13 Feb, 2015
34 commits
-
Currently, the isolate callback passed to the list_lru_walk family of
functions is supposed to just delete an item from the list upon returning
LRU_REMOVED or LRU_REMOVED_RETRY, while nr_items counter is fixed by
__list_lru_walk_one after the callback returns. Since the callback is
allowed to drop the lock after removing an item (it has to return
LRU_REMOVED_RETRY then), the nr_items can be less than the actual number
of elements on the list even if we check them under the lock. This makes
it difficult to move items from one list_lru_one to another, which is
required for per-memcg list_lru reparenting - we can't just splice the
lists, we have to move entries one by one.This patch therefore introduces helpers that must be used by callback
functions to isolate items instead of raw list_del/list_move. These are
list_lru_isolate and list_lru_isolate_move. They not only remove the
entry from the list, but also fix the nr_items counter, making sure
nr_items always reflects the actual number of elements on the list if
checked under the appropriate lock.Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Tejun Heo
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Dave Chinner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
We need to look up a kmem_cache in ->memcg_params.memcg_caches arrays only
on allocations, so there is no need to have the array entries set until
css free - we can clear them on css offline. This will allow us to reuse
array entries more efficiently and avoid costly array relocations.Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Tejun Heo
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Dave Chinner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Currently, we use mem_cgroup->kmemcg_id to guarantee kmem_cache->name
uniqueness. This is correct, because kmemcg_id is only released on css
free after destroying all per memcg caches.However, I am going to change that and release kmemcg_id on css offline,
because it is not wise to keep it for so long, wasting valuable entries of
memcg_cache_params->memcg_caches arrays. Therefore, to preserve cache
name uniqueness, let us switch to css->id.Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Tejun Heo
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Dave Chinner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Currently, we release css->id in css_release_work_fn, right before calling
css_free callback, so that when css_free is called, the id may have
already been reused for a new cgroup.I am going to use css->id to create unique names for per memcg kmem
caches. Since kmem caches are destroyed only on css_free, I need css->id
to be freed after css_free was called to avoid name clashes. This patch
therefore moves css->id removal to css_free_work_fn. To prevent
css_from_id from returning a pointer to a stale css, it makes
css_release_work_fn replace the css ptr at css_idr:css->id with NULL.Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Acked-by: Tejun Heo
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Dave Chinner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Sometimes, we need to iterate over all memcg copies of a particular root
kmem cache. Currently, we use memcg_cache_params->memcg_caches array for
that, because it contains all existing memcg caches.However, it's a bad practice to keep all caches, including those that
belong to offline cgroups, in this array, because it will be growing
beyond any bounds then. I'm going to wipe away dead caches from it to
save space. To still be able to perform iterations over all memcg caches
of the same kind, let us link them into a list.Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Tejun Heo
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Dave Chinner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Currently, kmem_cache stores a pointer to struct memcg_cache_params
instead of embedding it. The rationale is to save memory when kmem
accounting is disabled. However, the memcg_cache_params has shrivelled
drastically since it was first introduced:* Initially:
struct memcg_cache_params {
bool is_root_cache;
union {
struct kmem_cache *memcg_caches[0];
struct {
struct mem_cgroup *memcg;
struct list_head list;
struct kmem_cache *root_cache;
bool dead;
atomic_t nr_pages;
struct work_struct destroy;
};
};
};* Now:
struct memcg_cache_params {
bool is_root_cache;
union {
struct {
struct rcu_head rcu_head;
struct kmem_cache *memcg_caches[0];
};
struct {
struct mem_cgroup *memcg;
struct kmem_cache *root_cache;
};
};
};So the memory saving does not seem to be a clear win anymore.
OTOH, keeping a pointer to memcg_cache_params struct instead of embedding
it results in touching one more cache line on kmem alloc/free hot paths.
