int gnttab_post_map_adjust(const struct gnttab_map_grant_ref *map, unsigned int count)
{
unsigned int i;
int rc = 0;
for (i = 0; i < count && rc == 0; ++i, ++map) {
pte_t pte;
if (!(map->flags & GNTMAP_host_map)
|| !(map->flags & GNTMAP_application_map))
continue;
#ifdef CONFIG_X86
pte = __pte_ma((map->dev_bus_addr | _PAGE_PRESENT | _PAGE_USER
| _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_NX
| _PAGE_SPECIAL)
& __supported_pte_mask);
#else
#error Architecture not yet supported.
#endif
if (!(map->flags & GNTMAP_readonly))
pte = pte_mkwrite(pte);
if (map->flags & GNTMAP_contains_pte) {
mmu_update_t u;
u.ptr = map->host_addr;
u.val = __pte_val(pte);
rc = HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF);
} else
rc = HYPERVISOR_update_va_mapping(map->host_addr, pte, 0);
}
return rc;
}
static int __init
#else
int __devinit
#endif
gnttab_init(void)
{
int i;
unsigned int max_nr_glist_frames, nr_glist_frames;
unsigned int nr_init_grefs;
if (!is_running_on_xen())
return -ENODEV;
nr_grant_frames = 1;
boot_max_nr_grant_frames = __max_nr_grant_frames();
/* Determine the maximum number of frames required for the
* grant reference free list on the current hypervisor.
*/
max_nr_glist_frames = nr_freelist_frames(boot_max_nr_grant_frames);
gnttab_list = kmalloc(max_nr_glist_frames * sizeof(grant_ref_t *),
GFP_KERNEL);
if (gnttab_list == NULL)
return -ENOMEM;
nr_glist_frames = nr_freelist_frames(nr_grant_frames);
for (i = 0; i < nr_glist_frames; i++) {
gnttab_list[i] = (grant_ref_t *)__get_free_page(GFP_KERNEL);
if (gnttab_list[i] == NULL)
goto ini_nomem;
}
if (gnttab_resume() < 0)
return -ENODEV;
nr_init_grefs = nr_grant_frames * ENTRIES_PER_GRANT_FRAME;
for (i = NR_RESERVED_ENTRIES; i < nr_init_grefs - 1; i++)
gnttab_entry(i) = i + 1;
gnttab_entry(nr_init_grefs - 1) = GNTTAB_LIST_END;
gnttab_free_count = nr_init_grefs - NR_RESERVED_ENTRIES;
gnttab_free_head = NR_RESERVED_ENTRIES;
#if defined(CONFIG_XEN) && defined(__HAVE_ARCH_PTE_SPECIAL)
if (!xen_feature(XENFEAT_auto_translated_physmap)
&& xen_feature(XENFEAT_gnttab_map_avail_bits)) {
#ifdef CONFIG_X86
GNTMAP_pte_special = (__pte_val(pte_mkspecial(__pte_ma(0)))
>> _PAGE_BIT_UNUSED1) << _GNTMAP_guest_avail0;
#else
#error Architecture not yet supported.
