[PATCH v5 0/2] gnttab: Improve scaleability

[PATCH v5 0/2] gnttab: Improve scaleability

[Christoph Egger]

This patch series changes the grant table locking to
a more fain grained locking protocol. The result is
a performance boost measured with blkfront/blkback.
Document the locking protocol.

v5:
  * Addressed locking issue pointed out by Jan Beulich
  * Fixed git rebase merge issue introduced in v4
    (acquiring locking twice)
  * Change for ()-loop in grant_map_exists
  * Coding style fixes
v4:
  * Coding style nits from Jan Beulich
  * Fixup read locks pointed out by Jan Beulich
  * renamed double_gt_(un)lock to double_maptrack_(un)lock
    per request from Jan Beulich
  * Addressed ASSERT()'s from Jan Beulich

v3:
  * Addressed gnttab_swap_grant_ref() comment from Andrew Cooper
v2:
  * Add arm part per request from Julien Grall

Christoph Egger (1):
  gnttab: Introduce rwlock to protect updates to grant table state

Matt Wilson (1):
  gnttab: refactor locking for scalability

 docs/misc/grant-tables.txt    |   49 ++++++-
 xen/arch/arm/mm.c             |    4 +-
 xen/arch/x86/mm.c             |    4 +-
 xen/common/grant_table.c      |  321 +++++++++++++++++++++++++----------------
 xen/include/xen/grant_table.h |    9 +-
 5 files changed, 258 insertions(+), 129 deletions(-)

-- 
1.7.9.5

[PATCH v5 1/2] gnttab: Introduce rwlock to protect updates to grant table state

[Christoph Egger]

Split grant table lock into two separate locks. One to protect
maptrack state and change the other into a rwlock.

The rwlock is used to prevent readers from accessing
inconsistent grant table state such as current
version, partially initialized active table pages,
etc.

Signed-off-by: Matt Wilson <msw@amazon.com>
[chegger: ported to xen-staging, split into multiple commits]

v5:
  * Addressed locking issue pointed out by Jan Beulich
  * Fixed git rebase merge issue introduced in v4 
    (acquiring locking twice)
  * Coding style fixes
v4:
  * Coding style nits from Jan Beulich
  * Fixup read locks pointed out by Jan Beulich
  * renamed double_gt_(un)lock to double_maptrack_(un)lock
    per request from Jan Beulich
v3:
  * Addressed gnttab_swap_grant_ref() comment from Andrew Cooper
v2:
  * Add arm part per request from Julien Grall

Signed-off-by: Christoph Egger <chegger@amazon.de>
Cc: Jan Beulich <jbeulich@suse.com>
Cc: Keir Fraser <keir@xen.org>
Cc: Julien Grall <julien.grall@linaro.org>

Adjust locking
---
 docs/misc/grant-tables.txt    |   28 +++++++-
 xen/arch/arm/mm.c             |    4 +-
 xen/arch/x86/mm.c             |    4 +-
 xen/common/grant_table.c      |  146 ++++++++++++++++++++++++-----------------
 xen/include/xen/grant_table.h |   17 +++--
 5 files changed, 127 insertions(+), 72 deletions(-)

