE)
xendev->dev.parent = &bus->dev;
+#endif
xendev->dev.bus = &bus->bus;
xendev->dev.release = xenbus_dev_release;
@@ -563,15 +581,16 @@ int xenbus_probe_node(struct xen_bus_typ
err = device_create_file(&xendev->dev, &dev_attr_nodename);
if (err)
- goto unregister;
+ goto fail_unregister;
+
err = device_create_file(&xendev->dev, &dev_attr_devtype);
if (err)
- goto unregister;
+ goto fail_remove_file;
return 0;
-unregister:
+fail_remove_file:
device_remove_file(&xendev->dev, &dev_attr_nodename);
- device_remove_file(&xendev->dev, &dev_attr_devtype);
+fail_unregister:
device_unregister(&xendev->dev);
fail:
kfree(xendev);
@@ -587,7 +606,8 @@ static int xenbus_probe_frontend(const c
if (!strcmp(type, "console"))
return 0;
- nodename = kasprintf(GFP_KERNEL, "%s/%s/%s", xenbus_frontend.root, type, name);
+ nodename = kasprintf(GFP_KERNEL, "%s/%s/%s",
+ xenbus_frontend.root, type, name);
if (!nodename)
return -ENOMEM;
@@ -665,7 +685,7 @@ static int strsep_len(const char *str, c
return (len == 0) ? i : -ERANGE;
}
-void dev_changed(const char *node, struct xen_bus_type *bus)
+void xenbus_dev_changed(const char *node, struct xen_bus_type *bus)
{
int exists, rootlen;
struct xenbus_device *dev;
@@ -673,7 +693,7 @@ void dev_changed(const char *node, struc
const char *p, *root;
if (bus->error || char_count(node, '/') < 2)
- return;
+ return;
exists = xenbus_exists(XBT_NIL, node, "");
if (!exists) {
@@ -707,7 +727,7 @@ static void frontend_changed(struct xenb
{
DPRINTK("");
- dev_changed(vec[XS_WATCH_PATH], &xenbus_frontend);
+ xenbus_dev_changed(vec[XS_WATCH_PATH], &xenbus_frontend);
}
/* We watch for devices appearing and vanishing. */
@@ -1119,10 +1139,18 @@ fail0:
#endif
#endif /* CONFIG_XEN_PRIVILEGED_GUEST */
-static int xenbus_probe_init(void)
+static int
+#ifndef MODULE
+__init
+#else
+__devinit
+#endif
+xenbus_probe_init(void)
{
int err = 0;
+#if defined(CONFIG_XEN) || defined(MODULE)
unsigned long page = 0;
+#endif
DPRINTK("");
@@ -1141,6 +1169,7 @@ static int xenbus_probe_init(void)
* Domain0 doesn't have a store_evtchn or store_mfn yet.
*/
if (is_initial_xendomain()) {
+#if defined(CONFIG_XEN) || defined(MODULE)
struct evtchn_alloc_unbound alloc_unbound;
/* Allocate Xenstore page */
@@ -1179,10 +1208,13 @@ static int xenbus_probe_init(void)
if (xsd_port_intf)
xsd_port_intf->read_proc = xsd_port_read;
#endif
+#else
+ /* dom0 not yet supported */
+#endif
xen_store_interface = mfn_to_virt(xen_store_mfn);
} else {
atomic_set(&xenbus_xsd_state, XENBUS_XSD_FOREIGN_READY);
-#ifdef CONFIG_XEN
+#ifndef MODULE
xen_store_evtchn = xen_start_info->store_evtchn;
xen_store_mfn = xen_start_info->store_mfn;
xen_store_interface = mfn_to_virt(xen_store_mfn);
@@ -1198,7 +1230,9 @@ static int xenbus_probe_init(void)
goto err;
}
+#if defined(CONFIG_XEN) || defined(MODULE)
xenbus_dev_init();
+#endif
/* Initialize the interface to xenstore. */
err = xs_init();
@@ -1208,6 +1242,7 @@ static int xenbus_probe_init(void)
goto err;
}
+#if defined(CONFIG_XEN) || defined(MODULE)
/* Register ourselves with the kernel device subsystem */
if (!xenbus_frontend.error) {
xenbus_frontend.error = device_register(&xenbus_frontend.dev);
@@ -1218,6 +1253,7 @@ static int xenbus_probe_init(void)
xenbus_frontend.error);
}
}
+#endif
xenbus_backend_device_register();
if (!is_initial_xendomain())
@@ -1232,16 +1268,22 @@ static int xenbus_probe_init(void)
* registered.
*/
+#if defined(CONFIG_XEN) || defined(MODULE)
if (page != 0)
free_page(page);
+#endif
return err;
}
-#ifdef CONFIG_XEN
+#ifndef MODULE
postcore_initcall(xenbus_probe_init);
+#ifdef CONFIG_XEN
MODULE_LICENSE("Dual BSD/GPL");
#else
-int xenbus_init(void)
+MODULE_LICENSE("GPL");
+#endif
+#else
+int __devinit xenbus_init(void)
{
return xenbus_probe_init();
}
--- head.orig/drivers/xen/xenbus/xenbus_probe.h 2011-09-07 15:18:14.000000000 +0200
+++ head/drivers/xen/xenbus/xenbus_probe.h 2011-12-21 10:50:52.000000000 +0100
@@ -61,7 +61,9 @@ struct xen_bus_type {
int (*get_bus_id)(char bus_id[XEN_BUS_ID_SIZE], const char *nodename);
int (*probe)(const char *type, const char *dir);
struct bus_type bus;
+#if defined(CONFIG_XEN) || defined(MODULE)
struct device dev;
+#endif
};
extern int xenbus_match(struct device *_dev, struct device_driver *_drv);
@@ -74,6 +76,6 @@ extern int xenbus_probe_node(struct xen_
const char *nodename);
extern int xenbus_probe_devices(struct xen_bus_type *bus);
-extern void dev_changed(const char *node, struct xen_bus_type *bus);
+extern void xenbus_dev_changed(const char *node, struct xen_bus_type *bus);
#endif
--- head.orig/drivers/xen/xenbus/xenbus_probe_backend.c 2011-09-07 15:18:19.000000000 +0200
+++ head/drivers/xen/xenbus/xenbus_probe_backend.c 2011-12-21 10:51:23.000000000 +0100
@@ -234,7 +234,7 @@ static void backend_changed(struct xenbu
{
DPRINTK("");
- dev_changed(vec[XS_WATCH_PATH], &xenbus_backend);
+ xenbus_dev_changed(vec[XS_WATCH_PATH], &xenbus_backend);
}
static struct xenbus_watch be_watch = {
--- head.orig/drivers/xen/xenbus/xenbus_xs.c 2011-06-30 16:06:45.000000000 +0200
+++ head/drivers/xen/xenbus/xenbus_xs.c 2012-01-20 14:47:34.000000000 +0100
@@ -626,7 +626,7 @@ static struct xenbus_watch *find_watch(c
static void xs_reset_watches(void)
{
-#ifndef CONFIG_XEN
+#ifdef MODULE
int err, supported = 0;
err = xenbus_scanf(XBT_NIL, "control",
@@ -677,7 +677,9 @@ void unregister_xenbus_watch(struct xenb
char token[sizeof(watch) * 2 + 1];
int err;
+#if defined(CONFIG_XEN) || defined(MODULE)