#endif
}
void xen_arch_pre_suspend(void)
{
xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
xen_start_info->console.domU.mfn =
mfn_to_pfn(xen_start_info->console.domU.mfn);
BUG_ON(!irqs_disabled());
HYPERVISOR_shared_info = &xen_dummy_shared_info;
if (HYPERVISOR_update_va_mapping(fix_to_virt(FIX_PARAVIRT_BOOTMAP),
__pte_ma(0), 0))
BUG();
}
static int dealloc_pte_fn(
pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
{
unsigned long mfn = pte_mfn(*pte);
int ret;
struct xen_memory_reservation reservation = {
.nr_extents = 1,
.extent_order = 0,
.domid = DOMID_SELF
};
set_xen_guest_handle(reservation.extent_start, &mfn);
set_pte_at(&init_mm, addr, pte, __pte_ma(0));
set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY);
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
BUG_ON(ret != 1);
return 0;
}
#endif
struct page **alloc_empty_pages_and_pagevec(int nr_pages)
{
unsigned long vaddr, flags;
struct page *page, **pagevec;
int i, ret;
pagevec = kmalloc(sizeof(page) * nr_pages, GFP_KERNEL);
if (pagevec == NULL)
return NULL;
for (i = 0; i < nr_pages; i++) {
page = pagevec[i] = alloc_page(GFP_KERNEL);
if (page == NULL)
goto err;
vaddr = (unsigned long)page_address(page);
scrub_pages(vaddr, 1);
balloon_lock(flags);
if (xen_feature(XENFEAT_auto_translated_physmap)) {
unsigned long gmfn = page_to_pfn(page);
struct xen_memory_reservation reservation = {
.nr_extents = 1,
.extent_order = 0,
.domid = DOMID_SELF
};
set_xen_guest_handle(reservation.extent_start, &gmfn);
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
&reservation);
if (ret == 1)
ret = 0; /* success */
} else {
#ifdef CONFIG_XEN
ret = apply_to_page_range(&init_mm, vaddr, PAGE_SIZE,
dealloc_pte_fn, NULL);
#else
/* Cannot handle non-auto translate mode. */
ret = 1;
#endif
}
if (ret != 0) {
balloon_unlock(flags);
__free_page(page);
goto err;
}
totalram_pages = --current_pages;
balloon_unlock(flags);
}
out:
schedule_work(&balloon_worker);
#ifdef CONFIG_XEN
flush_tlb_all();
#endif
return pagevec;
err:
balloon_lock(flags);
while (--i >= 0)
balloon_append(pagevec[i]);
balloon_unlock(flags);
kfree(pagevec);
pagevec = NULL;
goto out;
}
static int increase_reservation(unsigned long nr_pages)
{
unsigned long pfn, i, flags;
struct page *page;
long rc;
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
if (nr_pages > ARRAY_SIZE(frame_list))
nr_pages = ARRAY_SIZE(frame_list);
balloon_lock(flags);
page = balloon_first_page();
for (i = 0; i < nr_pages; i++) {
BUG_ON(page == NULL);
frame_list[i] = page_to_pfn(page);;
page = balloon_next_page(page);
}
set_xen_guest_handle(reservation.extent_start, frame_list);
reservation.nr_extents = nr_pages;
rc = HYPERVISOR_memory_op(
XENMEM_populate_physmap, &reservation);
if (rc < nr_pages) {
int ret;
/* We hit the Xen hard limit: reprobe. */
set_xen_guest_handle(reservation.extent_start, frame_list);
reservation.nr_extents = rc;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
&reservation);
BUG_ON(ret != rc);
hard_limit = current_pages + rc - driver_pages;
goto out;
}
for (i = 0; i < nr_pages; i++) {
page = balloon_retrieve();
BUG_ON(page == NULL);
pfn = page_to_pfn(page);
BUG_ON(!xen_feature(XENFEAT_auto_translated_physmap) &&
phys_to_machine_mapping_valid(pfn));
/* Update P->M and M->P tables. */
set_phys_to_machine(pfn, frame_list[i]);
#ifdef CONFIG_XEN
xen_machphys_update(frame_list[i], pfn);
/* Link back into the page tables if not highmem. */
if (pfn < max_low_pfn) {
int ret;
ret = HYPERVISOR_update_va_mapping(
(unsigned long)__va(pfn << PAGE_SHIFT),
pfn_pte_ma(frame_list[i], PAGE_KERNEL),
0);
BUG_ON(ret);
}
#endif
/* Relinquish the page back to the allocator. */
ClearPageReserved(page);
set_page_count(page, 1);
__free_page(page);
}
current_pages += nr_pages;