diff --git a/docs/misc/grant-tables.txt b/docs/misc/grant-tables.txt
index 19db4ec..c9ae8f2 100644
--- a/docs/misc/grant-tables.txt
+++ b/docs/misc/grant-tables.txt
@@ -74,7 +74,33 @@ is complete.
  matching map track entry is then removed, as if unmap had been invoked.
  These are not used by the transfer mechanism.
   map->domid         : owner of the mapped frame
-  map->ref_and_flags : grant reference, ro/rw, mapped for host or device access
+  map->ref           : grant reference
+  map->flags         : ro/rw, mapped for host or device access
+
+********************************************************************************
+ Locking
+ ~~~~~~~
+ Xen uses several locks to serialize access to the internal grant table state.
+
+  grant_table->lock          : rwlock used to prevent readers from accessing
+                               inconsistent grant table state such as current
+                               version, partially initialized active table pages,
+                               etc.
+  grant_table->maptrack_lock : spinlock used to protect the maptrack state
+
+ The primary lock for the grant table is a read/write spinlock. All
+ functions that access members of struct grant_table must acquire a
+ read lock around critical sections. Any modification to the members
+ of struct grant_table (e.g., nr_status_frames, nr_grant_frames,
+ active frames, etc.) must only be made if the write lock is
+ held. These elements are read-mostly, and read critical sections can
+ be large, which makes a rwlock a good choice.
+
+ The maptrack state is protected by its own spinlock. Any access (read
+ or write) of struct grant_table members that have a "maptrack_"
+ prefix must be made while holding the maptrack lock. The maptrack
+ state can be rapidly modified under some workloads, and the critical
+ sections are very small, thus we use a spinlock to protect them.
 
 ********************************************************************************
 
diff --git a/xen/arch/arm/mm.c b/xen/arch/arm/mm.c
index 7d4ba0c..2765683 100644
--- a/xen/arch/arm/mm.c
+++ b/xen/arch/arm/mm.c
@@ -1037,7 +1037,7 @@ int xenmem_add_to_physmap_one(
     switch ( space )
     {
     case XENMAPSPACE_grant_table:
-        spin_lock(&d->grant_table->lock);
+        write_lock(&d->grant_table->lock);
 
         if ( d->grant_table->gt_version == 0 )
             d->grant_table->gt_version = 1;
@@ -1067,7 +1067,7 @@ int xenmem_add_to_physmap_one(
 
         t = p2m_ram_rw;
 
-        spin_unlock(&d->grant_table->lock);
+        write_unlock(&d->grant_table->lock);
         break;
     case XENMAPSPACE_shared_info:
         if ( idx != 0 )
diff --git a/xen/arch/x86/mm.c b/xen/arch/x86/mm.c
index d4965da..af4762d 100644
--- a/xen/arch/x86/mm.c
+++ b/xen/arch/x86/mm.c
@@ -4569,7 +4569,7 @@ int xenmem_add_to_physmap_one(
                 mfn = virt_to_mfn(d->shared_info);
             break;
         case XENMAPSPACE_grant_table:
-            spin_lock(&d->grant_table->lock);
+            write_lock(&d->grant_table->lock);
 
             if ( d->grant_table->gt_version == 0 )
                 d->grant_table->gt_version = 1;
@@ -4591,7 +4591,7 @@ int xenmem_add_to_physmap_one(
                     mfn = virt_to_mfn(d->grant_table->shared_raw[idx]);
             }
 
-            spin_unlock(&d->grant_table->lock);
+            write_unlock(&d->grant_table->lock);
             break;
         case XENMAPSPACE_gmfn_range:
         case XENMAPSPACE_gmfn:
diff --git a/xen/common/grant_table.c b/xen/common/grant_table.c
index 1a11766..f52a51d 100644
--- a/xen/common/grant_table.c
+++ b/xen/common/grant_table.c
@@ -229,27 +229,27 @@ static int __get_paged_frame(unsigned long gfn, unsigned long *frame, struct pag
 }
 
 static inline void
-double_gt_lock(struct grant_table *lgt, struct grant_table *rgt)
+double_maptrack_lock(struct grant_table *lgt, struct grant_table *rgt)
 {
     if ( lgt < rgt )
     {
-        spin_lock(&lgt->lock);
-        spin_lock(&rgt->lock);
+        spin_lock(&lgt->maptrack_lock);
+        spin_lock(&rgt->maptrack_lock);
     }
     else
     {
         if ( lgt != rgt )
-            spin_lock(&rgt->lock);
-        spin_lock(&lgt->lock);
+            spin_lock(&rgt->maptrack_lock);
+        spin_lock(&lgt->maptrack_lock);
     }
 }
 
 static inline void
-double_gt_unlock(struct grant_table *lgt, struct grant_table *rgt)
+double_maptrack_unlock(struct grant_table *lgt, struct grant_table *rgt)
 {
-    spin_unlock(&lgt->lock);
+    spin_unlock(&lgt->maptrack_lock);
     if ( lgt != rgt )
-        spin_unlock(&rgt->lock);
+        spin_unlock(&rgt->maptrack_lock);
 }
 
 static inline int
@@ -267,10 +267,10 @@ static inline void
 put_maptrack_handle(
     struct grant_table *t, int handle)
 {
-    spin_lock(&t->lock);
+    spin_lock(&t->maptrack_lock);
     maptrack_entry(t, handle).ref = t->maptrack_head;
     t->maptrack_head = handle;
-    spin_unlock(&t->lock);
+    spin_unlock(&t->maptrack_lock);
 }
 
 static inline int
@@ -282,7 +282,7 @@ get_maptrack_handle(
     struct grant_mapping *new_mt;
     unsigned int          new_mt_limit, nr_frames;
 
-    spin_lock(&lgt->lock);
+    spin_lock(&lgt->maptrack_lock);
 
     while ( unlikely((handle = __get_maptrack_handle(lgt)) == -1) )
     {
@@ -311,7 +311,7 @@ get_maptrack_handle(
                  nr_frames + 1);
     }
 
-    spin_unlock(&lgt->lock);
+    spin_unlock(&lgt->maptrack_lock);
 
     return handle;
 }
@@ -508,7 +508,7 @@ static int grant_map_exists(const struct domain *ld,
     const struct active_grant_entry *act;
     unsigned int ref, max_iter;
     
-    ASSERT(spin_is_locked(&rgt->lock));
+    ASSERT(rw_is_locked(&rgt->lock));
 
     max_iter = min(*ref_count + (1 << GNTTABOP_CONTINUATION_ARG_SHIFT),
                    nr_grant_entries(rgt));
@@ -629,7 +629,7 @@ __gnttab_map_grant_ref(
     }
 
     rgt = rd->grant_table;
-    spin_lock(&rgt->lock);
+    read_lock(&rgt->lock);
 
     if ( rgt->gt_version == 0 )
         PIN_FAIL(unlock_out, GNTST_general_error,
@@ -702,8 +702,6 @@ __gnttab_map_grant_ref(
 
     cache_flags = (shah->flags & (GTF_PAT | GTF_PWT | GTF_PCD) );
 
-    spin_unlock(&rgt->lock);
-
     /* pg may be set, with a refcount included, from __get_paged_frame */
     if ( !pg )
     {
@@ -778,7 +776,7 @@ __gnttab_map_grant_ref(
         goto undo_out;
     }
 
-    double_gt_lock(lgt, rgt);
+    double_maptrack_lock(lgt, rgt);
 
     if ( gnttab_need_iommu_mapping(ld) )
     {
@@ -801,7 +799,7 @@ __gnttab_map_grant_ref(
         }
         if ( err )
         {
-            double_gt_unlock(lgt, rgt);
+            double_maptrack_unlock(lgt, rgt);
             rc = GNTST_general_error;
             goto undo_out;
         }
@@ -814,7 +812,8 @@ __gnttab_map_grant_ref(
     mt->ref   = op->ref;
     mt->flags = op->flags;
 
-    double_gt_unlock(lgt, rgt);
+    double_maptrack_unlock(lgt, rgt);
+    read_unlock(&rgt->lock);
 
     op->dev_bus_addr = (u64)frame << PAGE_SHIFT;
     op->handle       = handle;
@@ -837,7 +836,7 @@ __gnttab_map_grant_ref(
         put_page(pg);
     }
 
-    spin_lock(&rgt->lock);
+    read_lock(&rgt->lock);
 
     act = &active_entry(rgt, op->ref);
 
@@ -857,7 +856,7 @@ __gnttab_map_grant_ref(
         gnttab_clear_flag(_GTF_reading, status);
 
  unlock_out:
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
     op->status = rc;
     put_maptrack_handle(lgt, handle);
     rcu_unlock_domain(rd);
@@ -907,18 +906,19 @@ __gnttab_unmap_common(
     }
 
     op->map = &maptrack_entry(lgt, op->handle);
-    spin_lock(&lgt->lock);
+
+    read_lock(&lgt->lock);
 
     if ( unlikely(!op->map->flags) )
     {
-        spin_unlock(&lgt->lock);
+        read_unlock(&lgt->lock);
         gdprintk(XENLOG_INFO, "Zero flags for handle (%d).\n", op->handle);
         op->status = GNTST_bad_handle;
         return;
     }
 
     dom = op->map->domid;
-    spin_unlock(&lgt->lock);
+    read_unlock(&lgt->lock);
 
     if ( unlikely((rd = rcu_lock_domain_by_id(dom)) == NULL) )
     {
@@ -939,7 +939,9 @@ __gnttab_unmap_common(
     TRACE_1D(TRC_MEM_PAGE_GRANT_UNMAP, dom);
 
     rgt = rd->grant_table;
-    double_gt_lock(lgt, rgt);
+
+    read_lock(&rgt->lock);
+    double_maptrack_lock(lgt, rgt);
 
     op->flags = op->map->flags;
     if ( unlikely(!op->flags) || unlikely(op->map->domid != dom) )
@@ -1009,7 +1011,9 @@ __gnttab_unmap_common(
          gnttab_mark_dirty(rd, op->frame);
 
  unmap_out:
-    double_gt_unlock(lgt, rgt);
+    double_maptrack_unlock(lgt, rgt);
+    read_unlock(&rgt->lock);
+
     op->status = rc;
     rcu_unlock_domain(rd);
 }
@@ -1039,8 +1043,8 @@ __gnttab_unmap_common_complete(struct gnttab_unmap_common *op)
 
     rcu_lock_domain(rd);
     rgt = rd->grant_table;
-    spin_lock(&rgt->lock);
 
+    read_lock(&rgt->lock);
     if ( rgt->gt_version == 0 )
         goto unmap_out;
 
@@ -1104,7 +1108,7 @@ __gnttab_unmap_common_complete(struct gnttab_unmap_common *op)
         gnttab_clear_flag(_GTF_reading, status);
 
  unmap_out:
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
     if ( put_handle )
     {
         op->map->flags = 0;
@@ -1290,10 +1294,14 @@ gnttab_unpopulate_status_frames(struct domain *d, struct grant_table *gt)
     gt->nr_status_frames = 0;
 }
 
+/* 
+ * Grow the grant table. The caller must hold the grant table's
+ * write lock before calling this function.
+ */
 int
 gnttab_grow_table(struct domain *d, unsigned int req_nr_frames)
 {
-    /* d's grant table lock must be held by the caller */
+    /* d's grant table write lock must be held by the caller */
 
     struct grant_table *gt = d->grant_table;
     unsigned int i;
@@ -1398,7 +1406,7 @@ gnttab_setup_table(
     }
 
     gt = d->grant_table;
-    spin_lock(&gt->lock);
+    write_lock(&gt->lock);
 
     if ( gt->gt_version == 0 )
         gt->gt_version = 1;
@@ -1426,7 +1434,7 @@ gnttab_setup_table(
     }
 
  out3:
-    spin_unlock(&gt->lock);
+    write_unlock(&gt->lock);
  out2:
     rcu_unlock_domain(d);
  out1:
@@ -1468,13 +1476,13 @@ gnttab_query_size(
         goto query_out_unlock;
     }
 
-    spin_lock(&d->grant_table->lock);
+    read_lock(&d->grant_table->lock);
 
     op.nr_frames     = nr_grant_frames(d->grant_table);
     op.max_nr_frames = max_grant_frames;
     op.status        = GNTST_okay;
 
-    spin_unlock(&d->grant_table->lock);
+    read_unlock(&d->grant_table->lock);
 
  
  query_out_unlock:
@@ -1500,7 +1508,7 @@ gnttab_prepare_for_transfer(
     union grant_combo   scombo, prev_scombo, new_scombo;
     int                 retries = 0;
 
-    spin_lock(&rgt->lock);
+    read_lock(&rgt->lock);
 
     if ( rgt->gt_version == 0 )
     {
@@ -1551,11 +1559,11 @@ gnttab_prepare_for_transfer(
         scombo = prev_scombo;
     }
 
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
     return 1;
 
  fail:
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
     return 0;
 }
 
@@ -1747,7 +1755,7 @@ gnttab_transfer(
         TRACE_1D(TRC_MEM_PAGE_GRANT_TRANSFER, e->domain_id);
 
         /* Tell the guest about its new page frame. */