BUG_ON(watch->flags & XBWF_new_thread);
+#endif
sprintf(token, "%lX", (long)watch);
@@ -696,6 +698,11 @@ void unregister_xenbus_watch(struct xenb
up_read(&xs_state.watch_mutex);
+ /* Make sure there are no callbacks running currently (unless
+ its us) */
+ if (current->pid != xenwatch_pid)
+ mutex_lock(&xenwatch_mutex);
+
/* Cancel pending watch events. */
spin_lock(&watch_events_lock);
list_for_each_entry_safe(msg, tmp, &watch_events, list) {
@@ -707,11 +714,8 @@ void unregister_xenbus_watch(struct xenb
}
spin_unlock(&watch_events_lock);
- /* Flush any currently-executing callback, unless we are it. :-) */
- if (current->pid != xenwatch_pid) {
- mutex_lock(&xenwatch_mutex);
+ if (current->pid != xenwatch_pid)
mutex_unlock(&xenwatch_mutex);
- }
}
EXPORT_SYMBOL_GPL(unregister_xenbus_watch);
@@ -749,6 +753,7 @@ void xs_suspend_cancel(void)
mutex_unlock(&xs_state.transaction_mutex);
}
+#if defined(CONFIG_XEN) || defined(MODULE)
static int xenwatch_handle_callback(void *data)
{
struct xs_stored_msg *msg = data;
@@ -766,6 +771,7 @@ static int xenwatch_handle_callback(void
return 0;
}
+#endif
static int xenwatch_thread(void *unused)
{
@@ -795,6 +801,7 @@ static int xenwatch_thread(void *unused)
msg = list_entry(ent, struct xs_stored_msg, list);
+#if defined(CONFIG_XEN) || defined(MODULE)
/*
* Unlock the mutex before running an XBWF_new_thread
* handler. kthread_run can block which can deadlock
@@ -811,6 +818,15 @@ static int xenwatch_thread(void *unused)
xenwatch_handle_callback(msg);
mutex_unlock(&xenwatch_mutex);
}
+#else
+ msg->u.watch.handle->callback(
+ msg->u.watch.handle,
+ (const char **)msg->u.watch.vec,
+ msg->u.watch.vec_size);
+ mutex_unlock(&xenwatch_mutex);
+ kfree(msg->u.watch.vec);
+ kfree(msg);
+#endif
}
return 0;
--- head.orig/drivers/xen/xenoprof/xenoprofile.c 2012-02-16 12:29:42.000000000 +0100
+++ head/drivers/xen/xenoprof/xenoprofile.c 2011-01-31 17:49:31.000000000 +0100
@@ -26,7 +26,6 @@
#include <asm/pgtable.h>
#include <xen/evtchn.h>
#include <xen/xenoprof.h>
-#include <xen/driver_util.h>
#include <xen/interface/xen.h>
#include <xen/interface/xenoprof.h>
#include "../../../drivers/oprofile/cpu_buffer.h"
--- head.orig/arch/x86/include/mach-xen/asm/fixmap_32.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/fixmap_32.h 2011-01-31 17:49:31.000000000 +0100
@@ -53,6 +53,8 @@ extern unsigned long __FIXADDR_TOP;
enum fixed_addresses {
FIX_HOLE,
FIX_VDSO,
+ FIX_DBGP_BASE,
+ FIX_EARLYCON_MEM_BASE,
#ifdef CONFIG_X86_LOCAL_APIC
FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
#endif
--- head.orig/arch/x86/include/mach-xen/asm/highmem.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/highmem.h 2011-01-31 17:49:31.000000000 +0100
@@ -75,8 +75,7 @@ struct page *kmap_atomic_to_page(void *p
#define kmap_atomic_pte(page, type) \
kmap_atomic_prot(page, type, \
- test_bit(PG_pinned, &(page)->flags) \
- ? PAGE_KERNEL_RO : kmap_prot)
+ PagePinned(page) ? PAGE_KERNEL_RO : kmap_prot)
#define flush_cache_kmaps() do { } while (0)
--- head.orig/arch/x86/include/mach-xen/asm/maddr_32.h 2011-08-08 12:54:10.000000000 +0200
+++ head/arch/x86/include/mach-xen/asm/maddr_32.h 2011-01-31 17:49:31.000000000 +0100
@@ -155,6 +155,7 @@ static inline paddr_t pte_machine_to_phy
#ifdef CONFIG_X86_PAE
#define __pte_ma(x) ((pte_t) { (x), (maddr_t)(x) >> 32 } )
+extern unsigned long long __supported_pte_mask;
static inline pte_t pfn_pte_ma(unsigned long page_nr, pgprot_t pgprot)
{
pte_t pte;
--- head.orig/arch/x86/include/mach-xen/asm/mmu_context_32.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/mmu_context_32.h 2011-01-31 17:49:31.000000000 +0100
@@ -16,7 +16,7 @@ void mm_pin_all(void);
static inline void xen_activate_mm(struct mm_struct *prev,
struct mm_struct *next)
{
- if (!test_bit(PG_pinned, &virt_to_page(next->pgd)->flags))
+ if (!PagePinned(virt_to_page(next->pgd)))
mm_pin(next);
}
@@ -51,6 +51,8 @@ static inline void __prepare_arch_switch
: : "r" (0) );
}
+void leave_mm(unsigned long cpu);
+
static inline void switch_mm(struct mm_struct *prev,
struct mm_struct *next,
struct task_struct *tsk)
@@ -60,7 +62,7 @@ static inline void switch_mm(struct mm_s
if (likely(prev != next)) {
BUG_ON(!xen_feature(XENFEAT_writable_page_tables) &&
- !test_bit(PG_pinned, &virt_to_page(next->pgd)->flags));
+ !PagePinned(virt_to_page(next->pgd)));
/* stop flush ipis for the previous mm */
cpu_clear(cpu, prev->cpu_vm_mask);
--- head.orig/arch/x86/include/mach-xen/asm/pgalloc_32.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/pgalloc_32.h 2011-01-31 17:49:31.000000000 +0100
@@ -5,7 +5,7 @@
#include <linux/mm.h> /* for struct page */
#include <asm/io.h> /* for phys_to_virt and page_to_pseudophys */
-#define paravirt_alloc_pt(pfn) do { } while (0)
+#define paravirt_alloc_pt(mm, pfn) do { } while (0)
#define paravirt_alloc_pd(pfn) do { } while (0)
#define paravirt_alloc_pd(pfn) do { } while (0)
#define paravirt_alloc_pd_clone(pfn, clonepfn, start, count) do { } while (0)
@@ -14,15 +14,15 @@
#define pmd_populate_kernel(mm, pmd, pte) \
do { \
- paravirt_alloc_pt(__pa(pte) >> PAGE_SHIFT); \
+ paravirt_alloc_pt(mm, __pa(pte) >> PAGE_SHIFT); \
set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte))); \
} while (0)
#define pmd_populate(mm, pmd, pte) \
do { \
unsigned long pfn = page_to_pfn(pte); \
- paravirt_alloc_pt(pfn); \
- if (test_bit(PG_pinned, &virt_to_page((mm)->pgd)->flags)) { \
+ paravirt_alloc_pt(mm, pfn); \
+ if (PagePinned(virt_to_page((mm)->pgd))) { \
