void copy_highpage(struct page *to, struct page *from)
{
void *vfrom, *vto;
if (likely(xen_feature(XENFEAT_highmem_assist))
&& (PageHighMem(from) || PageHighMem(to))) {
unsigned long from_pfn = page_to_pfn(from);
unsigned long to_pfn = page_to_pfn(to);
struct mmuext_op meo;
meo.cmd = MMUEXT_COPY_PAGE;
meo.arg1.mfn = pfn_to_mfn(to_pfn);
meo.arg2.src_mfn = pfn_to_mfn(from_pfn);
if (mfn_to_pfn(meo.arg2.src_mfn) == from_pfn
&& mfn_to_pfn(meo.arg1.mfn) == to_pfn
&& HYPERVISOR_mmuext_op(&meo, 1, NULL, DOMID_SELF) == 0)
return;
}
vfrom = kmap_atomic(from, KM_USER0);
vto = kmap_atomic(to, KM_USER1);
copy_page(vto, vfrom);
kunmap_atomic(vfrom, KM_USER0);
kunmap_atomic(vto, KM_USER1);
}
static void pre_suspend(void)
{
HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
clear_fixmap(FIX_SHARED_INFO);
xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
xen_start_info->console_mfn =
mfn_to_pfn(xen_start_info->console_mfn);
}
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();
}
CAMLprim value
stub_hypervisor_suspend(value unit)
{
CAMLparam0();
int cancelled;
printk("WARNING: stub_hypervisor_suspend not yet implemented\n");
cancelled = 1;
#if 0
/* Turn the store and console mfns to pfns - required because xc_domain_restore uses these values */
start_info.store_mfn = mfn_to_pfn(start_info.store_mfn);
start_info.console.domU.mfn = mfn_to_pfn(start_info.console.domU.mfn);
/* canonicalize_pagetables can't cope with pagetable entries that are outside of the guest's mfns,
so we must unmap anything outside of our space */
unmap_shared_info();
/* Actually do the suspend. When this function returns 0, we've been resumed */
cancelled = HYPERVISOR_suspend(virt_to_mfn(&start_info));
if(cancelled) {
start_info.store_mfn = pfn_to_mfn(start_info.store_mfn);
start_info.console.domU.mfn = pfn_to_mfn(start_info.console.domU.mfn);
}
/* Reinitialise several things */
trap_init();
init_events();
/* ENABLE EVENT DELIVERY. This is disabled at start of day. */
local_irq_enable();
setup_xen_features();
HYPERVISOR_shared_info = map_shared_info(start_info.shared_info);
/* Set up event and failsafe callback addresses. */
HYPERVISOR_set_callbacks(
(unsigned long)hypervisor_callback,
(unsigned long)failsafe_callback, 0);
init_time();
arch_rebuild_p2m();
unmask_evtchn(start_info.console.domU.evtchn);
unmask_evtchn(start_info.store_evtchn);
#endif
CAMLreturn(Val_int(cancelled));
}
/* Assume pteval_t is equivalent to all the other *val_t types. */
static pteval_t pte_mfn_to_pfn(pteval_t val)
{
if (val & _PAGE_PRESENT) {
unsigned long mfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT;
pteval_t flags = val & PTE_FLAGS_MASK;
val = ((pteval_t)mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
}
static int __init xen_hvc_init(void)
{
struct hvc_struct *hp;
struct hv_ops *ops;
if (!xen_pv_domain())
return -ENODEV;
if (xen_initial_domain()) {
ops = &dom0_hvc_ops;
xencons_irq = bind_virq_to_irq(VIRQ_CONSOLE, 0);
} else {
if (!xen_start_info->console.domU.evtchn)
return -ENODEV;
ops = &domU_hvc_ops;
xencons_irq = bind_evtchn_to_irq(xen_start_info->console.domU.evtchn);
}
if (xencons_irq < 0)
xencons_irq = 0; /* NO_IRQ */
else
irq_set_noprobe(xencons_irq);
hp = hvc_alloc(HVC_COOKIE, xencons_irq, ops, 256);
if (IS_ERR(hp))
return PTR_ERR(hp);
hvc = hp;
console_pfn = mfn_to_pfn(xen_start_info->console.domU.mfn);
return 0;
}
static int xsd_kva_mmap(struct file *file, struct vm_area_struct *vma)
{