-        spin_lock(&e->grant_table->lock);
+        read_lock(&e->grant_table->lock);
 
         if ( e->grant_table->gt_version == 1 )
         {
@@ -1765,7 +1773,7 @@ gnttab_transfer(
         shared_entry_header(e->grant_table, gop.ref)->flags |=
             GTF_transfer_completed;
 
-        spin_unlock(&e->grant_table->lock);
+        read_unlock(&e->grant_table->lock);
 
         rcu_unlock_domain(e);
 
@@ -1803,7 +1811,7 @@ __release_grant_for_copy(
     released_read = 0;
     released_write = 0;
 
-    spin_lock(&rgt->lock);
+    read_lock(&rgt->lock);
 
     act = &active_entry(rgt, gref);
     sha = shared_entry_header(rgt, gref);
@@ -1844,7 +1852,7 @@ __release_grant_for_copy(
         released_read = 1;
     }
 
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
 
     if ( td != rd )
     {
@@ -1902,7 +1910,7 @@ __acquire_grant_for_copy(
 
     *page = NULL;
 
-    spin_lock(&rgt->lock);
+    read_lock(&rgt->lock);
 
     if ( rgt->gt_version == 0 )
         PIN_FAIL(unlock_out, GNTST_general_error,
@@ -1971,17 +1979,23 @@ __acquire_grant_for_copy(
                 PIN_FAIL(unlock_out_clear, GNTST_general_error,
                          "transitive grant referenced bad domain %d\n",
                          trans_domid);
-            spin_unlock(&rgt->lock);
+
+            /* 
+             * __acquire_grant_for_copy() could take the read lock on
+             * the right table (if rd == td), so we have to drop the
+             * lock here and reacquire
+             */
+            read_unlock(&rgt->lock);
 
             rc = __acquire_grant_for_copy(td, trans_gref, rd->domain_id,
                                           readonly, &grant_frame, page,
                                           &trans_page_off, &trans_length, 0);
 
-            spin_lock(&rgt->lock);
+            read_lock(&rgt->lock);
             if ( rc != GNTST_okay ) {
                 __fixup_status_for_copy_pin(act, status);
+                read_unlock(&rgt->lock);
                 rcu_unlock_domain(td);
-                spin_unlock(&rgt->lock);
                 return rc;
             }
 
@@ -1993,7 +2007,7 @@ __acquire_grant_for_copy(
             {
                 __fixup_status_for_copy_pin(act, status);
                 rcu_unlock_domain(td);
-                spin_unlock(&rgt->lock);
+                read_unlock(&rgt->lock);
                 put_page(*page);
                 return __acquire_grant_for_copy(rd, gref, ldom, readonly,
                                                 frame, page, page_off, length,
@@ -2061,7 +2075,7 @@ __acquire_grant_for_copy(
     *length = act->length;
     *frame = act->frame;
 
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
     return rc;
  
  unlock_out_clear:
@@ -2073,7 +2087,8 @@ __acquire_grant_for_copy(
         gnttab_clear_flag(_GTF_reading, status);
 
  unlock_out:
-    spin_unlock(&rgt->lock);
+    read_unlock(&rgt->lock);
+
     return rc;
 }
 
@@ -2389,7 +2404,7 @@ gnttab_set_version(XEN_GUEST_HANDLE_PARAM(gnttab_set_version_t) uop)
     if ( gt->gt_version == op.version )
         goto out;
 
-    spin_lock(&gt->lock);
+    write_lock(&gt->lock);
     /* Make sure that the grant table isn't currently in use when we
        change the version number, except for the first 8 entries which
        are allowed to be in use (xenstore/xenconsole keeps them mapped).
@@ -2475,7 +2490,7 @@ gnttab_set_version(XEN_GUEST_HANDLE_PARAM(gnttab_set_version_t) uop)
     gt->gt_version = op.version;
 
 out_unlock:
-    spin_unlock(&gt->lock);
+    write_unlock(&gt->lock);
 
 out:
     op.version = gt->gt_version;
@@ -2531,7 +2546,7 @@ gnttab_get_status_frames(XEN_GUEST_HANDLE_PARAM(gnttab_get_status_frames_t) uop,
 
     op.status = GNTST_okay;
 
-    spin_lock(&gt->lock);
+    read_lock(&gt->lock);
 
     for ( i = 0; i < op.nr_frames; i++ )
     {
@@ -2540,7 +2555,7 @@ gnttab_get_status_frames(XEN_GUEST_HANDLE_PARAM(gnttab_get_status_frames_t) uop,
             op.status = GNTST_bad_virt_addr;
     }
 
-    spin_unlock(&gt->lock);
+    read_unlock(&gt->lock);
 out2:
     rcu_unlock_domain(d);
 out1:
@@ -2589,7 +2604,7 @@ __gnttab_swap_grant_ref(grant_ref_t ref_a, grant_ref_t ref_b)
     struct active_grant_entry *act;
     s16 rc = GNTST_okay;
 
-    spin_lock(&gt->lock);
+    write_lock(&gt->lock);
 
     /* Bounds check on the grant refs */
     if ( unlikely(ref_a >= nr_grant_entries(d->grant_table)))
@@ -2629,7 +2644,7 @@ __gnttab_swap_grant_ref(grant_ref_t ref_a, grant_ref_t ref_b)
     }
 
 out:
-    spin_unlock(&gt->lock);
+    write_unlock(&gt->lock);
 
     rcu_unlock_domain(d);
 
@@ -2700,12 +2715,12 @@ static int __gnttab_cache_flush(gnttab_cache_flush_t *cflush,
 
     if ( d != owner )
     {
-        spin_lock(&owner->grant_table->lock);
+        read_lock(&owner->grant_table->lock);
 
         ret = grant_map_exists(d, owner->grant_table, mfn, ref_count);
         if ( ret != 0 )
         {
-            spin_unlock(&owner->grant_table->lock);
+            read_unlock(&owner->grant_table->lock);
             rcu_unlock_domain(d);
             put_page(page);
             return ret;
@@ -2725,7 +2740,7 @@ static int __gnttab_cache_flush(gnttab_cache_flush_t *cflush,
         ret = 0;
 
     if ( d != owner )
-        spin_unlock(&owner->grant_table->lock);
+        read_unlock(&owner->grant_table->lock);
     unmap_domain_page(v);
     put_page(page);
 
@@ -2944,7 +2959,8 @@ grant_table_create(
         goto no_mem_0;
 
     /* Simple stuff. */
-    spin_lock_init(&t->lock);
+    rwlock_init(&t->lock);
+    spin_lock_init(&t->maptrack_lock);
     t->nr_grant_frames = INITIAL_NR_GRANT_FRAMES;
 
     /* Active grant table. */
@@ -3051,7 +3067,8 @@ gnttab_release_mappings(
         }
 
         rgt = rd->grant_table;
-        spin_lock(&rgt->lock);
+        read_lock(&rgt->lock);
+        double_maptrack_lock(gt, rgt);
 
         act = &active_entry(rgt, ref);
         sha = shared_entry_header(rgt, ref);
@@ -3111,7 +3128,8 @@ gnttab_release_mappings(
         if ( act->pin == 0 )
             gnttab_clear_flag(_GTF_reading, status);
 
-        spin_unlock(&rgt->lock);
+        double_maptrack_unlock(gt, rgt);
+        read_unlock(&rgt->lock);
 
         rcu_unlock_domain(rd);
 
@@ -3129,7 +3147,9 @@ grant_table_destroy(
 
     if ( t == NULL )
         return;
-    
+
+    write_lock(&t->lock);
+
     for ( i = 0; i < nr_grant_frames(t); i++ )
         free_xenheap_page(t->shared_raw[i]);
     xfree(t->shared_raw);
@@ -3146,6 +3166,8 @@ grant_table_destroy(
         free_xenheap_page(t->status[i]);
     xfree(t->status);
 
+    write_unlock(&t->lock);
+
     xfree(t);
     d->grant_table = NULL;
 }
@@ -3159,7 +3181,7 @@ static void gnttab_usage_print(struct domain *rd)
     printk("      -------- active --------       -------- shared --------\n");
     printk("[ref] localdom mfn      pin          localdom gmfn     flags\n");
 
-    spin_lock(&gt->lock);
+    read_lock(&gt->lock);
 
     if ( gt->gt_version == 0 )
         goto out;
@@ -3208,7 +3230,7 @@ static void gnttab_usage_print(struct domain *rd)
     }
 
  out:
-    spin_unlock(&gt->lock);
+    read_unlock(&gt->lock);
 
     if ( first )
         printk("grant-table for remote domain:%5d ... "
diff --git a/xen/include/xen/grant_table.h b/xen/include/xen/grant_table.h
index 32f5786..298031e 100644
--- a/xen/include/xen/grant_table.h
+++ b/xen/include/xen/grant_table.h
@@ -82,10 +82,17 @@ struct grant_table {
     struct grant_mapping **maptrack;
     unsigned int          maptrack_head;
     unsigned int          maptrack_limit;
-    /* Lock protecting updates to active and shared grant tables. */
-    spinlock_t            lock;
-    /* The defined versions are 1 and 2.  Set to 0 if we don't know
-       what version to use yet. */
+    /* Lock protecting the maptrack page list, head, and limit */
+    spinlock_t            maptrack_lock;
+    /* 
+     * Lock protecting updates to grant table state (version, active
+     * entry list, etc.)
+     */
+    rwlock_t              lock;
+    /* 
+     * The defined versions are 1 and 2.  Set to 0 if we don't know
+     * what version to use yet.
+     */
     unsigned              gt_version;
 };
 
@@ -101,7 +108,7 @@ gnttab_release_mappings(
     struct domain *d);
 
 /* Increase the size of a domain's grant table.
- * Caller must hold d's grant table lock.
+ * Caller must hold d's grant table write lock.
  */
 int
 gnttab_grow_table(struct domain *d, unsigned int req_nr_frames);
-- 
1.7.9.5

[PATCH v5 2/2] gnttab: refactor locking for scalability

[Christoph Egger]

From: Matt Wilson <msw@amazon.com>

This patch refactors grant table locking. It splits the previous
single spinlock per grant table into multiple locks. The heavily
modified components of the grant table (the maptrack state and the
active entries) are now protected by their own spinlocks. The
remaining elements of the grant table are read-mostly, so the main
grant table lock is modified to be a rwlock to improve concurrency.

Workloads with high grant table operation rates, especially map/unmap
operations as used by blkback/blkfront when persistent grants are not
supported, show marked improvement with these changes. A domU with 24
VBDs in a streaming 2M write workload achieved 1,400 MB/sec before
this change. Performance more than doubles with this patch, reaching
3,000 MB/sec before tuning and 3,600 MB/sec after adjusting event
channel vCPU bindings.

Signed-off-by: Matt Wilson <msw@amazon.com>
[chegger: ported to xen-staging, split into multiple commits]

v5:
  * Fixed git rebase merge issue introduced in v4 
    (acquiring locking twice)
  * Change for ()-loop in grant_map_exists
  * Coding style fixes
v4:
  * Addressed ASSERT()'s from Jan Beulich
  * Addressed locking issues in unmap_common pointed out
    by Jan Beulich
v3:
  * Addressed gnttab_swap_grant_ref() comment from Andrew Cooper

Signed-off-by: Christoph Egger <chegger@amazon.de>
Cc: Anthony Liguori <aliguori@amazon.com>
Cc: Jan Beulich <jbeulich@suse.com>
Cc: Keir Fraser <keir@xen.org>
---
 docs/misc/grant-tables.txt |   21 +++++
 xen/common/grant_table.c   |  215 +++++++++++++++++++++++++++-----------------
 2 files changed, 155 insertions(+), 81 deletions(-)

diff --git a/docs/misc/grant-tables.txt b/docs/misc/grant-tables.txt
index c9ae8f2..1ada018 100644
--- a/docs/misc/grant-tables.txt
+++ b/docs/misc/grant-tables.txt
@@ -63,6 +63,7 @@ is complete.
   act->domid : remote domain being granted rights
   act->frame : machine frame being granted
   act->pin   : used to hold reference counts
+  act->lock  : spinlock used to serialize access to active entry state
 
  Map tracking
  ~~~~~~~~~~~~
@@ -87,6 +88,8 @@ is complete.
                                version, partially initialized active table pages,
                                etc.
   grant_table->maptrack_lock : spinlock used to protect the maptrack state
+  active_grant_entry->lock   : spinlock used to serialize modifications to
+                               active entries
 