if (!PageHighMem(pte)) \
BUG_ON(HYPERVISOR_update_va_mapping( \
(unsigned long)__va(pfn << PAGE_SHIFT), \
--- head.orig/arch/x86/include/mach-xen/asm/pgtable_32.h 2011-02-07 15:38:07.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/pgtable_32.h 2011-02-07 15:38:23.000000000 +0100
@@ -26,9 +26,6 @@
#include <linux/spinlock.h>
#include <linux/sched.h>
-/* Is this pagetable pinned? */
-#define PG_pinned PG_arch_1
-
struct vm_area_struct;
/*
@@ -82,7 +79,7 @@ void paging_init(void);
* area for the same reason. ;)
*/
#define VMALLOC_OFFSET (8*1024*1024)
-#define VMALLOC_START (((unsigned long) high_memory + vmalloc_earlyreserve + \
+#define VMALLOC_START (((unsigned long) high_memory + \
2*VMALLOC_OFFSET-1) & ~(VMALLOC_OFFSET-1))
#ifdef CONFIG_HIGHMEM
# define VMALLOC_END (PKMAP_BASE-2*PAGE_SIZE)
@@ -231,8 +228,6 @@ extern unsigned long pg0[];
* The following only work if pte_present() is true.
* Undefined behaviour if not..
*/
-static inline int pte_user(pte_t pte) { return (pte).pte_low & _PAGE_USER; }
-static inline int pte_read(pte_t pte) { return (pte).pte_low & _PAGE_USER; }
static inline int pte_dirty(pte_t pte) { return (pte).pte_low & _PAGE_DIRTY; }
static inline int pte_young(pte_t pte) { return (pte).pte_low & _PAGE_ACCESSED; }
static inline int pte_write(pte_t pte) { return (pte).pte_low & _PAGE_RW; }
@@ -243,13 +238,9 @@ static inline int pte_huge(pte_t pte) {
*/
static inline int pte_file(pte_t pte) { return (pte).pte_low & _PAGE_FILE; }
-static inline pte_t pte_rdprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_USER; return pte; }
-static inline pte_t pte_exprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_USER; return pte; }
static inline pte_t pte_mkclean(pte_t pte) { (pte).pte_low &= ~_PAGE_DIRTY; return pte; }
static inline pte_t pte_mkold(pte_t pte) { (pte).pte_low &= ~_PAGE_ACCESSED; return pte; }
static inline pte_t pte_wrprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_RW; return pte; }
-static inline pte_t pte_mkread(pte_t pte) { (pte).pte_low |= _PAGE_USER; return pte; }
-static inline pte_t pte_mkexec(pte_t pte) { (pte).pte_low |= _PAGE_USER; return pte; }
static inline pte_t pte_mkdirty(pte_t pte) { (pte).pte_low |= _PAGE_DIRTY; return pte; }
static inline pte_t pte_mkyoung(pte_t pte) { (pte).pte_low |= _PAGE_ACCESSED; return pte; }
static inline pte_t pte_mkwrite(pte_t pte) { (pte).pte_low |= _PAGE_RW; return pte; }
@@ -295,22 +286,20 @@ static inline pte_t xen_local_ptep_get_a
#define ptep_set_access_flags(vma, address, ptep, entry, dirty) \
({ \
int __changed = !pte_same(*(ptep), entry); \
- if (__changed && (dirty)) \
- ptep_establish(vma, address, ptep, entry); \
+ if (__changed && (dirty)) { \
+ if ( likely((vma)->vm_mm == current->mm) ) { \
+ BUG_ON(HYPERVISOR_update_va_mapping(address, \
+ entry, \
+ (unsigned long)(vma)->vm_mm->cpu_vm_mask.bits| \
+ UVMF_INVLPG|UVMF_MULTI)); \
+ } else { \
+ xen_l1_entry_update(ptep, entry); \
+ flush_tlb_page(vma, address); \
+ } \
+ } \
__changed; \
})
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
-#define ptep_test_and_clear_dirty(vma, addr, ptep) ({ \
- int __ret = 0; \
- if (pte_dirty(*(ptep))) \
- __ret = test_and_clear_bit(_PAGE_BIT_DIRTY, \
- &(ptep)->pte_low); \
- if (__ret) \
- pte_update((vma)->vm_mm, addr, ptep); \
- __ret; \
-})
-
#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
#define ptep_test_and_clear_young(vma, addr, ptep) ({ \
int __ret = 0; \
@@ -322,44 +311,13 @@ static inline pte_t xen_local_ptep_get_a
__ret; \
})
-/*
- * Rules for using ptep_establish: the pte MUST be a user pte, and
- * must be a present->present transition.
- */
-#define __HAVE_ARCH_PTEP_ESTABLISH
-#define ptep_establish(vma, address, ptep, pteval) \
-do { \
- if ( likely((vma)->vm_mm == current->mm) ) { \
- BUG_ON(HYPERVISOR_update_va_mapping(address, \
- pteval, \
- (unsigned long)(vma)->vm_mm->cpu_vm_mask.bits| \
- UVMF_INVLPG|UVMF_MULTI)); \
- } else { \
- xen_l1_entry_update(ptep, pteval); \
- flush_tlb_page(vma, address); \
- } \
-} while (0)
-
-#define __HAVE_ARCH_PTEP_CLEAR_DIRTY_FLUSH
-#define ptep_clear_flush_dirty(vma, address, ptep) \
-({ \
- pte_t __pte = *(ptep); \
- int __dirty = pte_dirty(__pte); \
- __pte = pte_mkclean(__pte); \
- if (test_bit(PG_pinned, &virt_to_page((vma)->vm_mm->pgd)->flags)) \
- (void)ptep_set_access_flags(vma, address, ptep, __pte, __dirty); \
- else if (__dirty) \
- (ptep)->pte_low = __pte.pte_low; \
- __dirty; \
-})
-
#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
#define ptep_clear_flush_young(vma, address, ptep) \
({ \
pte_t __pte = *(ptep); \
int __young = pte_young(__pte); \
__pte = pte_mkold(__pte); \
- if (test_bit(PG_pinned, &virt_to_page((vma)->vm_mm->pgd)->flags)) \
+ if (PagePinned(virt_to_page((vma)->vm_mm->pgd))) \
(void)ptep_set_access_flags(vma, address, ptep, __pte, __young); \
else if (__young) \
(ptep)->pte_low = __pte.pte_low; \
@@ -383,7 +341,7 @@ static inline pte_t ptep_get_and_clear(s
#define ptep_get_and_clear_full(mm, addr, ptep, full) \
((full) ? ({ \
pte_t __res = *(ptep); \
- if (test_bit(PG_pinned, &virt_to_page((mm)->pgd)->flags)) \
+ if (PagePinned(virt_to_page((mm)->pgd))) \
xen_l1_entry_update(ptep, __pte(0)); \
else \
*(ptep) = __pte(0); \
--- head.orig/arch/x86/include/mach-xen/asm/pgtable-3level.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/pgtable-3level.h 2011-01-31 17:49:31.000000000 +0100
@@ -23,26 +23,11 @@
#define pud_present(pud) 1
/*
- * Is the pte executable?