totalram_pages = current_pages;
out:
balloon_unlock(flags);
return 0;
}
static int decrease_reservation(unsigned long nr_pages)
{
unsigned long pfn, i, flags;
struct page *page;
void *v;
int need_sleep = 0;
int ret;
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
if (nr_pages > ARRAY_SIZE(frame_list))
nr_pages = ARRAY_SIZE(frame_list);
for (i = 0; i < nr_pages; i++) {
if ((page = alloc_page(GFP_BALLOON)) == NULL) {
nr_pages = i;
need_sleep = 1;
break;
}
pfn = page_to_pfn(page);
frame_list[i] = pfn_to_mfn(pfn);
if (!PageHighMem(page)) {
v = phys_to_virt(pfn << PAGE_SHIFT);
scrub_pages(v, 1);
#ifdef CONFIG_XEN
ret = HYPERVISOR_update_va_mapping(
(unsigned long)v, __pte_ma(0), 0);
BUG_ON(ret);
#endif
}
#ifdef CONFIG_XEN_SCRUB_PAGES
else {
v = kmap(page);
scrub_pages(v, 1);
kunmap(page);
}
#endif
}
#ifdef CONFIG_XEN
/* Ensure that ballooned highmem pages don't have kmaps. */
kmap_flush_unused();
flush_tlb_all();
#endif
balloon_lock(flags);
/* No more mappings: invalidate P2M and add to balloon. */
for (i = 0; i < nr_pages; i++) {
pfn = mfn_to_pfn(frame_list[i]);
set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
balloon_append(pfn_to_page(pfn));
}
set_xen_guest_handle(reservation.extent_start, frame_list);
reservation.nr_extents = nr_pages;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
BUG_ON(ret != nr_pages);
current_pages -= nr_pages;
totalram_pages = current_pages;
balloon_unlock(flags);
return need_sleep;
}
/*
* We avoid multiple worker processes conflicting via the balloon mutex.
* We may of course race updates of the target counts (which are protected
* by the balloon lock), or with changes to the Xen hard limit, but we will
* recover from these in time.
*/
static void balloon_process(void *unused)
{
int need_sleep = 0;
long credit;
down(&balloon_mutex);
do {
credit = current_target() - current_pages;
if (credit > 0)
need_sleep = (increase_reservation(credit) != 0);
if (credit < 0)
need_sleep = (decrease_reservation(-credit) != 0);
#ifndef CONFIG_PREEMPT
if (need_resched())
schedule();
#endif
} while ((credit != 0) && !need_sleep);
/* Schedule more work if there is some still to be done. */
if (current_target() != current_pages)
mod_timer(&balloon_timer, jiffies + HZ);
up(&balloon_mutex);
}
static unsigned long __init xen_do_chunk(unsigned long start,
unsigned long end, bool release)
{
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
unsigned long len = 0;
int xlated_phys = xen_feature(XENFEAT_auto_translated_physmap);
unsigned long pfn;
int ret;
for (pfn = start; pfn < end; pfn++) {
unsigned long frame;
unsigned long mfn = pfn_to_mfn(pfn);
if (release) {
/* Make sure pfn exists to start with */
if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn)
continue;
frame = mfn;
} else {
if (!xlated_phys && mfn != INVALID_P2M_ENTRY)
continue;
frame = pfn;
}
set_xen_guest_handle(reservation.extent_start, &frame);
reservation.nr_extents = 1;
ret = HYPERVISOR_memory_op(release ? XENMEM_decrease_reservation : XENMEM_populate_physmap,
&reservation);
WARN(ret != 1, "Failed to %s pfn %lx err=%d\n",
release ? "release" : "populate", pfn, ret);
if (ret == 1) {
if (!early_set_phys_to_machine(pfn, release ? INVALID_P2M_ENTRY : frame)) {
if (release)
break;
set_xen_guest_handle(reservation.extent_start, &frame);
reservation.nr_extents = 1;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
&reservation);
break;
}
len++;
} else
break;
}
if (len)
printk(KERN_INFO "%s %lx-%lx pfn range: %lu pages %s\n",
release ? "Freeing" : "Populating",
start, end, len,
release ? "freed" : "added");
return len;
}
static unsigned long __init xen_release_chunk(unsigned long start,
unsigned long end)
{
/*
* Xen already ballooned out the E820 non RAM regions for us
* and set them up properly in EPT.