Besides, it makes linking kmem caches in a list chained by a field of
struct memcg_cache_params really painful due to a level of indirection,
while I want to make them linked in the following patch. That said, let
us embed it.Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Tejun Heo
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Dave Chinner
Cc: Dan Carpenter
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
In super_cache_scan() we divide the number of objects of particular type
by the total number of objects in order to distribute pressure among As a
result, in some corner cases we can get nr_to_scan=0 even if there are
some objects to reclaim, e.g. dentries=1, inodes=1, fs_objects=1,
nr_to_scan=1/3=0.This is unacceptable for per memcg kmem accounting, because this means
that some objects may never get reclaimed after memcg death, preventing it
from being freed.This patch therefore assures that super_cache_scan() will scan at least
one object of each type if any.[akpm@linux-foundation.org: add comment]
Signed-off-by: Vladimir Davydov
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Alexander Viro
Cc: Dave Chinner
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Now, to make any list_lru-based shrinker memcg aware we should only
initialize its list_lru as memcg aware. Let's do it for the general FS
shrinker (super_block::s_shrink).There are other FS-specific shrinkers that use list_lru for storing
objects, such as XFS and GFS2 dquot cache shrinkers, but since they
reclaim objects that are shared among different cgroups, there is no point
making them memcg aware. It's a big question whether we should account
them to memcg at all.Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
There are several FS shrinkers, including super_block::s_shrink, that
keep reclaimable objects in the list_lru structure. Hence to turn them
to memcg-aware shrinkers, it is enough to make list_lru per-memcg.This patch does the trick. It adds an array of lru lists to the
list_lru_node structure (per-node part of the list_lru), one for each
kmem-active memcg, and dispatches every item addition or removal to the
list corresponding to the memcg which the item is accounted to. So now
the list_lru structure is not just per node, but per node and per memcg.Not all list_lrus need this feature, so this patch also adds a new
method, list_lru_init_memcg, which initializes a list_lru as memcg
aware. Otherwise (i.e. if initialized with old list_lru_init), the
list_lru won't have per memcg lists.Just like per memcg caches arrays, the arrays of per-memcg lists are
indexed by memcg_cache_id, so we must grow them whenever
memcg_nr_cache_ids is increased. So we introduce a callback,
memcg_update_all_list_lrus, invoked by memcg_alloc_cache_id if the id
space is full.The locking is implemented in a manner similar to lruvecs, i.e. we have
one lock per node that protects all lists (both global and per cgroup) on
the node.Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
To make list_lru memcg aware, we need all list_lrus to be kept on a list
protected by a mutex, so that we could sleep while walking over the
list.Therefore after this change list_lru_destroy may sleep. Fortunately,
there is only one user that calls it from an atomic context - it's
put_super - and we can easily fix it by calling list_lru_destroy before
put_super in destroy_locked_super - anyway we don't longer need lrus by
that time.Another point that should be noted is that list_lru_destroy is allowed
to be called on an uninitialized zeroed-out object, in which case it is
a no-op. Before this patch this was guaranteed by kfree, but now we
need an explicit check there.Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The active_nodes mask allows us to skip empty nodes when walking over
list_lru items from all nodes in list_lru_count/walk. However, these
functions are never called from hot paths, so it doesn't seem we need
such kind of optimization there. OTOH, removing the mask will make it
easier to make list_lru per-memcg.Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
We need a stable value of memcg_nr_cache_ids in kmem_cache_create()
(memcg_alloc_cache_params() wants it for root caches), where we only
hold the slab_mutex and no memcg-related locks. As a result, we have to
update memcg_nr_cache_ids under the slab_mutex, which we can only take
on the slab's side (see memcg_update_array_size). This looks awkward
and will become even worse when per-memcg list_lru is introduced, which
also wants stable access to memcg_nr_cache_ids.To get rid of this dependency between the memcg_nr_cache_ids and the
slab_mutex, this patch introduces a special rwsem. The rwsem is held
for writing during memcg_caches arrays relocation and memcg_nr_cache_ids
updates. Therefore one can take it for reading to get a stable access
to memcg_caches arrays and/or memcg_nr_cache_ids.Currently the semaphore is taken for reading only from
kmem_cache_create, right before taking the slab_mutex, so right now
there's no much point in using rwsem instead of mutex. However, once
list_lru is made per-memcg it will allow list_lru initializations to
proceed concurrently.Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
memcg_limited_groups_array_size, which defines the size of memcg_caches
arrays, sounds rather cumbersome. Also it doesn't point anyhow that
it's related to kmem/caches stuff. So let's rename it to
memcg_nr_cache_ids. It's concise and points us directly to
memcg_cache_id.Also, rename kmem_limited_groups to memcg_cache_ida.
Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch adds SHRINKER_MEMCG_AWARE flag. If a shrinker has this flag
set, it will be called per memory cgroup. The memory cgroup to scan
objects from is passed in shrink_control->memcg. If the memory cgroup
is NULL, a memcg aware shrinker is supposed to scan objects from the
global list. Unaware shrinkers are only called on global pressure with
memcg=NULL.Signed-off-by: Vladimir Davydov
Cc: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
We are going to make FS shrinkers memcg-aware. To achieve that, we will
have to pass the memcg to scan to the nr_cached_objects and
free_cached_objects VFS methods, which currently take only the NUMA node
to scan. Since the shrink_control structure already holds the node, and
the memcg to scan will be added to it when we introduce memcg-aware
vmscan, let us consolidate the methods' arguments in this structure to
keep things clean.Signed-off-by: Vladimir Davydov
Suggested-by: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Kmem accounting of memcg is unusable now, because it lacks slab shrinker
support. That means when we hit the limit we will get ENOMEM w/o any
chance to recover. What we should do then is to call shrink_slab, which
would reclaim old inode/dentry caches from this cgroup. This is what
this patch set is intended to do.Basically, it does two things. First, it introduces the notion of
per-memcg slab shrinker. A shrinker that wants to reclaim objects per
cgroup should mark itself as SHRINKER_MEMCG_AWARE. Then it will be
passed the memory cgroup to scan from in shrink_control->memcg. For
such shrinkers shrink_slab iterates over the whole cgroup subtree under
the target cgroup and calls the shrinker for each kmem-active memory
cgroup.Secondly, this patch set makes the list_lru structure per-memcg. It's
done transparently to list_lru users - everything they have to do is to
tell list_lru_init that they want memcg-aware list_lru. Then the
list_lru will automatically distribute objects among per-memcg lists
basing on which cgroup the object is accounted to. This way to make FS
shrinkers (icache, dcache) memcg-aware we only need to make them use
memcg-aware list_lru, and this is what this patch set does.As before, this patch set only enables per-memcg kmem reclaim when the
pressure goes from memory.limit, not from memory.kmem.limit. Handling
memory.kmem.limit is going to be tricky due to GFP_NOFS allocations, and
it is still unclear whether we will have this knob in the unified
hierarchy.This patch (of 9):
NUMA aware slab shrinkers use the list_lru structure to distribute
objects coming from different NUMA nodes to different lists. Whenever
such a shrinker needs to count or scan objects from a particular node,
it issues commands like this:count = list_lru_count_node(lru, sc->nid);
freed = list_lru_walk_node(lru, sc->nid, isolate_func,
isolate_arg, &sc->nr_to_scan);where sc is an instance of the shrink_control structure passed to it
from vmscan.To simplify this, let's add special list_lru functions to be used by
shrinkers, list_lru_shrink_count() and list_lru_shrink_walk(), which
consolidate the nid and nr_to_scan arguments in the shrink_control
structure.This will also allow us to avoid patching shrinkers that use list_lru
when we make shrink_slab() per-memcg - all we will have to do is extend
the shrink_control structure to include the target memcg and make
list_lru_shrink_{count,walk} handle this appropriately.Signed-off-by: Vladimir Davydov
Suggested-by: Dave Chinner
Cc: Johannes Weiner
Cc: Michal Hocko
Cc: Greg Thelen
Cc: Glauber Costa
Cc: Alexander Viro
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: David Rientjes
Cc: Joonsoo Kim
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
If a PTE or PMD is already marked NUMA when scanning to mark entries for
NUMA hinting then it is not necessary to update the entry and incur a TLB
flush penalty. Avoid the avoidhead where possible.Signed-off-by: Mel Gorman
Cc: Aneesh Kumar K.V
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Sasha Levin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
pte_protnone_numa is only safe to use after VMA checks for PROT_NONE are
complete. Treating a real PROT_NONE PTE as a NUMA hinting fault is going
to result in strangeness so add a check for it. BUG_ON looks like
overkill but if this is hit then it's a serious bug that could result in
corruption so do not even try recovering. It would have been more
comprehensive to check VMA flags in pte_protnone_numa but it would have
made the API ugly just for a debugging check.Signed-off-by: Mel Gorman
Cc: Aneesh Kumar K.V
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Sasha Levin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Commit b38af4721f59 ("x86,mm: fix pte_special versus pte_numa") adjusted
the pte_special check to take into account that a special pte had
SPECIAL and neither PRESENT nor PROTNONE. Now that NUMA hinting PTEs
are no longer modifying _PAGE_PRESENT it should be safe to restore the
original pte_special behaviour.Signed-off-by: Mel Gorman
Cc: Aneesh Kumar K.V
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Sasha Levin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Faults on the huge zero page are pointless and there is a BUG_ON to catch