- */
-static inline int pte_x(pte_t pte)
-{
- return !(__pte_val(pte) & _PAGE_NX);
-}
-
-/*
- * All present user-pages with !NX bit are user-executable:
- */
-static inline int pte_exec(pte_t pte)
-{
- return pte_user(pte) && pte_x(pte);
-}
-/*
* All present pages with !NX bit are kernel-executable:
*/
static inline int pte_exec_kernel(pte_t pte)
{
- return pte_x(pte);
+ return !(__pte_val(pte) & _PAGE_NX);
}
/* Rules for using set_pte: the pte being assigned *must* be
--- head.orig/arch/x86/include/mach-xen/asm/processor_32.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/processor_32.h 2011-01-31 17:49:31.000000000 +0100
@@ -89,7 +89,6 @@ struct cpuinfo_x86 {
#define X86_VENDOR_UMC 3
#define X86_VENDOR_NEXGEN 4
#define X86_VENDOR_CENTAUR 5
-#define X86_VENDOR_RISE 6
#define X86_VENDOR_TRANSMETA 7
#define X86_VENDOR_NSC 8
#define X86_VENDOR_NUM 9
@@ -122,6 +121,7 @@ void __init cpu_detect(struct cpuinfo_x8
extern void identify_boot_cpu(void);
extern void identify_secondary_cpu(struct cpuinfo_x86 *);
extern void print_cpu_info(struct cpuinfo_x86 *);
+extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
extern unsigned short num_cache_leaves;
@@ -171,17 +171,6 @@ static inline void clear_in_cr4 (unsigne
write_cr4(cr4);
}
-/*
- * NSC/Cyrix CPU indexed register access macros
- */
-
-#define getCx86(reg) ({ outb((reg), 0x22); inb(0x23); })
-
-#define setCx86(reg, data) do { \
- outb((reg), 0x22); \
- outb((data), 0x23); \
-} while (0)
-
/* Stop speculative execution */
static inline void sync_core(void)
{
@@ -230,6 +219,10 @@ extern int bootloader_type;
#define HAVE_ARCH_PICK_MMAP_LAYOUT
+extern void hard_disable_TSC(void);
+extern void disable_TSC(void);
+extern void hard_enable_TSC(void);
+
/*
* Size of io_bitmap.
*/
--- head.orig/arch/x86/include/mach-xen/asm/tlbflush_32.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/tlbflush_32.h 2011-01-31 17:49:31.000000000 +0100
@@ -91,7 +91,11 @@ struct tlb_state
DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
#endif /* SMP */
-#define flush_tlb_kernel_range(start, end) flush_tlb_all()
+static inline void flush_tlb_kernel_range(unsigned long start,
+ unsigned long end)
+{
+ flush_tlb_all();
+}
static inline void flush_tlb_pgtables(struct mm_struct *mm,
unsigned long start, unsigned long end)
--- head.orig/arch/x86/include/mach-xen/asm/fixmap_64.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/fixmap_64.h 2011-01-31 17:49:31.000000000 +0100
@@ -23,9 +23,9 @@
* compile time, but to set the physical address only
* in the boot process.
*
- * these 'compile-time allocated' memory buffers are
- * fixed-size 4k pages. (or larger if used with an increment
- * highger than 1) use fixmap_set(idx,phys) to associate
+ * These 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages (or larger if used with an increment
+ * higher than 1). Use set_fixmap(idx,phys) to associate
* physical memory with fixmap indices.
*
* TLB entries of such buffers will not be flushed across
@@ -36,6 +36,8 @@ enum fixed_addresses {
VSYSCALL_LAST_PAGE,
VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
VSYSCALL_HPET,
+ FIX_DBGP_BASE,
+ FIX_EARLYCON_MEM_BASE,
FIX_HPET_BASE,
#ifdef CONFIG_X86_LOCAL_APIC
FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
@@ -105,7 +107,7 @@ static __always_inline unsigned long fix
if (idx >= __end_of_fixed_addresses)
__this_fixmap_does_not_exist();
- return __fix_to_virt(idx);
+ return __fix_to_virt(idx);
}
#endif
--- head.orig/arch/x86/include/mach-xen/asm/mmu_context_64.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/mmu_context_64.h 2011-01-31 17:49:31.000000000 +0100
@@ -76,7 +76,7 @@ static inline void switch_mm(struct mm_s
if (likely(prev != next)) {
BUG_ON(!xen_feature(XENFEAT_writable_page_tables) &&
- !next->context.pinned);
+ !PagePinned(virt_to_page(next->pgd)));
/* stop flush ipis for the previous mm */
cpu_clear(cpu, prev->cpu_vm_mask);
@@ -131,7 +131,7 @@ static inline void switch_mm(struct mm_s
static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
{
- if (!next->context.pinned)
+ if (!PagePinned(virt_to_page(next->pgd)))
mm_pin(next);
switch_mm(prev, next, NULL);
}
--- head.orig/arch/x86/include/mach-xen/asm/pgalloc_64.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/pgalloc_64.h 2011-01-31 17:49:31.000000000 +0100
@@ -21,7 +21,7 @@ static inline void pmd_populate_kernel(s
static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *pte)
{
- if (unlikely((mm)->context.pinned)) {
+ if (unlikely(PagePinned(virt_to_page((mm)->pgd)))) {
BUG_ON(HYPERVISOR_update_va_mapping(
(unsigned long)__va(page_to_pfn(pte) << PAGE_SHIFT),
pfn_pte(page_to_pfn(pte), PAGE_KERNEL_RO), 0));
@@ -33,7 +33,7 @@ static inline void pmd_populate(struct m
static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd)
{
- if (unlikely((mm)->context.pinned)) {
+ if (unlikely(PagePinned(virt_to_page((mm)->pgd)))) {
BUG_ON(HYPERVISOR_update_va_mapping(
(unsigned long)pmd,
pfn_pte(virt_to_phys(pmd)>>PAGE_SHIFT,
@@ -50,7 +50,7 @@ static inline void pud_populate(struct m
*/
static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, pud_t *pud)
{
- if (unlikely((mm)->context.pinned)) {
+ if (unlikely(PagePinned(virt_to_page((mm)->pgd)))) {
BUG_ON(HYPERVISOR_update_va_mapping(
(unsigned long)pud,
pfn_pte(virt_to_phys(pud)>>PAGE_SHIFT,
--- head.orig/arch/x86/include/mach-xen/asm/pgtable_64.h 2011-02-07 15:38:08.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/pgtable_64.h 2011-02-07 15:38:18.000000000 +0100
@@ -304,7 +304,7 @@ static inline pte_t ptep_get_and_clear_f
{
if (full) {
pte_t pte = *ptep;
- if (mm->context.pinned)
+ if (PagePinned(virt_to_page(mm->pgd)))
xen_l1_entry_update(ptep, __pte(0));
else
*ptep = __pte(0);
@@ -333,21 +333,15 @@ static inline pte_t ptep_get_and_clear_f
* Undefined behaviour if not..
