We don't need to free slab management object in rcu context,
because, from now on, we don't manage this slab anymore.
So put forward freeing.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index b378f91..607a9b8 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -1820,8 +1820,6
in the same
cache line. It is not good for performance.
We can do same thing through more easy way, like as the stack.
This patchset implement it and remove complex code for above algorithm.
This makes slab code much cleaner.
This patchset is based on v3.11-rc6, but tested on v3.10.
Than
Now there is no user colouroff, so remove it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 180f532..d9f81a0 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -219,7 +219,6 @@ struct slab {
union {
struct {
struct list_head list
ust
return 'struct page' of that object, not one of first page, since the SLAB
don't use __GFP_COMP when CONFIG_MMU. To get 'struct page' of first page,
we first get a slab and try to get it via virt_to_head_page(slab->s_mem).
Cc: Mel Gorman
Signed-off-by: Joonsoo Ki
mechanical ones and there is no functional change.
Signed-off-by: Joonsoo Kim
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index ace9a5f..66ee577 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -42,18 +42,22 @@ struct page {
/* First double
b1f9 mm/slab.o
* After
textdata bss dec hex filename
22074 23434 4 45512b1c8 mm/slab.o
And this help following patch to remove struct slab's colouroff.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index d9eae39..180f532 100644
--- a/mm/s
Now, bufctl is not proper name to this array.
So change it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 6abc069..e8ec4c5 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2538,7 +2538,7 @@ static struct freelist *alloc_slabmgmt(struct kmem_cache
*cachep,
return
Now, virt_to_page(page->s_mem) is same as the page,
because slab use this structure for management.
So remove useless statement.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index cf39309..6abc069 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -750,9 +750,7 @@ static str
It's useless now, so remove it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 7216ebe..98257e4 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -163,8 +163,6 @@
*/
static bool pfmemalloc_active __read_mostly;
-#defineSLAB_LIMIT (((unsigned int)(~0
We can get nodeid using address translation, so this field is not useful.
Therefore, remove it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index d9f81a0..69dc25a 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -222,7 +222,6 @@ struct slab {
void *s_mem
Now, free in struct slab is same meaning as inuse.
So, remove both and replace them with active.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 98257e4..9dcbb22 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -174,8 +174,7 @@ struct slab {
struct {
struct
is method.
struct slab's free = 0
kmem_bufctl_t array: 6 3 7 2 5 4 0 1
To get free objects, we access this array with following pattern.
0 -> 1 -> 2 -> 3 -> 4 -> 5 -> 6 -> 7
This may help cache line footprint if slab has many objects, and,
in addition, this makes code much
Now, we changed the management method of free objects of the slab and
there is no need to use special value, BUFCTL_END, BUFCTL_FREE and
BUFCTL_ACTIVE. So remove them.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 4551d57..7216ebe 100644
--- a/mm/slab.c
+++ b/mm/slab.c
We can get cachep using page in struct slab_rcu, so remove it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 69dc25a..b378f91 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -204,7 +204,6 @@ typedef unsigned int kmem_bufctl_t;
*/
struct slab_rcu {
struct rcu_head head
This is trivial change, just use well-defined macro.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 9e98ee0..ee03eba 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2853,7 +2853,6 @@ static inline void verify_redzone_free(struct kmem_cache
*cache, void *obj)
static void
: Joonsoo Kim
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 0c62175..b8d19b1 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -51,7 +51,14 @@
* }
* rcu_read_unlock();
*
- * See also the comment on struct slab_rcu in mm/slab.c.
+ * This is useful if we need to
If we use 'struct page' of first page as 'struct slab', there is no
advantage not to use __GFP_COMP. So use __GFP_COMP flag for all the cases.
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index ee03eba..855f481 100644
--- a/mm/slab.c
+++ b/mm/slab.c
On Thu, Aug 22, 2013 at 02:14:38PM +0530, Aneesh Kumar K.V wrote:
> Joonsoo Kim writes:
>
> > vma_has_reserves() can be substituted by using return value of
> > vma_needs_reservation(). If chg returned by vma_needs_reservation()
> > is 0, it means that vma has reserves.