them during fault time. This patch reintroduces a check that avoids
marking the zero page PAGE_NONE.Signed-off-by: Mel Gorman
Cc: Aneesh Kumar K.V
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Sasha Levin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch removes the NUMA PTE bits and associated helpers. As a
side-effect it increases the maximum possible swap space on x86-64.One potential source of problems is races between the marking of PTEs
PROT_NONE, NUMA hinting faults and migration. It must be guaranteed that
a PTE being protected is not faulted in parallel, seen as a pte_none and
corrupting memory. The base case is safe but transhuge has problems in
the past due to an different migration mechanism and a dependance on page
lock to serialise migrations and warrants a closer look.task_work hinting update parallel fault
------------------------ --------------
change_pmd_range
change_huge_pmd
__pmd_trans_huge_lock
pmdp_get_and_clear
__handle_mm_fault
pmd_none
do_huge_pmd_anonymous_page
read? pmd_lock blocks until hinting complete, fail !pmd_none test
write? __do_huge_pmd_anonymous_page acquires pmd_lock, checks pmd_none
pmd_modify
set_pmd_attask_work hinting update parallel migration
------------------------ ------------------
change_pmd_range
change_huge_pmd
__pmd_trans_huge_lock
pmdp_get_and_clear
__handle_mm_fault
do_huge_pmd_numa_page
migrate_misplaced_transhuge_page
pmd_lock waits for updates to complete, recheck pmd_same
pmd_modify
set_pmd_atBoth of those are safe and the case where a transhuge page is inserted
during a protection update is unchanged. The case where two processes try
migrating at the same time is unchanged by this series so should still be
ok. I could not find a case where we are accidentally depending on the
PTE not being cleared and flushed. If one is missed, it'll manifest as
corruption problems that start triggering shortly after this series is
merged and only happen when NUMA balancing is enabled.Signed-off-by: Mel Gorman
Tested-by: Sasha Levin
Cc: Aneesh Kumar K.V
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Mark Brown
Cc: Stephen Rothwell
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
With PROT_NONE, the traditional page table manipulation functions are
sufficient.[andre.przywara@arm.com: fix compiler warning in pmdp_invalidate()]
[akpm@linux-foundation.org: fix build with STRICT_MM_TYPECHECKS]
Signed-off-by: Mel Gorman
Acked-by: Linus Torvalds
Acked-by: Aneesh Kumar
Tested-by: Sasha Levin
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Paul Mackerras
Cc: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
ppc64 should not be depending on DSISR_PROTFAULT and it's unexpected if
they are triggered. This patch adds warnings just in case they are being
accidentally depended upon.Signed-off-by: Mel Gorman
Acked-by: Aneesh Kumar K.V
Tested-by: Sasha Levin
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Convert existing users of pte_numa and friends to the new helper. Note
that the kernel is broken after this patch is applied until the other page
table modifiers are also altered. This patch layout is to make review
easier.Signed-off-by: Mel Gorman
Acked-by: Linus Torvalds
Acked-by: Aneesh Kumar
Acked-by: Benjamin Herrenschmidt
Tested-by: Sasha Levin
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Sasha Levin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This is a preparatory patch that introduces protnone helpers for automatic
NUMA balancing.Signed-off-by: Mel Gorman
Acked-by: Linus Torvalds
Acked-by: Aneesh Kumar K.V
Tested-by: Sasha Levin
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Paul Mackerras
Cc: Rik van Riel
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Automatic NUMA balancing depends on being able to protect PTEs to trap a
fault and gather reference locality information. Very broadly speaking
it would mark PTEs as not present and use another bit to distinguish
between NUMA hinting faults and other types of faults. It was
universally loved by everybody and caused no problems whatsoever. That
last sentence might be a lie.This series is very heavily based on patches from Linus and Aneesh to
replace the existing PTE/PMD NUMA helper functions with normal change
protections. I did alter and add parts of it but I consider them
relatively minor contributions. At their suggestion, acked-bys are in
there but I've no problem converting them to Signed-off-by if requested.AFAIK, this has received no testing on ppc64 and I'm depending on Aneesh
for that. I tested trinity under kvm-tool and passed and ran a few
other basic tests. At the time of writing, only the short-lived tests
have completed but testing of V2 indicated that long-term testing had no
surprises. In most cases I'm leaving out detail as it's not that
interesting.specjbb single JVM: There was negligible performance difference in the
benchmark itself for short runs. However, system activity is
higher and interrupts are much higher over time -- possibly TLB
flushes. Migrations are also higher. Overall, this is more overhead
but considering the problems faced with the old approach I think
we just have to suck it up and find another way of reducing the