*/
#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
-static inline int pte_user(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
-static inline int pte_read(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
-static inline int pte_exec(pte_t pte) { return !(__pte_val(pte) & _PAGE_NX); }
static inline int pte_dirty(pte_t pte) { return __pte_val(pte) & _PAGE_DIRTY; }
static inline int pte_young(pte_t pte) { return __pte_val(pte) & _PAGE_ACCESSED; }
static inline int pte_write(pte_t pte) { return __pte_val(pte) & _PAGE_RW; }
static inline int pte_file(pte_t pte) { return __pte_val(pte) & _PAGE_FILE; }
static inline int pte_huge(pte_t pte) { return __pte_val(pte) & _PAGE_PSE; }
-static inline pte_t pte_rdprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_USER; return pte; }
-static inline pte_t pte_exprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_USER; return pte; }
static inline pte_t pte_mkclean(pte_t pte) { __pte_val(pte) &= ~_PAGE_DIRTY; return pte; }
static inline pte_t pte_mkold(pte_t pte) { __pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
static inline pte_t pte_wrprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_RW; return pte; }
-static inline pte_t pte_mkread(pte_t pte) { __pte_val(pte) |= _PAGE_USER; return pte; }
static inline pte_t pte_mkexec(pte_t pte) { __pte_val(pte) &= ~_PAGE_NX; return pte; }
static inline pte_t pte_mkdirty(pte_t pte) { __pte_val(pte) |= _PAGE_DIRTY; return pte; }
static inline pte_t pte_mkyoung(pte_t pte) { __pte_val(pte) |= _PAGE_ACCESSED; return pte; }
@@ -355,13 +349,6 @@ static inline pte_t pte_mkwrite(pte_t pt
static inline pte_t pte_mkhuge(pte_t pte) { __pte_val(pte) |= _PAGE_PSE; return pte; }
static inline pte_t pte_clrhuge(pte_t pte) { __pte_val(pte) &= ~_PAGE_PSE; return pte; }
-static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
-{
- if (!pte_dirty(*ptep))
- return 0;
- return test_and_clear_bit(_PAGE_BIT_DIRTY, &ptep->pte);
-}
-
static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
{
if (!pte_young(*ptep))
@@ -500,26 +487,13 @@ static inline pte_t pte_modify(pte_t pte
__changed; \
})
-#define __HAVE_ARCH_PTEP_CLEAR_DIRTY_FLUSH
-#define ptep_clear_flush_dirty(vma, address, ptep) \
-({ \
- pte_t __pte = *(ptep); \
- int __dirty = pte_dirty(__pte); \
- __pte = pte_mkclean(__pte); \
- if ((vma)->vm_mm->context.pinned) \
- (void)ptep_set_access_flags(vma, address, ptep, __pte, __dirty); \
- else if (__dirty) \
- set_pte(ptep, __pte); \
- __dirty; \
-})
-
#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
#define ptep_clear_flush_young(vma, address, ptep) \
({ \
pte_t __pte = *(ptep); \
int __young = pte_young(__pte); \
__pte = pte_mkold(__pte); \
- if ((vma)->vm_mm->context.pinned) \
+ if (PagePinned(virt_to_page((vma)->vm_mm->pgd))) \
(void)ptep_set_access_flags(vma, address, ptep, __pte, __young); \
else if (__young) \
set_pte(ptep, __pte); \
@@ -566,6 +540,8 @@ int xen_change_pte_range(struct mm_struc
#define arch_change_pte_range(mm, pmd, addr, end, newprot, dirty_accountable) \
xen_change_pte_range(mm, pmd, addr, end, newprot, dirty_accountable)
+pte_t *lookup_address(unsigned long addr);
+
#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
direct_remap_pfn_range(vma,vaddr,pfn,size,prot,DOMID_IO)
@@ -583,7 +559,6 @@ int xen_change_pte_range(struct mm_struc
(((o) & (1UL << (__VIRTUAL_MASK_SHIFT-1))) ? ((o) | (~__VIRTUAL_MASK)) : (o))
#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
#define __HAVE_ARCH_PTEP_CLEAR_FLUSH
--- head.orig/arch/x86/include/mach-xen/asm/processor_64.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/processor_64.h 2011-01-31 17:49:31.000000000 +0100
@@ -83,7 +83,6 @@ struct cpuinfo_x86 {
#define X86_VENDOR_UMC 3
#define X86_VENDOR_NEXGEN 4
#define X86_VENDOR_CENTAUR 5
-#define X86_VENDOR_RISE 6
#define X86_VENDOR_TRANSMETA 7
#define X86_VENDOR_NUM 8
#define X86_VENDOR_UNKNOWN 0xff
@@ -100,6 +99,7 @@ extern char ignore_irq13;
extern void identify_cpu(struct cpuinfo_x86 *);
extern void print_cpu_info(struct cpuinfo_x86 *);
+extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
extern unsigned short num_cache_leaves;
@@ -377,12 +377,10 @@ static inline void sync_core(void)
asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
}
-#define cpu_has_fpu 1