*/
if (xen_feature(XENFEAT_auto_translated_physmap))
return end - start;
return xen_do_chunk(start, end, true);
}
static unsigned long __init xen_populate_chunk(
const struct e820entry *list, size_t map_size,
unsigned long max_pfn, unsigned long *last_pfn,
unsigned long credits_left)
{
const struct e820entry *entry;
unsigned int i;
unsigned long done = 0;
unsigned long dest_pfn;
for (i = 0, entry = list; i < map_size; i++, entry++) {
unsigned long s_pfn;
unsigned long e_pfn;
unsigned long pfns;
long capacity;
if (credits_left <= 0)
break;
if (entry->type != E820_RAM)
continue;
e_pfn = PFN_DOWN(entry->addr + entry->size);
/* We only care about E820 after the xen_start_info->nr_pages */
if (e_pfn <= max_pfn)
continue;
s_pfn = PFN_UP(entry->addr);
/* If the E820 falls within the nr_pages, we want to start
* at the nr_pages PFN.
* If that would mean going past the E820 entry, skip it
*/
if (s_pfn <= max_pfn) {
capacity = e_pfn - max_pfn;
dest_pfn = max_pfn;
} else {
capacity = e_pfn - s_pfn;
dest_pfn = s_pfn;
}
if (credits_left < capacity)
capacity = credits_left;
pfns = xen_do_chunk(dest_pfn, dest_pfn + capacity, false);
done += pfns;
*last_pfn = (dest_pfn + pfns);
if (pfns < capacity)
break;
credits_left -= pfns;
}
return done;
}
static void __init xen_set_identity_and_release_chunk(
unsigned long start_pfn, unsigned long end_pfn, unsigned long nr_pages,
unsigned long *released, unsigned long *identity)
{
unsigned long pfn;
/*
* If the PFNs are currently mapped, clear the mappings
* (except for the ISA region which must be 1:1 mapped) to
* release the refcounts (in Xen) on the original frames.
*/
/*
* PVH E820 matches the hypervisor's P2M which means we need to
* account for the proper values of *release and *identity.
*/
for (pfn = start_pfn; !xen_feature(XENFEAT_auto_translated_physmap) &&
pfn <= max_pfn_mapped && pfn < end_pfn; pfn++) {
pte_t pte = __pte_ma(0);
if (pfn < PFN_UP(ISA_END_ADDRESS))
pte = mfn_pte(pfn, PAGE_KERNEL_IO);
(void)HYPERVISOR_update_va_mapping(
(unsigned long)__va(pfn << PAGE_SHIFT), pte, 0);
}
if (start_pfn < nr_pages)
*released += xen_release_chunk(
start_pfn, min(end_pfn, nr_pages));
*identity += set_phys_range_identity(start_pfn, end_pfn);
}
void __init setup_arch(char **cmdline_p)
{
unsigned long kernel_end;
#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
struct e820entry *machine_e820;
struct xen_memory_map memmap;
#endif
#ifdef CONFIG_XEN
/* Register a call for panic conditions. */
atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
ROOT_DEV = MKDEV(RAMDISK_MAJOR,0);
kernel_end = 0; /* dummy */
screen_info = SCREEN_INFO;
if (xen_start_info->flags & SIF_INITDOMAIN) {
/* This is drawn from a dump from vgacon:startup in
* standard Linux. */
screen_info.orig_video_mode = 3;
screen_info.orig_video_isVGA = 1;
screen_info.orig_video_lines = 25;
screen_info.orig_video_cols = 80;
screen_info.orig_video_ega_bx = 3;
screen_info.orig_video_points = 16;
} else
screen_info.orig_video_isVGA = 0;
edid_info = EDID_INFO;
saved_video_mode = SAVED_VIDEO_MODE;
bootloader_type = LOADER_TYPE;
#ifdef CONFIG_BLK_DEV_RAM
rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
#endif
setup_xen_features();
HYPERVISOR_vm_assist(VMASST_CMD_enable,
VMASST_TYPE_writable_pagetables);
ARCH_SETUP
#else
ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
screen_info = SCREEN_INFO;
edid_info = EDID_INFO;
saved_video_mode = SAVED_VIDEO_MODE;
bootloader_type = LOADER_TYPE;
#ifdef CONFIG_BLK_DEV_RAM
rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
#endif
#endif /* !CONFIG_XEN */
setup_memory_region();
copy_edd();
if (!MOUNT_ROOT_RDONLY)
root_mountflags &= ~MS_RDONLY;
init_mm.start_code = (unsigned long) &_text;
init_mm.end_code = (unsigned long) &_etext;
init_mm.end_data = (unsigned long) &_edata;
init_mm.brk = (unsigned long) &_end;
#ifndef CONFIG_XEN
code_resource.start = virt_to_phys(&_text);
code_resource.end = virt_to_phys(&_etext)-1;
data_resource.start = virt_to_phys(&_etext);
data_resource.end = virt_to_phys(&_edata)-1;
#endif
parse_cmdline_early(cmdline_p);
early_identify_cpu(&boot_cpu_data);
/*
* partially used pages are not usable - thus
* we are rounding upwards:
*/
end_pfn = e820_end_of_ram();
num_physpages = end_pfn; /* for pfn_valid */
check_efer();
#ifndef CONFIG_XEN
discover_ebda();
#endif
init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
#ifdef CONFIG_ACPI_NUMA
/*
* Parse SRAT to discover nodes.