overhead.specjbb multi JVM: Negligible performance difference to the actual benchmark
but like the single JVM case, the system overhead is noticeably
higher. Again, interrupts are a major factor.autonumabench: This was all over the place and about all that can be
reasonably concluded is that it's different but not necessarily
better or worse.autonumabench
3.18.0-rc5 3.18.0-rc5
mmotm-20141119 protnone-v3r3
User NUMA01 32380.24 ( 0.00%) 21642.92 ( 33.16%)
User NUMA01_THEADLOCAL 22481.02 ( 0.00%) 22283.22 ( 0.88%)
User NUMA02 3137.00 ( 0.00%) 3116.54 ( 0.65%)
User NUMA02_SMT 1614.03 ( 0.00%) 1543.53 ( 4.37%)
System NUMA01 322.97 ( 0.00%) 1465.89 (-353.88%)
System NUMA01_THEADLOCAL 91.87 ( 0.00%) 49.32 ( 46.32%)
System NUMA02 37.83 ( 0.00%) 14.61 ( 61.38%)
System NUMA02_SMT 7.36 ( 0.00%) 7.45 ( -1.22%)
Elapsed NUMA01 716.63 ( 0.00%) 599.29 ( 16.37%)
Elapsed NUMA01_THEADLOCAL 553.98 ( 0.00%) 539.94 ( 2.53%)
Elapsed NUMA02 83.85 ( 0.00%) 83.04 ( 0.97%)
Elapsed NUMA02_SMT 86.57 ( 0.00%) 79.15 ( 8.57%)
CPU NUMA01 4563.00 ( 0.00%) 3855.00 ( 15.52%)
CPU NUMA01_THEADLOCAL 4074.00 ( 0.00%) 4136.00 ( -1.52%)
CPU NUMA02 3785.00 ( 0.00%) 3770.00 ( 0.40%)
CPU NUMA02_SMT 1872.00 ( 0.00%) 1959.00 ( -4.65%)System CPU usage of NUMA01 is worse but it's an adverse workload on this
machine so I'm reluctant to conclude that it's a problem that matters. On
the other workloads that are sensible on this machine, system CPU usage is
great. Overall time to complete the benchmark is comparable3.18.0-rc5 3.18.0-rc5
mmotm-20141119protnone-v3r3
User 59612.50 48586.44
System 460.22 1537.45
Elapsed 1442.20 1304.29NUMA alloc hit 5075182 5743353
NUMA alloc miss 0 0
NUMA interleave hit 0 0
NUMA alloc local 5075174 5743339
NUMA base PTE updates 637061448 443106883
NUMA huge PMD updates 1243434 864747
NUMA page range updates 1273699656 885857347
NUMA hint faults 1658116 1214277
NUMA hint local faults 959487 754113
NUMA hint local percent 57 62
NUMA pages migrated 5467056 61676398The NUMA pages migrated look terrible but when I looked at a graph of the
activity over time I see that the massive spike in migration activity was
during NUMA01. This correlates with high system CPU usage and could be
simply down to bad luck but any modifications that affect that workload
would be related to scan rates and migrations, not the protection
mechanism. For all other workloads, migration activity was comparable.Overall, headline performance figures are comparable but the overhead is
higher, mostly in interrupts. To some extent, higher overhead from this
approach was anticipated but not to this degree. It's going to be
necessary to reduce this again with a separate series in the future. It's
still worth going ahead with this series though as it's likely to avoid
constant headaches with Xen and is probably easier to maintain.This patch (of 10):
A transhuge NUMA hinting fault may find the page is migrating and should
wait until migration completes. The check is race-prone because the pmd
is deferenced outside of the page lock and while the race is tiny, it'll
be larger if the PMD is cleared while marking PMDs for hinting fault.
This patch closes the race.Signed-off-by: Mel Gorman
Cc: Aneesh Kumar K.V
Cc: Benjamin Herrenschmidt
Cc: Dave Jones
Cc: Hugh Dickins
Cc: Ingo Molnar
Cc: Kirill Shutemov
Cc: Linus Torvalds
Cc: Paul Mackerras
Cc: Rik van Riel
Cc: Sasha Levin
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Pull jfs updates from David Kleikamp:
"A couple cleanups for jfs"* tag 'jfs-3.20' of git://github.com/kleikamp/linux-shaggy:
jfs: Deletion of an unnecessary check before the function call "unload_nls"
jfs: get rid of homegrown endianness helpers -
Pull nfsd updates from Bruce Fields:
"The main change is the pNFS block server support from Christoph, which
allows an NFS client connected to shared disk to do block IO to the
shared disk in place of NFS reads and writes. This also requires xfs
patches, which should arrive soon through the xfs tree, barring
unexpected problems. Support for other filesystems is also possible
if there's interest.Thanks also to Chuck Lever for continuing work to get NFS/RDMA into
shape"* 'for-3.20' of git://linux-nfs.org/~bfields/linux: (32 commits)
nfsd: default NFSv4.2 to on
nfsd: pNFS block layout driver
exportfs: add methods for block layout exports
nfsd: add trace events
nfsd: update documentation for pNFS support
nfsd: implement pNFS layout recalls
nfsd: implement pNFS operations
nfsd: make find_any_file available outside nfs4state.c
nfsd: make find/get/put file available outside nfs4state.c
nfsd: make lookup/alloc/unhash_stid available outside nfs4state.c
nfsd: add fh_fsid_match helper
nfsd: move nfsd_fh_match to nfsfh.h
fs: add FL_LAYOUT lease type
fs: track fl_owner for leases
nfs: add LAYOUT_TYPE_MAX enum value
nfsd: factor out a helper to decode nfstime4 values
sunrpc/lockd: fix references to the BKL
nfsd: fix year-2038 nfs4 state problem
svcrdma: Handle additional inline content
svcrdma: Move read list XDR round-up logic
... -
Pull IOMMU updates from Joerg Roedel:
"This time with:- Generic page-table framework for ARM IOMMUs using the LPAE
page-table format, ARM-SMMU and Renesas IPMMU make use of it
already.- Break out the IO virtual address allocator from the Intel IOMMU so