-
#define ARCH_HAS_PREFETCH
static inline void prefetch(void *x)
{
- asm volatile("prefetcht0 %0" :: "m" (*(unsigned long *)x));
+ asm volatile("prefetcht0 (%0)" :: "r" (x));
}
#define ARCH_HAS_PREFETCHW 1
@@ -400,17 +398,6 @@ static inline void prefetchw(void *x)
#define cpu_relax() rep_nop()
-/*
- * NSC/Cyrix CPU indexed register access macros
- */
-
-#define getCx86(reg) ({ outb((reg), 0x22); inb(0x23); })
-
-#define setCx86(reg, data) do { \
- outb((reg), 0x22); \
- outb((data), 0x23); \
-} while (0)
-
static inline void serialize_cpu(void)
{
__asm__ __volatile__ ("cpuid" : : : "ax", "bx", "cx", "dx");
--- head.orig/arch/x86/include/mach-xen/asm/tlbflush_64.h 2011-01-31 17:32:29.000000000 +0100
+++ head/arch/x86/include/mach-xen/asm/tlbflush_64.h 2011-01-31 17:49:31.000000000 +0100
@@ -89,7 +89,11 @@ static inline void flush_tlb_range(struc
#endif
-#define flush_tlb_kernel_range(start, end) flush_tlb_all()
+static inline void flush_tlb_kernel_range(unsigned long start,
+ unsigned long end)
+{
+ flush_tlb_all();
+}
static inline void flush_tlb_pgtables(struct mm_struct *mm,
unsigned long start, unsigned long end)
--- head.orig/arch/x86/include/asm/thread_info.h 2012-04-10 16:11:24.000000000 +0200
+++ head/arch/x86/include/asm/thread_info.h 2012-04-10 16:36:11.000000000 +0200
@@ -154,7 +154,8 @@ struct thread_info {
#define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY)
#define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW|_TIF_DEBUG)
#else
-#define _TIF_WORK_CTXSW _TIF_DEBUG
+#define _TIF_WORK_CTXSW_NEXT (_TIF_NOTSC | _TIF_DEBUG)
+#define _TIF_WORK_CTXSW_PREV (_TIF_NOTSC)
#endif
#define PREEMPT_ACTIVE 0x10000000
--- head.orig/arch/x86/include/asm/xen/interface.h 2012-06-12 15:06:23.000000000 +0200
+++ head/arch/x86/include/asm/xen/interface.h 2012-04-10 16:35:52.000000000 +0200
@@ -10,17 +10,20 @@
#define _ASM_X86_XEN_INTERFACE_H
#ifdef __XEN__
-#define __DEFINE_GUEST_HANDLE(name, type) \
+#define ___DEFINE_XEN_GUEST_HANDLE(name, type) \
typedef struct { type *p; } __guest_handle_ ## name
#else
-#define __DEFINE_GUEST_HANDLE(name, type) \
+#define ___DEFINE_XEN_GUEST_HANDLE(name, type) \
typedef type * __guest_handle_ ## name
#endif
+#define __DEFINE_XEN_GUEST_HANDLE(name, type) \
+ ___DEFINE_XEN_GUEST_HANDLE(name, type); \
+ ___DEFINE_XEN_GUEST_HANDLE(const_##name, const type)
#define DEFINE_GUEST_HANDLE_STRUCT(name) \
- __DEFINE_GUEST_HANDLE(name, struct name)
-#define DEFINE_GUEST_HANDLE(name) __DEFINE_GUEST_HANDLE(name, name)
-#define GUEST_HANDLE(name) __guest_handle_ ## name
+ __DEFINE_XEN_GUEST_HANDLE(name, struct name)
+#define DEFINE_XEN_GUEST_HANDLE(name) __DEFINE_XEN_GUEST_HANDLE(name, name)
+#define XEN_GUEST_HANDLE(name) __guest_handle_ ## name
#ifdef __XEN__
#if defined(__i386__)
@@ -47,16 +50,8 @@
#endif
#ifndef __ASSEMBLY__
-/* Guest handles for primitive C types. */
-__DEFINE_GUEST_HANDLE(uchar, unsigned char);
-__DEFINE_GUEST_HANDLE(uint, unsigned int);
-__DEFINE_GUEST_HANDLE(ulong, unsigned long);
-DEFINE_GUEST_HANDLE(char);
-DEFINE_GUEST_HANDLE(int);
-DEFINE_GUEST_HANDLE(long);
-DEFINE_GUEST_HANDLE(void);
-DEFINE_GUEST_HANDLE(uint64_t);
-DEFINE_GUEST_HANDLE(uint32_t);
+typedef unsigned long xen_pfn_t;
+typedef unsigned long xen_ulong_t;
#endif
#ifndef HYPERVISOR_VIRT_START
@@ -68,7 +63,7 @@ DEFINE_GUEST_HANDLE(uint32_t);
#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>__MACH2PHYS_SHIFT)
/* Maximum number of virtual CPUs in multi-processor guests. */
-#define MAX_VIRT_CPUS 32
+#define XEN_LEGACY_MAX_VCPUS 32
/*
* SEGMENT DESCRIPTOR TABLES
--- head.orig/include/linux/elfnote.h 2012-06-12 15:06:23.000000000 +0200
+++ head/include/linux/elfnote.h 2011-01-31 17:49:31.000000000 +0100
@@ -52,7 +52,7 @@
4484:.balign 4 ; \
.popsection ;
-#define ELFNOTE(name, type, desc) \
+#define ELFNOTE(name, type, desc...) \
ELFNOTE_START(name, type, "") \
desc ; \
ELFNOTE_END
--- head.orig/include/linux/page-flags.h 2012-04-10 16:07:40.000000000 +0200
+++ head/include/linux/page-flags.h 2012-04-10 16:36:02.000000000 +0200
@@ -125,8 +125,15 @@ enum pageflags {
PG_fscache = PG_private_2, /* page backed by cache */
/* XEN */
+#ifdef CONFIG_XEN
+ PG_pinned = PG_locked, /* Cannot alias with PG_owner_priv_1 since
+ * bad_page() checks should include this bit.