*/
acpi_numa_init();
#endif
#ifdef CONFIG_NUMA
numa_initmem_init(0, end_pfn);
#else
contig_initmem_init(0, end_pfn);
#endif
/* Reserve direct mapping */
reserve_bootmem_generic(table_start << PAGE_SHIFT,
(table_end - table_start) << PAGE_SHIFT);
/* reserve kernel */
kernel_end = round_up(__pa_symbol(&_end),PAGE_SIZE);
reserve_bootmem_generic(HIGH_MEMORY, kernel_end - HIGH_MEMORY);
#ifdef CONFIG_XEN
/* reserve physmap, start info and initial page tables */
reserve_bootmem(kernel_end, (table_start<<PAGE_SHIFT)-kernel_end);
#else
/*
* reserve physical page 0 - it's a special BIOS page on many boxes,
* enabling clean reboots, SMP operation, laptop functions.
*/
reserve_bootmem_generic(0, PAGE_SIZE);
/* reserve ebda region */
if (ebda_addr)
reserve_bootmem_generic(ebda_addr, ebda_size);
#endif
#ifdef CONFIG_SMP
/*
* But first pinch a few for the stack/trampoline stuff
* FIXME: Don't need the extra page at 4K, but need to fix
* trampoline before removing it. (see the GDT stuff)
*/
reserve_bootmem_generic(PAGE_SIZE, PAGE_SIZE);
/* Reserve SMP trampoline */
reserve_bootmem_generic(SMP_TRAMPOLINE_BASE, PAGE_SIZE);
#endif
#ifdef CONFIG_ACPI_SLEEP
/*
* Reserve low memory region for sleep support.
*/
acpi_reserve_bootmem();
#endif
#ifdef CONFIG_XEN
#ifdef CONFIG_BLK_DEV_INITRD
if (xen_start_info->mod_start) {
if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
/*reserve_bootmem_generic(INITRD_START, INITRD_SIZE);*/
initrd_start = INITRD_START + PAGE_OFFSET;
initrd_end = initrd_start+INITRD_SIZE;
initrd_below_start_ok = 1;
} else {
printk(KERN_ERR "initrd extends beyond end of memory "
"(0x%08lx > 0x%08lx)\ndisabling initrd\n",
(unsigned long)(INITRD_START + INITRD_SIZE),
(unsigned long)(end_pfn << PAGE_SHIFT));
initrd_start = 0;
}
}
#endif
#else /* CONFIG_XEN */
#ifdef CONFIG_BLK_DEV_INITRD
if (LOADER_TYPE && INITRD_START) {
if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
reserve_bootmem_generic(INITRD_START, INITRD_SIZE);
initrd_start =
INITRD_START ? INITRD_START + PAGE_OFFSET : 0;
initrd_end = initrd_start+INITRD_SIZE;
}
else {
printk(KERN_ERR "initrd extends beyond end of memory "
"(0x%08lx > 0x%08lx)\ndisabling initrd\n",
(unsigned long)(INITRD_START + INITRD_SIZE),
(unsigned long)(end_pfn << PAGE_SHIFT));
initrd_start = 0;
}
}
#endif
#endif /* !CONFIG_XEN */
#ifdef CONFIG_KEXEC
if (crashk_res.start != crashk_res.end) {
reserve_bootmem(crashk_res.start,
crashk_res.end - crashk_res.start + 1);
}
#endif
paging_init();
#ifdef CONFIG_X86_LOCAL_APIC
/*
* Find and reserve possible boot-time SMP configuration:
*/
find_smp_config();
#endif
#ifdef CONFIG_XEN
{
int i, j, k, fpp;
unsigned long va;
/* 'Initial mapping' of initrd must be destroyed. */
