/**
* p2m_mem_paging_nominate - Mark a guest page as to-be-paged-out
* @d: guest domain
* @gfn: guest page to nominate
*
* Returns 0 for success or negative errno values if gfn is not pageable.
*
* p2m_mem_paging_nominate() is called by the pager and checks if a guest page
* can be paged out. If the following conditions are met the p2mt will be
* changed:
* - the gfn is backed by a mfn
* - the p2mt of the gfn is pageable
* - the mfn is not used for IO
* - the mfn has exactly one user and has no special meaning
*
* Once the p2mt is changed the page is readonly for the guest. On success the
* pager can write the page contents to disk and later evict the page.
*/
int p2m_mem_paging_nominate(struct domain *d, unsigned long gfn)
{
struct page_info *page;
struct p2m_domain *p2m = p2m_get_hostp2m(d);
p2m_type_t p2mt;
p2m_access_t a;
mfn_t mfn;
int ret = -EBUSY;
gfn_lock(p2m, gfn, 0);
mfn = p2m->get_entry(p2m, gfn, &p2mt, &a, 0, NULL);
/* Check if mfn is valid */
if ( !mfn_valid(mfn) )
goto out;
/* Check p2m type */
if ( !p2m_is_pageable(p2mt) )
goto out;
/* Check for io memory page */
if ( is_iomem_page(mfn_x(mfn)) )
goto out;
/* Check page count and type */
page = mfn_to_page(mfn);
if ( (page->count_info & (PGC_count_mask | PGC_allocated)) !=
(1 | PGC_allocated) )
goto out;
if ( (page->u.inuse.type_info & PGT_count_mask) != 0 )
goto out;
/* Fix p2m entry */
set_p2m_entry(p2m, gfn, mfn, PAGE_ORDER_4K, p2m_ram_paging_out, a);
ret = 0;
out:
gfn_unlock(p2m, gfn, 0);
return ret;
}
/*
* Returns 0 if TLB flush / invalidate required by caller.
* va will indicate the address to be invalidated.
*
* addr is _either_ a host virtual address, or the address of the pte to
* update, as indicated by the GNTMAP_contains_pte flag.
*/
static void
__gnttab_map_grant_ref(
struct gnttab_map_grant_ref *op)
{
struct domain *ld, *rd;
struct vcpu *led;
int handle;
unsigned long frame = 0;
int rc = GNTST_okay;
unsigned int cache_flags;
struct active_grant_entry *act;
struct grant_mapping *mt;
grant_entry_t *sha;
union grant_combo scombo, prev_scombo, new_scombo;
/*
* We bound the number of times we retry CMPXCHG on memory locations that
* we share with a guest OS. The reason is that the guest can modify that
* location at a higher rate than we can read-modify-CMPXCHG, so the guest
* could cause us to livelock. There are a few cases where it is valid for
* the guest to race our updates (e.g., to change the GTF_readonly flag),
* so we allow a few retries before failing.
*/
int retries = 0;
led = current;
ld = led->domain;
if ( unlikely((op->flags & (GNTMAP_device_map|GNTMAP_host_map)) == 0) )
{
gdprintk(XENLOG_INFO, "Bad flags in grant map op (%x).\n", op->flags);
op->status = GNTST_bad_gntref;
return;
}
if ( unlikely((rd = rcu_lock_domain_by_id(op->dom)) == NULL) )
{
gdprintk(XENLOG_INFO, "Could not find domain %d\n", op->dom);
op->status = GNTST_bad_domain;
return;
}
rc = xsm_grant_mapref(ld, rd, op->flags);
if ( rc )
{
rcu_unlock_domain(rd);
op->status = GNTST_permission_denied;
return;
}
if ( unlikely((handle = get_maptrack_handle(ld->grant_table)) == -1) )
{
rcu_unlock_domain(rd);
gdprintk(XENLOG_INFO, "Failed to obtain maptrack handle.\n");
op->status = GNTST_no_device_space;
return;
}
spin_lock(&rd->grant_table->lock);
/* Bounds check on the grant ref */
if ( unlikely(op->ref >= nr_grant_entries(rd->grant_table)))
PIN_FAIL(unlock_out, GNTST_bad_gntref, "Bad ref (%d).\n", op->ref);
act = &active_entry(rd->grant_table, op->ref);
sha = &shared_entry(rd->grant_table, op->ref);
/* If already pinned, check the active domid and avoid refcnt overflow. */
if ( act->pin &&
((act->domid != ld->domain_id) ||
(act->pin & 0x80808080U) != 0) )
PIN_FAIL(unlock_out, GNTST_general_error,
"Bad domain (%d != %d), or risk of counter overflow %08x\n",
act->domid, ld->domain_id, act->pin);
if ( !act->pin ||
(!(op->flags & GNTMAP_readonly) &&
!(act->pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask))) )
{
scombo.word = *(u32 *)&sha->flags;
/*
* This loop attempts to set the access (reading/writing) flags
* in the grant table entry. It tries a cmpxchg on the field
* up to five times, and then fails under the assumption that
* the guest is misbehaving.