+ * Also cannot use PG_arch_1 since that now
+ * has a different purpose on x86. */
+#else
PG_pinned = PG_owner_priv_1,
PG_savepinned = PG_dirty,
+#endif
/* SLOB */
PG_slob_free = PG_private,
@@ -493,10 +500,8 @@ static inline int PageTransTail(struct p
#define __PG_COMPOUND_LOCK 0
#endif
-#if !defined(CONFIG_XEN)
+#ifndef CONFIG_XEN
# define __PG_XEN 0
-#elif defined(CONFIG_X86)
-# define __PG_XEN ((1 << PG_pinned) | (1 << PG_foreign))
#else
# define __PG_XEN (1 << PG_foreign)
#endif
--- head.orig/include/xen/driver_util.h 2007-06-12 13:14:19.000000000 +0200
+++ head/include/xen/driver_util.h 2011-01-31 17:49:31.000000000 +0100
@@ -1,14 +1,8 @@
+#ifndef __XEN_DRIVER_UTIL_H__
+#define __XEN_DRIVER_UTIL_H__
-#ifndef __ASM_XEN_DRIVER_UTIL_H__
-#define __ASM_XEN_DRIVER_UTIL_H__
-
-#include <linux/vmalloc.h>
#include <linux/device.h>
-/* Allocate/destroy a 'vmalloc' VM area. */
-extern struct vm_struct *alloc_vm_area(unsigned long size);
-extern void free_vm_area(struct vm_struct *area);
-
extern struct class *get_xen_class(void);
-#endif /* __ASM_XEN_DRIVER_UTIL_H__ */
+#endif /* __XEN_DRIVER_UTIL_H__ */
--- head.orig/include/xen/features.h 2012-06-12 15:06:23.000000000 +0200
+++ head/include/xen/features.h 2011-01-31 17:49:31.000000000 +0100
@@ -10,6 +10,7 @@
#define __XEN_FEATURES_H__
#include <xen/interface/features.h>
+#include <xen/interface/version.h>
void xen_setup_features(void);
@@ -20,4 +21,4 @@ static inline int xen_feature(int flag)
return xen_features[flag];
}
-#endif /* __ASM_XEN_FEATURES_H__ */
+#endif /* __XEN_FEATURES_H__ */
--- head.orig/include/xen/interface/arch-x86/xen.h 2012-02-16 11:09:01.000000000 +0100
+++ head/include/xen/interface/arch-x86/xen.h 2011-03-17 14:11:48.000000000 +0100
@@ -50,6 +50,9 @@
#endif
#define set_xen_guest_handle(hnd, val) set_xen_guest_handle_raw(hnd, val)
+/* Allow co-existing Linux 2.6.23+ Xen interface definitions. */
+#define DEFINE_GUEST_HANDLE_STRUCT(name) struct name
+
#if defined(__i386__)
#include "xen-x86_32.h"
#elif defined(__x86_64__)
--- head.orig/include/xen/interface/event_channel.h 2012-04-04 08:57:09.000000000 +0200
+++ head/include/xen/interface/event_channel.h 2012-04-04 10:14:18.000000000 +0200
@@ -278,6 +278,7 @@ struct evtchn_op {
struct evtchn_unmask unmask;
} u;
};
+DEFINE_GUEST_HANDLE_STRUCT(evtchn_op);
typedef struct evtchn_op evtchn_op_t;
DEFINE_XEN_GUEST_HANDLE(evtchn_op_t);
--- head.orig/include/xen/interface/io/netif.h 2011-01-31 15:14:12.000000000 +0100
+++ head/include/xen/interface/io/netif.h 2011-01-31 17:49:31.000000000 +0100
@@ -187,8 +187,22 @@ typedef struct netif_rx_response netif_r
* Generate netif ring structures and types.
*/
+#if defined(CONFIG_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
DEFINE_RING_TYPES(netif_tx, struct netif_tx_request, struct netif_tx_response);
DEFINE_RING_TYPES(netif_rx, struct netif_rx_request, struct netif_rx_response);
+#else
+#define xen_netif_tx_request netif_tx_request
+#define xen_netif_rx_request netif_rx_request
+#define xen_netif_tx_response netif_tx_response
+#define xen_netif_rx_response netif_rx_response
+DEFINE_RING_TYPES(xen_netif_tx,
+ struct xen_netif_tx_request,
+ struct xen_netif_tx_response);
+DEFINE_RING_TYPES(xen_netif_rx,
+ struct xen_netif_rx_request,
+ struct xen_netif_rx_response);
+#define xen_netif_extra_info netif_extra_info
+#endif
#define NETIF_RSP_DROPPED -2
#define NETIF_RSP_ERROR -1
--- head.orig/include/xen/interface/memory.h 2012-04-04 08:57:09.000000000 +0200
+++ head/include/xen/interface/memory.h 2012-02-16 12:30:17.000000000 +0100
@@ -94,6 +94,7 @@ struct xen_memory_reservation {
*/
domid_t domid;
};
+DEFINE_GUEST_HANDLE_STRUCT(xen_memory_reservation);
typedef struct xen_memory_reservation xen_memory_reservation_t;
DEFINE_XEN_GUEST_HANDLE(xen_memory_reservation_t);
@@ -187,6 +188,7 @@ struct xen_machphys_mfn_list {
*/
unsigned int nr_extents;
};
+DEFINE_GUEST_HANDLE_STRUCT(xen_machphys_mfn_list);
typedef struct xen_machphys_mfn_list xen_machphys_mfn_list_t;
DEFINE_XEN_GUEST_HANDLE(xen_machphys_mfn_list_t);
@@ -232,6 +234,7 @@ struct xen_add_to_physmap {
/* GPFN where the source mapping page should appear. */
xen_pfn_t gpfn;
};
+DEFINE_GUEST_HANDLE_STRUCT(xen_add_to_physmap);
typedef struct xen_add_to_physmap xen_add_to_physmap_t;
DEFINE_XEN_GUEST_HANDLE(xen_add_to_physmap_t);
--- head.orig/include/xen/interface/sched.h 2011-01-31 15:14:12.000000000 +0100
+++ head/include/xen/interface/sched.h 2011-01-31 17:49:31.000000000 +0100
@@ -67,6 +67,7 @@
struct sched_shutdown {
unsigned int reason; /* SHUTDOWN_* */
};
+DEFINE_GUEST_HANDLE_STRUCT(sched_shutdown);
typedef struct sched_shutdown sched_shutdown_t;
DEFINE_XEN_GUEST_HANDLE(sched_shutdown_t);
@@ -81,6 +82,7 @@ struct sched_poll {
unsigned int nr_ports;
uint64_t timeout;
};
+DEFINE_GUEST_HANDLE_STRUCT(sched_poll);
typedef struct sched_poll sched_poll_t;
DEFINE_XEN_GUEST_HANDLE(sched_poll_t);
--- head.orig/include/xen/interface/version.h 2011-01-31 15:14:12.000000000 +0100
+++ head/include/xen/interface/version.h 2011-01-31 17:49:31.000000000 +0100
@@ -36,6 +36,9 @@
/* arg == xen_extraversion_t. */
#define XENVER_extraversion 1
typedef char xen_extraversion_t[16];
+struct xen_extraversion {
+ xen_extraversion_t extraversion;
+};
#define XEN_EXTRAVERSION_LEN (sizeof(xen_extraversion_t))
/* arg == xen_compile_info_t. */
@@ -50,10 +53,16 @@ typedef struct xen_compile_info xen_comp
#define XENVER_capabilities 3
typedef char xen_capabilities_info_t[1024];
+struct xen_capabilities_info {
+ xen_capabilities_info_t info;
+};
#define XEN_CAPABILITIES_INFO_LEN (sizeof(xen_capabilities_info_t))
#define XENVER_changeset 4
typedef char xen_changeset_info_t[64];
+struct xen_changeset_info {
+ xen_changeset_info_t info;
+};
#define XEN_CHANGESET_INFO_LEN (sizeof(xen_changeset_info_t))
#define XENVER_platform_parameters 5
--- head.orig/include/xen/interface/xen.h 2012-04-04 08:57:09.000000000 +0200