that it can be used by other DMA-API implementations too. The
first user will be the ARM64 common DMA-API implementation for
IOMMUs- Device tree support for Renesas IPMMU
- Various fixes and cleanups all over the place"
* tag 'iommu-updates-v3.20' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (36 commits)
iommu/amd: Convert non-returned local variable to boolean when relevant
iommu: Update my email address
iommu/amd: Use wait_event in put_pasid_state_wait
iommu/amd: Fix amd_iommu_free_device()
iommu/arm-smmu: Avoid build warning
iommu/fsl: Various cleanups
iommu/fsl: Use %pa to print phys_addr_t
iommu/omap: Print phys_addr_t using %pa
iommu: Make more drivers depend on COMPILE_TEST
iommu/ipmmu-vmsa: Fix IOMMU lookup when multiple IOMMUs are registered
iommu: Disable on !MMU builds
iommu/fsl: Remove unused fsl_of_pamu_ids[]
iommu/fsl: Fix section mismatch
iommu/ipmmu-vmsa: Use the ARM LPAE page table allocator
iommu: Fix trace_map() to report original iova and original size
iommu/arm-smmu: add support for iova_to_phys through ATS1PR
iopoll: Introduce memory-mapped IO polling macros
iommu/arm-smmu: don't touch the secure STLBIALL register
iommu/arm-smmu: make use of generic LPAE allocator
iommu: io-pgtable-arm: add non-secure quirk
... -
Pull DeviceTree changes from Rob Herring:
- DT unittests for I2C probing and overlays from Pantelis Antoniou
- Remove DT unittest dependency on OF_DYNAMIC from Gaurav Minocha
- Add Tegra compatible strings missing for newer parts from Paul
Walmsley- Various vendor prefix additions
* tag 'devicetree-for-3.20' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
of: Add vendor prefix for OmniVision Technologies
of: Use ovti for Omnivision
of: Add vendor prefix for Truly Semiconductors Limited
of: Add vendor prefix for Himax Technologies Inc.
of/fdt: fix sparse warning
of: unitest: Add I2C overlay unit tests.
Documentation: DT: document compatible string existence requirement
Documentation: DT bindings: add nvidia, tegra132-denver compatible string
Documentation: DT bindings: add more Tegra chip compatible strings
of: EXPORT_SYMBOL_GPL of_property_read_u64_array
of: Fix brace position for struct of_device_id definition
of/unittest: Remove obsolete code
dt-bindings: use isil prefix for Intersil in vendor-prefixes.txt
Add AD Holdings Plc. to vendor-prefixes.
dt-bindings: Add Silicon Mitus vendor prefix
Removes OF_UNITTEST dependency on OF_DYNAMIC config symbol
pinctrl: fix up device tree bindings
DT: Vendors: Add Everspin
doc: add bindings document for altera fpga manager
drivers: of: Export of_reserved_mem_device_{init,release} -
Pull ARM updates from Russell King:
- clang assembly fixes from Ard
- optimisations and cleanups for Aurora L2 cache support
- efficient L2 cache support for secure monitor API on Exynos SoCs
- debug menu cleanup from Daniel Thompson to allow better behaviour for
multiplatform kernels- StrongARM SA11x0 conversion to irq domains, and pxa_timer
- kprobes updates for older ARM CPUs
- move probes support out of arch/arm/kernel to arch/arm/probes
- add inline asm support for the rbit (reverse bits) instruction
- provide an ARM mode secondary CPU entry point (for Qualcomm CPUs)
- remove the unused ARMv3 user access code
- add driver_override support to AMBA Primecell bus
* 'for-linus' of git://ftp.arm.linux.org.uk/~rmk/linux-arm: (55 commits)
ARM: 8256/1: driver coamba: add device binding path 'driver_override'
ARM: 8301/1: qcom: Use secondary_startup_arm()
ARM: 8302/1: Add a secondary_startup that assumes ARM mode
ARM: 8300/1: teach __asmeq that r11 == fp and r12 == ip
ARM: kprobes: Fix compilation error caused by superfluous '*'
ARM: 8297/1: cache-l2x0: optimize aurora range operations
ARM: 8296/1: cache-l2x0: clean up aurora cache handling
ARM: 8284/1: sa1100: clear RCSR_SMR on resume
ARM: 8283/1: sa1100: collie: clear PWER register on machine init
ARM: 8282/1: sa1100: use handle_domain_irq
ARM: 8281/1: sa1100: move GPIO-related IRQ code to gpio driver
ARM: 8280/1: sa1100: switch to irq_domain_add_simple()
ARM: 8279/1: sa1100: merge both GPIO irqdomains
ARM: 8278/1: sa1100: split irq handling for low GPIOs
ARM: 8291/1: replace magic number with PAGE_SHIFT macro in fixup_pv code
ARM: 8290/1: decompressor: fix a wrong comment
ARM: 8286/1: mm: Fix dma_contiguous_reserve comment
ARM: 8248/1: pm: remove outdated comment
ARM: 8274/1: Fix DEBUG_LL for multi-platform kernels (without PL01X)
ARM: 8273/1: Seperate DEBUG_UART_PHYS from DEBUG_LL on EP93XX
... -
Pull AVR32 update from Hans-Christian Egtvedt.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/egtvedt/linux-avr32:
avr32: update all default configurations
avr32: remove fake at91 cpu identification
avr32: wire up missing syscalls -
Pull tracing updates from Steven Rostedt:
"The updates included in this pull request for ftrace are:o Several clean ups to the code
One such clean up was to convert to 64 bit time keeping, in the
ring buffer benchmark code.o Adding of __print_array() helper macro for TRACE_EVENT()
o Updating the sample/trace_events/ to add samples of different ways
to make trace events. Lots of features have been added since the
sample code was made, and these features are mostly unknown.