  The primary lock for the grant table is a read/write spinlock. All
  functions that access members of struct grant_table must acquire a
@@ -102,6 +105,24 @@ is complete.
  state can be rapidly modified under some workloads, and the critical
  sections are very small, thus we use a spinlock to protect them.
 
+ Active entries are obtained by calling active_entry_acquire(gt, ref).
+ This function returns a pointer to the active entry after locking its
+ spinlock. The caller must hold the rwlock for the gt in question
+ before calling active_entry_acquire(). This is because the grant
+ table can be dynamically extended via gnttab_grow_table() while a
+ domain is running and must be fully initialized. Once all access to
+ the active entry is complete, release the lock by calling
+ active_entry_release(act).
+
+ Summary of rules for locking:
+  active_entry_acquire() and active_entry_release() can only be
+  called when holding the relevant grant table's lock. I.e.:
+    read_lock(&gt->lock);
+    act = active_entry_acquire(gt, ref);
+    ...
+    active_entry_release(act);
+    read_unlock(&gt->lock);
+
 ********************************************************************************
 
  Granting a foreign domain access to frames
diff --git a/xen/common/grant_table.c b/xen/common/grant_table.c
index f52a51d..94c9864 100644
--- a/xen/common/grant_table.c
+++ b/xen/common/grant_table.c
@@ -157,10 +157,13 @@ struct active_grant_entry {
                                in the page.                           */
     unsigned      length:16; /* For sub-page grants, the length of the
                                 grant.                                */
+    spinlock_t    lock;      /* lock to protect access of this entry.
+                                see docs/misc/grant-tables.txt for
+                                locking protocol                      */
 };
 