for (va = xen_start_info->mod_start;
va < (xen_start_info->mod_start+xen_start_info->mod_len);
va += PAGE_SIZE) {
HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0);
}
if (!xen_feature(XENFEAT_auto_translated_physmap)) {
/* Make sure we have a large enough P->M table. */
phys_to_machine_mapping = alloc_bootmem(
end_pfn * sizeof(unsigned long));
memset(phys_to_machine_mapping, ~0,
end_pfn * sizeof(unsigned long));
memcpy(phys_to_machine_mapping,
(unsigned long *)xen_start_info->mfn_list,
xen_start_info->nr_pages * sizeof(unsigned long));
free_bootmem(
__pa(xen_start_info->mfn_list),
PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
sizeof(unsigned long))));
/* Destroyed 'initial mapping' of old p2m table. */
for (va = xen_start_info->mfn_list;
va < (xen_start_info->mfn_list +
(xen_start_info->nr_pages*sizeof(unsigned long)));
va += PAGE_SIZE) {
HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0);
}
/*
* Initialise the list of the frames that specify the
* list of frames that make up the p2m table. Used by
* save/restore.
*/
pfn_to_mfn_frame_list_list = alloc_bootmem(PAGE_SIZE);
HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
virt_to_mfn(pfn_to_mfn_frame_list_list);
fpp = PAGE_SIZE/sizeof(unsigned long);
for (i=0, j=0, k=-1; i< end_pfn; i+=fpp, j++) {
if ((j % fpp) == 0) {
k++;
BUG_ON(k>=fpp);
pfn_to_mfn_frame_list[k] =
alloc_bootmem(PAGE_SIZE);
pfn_to_mfn_frame_list_list[k] =
virt_to_mfn(pfn_to_mfn_frame_list[k]);
j=0;
}
pfn_to_mfn_frame_list[k][j] =
virt_to_mfn(&phys_to_machine_mapping[i]);
}
HYPERVISOR_shared_info->arch.max_pfn = end_pfn;
}
}
if (xen_start_info->flags & SIF_INITDOMAIN)
dmi_scan_machine();
if ( ! (xen_start_info->flags & SIF_INITDOMAIN))
{
acpi_disabled = 1;
#ifdef CONFIG_ACPI
acpi_ht = 0;
#endif
}
#endif
#ifndef CONFIG_XEN
check_ioapic();
#endif
zap_low_mappings(0);
/*
* set this early, so we dont allocate cpu0
* if MADT list doesnt list BSP first
* mpparse.c/MP_processor_info() allocates logical cpu numbers.
*/
cpu_set(0, cpu_present_map);
#ifdef CONFIG_ACPI
/*
* Initialize the ACPI boot-time table parser (gets the RSDP and SDT).
* Call this early for SRAT node setup.
*/
acpi_boot_table_init();
/*
* Read APIC and some other early information from ACPI tables.
*/
acpi_boot_init();
#endif
init_cpu_to_node();
#ifdef CONFIG_X86_LOCAL_APIC
/*
* get boot-time SMP configuration:
*/
if (smp_found_config)
get_smp_config();
#ifndef CONFIG_XEN
init_apic_mappings();
#endif
#endif
#if defined(CONFIG_XEN) && defined(CONFIG_SMP) && !defined(CONFIG_HOTPLUG_CPU)
prefill_possible_map();
#endif
/*
* Request address space for all standard RAM and ROM resources
* and also for regions reported as reserved by the e820.
