|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Xen-devel] [PATCH v5 4/4] x86/ioreq server: Reset outstanding p2m_ioreq_server entries when an ioreq server unmaps.
This patch resets p2m_ioreq_server entries back to p2m_ram_rw,
after an ioreq server has unmapped. The resync is done both
asynchronously with the current p2m_change_entry_type_global()
interface, and synchronously by iterating the p2m table. The
synchronous resetting is necessary because we need to guarantee
the p2m table is clean before another ioreq server is mapped.
And since the sweeping of p2m table could be time consuming,
it is done with hypercall continuation. Asynchronous approach
is also taken so that p2m_ioreq_server entries can also be reset
when the HVM is scheduled between hypercall continuations.
This patch also disallows live migration, when there's still any
outstanding p2m_ioreq_server entry left. The core reason is our
current implementation of p2m_change_entry_type_global() can not
tell the state of p2m_ioreq_server entries(can not decide if an
entry is to be emulated or to be resynced).
Signed-off-by: Yu Zhang <yu.c.zhang@xxxxxxxxxxxxxxx>
---
Cc: Paul Durrant <paul.durrant@xxxxxxxxxx>
Cc: Jan Beulich <jbeulich@xxxxxxxx>
Cc: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
Cc: George Dunlap <george.dunlap@xxxxxxxxxxxxx>
Cc: Jun Nakajima <jun.nakajima@xxxxxxxxx>
Cc: Kevin Tian <kevin.tian@xxxxxxxxx>
---
xen/arch/x86/hvm/hvm.c | 52 ++++++++++++++++++++++++++++++++++++++++++++---
xen/arch/x86/mm/hap/hap.c | 9 ++++++++
xen/arch/x86/mm/p2m-ept.c | 6 +++++-
xen/arch/x86/mm/p2m-pt.c | 10 +++++++--
xen/arch/x86/mm/p2m.c | 3 +++
xen/include/asm-x86/p2m.h | 5 ++++-
6 files changed, 78 insertions(+), 7 deletions(-)
diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
index 4d98cc6..e57c8b9 100644
--- a/xen/arch/x86/hvm/hvm.c
+++ b/xen/arch/x86/hvm/hvm.c
@@ -5485,6 +5485,7 @@ static int hvmop_set_mem_type(
{
unsigned long pfn = a.first_pfn + start_iter;
p2m_type_t t;
+ struct p2m_domain *p2m = p2m_get_hostp2m(d);
get_gfn_unshare(d, pfn, &t);
if ( p2m_is_paging(t) )
@@ -5512,6 +5513,12 @@ static int hvmop_set_mem_type(
if ( rc )
goto out;
+ if ( t == p2m_ram_rw && memtype[a.hvmmem_type] == p2m_ioreq_server )
+ p2m->ioreq.entry_count++;
+
+ if ( t == p2m_ioreq_server && memtype[a.hvmmem_type] == p2m_ram_rw )
+ p2m->ioreq.entry_count--;
+
/* Check for continuation if it's not the last interation */
if ( a.nr > ++start_iter && !(start_iter & HVMOP_op_mask) &&
hypercall_preempt_check() )
@@ -5530,11 +5537,13 @@ static int hvmop_set_mem_type(
}
static int hvmop_map_mem_type_to_ioreq_server(
- XEN_GUEST_HANDLE_PARAM(xen_hvm_map_mem_type_to_ioreq_server_t) uop)
+ XEN_GUEST_HANDLE_PARAM(xen_hvm_map_mem_type_to_ioreq_server_t) uop,
+ unsigned long *iter)
{
xen_hvm_map_mem_type_to_ioreq_server_t op;
struct domain *d;
int rc;
+ unsigned long gfn = *iter;
if ( copy_from_guest(&op, uop, 1) )
return -EFAULT;
@@ -5559,7 +5568,42 @@ static int hvmop_map_mem_type_to_ioreq_server(
if ( rc != 0 )
goto out;
- rc = hvm_map_mem_type_to_ioreq_server(d, op.id, op.type, op.flags);
+ if ( gfn == 0 || op.flags != 0 )
+ rc = hvm_map_mem_type_to_ioreq_server(d, op.id, op.type, op.flags);
+
+ /*
+ * Iterate p2m table when an ioreq server unmaps from p2m_ioreq_server,
+ * and reset the remaining p2m_ioreq_server entries back to p2m_ram_rw.