 #define ACGNT_PER_PAGE (PAGE_SIZE / sizeof(struct active_grant_entry))
-#define active_entry(t, e) \
+#define _active_entry(t, e) \
     ((t)->active[(e)/ACGNT_PER_PAGE][(e)%ACGNT_PER_PAGE])
 
 static inline void gnttab_flush_tlb(const struct domain *d)
@@ -188,6 +191,28 @@ nr_active_grant_frames(struct grant_table *gt)
     return num_act_frames_from_sha_frames(nr_grant_frames(gt));
 }
 
+static inline struct active_grant_entry *
+active_entry_acquire(struct grant_table *t, grant_ref_t e)
+{
+    struct active_grant_entry *act;
+
+    /* 
+     * not perfect, but better than nothing for a debug build
+     * sanity check
+     */
+    ASSERT(rw_is_locked(&t->lock));
+
+    act = &_active_entry(t, e);
+    spin_lock(&act->lock);
+
+    return act;
+}
+
+static inline void active_entry_release(struct active_grant_entry *act)
+{
+    spin_unlock(&act->lock);
+}
+    
 /* Check if the page has been paged out, or needs unsharing. 
    If rc == GNTST_okay, *page contains the page struct with a ref taken.
    Caller must do put_page(*page).
@@ -228,30 +253,6 @@ static int __get_paged_frame(unsigned long gfn, unsigned long *frame, struct pag
     return rc;
 }
 
-static inline void
-double_maptrack_lock(struct grant_table *lgt, struct grant_table *rgt)
-{
-    if ( lgt < rgt )
-    {
-        spin_lock(&lgt->maptrack_lock);
-        spin_lock(&rgt->maptrack_lock);
-    }
-    else
-    {
-        if ( lgt != rgt )
-            spin_lock(&rgt->maptrack_lock);
-        spin_lock(&lgt->maptrack_lock);
-    }
-}
-
-static inline void
-double_maptrack_unlock(struct grant_table *lgt, struct grant_table *rgt)
-{
-    spin_unlock(&lgt->maptrack_lock);
-    if ( lgt != rgt )
-        spin_unlock(&rgt->maptrack_lock);
-}
-
 static inline int
 __get_maptrack_handle(
     struct grant_table *t)
@@ -316,7 +317,10 @@ get_maptrack_handle(
     return handle;
 }
 
-/* Number of grant table entries. Caller must hold d's grant table lock. */
+/* 
+ * Number of grant table entries. Caller must hold d's grant table
+ * read lock.
+ */
 static unsigned int nr_grant_entries(struct grant_table *gt)
 {
     ASSERT(gt->gt_version != 0);
@@ -505,26 +509,23 @@ static int grant_map_exists(const struct domain *ld,
                             unsigned long mfn,
                             unsigned int *ref_count)
 {
-    const struct active_grant_entry *act;
+    struct active_grant_entry *act;
     unsigned int ref, max_iter;
     
     ASSERT(rw_is_locked(&rgt->lock));
 
     max_iter = min(*ref_count + (1 << GNTTABOP_CONTINUATION_ARG_SHIFT),
                    nr_grant_entries(rgt));
-    for ( ref = *ref_count; ref < max_iter; ref++ )
+    for ( ref = *ref_count; ref < max_iter; active_entry_release(act), ref++ )
     {
-        act = &active_entry(rgt, ref);
+        act = active_entry_acquire(rgt, ref);
 
-        if ( !act->pin )
-            continue;
-
-        if ( act->domid != ld->domain_id )
-            continue;
-
-        if ( act->frame != mfn )
+        if ( !act->pin ||
+             act->domid != ld->domain_id ||
+             act->frame != mfn )
             continue;
 
+        active_entry_release(act);
         return 0;
     }
 
@@ -537,6 +538,13 @@ static int grant_map_exists(const struct domain *ld,
     return -EINVAL;
 }
 
+/* 
+ * Count the number of mapped ro or rw entries tracked in the left
+ * grant table given a provided mfn provided by a foreign domain. 
+ *
+ * This function takes the maptrack lock from the left grant table and
+ * must be called with the right grant table's rwlock held.
+ */
 static void mapcount(
     struct grant_table *lgt, struct domain *rd, unsigned long mfn,
     unsigned int *wrc, unsigned int *rdc)
@@ -546,15 +554,32 @@ static void mapcount(
 
     *wrc = *rdc = 0;
 
+    /* 
+     * N.B.: while taking the left side maptrack spinlock prevents
+     * any mapping changes, the right side active entries could be
+     * changing while we are counting. The caller has to hold the
+     * grant table write lock, or some other mechanism should be
+     * used to prevent concurrent changes during this operation.
+     * This is tricky because we can't promote a read lock into
+     * a write lock.
+     */
+    ASSERT(rw_is_locked(&rd->grant_table->lock));
+    spin_lock(&lgt->maptrack_lock);
+
     for ( handle = 0; handle < lgt->maptrack_limit; handle++ )
     {
+        struct active_grant_entry *act;
+
         map = &maptrack_entry(lgt, handle);
         if ( !(map->flags & (GNTMAP_device_map|GNTMAP_host_map)) ||
              map->domid != rd->domain_id )
             continue;
-        if ( active_entry(rd->grant_table, map->ref).frame == mfn )
+        act = &_active_entry(rd->grant_table, map->ref);
+        if ( act->frame == mfn )
             (map->flags & GNTMAP_readonly) ? (*rdc)++ : (*wrc)++;
     }
+
+    spin_unlock(&lgt->maptrack_lock);
 }
 
 /*
@@ -576,7 +601,6 @@ __gnttab_map_grant_ref(
     struct page_info *pg = NULL;
     int            rc = GNTST_okay;
     u32            old_pin;
-    u32            act_pin;
     unsigned int   cache_flags;
     struct active_grant_entry *act = NULL;
     struct grant_mapping *mt;
@@ -639,7 +663,7 @@ __gnttab_map_grant_ref(
     if ( unlikely(op->ref >= nr_grant_entries(rgt)))
         PIN_FAIL(unlock_out, GNTST_bad_gntref, "Bad ref (%d).\n", op->ref);
 
-    act = &active_entry(rgt, op->ref);
+    act = active_entry_acquire(rgt, op->ref);
     shah = shared_entry_header(rgt, op->ref);
     if (rgt->gt_version == 1) {
         sha1 = &shared_entry_v1(rgt, op->ref);
@@ -656,7 +680,7 @@ __gnttab_map_grant_ref(
          ((act->domid != ld->domain_id) ||
           (act->pin & 0x80808080U) != 0 ||
           (act->is_sub_page)) )
-        PIN_FAIL(unlock_out, GNTST_general_error,
+        PIN_FAIL(act_release_out, GNTST_general_error,
                  "Bad domain (%d != %d), or risk of counter overflow %08x, or subpage %d\n",
                  act->domid, ld->domain_id, act->pin, act->is_sub_page);
 
@@ -667,7 +691,7 @@ __gnttab_map_grant_ref(
         if ( (rc = _set_status(rgt->gt_version, ld->domain_id,
                                op->flags & GNTMAP_readonly,
                                1, shah, act, status) ) != GNTST_okay )
-             goto unlock_out;
+            goto act_release_out;
 
         if ( !act->pin )
         {
@@ -698,7 +722,6 @@ __gnttab_map_grant_ref(
             GNTPIN_hstr_inc : GNTPIN_hstw_inc;
 
     frame = act->frame;
-    act_pin = act->pin;
 
     cache_flags = (shah->flags & (GTF_PAT | GTF_PWT | GTF_PCD) );
 