*/
#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
probe_roms();
if (xen_start_info->flags & SIF_INITDOMAIN) {
machine_e820 = alloc_bootmem_low_pages(PAGE_SIZE);
memmap.nr_entries = E820MAX;
set_xen_guest_handle(memmap.buffer, machine_e820);
BUG_ON(HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap));
e820_reserve_resources(machine_e820, memmap.nr_entries);
}
#elif !defined(CONFIG_XEN)
probe_roms();
e820_reserve_resources(e820.map, e820.nr_map);
#endif
request_resource(&iomem_resource, &video_ram_resource);
{
unsigned i;
/* request I/O space for devices used on all i[345]86 PCs */
for (i = 0; i < STANDARD_IO_RESOURCES; i++)
request_resource(&ioport_resource, &standard_io_resources[i]);
}
#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
if (xen_start_info->flags & SIF_INITDOMAIN) {
e820_setup_gap(machine_e820, memmap.nr_entries);
free_bootmem(__pa(machine_e820), PAGE_SIZE);
}
#elif !defined(CONFIG_XEN)
e820_setup_gap(e820.map, e820.nr_map);
#endif
#ifdef CONFIG_GART_IOMMU
iommu_hole_init();
#endif
#ifdef CONFIG_XEN
{
struct physdev_set_iopl set_iopl;
set_iopl.iopl = 1;
HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
if (xen_start_info->flags & SIF_INITDOMAIN) {
if (!(xen_start_info->flags & SIF_PRIVILEGED))
panic("Xen granted us console access "
"but not privileged status");
#ifdef CONFIG_VT
#if defined(CONFIG_VGA_CONSOLE)
conswitchp = &vga_con;
#elif defined(CONFIG_DUMMY_CONSOLE)
conswitchp = &dummy_con;
#endif
#endif
} else {
extern int console_use_vt;
console_use_vt = 0;
}
}
#else /* CONFIG_XEN */
#ifdef CONFIG_VT
#if defined(CONFIG_VGA_CONSOLE)
conswitchp = &vga_con;
#elif defined(CONFIG_DUMMY_CONSOLE)
conswitchp = &dummy_con;
#endif
#endif
#endif /* !CONFIG_XEN */
}
static int increase_reservation(unsigned long nr_pages)
{
unsigned long pfn, i, flags;
struct page *page;
long rc;
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
if (nr_pages > ARRAY_SIZE(frame_list))
nr_pages = ARRAY_SIZE(frame_list);
spin_lock_irqsave(&balloon_lock, flags);
page = balloon_first_page();
for (i = 0; i < nr_pages; i++) {
BUG_ON(page == NULL);
frame_list[i] = page_to_pfn(page);
page = balloon_next_page(page);
}
set_xen_guest_handle(reservation.extent_start, frame_list);
reservation.nr_extents = nr_pages;
rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
if (rc < 0)
goto out;
for (i = 0; i < rc; i++) {
page = balloon_retrieve();
BUG_ON(page == NULL);
pfn = page_to_pfn(page);
BUG_ON(!xen_feature(XENFEAT_auto_translated_physmap) &&
phys_to_machine_mapping_valid(pfn));
set_phys_to_machine(pfn, frame_list[i]);
/* Link back into the page tables if not highmem. */
#ifdef CONFIG_PVM
if (!xen_hvm_domain() && pfn < max_low_pfn) {
int ret;
ret = HYPERVISOR_update_va_mapping(
(unsigned long)__va(pfn << PAGE_SHIFT),
mfn_pte(frame_list[i], PAGE_KERNEL),
0);
BUG_ON(ret);
}
#endif
/* Relinquish the page back to the allocator. */
ClearPageReserved(page);
init_page_count(page);
__free_page(page);
}
balloon_stats.current_pages += rc;
if (old_totalram_pages + rc < totalram_pages)
{
printk(KERN_INFO "old_totalram=%luKB, totalram_pages=%luKB\n", old_totalram_pages*4, totalram_pages*4);
balloon_stats.current_pages = totalram_pages + totalram_bias;
printk(KERN_INFO "when ballooning, the mem online! totalram=%luKB, current=%luKB\n", totalram_pages*4, balloon_stats.current_pages*4);
}
old_totalram_pages = totalram_pages;
out:
spin_unlock_irqrestore(&balloon_lock, flags);
return rc < 0 ? rc : rc != nr_pages;
}