+++ head/include/xen/interface/xen.h 2012-04-04 10:14:21.000000000 +0200
@@ -32,7 +32,9 @@
#include <asm/pvclock-abi.h>
#endif
-#if defined(__i386__) || defined(__x86_64__)
+#if defined(CONFIG_PARAVIRT_XEN) && !defined(HAVE_XEN_PLATFORM_COMPAT_H)
+#include <asm/xen/interface.h>
+#elif defined(__i386__) || defined(__x86_64__)
#include "arch-x86/xen.h"
#elif defined(__ia64__)
#include "arch-ia64.h"
@@ -121,7 +123,7 @@ DEFINE_XEN_GUEST_HANDLE(xen_pfn_t);
*/
/* New sched_op hypercall introduced in 0x00030101. */
-#if __XEN_INTERFACE_VERSION__ < 0x00030101
+#if __XEN_INTERFACE_VERSION__ < 0x00030101 || (defined(CONFIG_PARAVIRT_XEN) && !defined(HAVE_XEN_PLATFORM_COMPAT_H))
#define __HYPERVISOR_sched_op __HYPERVISOR_sched_op_compat
#else
#define __HYPERVISOR_sched_op __HYPERVISOR_sched_op_new
@@ -136,7 +138,7 @@ DEFINE_XEN_GUEST_HANDLE(xen_pfn_t);
#endif
/* New platform_op hypercall introduced in 0x00030204. */
-#if __XEN_INTERFACE_VERSION__ < 0x00030204
+#if __XEN_INTERFACE_VERSION__ < 0x00030204 || (defined(CONFIG_PARAVIRT_XEN) && !defined(HAVE_XEN_PLATFORM_COMPAT_H))
#define __HYPERVISOR_dom0_op __HYPERVISOR_platform_op
#endif
@@ -422,6 +424,7 @@ struct mmuext_op {
xen_pfn_t src_mfn;
} arg2;
};
+DEFINE_GUEST_HANDLE_STRUCT(mmuext_op);
typedef struct mmuext_op mmuext_op_t;
DEFINE_XEN_GUEST_HANDLE(mmuext_op_t);
#endif
@@ -514,6 +517,7 @@ struct mmu_update {
uint64_t ptr; /* Machine address of PTE. */
uint64_t val; /* New contents of PTE. */
};
+DEFINE_GUEST_HANDLE_STRUCT(mmu_update);
typedef struct mmu_update mmu_update_t;
DEFINE_XEN_GUEST_HANDLE(mmu_update_t);
@@ -522,9 +526,15 @@ DEFINE_XEN_GUEST_HANDLE(mmu_update_t);
* NB. The fields are natural register size for this architecture.
*/
struct multicall_entry {
- unsigned long op, result;
+ unsigned long op;
+#if !defined(CONFIG_PARAVIRT_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
+ unsigned long result;
+#else
+ long result;
+#endif
unsigned long args[6];
};
+DEFINE_GUEST_HANDLE_STRUCT(multicall_entry);
typedef struct multicall_entry multicall_entry_t;
DEFINE_XEN_GUEST_HANDLE(multicall_entry_t);
--- head.orig/include/xen/xenbus.h 2011-12-21 10:42:02.000000000 +0100
+++ head/include/xen/xenbus.h 2011-12-21 10:50:40.000000000 +0100
@@ -58,16 +58,20 @@ struct xenbus_watch
void (*callback)(struct xenbus_watch *,
const char **vec, unsigned int len);
+#if defined(CONFIG_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
/* See XBWF_ definitions below. */
unsigned long flags;
+#endif
};
+#if defined(CONFIG_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
/*
* Execute callback in its own kthread. Useful if the callback is long
* running or heavily serialised, to avoid taking out the main xenwatch thread
* for a long period of time (or even unwittingly causing a deadlock).
*/
#define XBWF_new_thread 1
+#endif
/* A xenbus device. */
struct xenbus_device {
@@ -181,6 +185,9 @@ void xs_suspend_cancel(void);
/* Used by xenbus_dev to borrow kernel's store connection. */
void *xenbus_dev_request_and_reply(struct xsd_sockmsg *msg);
+struct work_struct;
+void xenbus_probe(struct work_struct *);
+
/* Prepare for domain suspend: then resume or cancel the suspend. */
void xenbus_suspend(void);
void xenbus_resume(void);
@@ -211,6 +218,7 @@ int xenbus_watch_path(struct xenbus_devi
const char **, unsigned int));
+#if defined(CONFIG_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
/**
* Register a watch on the given path/path2, using the given xenbus_watch
* structure for storage, and the given callback function as the callback.
@@ -224,7 +232,13 @@ int xenbus_watch_path2(struct xenbus_dev
const char *path2, struct xenbus_watch *watch,
void (*callback)(struct xenbus_watch *,
const char **, unsigned int));
-
+#else
+int xenbus_watch_pathfmt(struct xenbus_device *dev, struct xenbus_watch *watch,
+ void (*callback)(struct xenbus_watch *,
+ const char **, unsigned int),
+ const char *pathfmt, ...)
+ __attribute__ ((format (printf, 4, 5)));
+#endif
/**
* Advertise in the store a change of the given driver to the given new_state.
@@ -233,7 +247,6 @@ int xenbus_watch_path2(struct xenbus_dev
*/
int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state new_state);
-
/**
* Grant access to the given ring_mfn to the peer of the given device. Return
* 0 on success, or -errno on error. On error, the device will switch to
@@ -241,7 +254,6 @@ int xenbus_switch_state(struct xenbus_de
*/
int xenbus_grant_ring(struct xenbus_device *dev, unsigned long ring_mfn);
-
/**
* Map a page of memory into this domain from another domain's grant table.
* xenbus_map_ring_valloc allocates a page of virtual address space, maps the
@@ -250,8 +262,15 @@ int xenbus_grant_ring(struct xenbus_devi
* or -ENOMEM on error. If an error is returned, device will switch to
* XenbusStateClosing and the error message will be saved in XenStore.
*/
+#if defined(CONFIG_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
struct vm_struct *xenbus_map_ring_valloc(struct xenbus_device *dev,
grant_ref_t ref);
+#else
+int xenbus_map_ring_valloc(struct xenbus_device *dev,
+ grant_ref_t gnt_ref, void **vaddr);
+#endif
+int xenbus_map_ring(struct xenbus_device *dev, grant_ref_t gnt_ref,
+ grant_handle_t *handle, void *vaddr);
/**
* Unmap a page of memory in this domain that was imported from another domain
@@ -259,8 +278,13 @@ struct vm_struct *xenbus_map_ring_valloc
* Returns 0 on success and returns GNTST_* on error
* (see xen/include/interface/grant_table.h).
*/
+#if defined(CONFIG_XEN) || defined(HAVE_XEN_PLATFORM_COMPAT_H)
int xenbus_unmap_ring_vfree(struct xenbus_device *dev, struct vm_struct *);
-
+#else
+int xenbus_unmap_ring_vfree(struct xenbus_device *dev, void *vaddr);
+#endif
+int xenbus_unmap_ring(struct xenbus_device *dev,
+ grant_handle_t handle, void *vaddr);
/**
* Allocate an event channel for the given xenbus_device, assigning the newly