Developers have been making their own hacks to do things that are
already available.o Performance improvements. Most notably, I found a performance bug
where a waiter that is waiting for a full page from the ring buffer
will see that a full page is not available, and go to sleep. The
sched event caused by it going to sleep would cause it to wake up
again. It would see that there was still not a full page, and go
back to sleep again, and that would wake it up again, until finally
it would see a full page. This change has been marked for stable.Other improvements include removing global locks from fast paths"
* tag 'trace-v3.20' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ring-buffer: Do not wake up a splice waiter when page is not full
tracing: Fix unmapping loop in tracing_mark_write
tracing: Add samples of DECLARE_EVENT_CLASS() and DEFINE_EVENT()
tracing: Add TRACE_EVENT_FN example
tracing: Add TRACE_EVENT_CONDITION sample
tracing: Update the TRACE_EVENT fields available in the sample code
tracing: Separate out initializing top level dir from instances
tracing: Make tracing_init_dentry_tr() static
trace: Use 64-bit timekeeping
tracing: Add array printing helper
tracing: Remove newline from trace_printk warning banner
tracing: Use IS_ERR() check for return value of tracing_init_dentry()
tracing: Remove unneeded includes of debugfs.h and fs.h
tracing: Remove taking of trace_types_lock in pipe files
tracing: Add ref count to tracer for when they are being read by pipe -
Pull ktest updates from Steven Rostedt:
"The following ktest updates were done:o Added timings to various parts of the test (build, install, boot,
tests) and report them so that the users can keep track of changes.o Josh Poimboeuf fixed the console output to work better with virtual
machine targets.o Various clean ups and fixes"
* tag 'ktest-v3.20' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest:
ktest: Place quotes around item variable
ktest: Cleanup terminal on dodie() failure
ktest: Print build,install,boot,test times at success and failure
ktest: Enable user input to the console
ktest: Give console process a dedicated tty
ktest: Rename start_monitor_and_boot to start_monitor_and_install
ktest: Show times for build, install, boot and test
ktest: Restore tty settings after closing console
ktest: Add timings for commands
12 Feb, 2015
6 commits
-
Pull security layer updates from James Morris:
"Highlights:- Smack adds secmark support for Netfilter
- /proc/keys is now mandatory if CONFIG_KEYS=y
- TPM gets its own device class
- Added TPM 2.0 support
- Smack file hook rework (all Smack users should review this!)"* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (64 commits)
cipso: don't use IPCB() to locate the CIPSO IP option
SELinux: fix error code in policydb_init()
selinux: add security in-core xattr support for pstore and debugfs
selinux: quiet the filesystem labeling behavior message
selinux: Remove unused function avc_sidcmp()
ima: /proc/keys is now mandatory
Smack: Repair netfilter dependency
X.509: silence asn1 compiler debug output
X.509: shut up about included cert for silent build
KEYS: Make /proc/keys unconditional if CONFIG_KEYS=y
MAINTAINERS: email update
tpm/tpm_tis: Add missing ifdef CONFIG_ACPI for pnp_acpi_device
smack: fix possible use after frees in task_security() callers
smack: Add missing logging in bidirectional UDS connect check
Smack: secmark support for netfilter
Smack: Rework file hooks
tpm: fix format string error in tpm-chip.c
char/tpm/tpm_crb: fix build error
smack: Fix a bidirectional UDS connect check typo
smack: introduce a special case for tmpfs in smack_d_instantiate()
... -
Pull audit fix from Paul Moore:
"Just one patch from the audit tree for v3.20, and a very minor one at
that.The patch simply removes an old, unused field from the audit_krule
structure, a private audit-only struct. In audit related news, we did
a proper overhaul of the audit pathname code and removed the nasty
getname()/putname() hacks for audit, you should see those patches in
Al's vfs tree if you haven't already.That's it for audit this time, let's hope for a quiet -rcX series"
* 'upstream' of git://git.infradead.org/users/pcmoore/audit:
audit: remove vestiges of vers_ops -
Merge second set of updates from Andrew Morton:
"More of MM"* emailed patches from Andrew Morton : (83 commits)
mm/nommu.c: fix arithmetic overflow in __vm_enough_memory()
mm/mmap.c: fix arithmetic overflow in __vm_enough_memory()
vmstat: Reduce time interval to stat update on idle cpu
mm/page_owner.c: remove unnecessary stack_trace field
Documentation/filesystems/proc.txt: describe /proc//map_files