static int decrease_reservation(unsigned long nr_pages)
{
unsigned long pfn, i, flags;
struct page *page;
int need_sleep = 0;
int ret;
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
if (nr_pages > ARRAY_SIZE(frame_list))
nr_pages = ARRAY_SIZE(frame_list);
for (i = 0; i < nr_pages; i++) {
if ((page = alloc_page(GFP_BALLOON)) == NULL) {
nr_pages = i;
need_sleep = 1;
break;
}
pfn = page_to_pfn(page);
frame_list[i] = pfn_to_mfn(pfn);
scrub_page(page);
if (!xen_hvm_domain() && !PageHighMem(page)) {
ret = HYPERVISOR_update_va_mapping(
(unsigned long)__va(pfn << PAGE_SHIFT),
__pte_ma(0), 0);
BUG_ON(ret);
}
}
/* Ensure that ballooned highmem pages don't have kmaps. */
#ifdef CONFIG_PVM
kmap_flush_unused();
flush_tlb_all();
#endif
spin_lock_irqsave(&balloon_lock, flags);
/* No more mappings: invalidate P2M and add to balloon. */
for (i = 0; i < nr_pages; i++) {
pfn = mfn_to_pfn(frame_list[i]);
set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
balloon_append(pfn_to_page(pfn));
}
set_xen_guest_handle(reservation.extent_start, frame_list);
reservation.nr_extents = nr_pages;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
BUG_ON(ret != nr_pages);
balloon_stats.current_pages -= nr_pages;
if(old_totalram_pages < totalram_pages + nr_pages)
{
printk(KERN_INFO "old_totalram=%luKB, totalram_pages=%luKB\n", old_totalram_pages*4, totalram_pages*4);
balloon_stats.current_pages = totalram_pages + totalram_bias;
printk(KERN_INFO "when ballooning, the mem online! totalram=%luKB, current=%luKB\n", totalram_pages*4, balloon_stats.current_pages*4);
}
old_totalram_pages = totalram_pages;
spin_unlock_irqrestore(&balloon_lock, flags);
return need_sleep;
}
/*
* We avoid multiple worker processes conflicting via the balloon mutex.
* We may of course race updates of the target counts (which are protected
* by the balloon lock), or with changes to the Xen hard limit, but we will
* recover from these in time.
*/
static void balloon_process(struct work_struct *work)
{
int need_sleep = 0;
long credit;
long total_increase = 0;
char buffer[16];
mutex_lock(&balloon_mutex);
printk(KERN_INFO "totalram_pages=%luKB, current_pages=%luKB,totalram_bias=%luKB\n", totalram_pages*4, balloon_stats.current_pages*4, totalram_bias*4);
if (totalram_pages > old_totalram_pages)
{
//TODO:Just know that totalram_pages will increase.
total_increase = (totalram_pages - old_totalram_pages) % GB2PAGE;
if (totalram_bias > total_increase )
{
totalram_bias = totalram_bias - total_increase;
}
balloon_stats.current_pages = totalram_pages + totalram_bias;
old_totalram_pages = totalram_pages;
}
printk(KERN_INFO "totalram_pages=%luKB, current_pages=%luKB, totalram_bias=%luKB,total_increase=%ld\n", totalram_pages*4, balloon_stats.current_pages*4, totalram_bias*4, total_increase*4);
xenbus_write(XBT_NIL, "control/uvp", "Balloon_flag", "1");
do {
credit = current_target() - balloon_stats.current_pages;
if (credit > 0)
need_sleep = (increase_reservation(credit) != 0);
if (credit < 0)
need_sleep = (decrease_reservation(-credit) != 0);
#ifndef CONFIG_PREEMPT
if (need_resched())
schedule();
#endif
} while ((credit != 0) && !need_sleep);
/* Schedule more work if there is some still to be done. */
if (current_target() != balloon_stats.current_pages)
{
mod_timer(&balloon_timer, jiffies + HZ);
sprintf(buffer,"%lu",balloon_stats.current_pages<<(PAGE_SHIFT-10));
xenbus_write(XBT_NIL, "memory", "target", buffer);
}
xenbus_write(XBT_NIL, "control/uvp", "Balloon_flag", "0");
mutex_unlock(&balloon_mutex);
}