size_t size = vma->vm_end - vma->vm_start;
int old;
int rc;
old = atomic_cmpxchg(&xenbus_xsd_state,
XENBUS_XSD_UNCOMMITTED,
XENBUS_XSD_LOCAL_INIT);
switch (old) {
case XENBUS_XSD_UNCOMMITTED:
rc = xb_init_comms();
if (rc != 0)
return rc;
break;
case XENBUS_XSD_FOREIGN_INIT:
case XENBUS_XSD_FOREIGN_READY:
return -EBUSY;
case XENBUS_XSD_LOCAL_INIT:
case XENBUS_XSD_LOCAL_READY:
default:
break;
}
if ((size > PAGE_SIZE) || (vma->vm_pgoff != 0))
return -EINVAL;
if (remap_pfn_range(vma, vma->vm_start, mfn_to_pfn(xen_store_mfn),
size, vma->vm_page_prot))
return -EAGAIN;
return 0;
}
static inline phys_addr_t xen_bus_to_phys(dma_addr_t baddr)
{
unsigned long pfn = mfn_to_pfn(PFN_DOWN(baddr));
dma_addr_t dma = (dma_addr_t)pfn << PAGE_SHIFT;
phys_addr_t paddr = dma;
paddr |= baddr & ~PAGE_MASK;
return paddr;
}
static inline phys_addr_t xen_bus_to_phys(dma_addr_t baddr)
{
unsigned long pfn = mfn_to_pfn(PFN_DOWN(baddr));
dma_addr_t dma = (dma_addr_t)pfn << PAGE_SHIFT;
phys_addr_t paddr = dma;
BUG_ON(paddr != dma); /* truncation has occurred, should never happen */
paddr |= baddr & ~PAGE_MASK;
return paddr;
}
static pteval_t pte_mfn_to_pfn(pteval_t val)
{
if (val & _PAGE_PRESENT) {
unsigned long mfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT;
unsigned long pfn = mfn_to_pfn(mfn);
pteval_t flags = val & PTE_FLAGS_MASK;
if (unlikely(pfn == ~0))
val = flags & ~_PAGE_PRESENT;
else
val = ((pteval_t)pfn << PAGE_SHIFT) | flags;
}
static int xsd_kva_mmap(struct file *file, struct vm_area_struct *vma)
{
size_t size = vma->vm_end - vma->vm_start;
if ((size > PAGE_SIZE) || (vma->vm_pgoff != 0))
return -EINVAL;
if (remap_pfn_range(vma, vma->vm_start, mfn_to_pfn(xen_store_mfn),
size, vma->vm_page_prot))
return -EAGAIN;
return 0;
}
static int __init xen_free_mfn(unsigned long mfn)
{
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
set_xen_guest_handle(reservation.extent_start, &mfn);
reservation.nr_extents = 1;
return HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
}
/*
* This releases a chunk of memory and then does the identity map. It's used
* as a fallback if the remapping fails.
*/
static void __init xen_set_identity_and_release_chunk(unsigned long start_pfn,
unsigned long end_pfn, unsigned long nr_pages)
{
unsigned long pfn, end;
int ret;
WARN_ON(start_pfn > end_pfn);
/* Release pages first. */
end = min(end_pfn, nr_pages);
for (pfn = start_pfn; pfn < end; pfn++) {
unsigned long mfn = pfn_to_mfn(pfn);
/* Make sure pfn exists to start with */
if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn)
continue;
ret = xen_free_mfn(mfn);
WARN(ret != 1, "Failed to release pfn %lx err=%d\n", pfn, ret);
if (ret == 1) {
xen_released_pages++;
if (!__set_phys_to_machine(pfn, INVALID_P2M_ENTRY))
break;
} else
break;
}
set_phys_range_identity(start_pfn, end_pfn);
}
static int __init xen_init(void)
{
struct hvc_struct *hp;
if (!xen_pv_domain() ||
xen_initial_domain() ||
!xen_start_info->console.domU.evtchn)
return -ENODEV;
xencons_irq = bind_evtchn_to_irq(xen_start_info->console.domU.evtchn);
if (xencons_irq < 0)
xencons_irq = 0; /* NO_IRQ */
hp = hvc_alloc(HVC_COOKIE, xencons_irq, &hvc_ops, 256);
if (IS_ERR(hp))
return PTR_ERR(hp);
hvc = hp;
console_pfn = mfn_to_pfn(xen_start_info->console.domU.mfn);
return 0;
}
/*
* blkif_queue_request
*
* request block io
*
* id: for guest use only.
* operation: BLKIF_OP_{READ,WRITE,PROBE}
* buffer: buffer to read/write into. this should be a
* virtual address in the guest os.