On Thu, Aug 22, 2013 at 12:42:57PM +0200, Peter Zijlstra wrote:
> > >
> > > +redo:
> >
> > One behavioral change worth noting here is that in the redo case if a
> > CPU has become idle we'll continue trying to load-balance in the
> > !new-idle case.
> >
> > This could be unpleasant in the case wh
On Thu, Aug 22, 2013 at 04:34:22PM +0530, Aneesh Kumar K.V wrote:
> Joonsoo Kim writes:
>
> > On Thu, Aug 22, 2013 at 02:14:38PM +0530, Aneesh Kumar K.V wrote:
> >> Joonsoo Kim writes:
> >>
> >> > vma_has_reserves() can be substituted by using return v
On Thu, Aug 22, 2013 at 04:47:25PM +, Christoph Lameter wrote:
> On Thu, 22 Aug 2013, Joonsoo Kim wrote:
>
> > And this patchset change a management method of free objects of a slab.
> > Current free objects management method of the slab is weird, because
> > it touc
On Thu, Aug 22, 2013 at 05:49:43PM +, Christoph Lameter wrote:
> On Thu, 22 Aug 2013, Joonsoo Kim wrote:
>
> > @@ -2042,7 +2042,7 @@ static void slab_destroy_debugcheck(struct kmem_cache
> > *cachep, struct slab *slab
> > */
> > static void slab_destroy(str
On Thu, Aug 22, 2013 at 05:51:58PM +, Christoph Lameter wrote:
> On Thu, 22 Aug 2013, Joonsoo Kim wrote:
>
> > @@ -1099,8 +1098,7 @@ static void drain_alien_cache(struct kmem_cache
> > *cachep,
> >
> > static inline int cache_free_alien(struct
On Thu, Aug 22, 2013 at 05:53:00PM +, Christoph Lameter wrote:
> On Thu, 22 Aug 2013, Joonsoo Kim wrote:
>
> > We can get cachep using page in struct slab_rcu, so remove it.
>
> Ok but this means that we need to touch struct page. Additional cacheline
> in cache foo
On Thu, Aug 22, 2013 at 06:00:56PM +, Christoph Lameter wrote:
> On Thu, 22 Aug 2013, Joonsoo Kim wrote:
>
> > If we use 'struct page' of first page as 'struct slab', there is no
> > advantage not to use __GFP_COMP. So use __GFP_COMP flag for all the case
2013/8/23 Christoph Lameter :
> On Fri, 23 Aug 2013, Joonsoo Kim wrote:
>
>> On Thu, Aug 22, 2013 at 05:53:00PM +, Christoph Lameter wrote:
>> > On Thu, 22 Aug 2013, Joonsoo Kim wrote:
>> >
>> > > We can get cachep using page in struct slab_rcu, so r
2013/8/24 Christoph Lameter :
> On Fri, 23 Aug 2013, JoonSoo Kim wrote:
>
>> I don't get it. This patch only affect to the rcu case, because it
>> change the code
>> which is in kmem_rcu_free(). It doesn't touch anything in standard case.
>
> In general th
tches, because I'm not sure they are right.
Joonsoo Kim (3):
sched: remove one division operation in find_buiest_queue()
sched: factor out code to should_we_balance()
sched: clean-up struct sd_lb_stat
kernel/sched/fair.c | 326 +--
1 file ch
d-off-by: Joonsoo Kim
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c6732d2..f8a9660 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4232,36 +4232,6 @@ static unsigned long task_h_load(struct task_struct *p)
/** Helpers for find_busiest_
PageSwapCache() is always false when !CONFIG_SWAP, so compiler
properly discard related code. Therefore, we don't need #ifdef explicitly.