@@ -776,8 +799,6 @@ __gnttab_map_grant_ref(
         goto undo_out;
     }
 
-    double_maptrack_lock(lgt, rgt);
-
     if ( gnttab_need_iommu_mapping(ld) )
     {
         unsigned int wrc, rdc;
@@ -785,21 +806,20 @@ __gnttab_map_grant_ref(
         /* We're not translated, so we know that gmfns and mfns are
            the same things, so the IOMMU entry is always 1-to-1. */
         mapcount(lgt, rd, frame, &wrc, &rdc);
-        if ( (act_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) &&
+        if ( (act->pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) &&
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( wrc == 0 )
                 err = iommu_map_page(ld, frame, frame,
                                      IOMMUF_readable|IOMMUF_writable);
         }
-        else if ( act_pin && !old_pin )
+        else if ( act->pin && !old_pin )
         {
             if ( (wrc + rdc) == 0 )
                 err = iommu_map_page(ld, frame, frame, IOMMUF_readable);
         }
         if ( err )
         {
-            double_maptrack_unlock(lgt, rgt);
             rc = GNTST_general_error;
             goto undo_out;
         }
@@ -812,7 +832,7 @@ __gnttab_map_grant_ref(
     mt->ref   = op->ref;
     mt->flags = op->flags;
 
-    double_maptrack_unlock(lgt, rgt);
+    active_entry_release(act);
     read_unlock(&rgt->lock);
 
     op->dev_bus_addr = (u64)frame << PAGE_SHIFT;
@@ -836,10 +856,6 @@ __gnttab_map_grant_ref(
         put_page(pg);
     }
 
-    read_lock(&rgt->lock);
-
-    act = &active_entry(rgt, op->ref);
-
     if ( op->flags & GNTMAP_device_map )
         act->pin -= (op->flags & GNTMAP_readonly) ?
             GNTPIN_devr_inc : GNTPIN_devw_inc;
@@ -855,6 +871,9 @@ __gnttab_map_grant_ref(
     if ( !act->pin )
         gnttab_clear_flag(_GTF_reading, status);
 
+ act_release_out:
+    active_entry_release(act);
+
  unlock_out:
     read_unlock(&rgt->lock);
     op->status = rc;
@@ -941,18 +960,17 @@ __gnttab_unmap_common(
     rgt = rd->grant_table;
 
     read_lock(&rgt->lock);
-    double_maptrack_lock(lgt, rgt);
+    act = active_entry_acquire(rgt, op->map->ref);
 
     op->flags = op->map->flags;
     if ( unlikely(!op->flags) || unlikely(op->map->domid != dom) )
     {
         gdprintk(XENLOG_WARNING, "Unstable handle %u\n", op->handle);
         rc = GNTST_bad_handle;
-        goto unmap_out;
+        goto act_release_out;
     }
 
     op->rd = rd;
-    act = &active_entry(rgt, op->map->ref);
 
     if ( op->frame == 0 )
     {
@@ -961,7 +979,7 @@ __gnttab_unmap_common(
     else
     {
         if ( unlikely(op->frame != act->frame) )
-            PIN_FAIL(unmap_out, GNTST_general_error,
+            PIN_FAIL(act_release_out, GNTST_general_error,
                      "Bad frame number doesn't match gntref. (%lx != %lx)\n",
                      op->frame, act->frame);
         if ( op->flags & GNTMAP_device_map )
@@ -980,7 +998,7 @@ __gnttab_unmap_common(
         if ( (rc = replace_grant_host_mapping(op->host_addr,
                                               op->frame, op->new_addr, 
                                               op->flags)) < 0 )
-            goto unmap_out;
+            goto act_release_out;
 
         ASSERT(act->pin & (GNTPIN_hstw_mask | GNTPIN_hstr_mask));
         op->map->flags &= ~GNTMAP_host_map;
@@ -1002,7 +1020,7 @@ __gnttab_unmap_common(
         if ( err )
         {
             rc = GNTST_general_error;
-            goto unmap_out;
+            goto act_release_out;
         }
     }
 
@@ -1010,8 +1028,8 @@ __gnttab_unmap_common(
     if ( !(op->flags & GNTMAP_readonly) )
          gnttab_mark_dirty(rd, op->frame);
 
- unmap_out:
-    double_maptrack_unlock(lgt, rgt);
+ act_release_out:
+    active_entry_release(act);
     read_unlock(&rgt->lock);
 
     op->status = rc;
@@ -1046,9 +1064,9 @@ __gnttab_unmap_common_complete(struct gnttab_unmap_common *op)
 
     read_lock(&rgt->lock);
     if ( rgt->gt_version == 0 )
-        goto unmap_out;
+        goto unlock_out;
 
-    act = &active_entry(rgt, op->map->ref);
+    act = active_entry_acquire(rgt, op->map->ref);
     sha = shared_entry_header(rgt, op->map->ref);
 
     if ( rgt->gt_version == 1 )
@@ -1062,7 +1080,7 @@ __gnttab_unmap_common_complete(struct gnttab_unmap_common *op)
          * Suggests that __gntab_unmap_common failed early and so
          * nothing further to do
          */
-        goto unmap_out;
+        goto act_release_out;
     }
 
     pg = mfn_to_page(op->frame);
@@ -1086,7 +1104,7 @@ __gnttab_unmap_common_complete(struct gnttab_unmap_common *op)
              * Suggests that __gntab_unmap_common failed in
              * replace_grant_host_mapping() so nothing further to do
              */
-            goto unmap_out;
+            goto act_release_out;
         }
 
         if ( !is_iomem_page(op->frame) ) 
@@ -1107,8 +1125,12 @@ __gnttab_unmap_common_complete(struct gnttab_unmap_common *op)
     if ( act->pin == 0 )
         gnttab_clear_flag(_GTF_reading, status);
 
- unmap_out:
+ act_release_out:
+    active_entry_release(act);
+    
+ unlock_out:
     read_unlock(&rgt->lock);
+
     if ( put_handle )
     {
         op->map->flags = 0;
@@ -1304,7 +1326,7 @@ gnttab_grow_table(struct domain *d, unsigned int req_nr_frames)
     /* d's grant table write lock must be held by the caller */
 
     struct grant_table *gt = d->grant_table;
-    unsigned int i;
+    unsigned int i, j;
 
     ASSERT(req_nr_frames <= max_grant_frames);
 
@@ -1319,6 +1341,8 @@ gnttab_grow_table(struct domain *d, unsigned int req_nr_frames)
         if ( (gt->active[i] = alloc_xenheap_page()) == NULL )
             goto active_alloc_failed;
         clear_page(gt->active[i]);
+        for ( j = 0; j < ACGNT_PER_PAGE; j++ )
+            spin_lock_init(&gt->active[i][j].lock);
     }
 
     /* Shared */
@@ -1813,7 +1837,7 @@ __release_grant_for_copy(
 
     read_lock(&rgt->lock);
 
-    act = &active_entry(rgt, gref);
+    act = active_entry_acquire(rgt, gref);
     sha = shared_entry_header(rgt, gref);
     r_frame = act->frame;
 
@@ -1852,6 +1876,7 @@ __release_grant_for_copy(
         released_read = 1;
     }
 