*/
int ixp_queue_request(struct app_request *app_req, void *metadata)
{
struct ixpfront_info *info = (struct ixpfront_info *) metadata;
unsigned long buffer_mfn;
struct ixp_request *ring_req;
char *req_page = 0, *curr_pos;
unsigned long id;
int ref, err;
grant_ref_t gref_head;
if (unlikely(info->connected != IXP_STATE_CONNECTED))
return 1;
if (RING_FULL(&info->ring)) {
printk(KERN_ERR "%s:Ring full - returning backpressure\n", __FUNCTION__);
return 1;
}
if (gnttab_alloc_grant_references(
IXPIF_MAX_SEGMENTS_PER_REQUEST, &gref_head) < 0) {
/*gnttab_request_free_callback(
&info->callback,
ixp_restart_queue_callback,
info,
IXP_MAX_SEGMENTS_PER_REQUEST);*/
return 1;
}
/* Fill out a communications ring structure. */
ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt);
id = get_id_from_freelist(info);
ring_req->id = id;
ring_req->handle = info->handle;
ring_req->operation = IXP_OP_3DES_ENCRYPT;
ring_req->nr_segments = 1;
BUG_ON(ring_req->nr_segments > IXPIF_MAX_SEGMENTS_PER_REQUEST);
req_page = (char *)__get_free_page(GFP_NOIO | __GFP_HIGH);
if(req_page == 0) {
printk(KERN_ERR "ixp_queue_request:Error allocating memory");
return 1;
}
((struct des_request *)req_page)->key_size = app_req->key_size;
((struct des_request *)req_page)->iv_size = app_req->iv_size;
((struct des_request *)req_page)->msg_size = app_req->msg_size;
curr_pos = req_page + sizeof(struct des_request);
memcpy(curr_pos, app_req->key, app_req->key_size);
curr_pos += app_req->key_size;
memcpy(curr_pos, app_req->iv, app_req->iv_size);
curr_pos += app_req->iv_size;
memcpy(curr_pos, app_req->msg, app_req->msg_size);
curr_pos += app_req->msg_size;
buffer_mfn = virt_to_mfn(req_page);
/* install a grant reference. */
ref = gnttab_claim_grant_reference(&gref_head);
BUG_ON(ref == -ENOSPC);
gnttab_grant_foreign_access_ref(
ref,
info->xbdev->otherend_id,
buffer_mfn,
0);
info->shadow[id].r_params.presp = app_req->presp;
info->shadow[id].r_params.callbk_tag = app_req->callbk_tag;
info->shadow[id].frame[0] = mfn_to_pfn(buffer_mfn);
info->shadow[id].req_page = req_page;
ring_req->seg[0] =
(struct ixp_request_segment) {
.gref = ref
};
info->ring.req_prod_pvt++;
/* Keep a private copy so we can reissue requests when recovering. */
info->shadow[id].req = *ring_req;
flush_requests(info);
//gnttab_free_grant_references(gref_head);
return 0;
}
static unsigned long __init xen_release_chunk(phys_addr_t start_addr,
phys_addr_t end_addr)
{
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
unsigned long start, end;
unsigned long len = 0;
unsigned long pfn;
int ret;
start = PFN_UP(start_addr);
end = PFN_DOWN(end_addr);
if (end <= start)
return 0;
printk(KERN_INFO "xen_release_chunk: looking at area pfn %lx-%lx: ",
start, end);
for(pfn = start; pfn < end; pfn++) {
unsigned long mfn = pfn_to_mfn(pfn);
/* Make sure pfn exists to start with */
if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn)
continue;
set_xen_guest_handle(reservation.extent_start, &mfn);
reservation.nr_extents = 1;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
&reservation);
WARN(ret != 1, "Failed to release memory %lx-%lx err=%d\n",
start, end, ret);
if (ret == 1) {
__set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
len++;
}
}
printk(KERN_CONT "%ld pages freed\n", len);
return len;
}
static unsigned long __init xen_return_unused_memory(unsigned long max_pfn,
const struct e820map *e820)
{
phys_addr_t max_addr = PFN_PHYS(max_pfn);
phys_addr_t last_end = ISA_END_ADDRESS;
unsigned long released = 0;
int i;
/* Free any unused memory above the low 1Mbyte. */
for (i = 0; i < e820->nr_map && last_end < max_addr; i++) {
phys_addr_t end = e820->map[i].addr;
end = min(max_addr, end);
if (last_end < end)
released += xen_release_chunk(last_end, end);
last_end = max(last_end, e820->map[i].addr + e820->map[i].size);
}
if (last_end < max_addr)
released += xen_release_chunk(last_end, max_addr);
printk(KERN_INFO "released %ld pages of unused memory\n", released);
return released;
}
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;
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 (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)
{
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, the VA mapping also needs
* to be updated to be 1:1.