+ */
+ if ( op.flags == 0 && rc == 0 )
+ {
+ struct p2m_domain *p2m = p2m_get_hostp2m(d);
+
+ while ( gfn <= p2m->max_mapped_pfn )
+ {
+ p2m_type_t t;
+
+ if ( p2m->ioreq.entry_count == 0 )
+ break;
+
+ get_gfn_unshare(d, gfn, &t);
+
+ if ( (t == p2m_ioreq_server) &&
+ !(p2m_change_type_one(d, gfn, t, p2m_ram_rw)) )
+ p2m->ioreq.entry_count--;
+
+ put_gfn(d, gfn);
+
+ /* Check for continuation if it's not the last iteration. */
+ if ( ++gfn <= p2m->max_mapped_pfn &&
+ !(gfn & HVMOP_op_mask) &&
+ hypercall_preempt_check() )
+ {
+ rc = -ERESTART;
+ goto out;
+ }
+ }
+ }
out:
rcu_unlock_domain(d);
@@ -5578,6 +5622,7 @@ long do_hvm_op(unsigned long op,
XEN_GUEST_HANDLE_PARAM(void) arg)
break;
case HVMOP_modified_memory:
case HVMOP_set_mem_type:
+ case HVMOP_map_mem_type_to_ioreq_server:
mask = HVMOP_op_mask;
break;
}
@@ -5607,7 +5652,8 @@ long do_hvm_op(unsigned long op,
XEN_GUEST_HANDLE_PARAM(void) arg)
case HVMOP_map_mem_type_to_ioreq_server:
rc = hvmop_map_mem_type_to_ioreq_server(
- guest_handle_cast(arg, xen_hvm_map_mem_type_to_ioreq_server_t));
+ guest_handle_cast(arg, xen_hvm_map_mem_type_to_ioreq_server_t),
+ &start_iter);
break;
case HVMOP_set_ioreq_server_state:
diff --git a/xen/arch/x86/mm/hap/hap.c b/xen/arch/x86/mm/hap/hap.c
index 9c2cd49..0442b1d 100644
--- a/xen/arch/x86/mm/hap/hap.c
+++ b/xen/arch/x86/mm/hap/hap.c
@@ -190,6 +190,15 @@ out:
*/
static int hap_enable_log_dirty(struct domain *d, bool_t log_global)
{
+ struct p2m_domain *p2m = p2m_get_hostp2m(d);
+
+ /*
+ * Refuse to turn on global log-dirty mode if
+ * there's outstanding p2m_ioreq_server pages.
+ */
+ if ( log_global && p2m->ioreq.entry_count > 0 )
+ return -EBUSY;
+
/* turn on PG_log_dirty bit in paging mode */
paging_lock(d);
d->arch.paging.mode |= PG_log_dirty;
diff --git a/xen/arch/x86/mm/p2m-ept.c b/xen/arch/x86/mm/p2m-ept.c
index 5f06d40..5d4d804 100644
--- a/xen/arch/x86/mm/p2m-ept.c
+++ b/xen/arch/x86/mm/p2m-ept.c
@@ -545,6 +545,9 @@ static int resolve_misconfig(struct p2m_domain *p2m,
unsigned long gfn)
e.ipat = ipat;
if ( e.recalc && p2m_is_changeable(e.sa_p2mt) )
{
+ if ( e.sa_p2mt == p2m_ioreq_server )
+ p2m->ioreq.entry_count--;
+
e.sa_p2mt = p2m_is_logdirty_range(p2m, gfn + i, gfn +
i)
? p2m_ram_logdirty : p2m_ram_rw;
ept_p2m_type_to_flags(p2m, &e, e.sa_p2mt, e.access);
@@ -965,7 +968,8 @@ static mfn_t ept_get_entry(struct p2m_domain *p2m,
if ( is_epte_valid(ept_entry) )
{
if ( (recalc || ept_entry->recalc) &&
- p2m_is_changeable(ept_entry->sa_p2mt) )
+ p2m_is_changeable(ept_entry->sa_p2mt) &&
+ (ept_entry->sa_p2mt != p2m_ioreq_server) )
*t = p2m_is_logdirty_range(p2m, gfn, gfn) ? p2m_ram_logdirty