+    active_entry_release(act);
     read_unlock(&rgt->lock);
 
     if ( td != rd )
@@ -1913,14 +1938,14 @@ __acquire_grant_for_copy(
     read_lock(&rgt->lock);
 
     if ( rgt->gt_version == 0 )
-        PIN_FAIL(unlock_out, GNTST_general_error,
+        PIN_FAIL(gnt_unlock_out, GNTST_general_error,
                  "remote grant table not ready\n");
 
     if ( unlikely(gref >= nr_grant_entries(rgt)) )
-        PIN_FAIL(unlock_out, GNTST_bad_gntref,
+        PIN_FAIL(gnt_unlock_out, GNTST_bad_gntref,
                  "Bad grant reference %ld\n", gref);
 
-    act = &active_entry(rgt, gref);
+    act = active_entry_acquire(rgt, gref);
     shah = shared_entry_header(rgt, gref);
     if ( rgt->gt_version == 1 )
     {
@@ -1985,6 +2010,7 @@ __acquire_grant_for_copy(
              * the right table (if rd == td), so we have to drop the
              * lock here and reacquire
              */
+            active_entry_release(act);
             read_unlock(&rgt->lock);
 
             rc = __acquire_grant_for_copy(td, trans_gref, rd->domain_id,
@@ -1992,8 +2018,11 @@ __acquire_grant_for_copy(
                                           &trans_page_off, &trans_length, 0);
 
             read_lock(&rgt->lock);
+            act = active_entry_acquire(rgt, gref);
+
             if ( rc != GNTST_okay ) {
                 __fixup_status_for_copy_pin(act, status);
+                active_entry_release(act);
                 read_unlock(&rgt->lock);
                 rcu_unlock_domain(td);
                 return rc;
@@ -2007,6 +2036,7 @@ __acquire_grant_for_copy(
             {
                 __fixup_status_for_copy_pin(act, status);
                 rcu_unlock_domain(td);
+                active_entry_release(act);
                 read_unlock(&rgt->lock);
                 put_page(*page);
                 return __acquire_grant_for_copy(rd, gref, ldom, readonly,
@@ -2075,6 +2105,7 @@ __acquire_grant_for_copy(
     *length = act->length;
     *frame = act->frame;
 
+    active_entry_release(act);
     read_unlock(&rgt->lock);
     return rc;
  
@@ -2087,6 +2118,9 @@ __acquire_grant_for_copy(
         gnttab_clear_flag(_GTF_reading, status);
 
  unlock_out:
+    active_entry_release(act);
+
+ gnt_unlock_out:
     read_unlock(&rgt->lock);
 
     return rc;
@@ -2413,16 +2447,18 @@ gnttab_set_version(XEN_GUEST_HANDLE_PARAM(gnttab_set_version_t) uop)
     {
         for ( i = GNTTAB_NR_RESERVED_ENTRIES; i < nr_grant_entries(gt); i++ )
         {
-            act = &active_entry(gt, i);
+            act = active_entry_acquire(gt, i);
             if ( act->pin != 0 )
             {
                 gdprintk(XENLOG_WARNING,
                          "tried to change grant table version from %d to %d, but some grant entries still in use\n",
                          gt->gt_version,
                          op.version);
+                active_entry_release(act);
                 res = -EBUSY;
                 goto out_unlock;
             }
+            active_entry_release(act);
         }
     }
 
@@ -2601,9 +2637,17 @@ __gnttab_swap_grant_ref(grant_ref_t ref_a, grant_ref_t ref_b)
 {
     struct domain *d = rcu_lock_current_domain();
     struct grant_table *gt = d->grant_table;
-    struct active_grant_entry *act;
+    struct active_grant_entry *act_a = NULL;
+    struct active_grant_entry *act_b = NULL;
     s16 rc = GNTST_okay;
 
+    if ( ref_a == ref_b )
+        /*
+         * noop, so avoid acquiring the same active entry
+         * twice which would case a deadlock.
+         */
+        return rc;
+
     write_lock(&gt->lock);
 
     /* Bounds check on the grant refs */
@@ -2612,12 +2656,12 @@ __gnttab_swap_grant_ref(grant_ref_t ref_a, grant_ref_t ref_b)
     if ( unlikely(ref_b >= nr_grant_entries(d->grant_table)))
         PIN_FAIL(out, GNTST_bad_gntref, "Bad ref-b (%d).\n", ref_b);
 
-    act = &active_entry(gt, ref_a);
-    if ( act->pin )
+    act_a = active_entry_acquire(gt, ref_a);
+    if ( act_a->pin )
         PIN_FAIL(out, GNTST_eagain, "ref a %ld busy\n", (long)ref_a);
 
-    act = &active_entry(gt, ref_b);
-    if ( act->pin )
+    act_b = active_entry_acquire(gt, ref_b);
+    if ( act_b->pin )
         PIN_FAIL(out, GNTST_eagain, "ref b %ld busy\n", (long)ref_b);
 
     if ( gt->gt_version == 1 )
@@ -2644,6 +2688,10 @@ __gnttab_swap_grant_ref(grant_ref_t ref_a, grant_ref_t ref_b)
     }
 
 out:
+    if ( act_b != NULL )
+        active_entry_release(act_b);
+    if ( act_a != NULL )
+        active_entry_release(act_a);
     write_unlock(&gt->lock);
 
     rcu_unlock_domain(d);
@@ -2953,7 +3001,7 @@ grant_table_create(
     struct domain *d)
 {
     struct grant_table *t;
-    int                 i;
+    int                 i, j;
 
     if ( (t = xzalloc(struct grant_table)) == NULL )
         goto no_mem_0;
@@ -2973,6 +3021,8 @@ grant_table_create(
         if ( (t->active[i] = alloc_xenheap_page()) == NULL )
             goto no_mem_2;
         clear_page(t->active[i]);
+        for ( j = 0; j < ACGNT_PER_PAGE; j++ )
+            spin_lock_init(&t->active[i][j].lock);
     }
 
     /* Tracking of mapped foreign frames table */
@@ -3068,9 +3118,8 @@ gnttab_release_mappings(
 
         rgt = rd->grant_table;
         read_lock(&rgt->lock);
-        double_maptrack_lock(gt, rgt);
 
-        act = &active_entry(rgt, ref);
+        act = active_entry_acquire(rgt, ref);
         sha = shared_entry_header(rgt, ref);
         if (rgt->gt_version == 1)
             status = &sha->flags;
@@ -3128,7 +3177,7 @@ gnttab_release_mappings(
         if ( act->pin == 0 )
             gnttab_clear_flag(_GTF_reading, status);
 
-        double_maptrack_unlock(gt, rgt);
+        active_entry_release(act);
         read_unlock(&rgt->lock);
 
         rcu_unlock_domain(rd);
@@ -3195,9 +3244,12 @@ static void gnttab_usage_print(struct domain *rd)
         uint16_t status;
         uint64_t frame;
 
-        act = &active_entry(gt, ref);
+        act = active_entry_acquire(gt, ref);
         if ( !act->pin )
+        {
+            active_entry_release(act);
             continue;
+        }
 
         sha = shared_entry_header(gt, ref);
 
@@ -3227,6 +3279,7 @@ static void gnttab_usage_print(struct domain *rd)
         printk("[%3d]    %5d 0x%06lx 0x%08x      %5d 0x%06"PRIx64" 0x%02x\n",
                ref, act->domid, act->frame, act->pin,
                sha->domid, frame, status);
+        active_entry_release(act);
     }
 
  out:
-- 
1.7.9.5

Re: [PATCH v5 1/2] gnttab: Introduce rwlock to protect updates to grant table state

[Jan Beulich]

>>> On 10.02.15 at 13:51, <chegger@amazon.de> wrote:
> @@ -629,7 +629,7 @@ __gnttab_map_grant_ref(
>      }
>  
>      rgt = rd->grant_table;
> -    spin_lock(&rgt->lock);
> +    read_lock(&rgt->lock);
>  
>      if ( rgt->gt_version == 0 )
>          PIN_FAIL(unlock_out, GNTST_general_error,
> @@ -702,8 +702,6 @@ __gnttab_map_grant_ref(
>  
>      cache_flags = (shah->flags & (GTF_PAT | GTF_PWT | GTF_PCD) );
>  
> -    spin_unlock(&rgt->lock);
> -
>      /* pg may be set, with a refcount included, from __get_paged_frame */
>      if ( !pg )
>      {
> @@ -778,7 +776,7 @@ __gnttab_map_grant_ref(
>          goto undo_out;

With the unlock above dropped, this and other similar goto-s now
end up acquiring the lock a second time, but then dropping it only
once. As said before (on a different piece of code) altering the
regions which get locked should be made explicit in the patch
description, allowing - if not spotted during review - later bug
fixing to be done with proper understanding of the intentions.