*/
for (pfn = start_pfn; pfn <= max_pfn_mapped && pfn < end_pfn; pfn++)
(void)HYPERVISOR_update_va_mapping(
(unsigned long)__va(pfn << PAGE_SHIFT),
mfn_pte(pfn, PAGE_KERNEL_IO), 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);
}
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);
}
static unsigned long __init xen_release_chunk(unsigned long start,
unsigned long end)
{
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
unsigned long len = 0;
unsigned long pfn;
int ret;
for(pfn = start; pfn < end; pfn++) {
unsigned long mfn = pfn_to_mfn(pfn);
/* Make sure pfn exists to start with */
if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn)
continue;
set_xen_guest_handle(reservation.extent_start, &mfn);
reservation.nr_extents = 1;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
&reservation);
WARN(ret != 1, "Failed to release pfn %lx err=%d\n", pfn, ret);
if (ret == 1) {
__set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
len++;
}
}
printk(KERN_INFO "Freeing %lx-%lx pfn range: %lu pages freed\n",
start, end, len);
return len;
}
static unsigned long __init xen_set_identity_and_release(
const struct e820entry *list, size_t map_size, unsigned long nr_pages)
{
phys_addr_t start = 0;
unsigned long released = 0;
unsigned long identity = 0;
const struct e820entry *entry;
int i;
/*
* Combine non-RAM regions and gaps until a RAM region (or the
* end of the map) is reached, then set the 1:1 map and
* release the pages (if available) in those non-RAM regions.
*
* The combined non-RAM regions are rounded to a whole number
* of pages so any partial pages are accessible via the 1:1
* mapping. This is needed for some BIOSes that put (for
* example) the DMI tables in a reserved region that begins on
* a non-page boundary.
*/
for (i = 0, entry = list; i < map_size; i++, entry++) {
phys_addr_t end = entry->addr + entry->size;
if (entry->type == E820_RAM || i == map_size - 1) {
unsigned long start_pfn = PFN_DOWN(start);
unsigned long end_pfn = PFN_UP(end);
if (entry->type == E820_RAM)
end_pfn = PFN_UP(entry->addr);
if (start_pfn < end_pfn) {
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);
}
start = end;
}
}
printk(KERN_INFO "Released %lu pages of unused memory\n", released);
printk(KERN_INFO "Set %ld page(s) to 1-1 mapping\n", identity);
return released;
}
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_release_chunk(unsigned long start,
unsigned long end)
{
struct xen_memory_reservation reservation = {
.address_bits = 0,
.extent_order = 0,
.domid = DOMID_SELF
};
unsigned long len = 0;
unsigned long pfn;
int ret;
for(pfn = start; pfn < end; pfn++) {
unsigned long mfn = pfn_to_mfn(pfn);
/* */
if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn)
continue;
set_xen_guest_handle(reservation.extent_start, &mfn);
reservation.nr_extents = 1;
ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
&reservation);
WARN(ret != 1, "Failed to release pfn %lx err=%d\n", pfn, ret);
if (ret == 1) {
__set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
len++;
}
}
printk(KERN_INFO "Freeing %lx-%lx pfn range: %lu pages freed\n",
start, end, len);
return len;
}
static unsigned long __init xen_set_identity_and_release(
const struct e820entry *list, size_t map_size, unsigned long nr_pages)
{
phys_addr_t start = 0;
unsigned long released = 0;
unsigned long identity = 0;
const struct e820entry *entry;
int i;
/*
*/
for (i = 0, entry = list; i < map_size; i++, entry++) {
phys_addr_t end = entry->addr + entry->size;
if (entry->type == E820_RAM || i == map_size - 1) {
unsigned long start_pfn = PFN_DOWN(start);
unsigned long end_pfn = PFN_UP(end);
if (entry->type == E820_RAM)
end_pfn = PFN_UP(entry->addr);
if (start_pfn < end_pfn) {
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);
}
start = end;
}
}
printk(KERN_INFO "Released %lu pages of unused memory\n", released);
printk(KERN_INFO "Set %ld page(s) to 1-1 mapping\n", identity);
return released;
}
static int __do_suspend(void *ignore)
{
int i, j, k, fpp, err;
extern unsigned long max_pfn;
extern unsigned long *pfn_to_mfn_frame_list_list;
extern unsigned long *pfn_to_mfn_frame_list[];
extern void time_resume(void);
BUG_ON(smp_processor_id() != 0);
BUG_ON(in_interrupt());
if (xen_feature(XENFEAT_auto_translated_physmap)) {
printk(KERN_WARNING "Cannot suspend in "
"auto_translated_physmap mode.\n");
return -EOPNOTSUPP;
}
err = smp_suspend();
if (err)
return err;
xenbus_suspend();
preempt_disable();
#ifdef __i386__
kmem_cache_shrink(pgd_cache);
#endif
mm_pin_all();
__cli();
preempt_enable();
gnttab_suspend();
HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
clear_fixmap(FIX_SHARED_INFO);
xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
xen_start_info->console_mfn = mfn_to_pfn(xen_start_info->console_mfn);
/*
* We'll stop somewhere inside this hypercall. When it returns,
* we'll start resuming after the restore.