: p2m_ram_rw;
else
diff --git a/xen/arch/x86/mm/p2m-pt.c b/xen/arch/x86/mm/p2m-pt.c
index 6209e7b..7bebfd1 100644
--- a/xen/arch/x86/mm/p2m-pt.c
+++ b/xen/arch/x86/mm/p2m-pt.c
@@ -439,11 +439,13 @@ static int do_recalc(struct p2m_domain *p2m, unsigned
long gfn)
needs_recalc(l1, *pent) )
{
l1_pgentry_t e = *pent;
+ p2m_type_t p2mt_old;
if ( !valid_recalc(l1, e) )
P2M_DEBUG("bogus recalc leaf at d%d:%lx:%u\n",
p2m->domain->domain_id, gfn, level);
- if ( p2m_is_changeable(p2m_flags_to_type(l1e_get_flags(e))) )
+ p2mt_old = p2m_flags_to_type(l1e_get_flags(e));
+ if ( p2m_is_changeable(p2mt_old) )
{
unsigned long mask = ~0UL << (level * PAGETABLE_ORDER);
p2m_type_t p2mt = p2m_is_logdirty_range(p2m, gfn & mask, gfn |
~mask)
@@ -463,6 +465,10 @@ static int do_recalc(struct p2m_domain *p2m, unsigned long
gfn)
mfn &= ~(_PAGE_PSE_PAT >> PAGE_SHIFT);
flags |= _PAGE_PSE;
}
+
+ if ( p2mt_old == p2m_ioreq_server )
+ p2m->ioreq.entry_count--;
+
e = l1e_from_pfn(mfn, flags);
p2m_add_iommu_flags(&e, level,
(p2mt == p2m_ram_rw)
@@ -729,7 +735,7 @@ p2m_pt_set_entry(struct p2m_domain *p2m, unsigned long gfn,
mfn_t mfn,
static inline p2m_type_t recalc_type(bool_t recalc, p2m_type_t t,
struct p2m_domain *p2m, unsigned long gfn)
{
- if ( !recalc || !p2m_is_changeable(t) )
+ if ( !recalc || !p2m_is_changeable(t) || (t == p2m_ioreq_server) )
return t;
return p2m_is_logdirty_range(p2m, gfn, gfn) ? p2m_ram_logdirty
: p2m_ram_rw;
diff --git a/xen/arch/x86/mm/p2m.c b/xen/arch/x86/mm/p2m.c
index 5567181..e1c3e31 100644
--- a/xen/arch/x86/mm/p2m.c
+++ b/xen/arch/x86/mm/p2m.c
@@ -313,6 +313,9 @@ int p2m_set_ioreq_server(struct domain *d,
p2m->ioreq.server = NULL;
p2m->ioreq.flags = 0;
+
+ if ( p2m->ioreq.entry_count > 0 )
+ p2m_change_entry_type_global(d, p2m_ioreq_server, p2m_ram_rw);
}
else
{
diff --git a/xen/include/asm-x86/p2m.h b/xen/include/asm-x86/p2m.h
index 0950a91..8e5b4f5 100644
--- a/xen/include/asm-x86/p2m.h
+++ b/xen/include/asm-x86/p2m.h
@@ -120,7 +120,8 @@ typedef unsigned int p2m_query_t;
/* Types that can be subject to bulk transitions. */
#define P2M_CHANGEABLE_TYPES (p2m_to_mask(p2m_ram_rw) \
- | p2m_to_mask(p2m_ram_logdirty) )
+ | p2m_to_mask(p2m_ram_logdirty) \
+ | p2m_to_mask(p2m_ioreq_server))
#define P2M_POD_TYPES (p2m_to_mask(p2m_populate_on_demand))
@@ -352,6 +353,8 @@ struct p2m_domain {
#define P2M_IOREQ_HANDLE_WRITE_ACCESS XEN_HVMOP_IOREQ_MEM_ACCESS_WRITE
#define P2M_IOREQ_HANDLE_READ_ACCESS XEN_HVMOP_IOREQ_MEM_ACCESS_READ
+
+ unsigned int entry_count;
} ioreq;
};
--
1.9.1
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel
|
 |
Lists.xenproject.org is hosted with RackSpace, monitoring our |