*/
for ( ; ; )
{
/* If not already pinned, check the grant domid and type. */
if ( !act->pin &&
(((scombo.shorts.flags & GTF_type_mask) !=
GTF_permit_access) ||
(scombo.shorts.domid != ld->domain_id)) )
PIN_FAIL(unlock_out, GNTST_general_error,
"Bad flags (%x) or dom (%d). (expected dom %d)\n",
scombo.shorts.flags, scombo.shorts.domid,
ld->domain_id);
new_scombo = scombo;
new_scombo.shorts.flags |= GTF_reading;
if ( !(op->flags & GNTMAP_readonly) )
{
new_scombo.shorts.flags |= GTF_writing;
if ( unlikely(scombo.shorts.flags & GTF_readonly) )
PIN_FAIL(unlock_out, GNTST_general_error,
"Attempt to write-pin a r/o grant entry.\n");
}
prev_scombo.word = cmpxchg((u32 *)&sha->flags,
scombo.word, new_scombo.word);
if ( likely(prev_scombo.word == scombo.word) )
break;
if ( retries++ == 4 )
PIN_FAIL(unlock_out, GNTST_general_error,
"Shared grant entry is unstable.\n");
scombo = prev_scombo;
}
if ( !act->pin )
{
act->domid = scombo.shorts.domid;
act->frame = gmfn_to_mfn(rd, sha->frame);
}
}
if ( op->flags & GNTMAP_device_map )
act->pin += (op->flags & GNTMAP_readonly) ?
GNTPIN_devr_inc : GNTPIN_devw_inc;
if ( op->flags & GNTMAP_host_map )
act->pin += (op->flags & GNTMAP_readonly) ?
GNTPIN_hstr_inc : GNTPIN_hstw_inc;
frame = act->frame;
cache_flags = (sha->flags & (GTF_PAT | GTF_PWT | GTF_PCD) );
spin_unlock(&rd->grant_table->lock);
if ( is_iomem_page(frame) )
{
if ( !iomem_access_permitted(rd, frame, frame) )
{
gdprintk(XENLOG_WARNING,
"Iomem mapping not permitted %lx (domain %d)\n",
frame, rd->domain_id);
rc = GNTST_general_error;
goto undo_out;
}
rc = create_grant_host_mapping(
op->host_addr, frame, op->flags, cache_flags);
if ( rc != GNTST_okay )
goto undo_out;
}
else
{
if ( unlikely(!mfn_valid(frame)) ||
unlikely(!(gnttab_host_mapping_get_page_type(op, ld, rd) ?
get_page_and_type(mfn_to_page(frame), rd,
PGT_writable_page) :
get_page(mfn_to_page(frame), rd))) )
{
if ( !rd->is_dying )
gdprintk(XENLOG_WARNING, "Could not pin grant frame %lx\n",
frame);
rc = GNTST_general_error;
goto undo_out;
}
if ( op->flags & GNTMAP_host_map )
{
rc = create_grant_host_mapping(op->host_addr, frame, op->flags, 0);
if ( rc != GNTST_okay )
{
if ( gnttab_host_mapping_get_page_type(op, ld, rd) )
put_page_type(mfn_to_page(frame));
put_page(mfn_to_page(frame));
goto undo_out;
}
if ( op->flags & GNTMAP_device_map )
{
(void)get_page(mfn_to_page(frame), rd);
if ( !(op->flags & GNTMAP_readonly) )
get_page_type(mfn_to_page(frame), PGT_writable_page);
}
}
}
TRACE_1D(TRC_MEM_PAGE_GRANT_MAP, op->dom);
mt = &maptrack_entry(ld->grant_table, handle);
mt->domid = op->dom;
mt->ref = op->ref;
mt->flags = op->flags;
op->dev_bus_addr = (u64)frame << PAGE_SHIFT;
op->handle = handle;
op->status = GNTST_okay;
rcu_unlock_domain(rd);
return;
undo_out:
spin_lock(&rd->grant_table->lock);
act = &active_entry(rd->grant_table, op->ref);
sha = &shared_entry(rd->grant_table, op->ref);
if ( op->flags & GNTMAP_device_map )
act->pin -= (op->flags & GNTMAP_readonly) ?
GNTPIN_devr_inc : GNTPIN_devw_inc;
if ( op->flags & GNTMAP_host_map )
act->pin -= (op->flags & GNTMAP_readonly) ?
GNTPIN_hstr_inc : GNTPIN_hstw_inc;
if ( !(op->flags & GNTMAP_readonly) &&
!(act->pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
gnttab_clear_flag(_GTF_writing, &sha->flags);
if ( !act->pin )
gnttab_clear_flag(_GTF_reading, &sha->flags);
unlock_out:
spin_unlock(&rd->grant_table->lock);
op->status = rc;
put_maptrack_handle(ld->grant_table, handle);
rcu_unlock_domain(rd);
}