*/
HYPERVISOR_suspend(virt_to_mfn(xen_start_info));
shutting_down = SHUTDOWN_INVALID;
set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
memset(empty_zero_page, 0, 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 < max_pfn; i += fpp, j++) {
if ((j % fpp) == 0) {
k++;
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 = max_pfn;
gnttab_resume();
irq_resume();
time_resume();
switch_idle_mm();
__sti();
xencons_resume();
xenbus_resume();
smp_resume();
return err;
}
//void free_pte_page(struct mmu_gather* tlb, struct page *pte)
void free_pte_page(struct page *pte)
{
struct ptrpte_t *newstruct = NULL;
struct ptrpte_t *temp_head = NULL;
int i = 0;
int counter = 0;
newstruct = (struct ptrpte_t *)kmalloc(sizeof(struct ptrpte_t), GFP_KERNEL);
newstruct -> content = pte;
//newstruct -> mmu_tlb = tlb;
spin_lock(&pte_cache_lock);
newstruct -> next = pte_head;
pte_head = newstruct;
temp_head = pte_head;
/*free node */
if(pte_used_counter)
pte_used_counter--;
pte_free_counter++;
//spin_unlock(&pte_cache_lock);
if(pte_used_counter)
{
//if((pte_free_counter/pte_used_counter>=8) && ((pte_used_counter + pte_free_counter) >= 2100))
//if(pte_used_counter + pte_free_counter >= 2100)
//if((pte_used_counter/pte_free_counter < 1) && (pte_used_counter >= 63))
if((pte_free_counter/pte_used_counter > 4) && ((pte_used_counter + pte_free_counter) >= 320))
//if((pte_free_counter/pte_used_counter >= 5) && ((pte_used_counter + pte_free_counter) >= 320))
{
printk("pte free counter is %ld\n", pte_free_counter);
printk("pte used counter is %ld\n", pte_used_counter);
counter = pte_free_counter * 3 / 10;
//counter = 0;
for(i=0;i<counter;i++)
{
pte_head = pte_head->next;
}
pte_free_counter -= counter;
}
}
spin_unlock(&pte_cache_lock);
if(counter != 0)
{
struct ptrpte * newstructarray = NULL;
struct ptrpte * newstructarray_head = NULL;
int rc = 1;
newstructarray = (struct ptrpte *)kmalloc(sizeof(struct ptrpte) * counter, GFP_KERNEL);
newstructarray_head = newstructarray;
for (i=0;i<counter;i++)
{
newstruct = temp_head;
temp_head = temp_head->next;
//newstructarray[i].content = (unsigned long)page_address(newstruct->content);
newstructarray[i].content = pfn_to_mfn(page_to_pfn(newstruct->content));
//newstructarray[i].mmu_tlb = newstruct->mmu_tlb;
kfree(newstruct);
}
//hypercall newstructarray
rc = HYPERVISOR_pte_op(newstructarray, counter);
//if (rc == 0)
//else
//printk("pte cache free error\n");
//free page to the buddy system
newstructarray = newstructarray_head;
for(i=0;i<counter;i++)
{
//tlb_remove_page(newstructarray[i].mmu_tlb,newstructarray[i].content);
//__free_page(newstructarray[i].content);
__free_page(pfn_to_page(mfn_to_pfn(newstructarray[i].content)));
}
//free newstructarray
kfree(newstructarray);
}
return;
}
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);
}