>      }
>  
> -    double_gt_lock(lgt, rgt);
> +    double_maptrack_lock(lgt, rgt);

Looking at this nested locking I find that you _still_ don't say
anything about lock hierarchy in the doc change.

> @@ -1971,17 +1979,23 @@ __acquire_grant_for_copy(
>                  PIN_FAIL(unlock_out_clear, GNTST_general_error,
>                           "transitive grant referenced bad domain %d\n",
>                           trans_domid);
> -            spin_unlock(&rgt->lock);
> +
> +            /* 
> +             * __acquire_grant_for_copy() could take the read lock on
> +             * the right table (if rd == td), so we have to drop the
> +             * lock here and reacquire
> +             */

Almost - there's still a full stop missing here.

> @@ -3129,7 +3147,9 @@ grant_table_destroy(
>  
>      if ( t == NULL )
>          return;
> -    
> +
> +    write_lock(&t->lock);
> +
>      for ( i = 0; i < nr_grant_frames(t); i++ )
>          free_xenheap_page(t->shared_raw[i]);
>      xfree(t->shared_raw);
> @@ -3146,6 +3166,8 @@ grant_table_destroy(
>          free_xenheap_page(t->status[i]);
>      xfree(t->status);
>  
> +    write_unlock(&t->lock);

Do we really need the lock here? If so (as also said before) this
ought to be a separate fix (needing backporting).

> --- a/xen/include/xen/grant_table.h
> +++ b/xen/include/xen/grant_table.h
> @@ -82,10 +82,17 @@ struct grant_table {
>      struct grant_mapping **maptrack;
>      unsigned int          maptrack_head;
>      unsigned int          maptrack_limit;
> -    /* Lock protecting updates to active and shared grant tables. */
> -    spinlock_t            lock;
> -    /* The defined versions are 1 and 2.  Set to 0 if we don't know
> -       what version to use yet. */
> +    /* Lock protecting the maptrack page list, head, and limit */
> +    spinlock_t            maptrack_lock;
> +    /* 
> +     * Lock protecting updates to grant table state (version, active
> +     * entry list, etc.)
> +     */
> +    rwlock_t              lock;

Repeating the recently raised similar question (in another context):
Is it really a good idea to have two locks on the same cache line?

Jan

Re: [PATCH v5 2/2] gnttab: refactor locking for scalability

[Jan Beulich]

>>> On 10.02.15 at 13:51, <chegger@amazon.de> wrote:
> @@ -188,6 +191,28 @@ nr_active_grant_frames(struct grant_table *gt)
>      return num_act_frames_from_sha_frames(nr_grant_frames(gt));
>  }
>  
> +static inline struct active_grant_entry *
> +active_entry_acquire(struct grant_table *t, grant_ref_t e)
> +{
> +    struct active_grant_entry *act;
> +
> +    /* 
> +     * not perfect, but better than nothing for a debug build
> +     * sanity check
> +     */

Better, but still not matching ./CODING_STYLE.

> @@ -316,7 +317,10 @@ get_maptrack_handle(
>      return handle;
>  }
>  
> -/* Number of grant table entries. Caller must hold d's grant table lock. */
> +/* 
> + * Number of grant table entries. Caller must hold d's grant table
> + * read lock.
> + */
>  static unsigned int nr_grant_entries(struct grant_table *gt)

Doesn't this belong in patch 1?

> @@ -505,26 +509,23 @@ static int grant_map_exists(const struct domain *ld,
>                              unsigned long mfn,
>                              unsigned int *ref_count)
>  {
> -    const struct active_grant_entry *act;
> +    struct active_grant_entry *act;
>      unsigned int ref, max_iter;
>      
>      ASSERT(rw_is_locked(&rgt->lock));
>  
>      max_iter = min(*ref_count + (1 << GNTTABOP_CONTINUATION_ARG_SHIFT),
>                     nr_grant_entries(rgt));
> -    for ( ref = *ref_count; ref < max_iter; ref++ )
> +    for ( ref = *ref_count; ref < max_iter; active_entry_release(act), ref++ )
>      {
> -        act = &active_entry(rgt, ref);
> +        act = active_entry_acquire(rgt, ref);
>  
> -        if ( !act->pin )
> -            continue;
> -
> -        if ( act->domid != ld->domain_id )
> -            continue;
> -
> -        if ( act->frame != mfn )
> +        if ( !act->pin ||
> +             act->domid != ld->domain_id ||
> +             act->frame != mfn )
>              continue;

I don't mind this consolidation, but it does grow the patch size
without real need.

>      for ( handle = 0; handle < lgt->maptrack_limit; handle++ )
>      {
> +        struct active_grant_entry *act;
> +
>          map = &maptrack_entry(lgt, handle);
>          if ( !(map->flags & (GNTMAP_device_map|GNTMAP_host_map)) ||
>               map->domid != rd->domain_id )
>              continue;
> -        if ( active_entry(rd->grant_table, map->ref).frame == mfn )
> +        act = &_active_entry(rd->grant_table, map->ref);
> +        if ( act->frame == mfn )
>              (map->flags & GNTMAP_readonly) ? (*rdc)++ : (*wrc)++;
>      }

Yet here I don't understand at all what the change is good for - the
only thing you appear to need to change is to insert an underscore.

> @@ -2601,9 +2637,17 @@ __gnttab_swap_grant_ref(grant_ref_t ref_a, grant_ref_t ref_b)
>  {
>      struct domain *d = rcu_lock_current_domain();
>      struct grant_table *gt = d->grant_table;
> -    struct active_grant_entry *act;
> +    struct active_grant_entry *act_a = NULL;
> +    struct active_grant_entry *act_b = NULL;
>      s16 rc = GNTST_okay;
>  
> +    if ( ref_a == ref_b )
> +        /*
> +         * noop, so avoid acquiring the same active entry
> +         * twice which would case a deadlock.

cause

> @@ -2953,7 +3001,7 @@ grant_table_create(
>      struct domain *d)
>  {
>      struct grant_table *t;
> -    int                 i;
> +    int                 i, j;

unsigned int

> @@ -3195,9 +3244,12 @@ static void gnttab_usage_print(struct domain *rd)
>          uint16_t status;
>          uint64_t frame;
>  
> -        act = &active_entry(gt, ref);
> +        act = active_entry_acquire(gt, ref);

As (again in a different context) said recently, debug key handlers
would better use try-lock variants, skipping the respective printing
if the lock is being held.

And to come back to a remark on v3 and v4: Considering that you
drop double_maptrack_lock() here altogether, is patch 1 really
correct/sensible in introducing that function in place of
double_gt_lock()? Wouldn't the new functionality want the local
maptrack and the remote grant table locked in such cases? I.e.
isn't one of the locks taken at least pointless?

Jan