mm: incorporate read-only pages into transparent huge pages
vmstat: do not use deferrable delayed work for vmstat_update
mm: more aggressive page stealing for UNMOVABLE allocations
mm: always steal split buddies in fallback allocations
mm: when stealing freepages, also take pages created by splitting buddy page
mincore: apply page table walker on do_mincore()
mm: /proc/pid/clear_refs: avoid split_huge_page()
mm: pagewalk: fix misbehavior of walk_page_range for vma(VM_PFNMAP)
mempolicy: apply page table walker on queue_pages_range()
arch/powerpc/mm/subpage-prot.c: use walk->vma and walk_page_vma()
memcg: cleanup preparation for page table walk
numa_maps: remove numa_maps->vma
numa_maps: fix typo in gather_hugetbl_stats
pagemap: use walk->vma instead of calling find_vma()
clear_refs: remove clear_refs_private->vma and introduce clear_refs_test_walk()
... -
Pull powerpc updates from Michael Ellerman:
- Update of all defconfigs
- Addition of a bunch of config options to modernise our defconfigs
- Some PS3 updates from Geoff
- Optimised memcmp for 64 bit from Anton
- Fix for kprobes that allows 'perf probe' to work from Naveen
- Several cxl updates from Ian & Ryan
- Expanded support for the '24x7' PMU from Cody & Sukadev
- Freescale updates from Scott:
"Highlights include 8xx optimizations, some more work on datapath
device tree content, e300 machine check support, t1040 corenet
error reporting, and various cleanups and fixes"* tag 'powerpc-3.20-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux: (102 commits)
cxl: Add missing return statement after handling AFU errror
cxl: Fail AFU initialisation if an invalid configuration record is found
cxl: Export optional AFU configuration record in sysfs
powerpc/mm: Warn on flushing tlb page in kernel context
powerpc/powernv: Add OPAL soft-poweroff routine
powerpc/perf/hv-24x7: Document sysfs event description entries
powerpc/perf/hv-gpci: add the remaining gpci requests
powerpc/perf/{hv-gpci, hv-common}: generate requests with counters annotated
powerpc/perf/hv-24x7: parse catalog and populate sysfs with events
perf: define EVENT_DEFINE_RANGE_FORMAT_LITE helper
perf: add PMU_EVENT_ATTR_STRING() helper
perf: provide sysfs_show for struct perf_pmu_events_attr
powerpc/kernel: Avoid initializing device-tree pointer twice
powerpc: Remove old compile time disabled syscall tracing code
powerpc/kernel: Make syscall_exit a local label
cxl: Fix device_node reference counting
powerpc/mm: bail out early when flushing TLB page
powerpc: defconfigs: add MTD_SPI_NOR (new dependency for M25P80)
perf/powerpc: reset event hw state when adding it to the PMU
powerpc/qe: Use strlcpy()
... -
Pull arm64 updates from Catalin Marinas:
"arm64 updates for 3.20:- reimplementation of the virtual remapping of UEFI Runtime Services
in a way that is stable across kexec
- emulation of the "setend" instruction for 32-bit tasks (user
endianness switching trapped in the kernel, SCTLR_EL1.E0E bit set
accordingly)
- compat_sys_call_table implemented in C (from asm) and made it a
constant array together with sys_call_table
- export CPU cache information via /sys (like other architectures)
- DMA API implementation clean-up in preparation for IOMMU support
- macros clean-up for KVM
- dropped some unnecessary cache+tlb maintenance
- CONFIG_ARM64_CPU_SUSPEND clean-up
- defconfig update (CPU_IDLE)The EFI changes going via the arm64 tree have been acked by Matt
Fleming. There is also a patch adding sys_*stat64 prototypes to
include/linux/syscalls.h, acked by Andrew Morton"* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (47 commits)
arm64: compat: Remove incorrect comment in compat_siginfo
arm64: Fix section mismatch on alloc_init_p[mu]d()
arm64: Avoid breakage caused by .altmacro in fpsimd save/restore macros
arm64: mm: use *_sect to check for section maps
arm64: drop unnecessary cache+tlb maintenance
arm64:mm: free the useless initial page table
arm64: Enable CPU_IDLE in defconfig
arm64: kernel: remove ARM64_CPU_SUSPEND config option
arm64: make sys_call_table const
arm64: Remove asm/syscalls.h
arm64: Implement the compat_sys_call_table in C
syscalls: Declare sys_*stat64 prototypes if __ARCH_WANT_(COMPAT_)STAT64
compat: Declare compat_sys_sigpending and compat_sys_sigprocmask prototypes
arm64: uapi: expose our struct ucontext to the uapi headers
smp, ARM64: Kill SMP single function call interrupt
arm64: Emulate SETEND for AArch32 tasks
arm64: Consolidate hotplug notifier for instruction emulation
arm64: Track system support for mixed endian EL0
arm64: implement generic IOMMU configuration
arm64: Combine coherent and non-coherent swiotlb dma_ops
...