Acked-by: Johannes Weiner
Signed-off-by: Joonsoo Kim
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 24db914..c03c139 100644
--- a/in
does right and nobody
re-evaluate if gcc do proper optimization with their change, for example,
it is not optimized properly on v3.10. So adding compiler hint here
is reasonable.
Acked-by: Johannes Weiner
Signed-off-by: Joonsoo Kim
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f5c549c..04
354958aa7 kernel/sched/fair.o
In addition, rename @balance to @should_balance in order to represent
its purpose more clearly.
Signed-off-by: Joonsoo Kim
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 52898dc..c6732d2 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
pgtable related functions are mostly in pgtable-generic.c.
So move remaining functions from memory.c to pgtable-generic.c.
Signed-off-by: Joonsoo Kim
diff --git a/mm/memory.c b/mm/memory.c
index f2ab2a8..8fd4d42 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -374,30 +374,6 @@ void
Remove one division operation in find_buiest_queue().
Signed-off-by: Joonsoo Kim
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 9565645..52898dc 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4968,7 +4968,7 @@ static struct rq *find_busiest_queue(struct lb_env *env
If we use a division operation, we can compute a batch count more closed
to ideal value. With this value, we can finish our job within
MIGRATE_PCPTYPES iteration. In addition, batching to free more pages
may be helpful to cache usage.
Signed-off-by: Joonsoo Kim
diff --git a/mm/page_alloc.c b/mm
If we use a division operation, we can compute a batch count more closed
to ideal value. With this value, we can finish our job within
MIGRATE_PCPTYPES iteration. In addition, batching to free more pages
may be helpful to cache usage.
Signed-off-by: Joonsoo Kim
diff --git a/mm/page_alloc.c b/mm
Currently, we free the avc objects with holding a lock. To minimize
lock hold time, we just move the avc objects to another list
with holding a lock. Then, iterate them and free objects without holding
a lock. This makes lock hold time minimized.
Signed-off-by: Joonsoo Kim
diff --git a/mm
If we fail due to some errorous situation, it is better to quit
without doing heavy work. So changing order of execution.
Signed-off-by: Joonsoo Kim
diff --git a/mm/rmap.c b/mm/rmap.c
index a149e3a..c2f51cb 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -278,19 +278,19 @@ int anon_vma_fork(struct
If we allocate anon_vma_chain before starting to link, we can reduce
the lock hold time. This patch implement it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/rmap.c b/mm/rmap.c
index c2f51cb..1603f64 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -240,18 +240,21 @@ int anon_vma_clone(struct
On Tue, Aug 06, 2013 at 05:43:40PM +0900, Joonsoo Kim wrote:
> If we use a division operation, we can compute a batch count more closed
> to ideal value. With this value, we can finish our job within
> MIGRATE_PCPTYPES iteration. In addition, batching to free more pages
> may be help
Hello, Johannes.
On Tue, Aug 06, 2013 at 08:58:54AM -0400, Johannes Weiner wrote:
> > if (anon_vma_clone(vma, pvma))
> > - return -ENOMEM;
> > -
> > - /* Then add our own anon_vma. */
> > - anon_vma = anon_vma_alloc();
> > - if (!anon_vma)
> > - goto out_error;
> > -
On Wed, Aug 07, 2013 at 02:08:03AM -0400, Johannes Weiner wrote:
> >
> > list_for_each_entry_reverse(pavc, &src->anon_vma_chain, same_vma) {
> > struct anon_vma *anon_vma;
> >
> > - avc = anon_vma_chain_alloc(GFP_NOWAIT | __GFP_NOWARN);
> > - if (unlikely(!av
On Wed, Aug 07, 2013 at 02:11:38AM -0400, Johannes Weiner wrote:
> On Tue, Aug 06, 2013 at 05:43:39PM +0900, Joonsoo Kim wrote:
> > Currently, we free the avc objects with holding a lock. To minimize
> > lock hold time, we just move the avc objects to another list
> > with
On Tue, Aug 06, 2013 at 06:38:49PM -0700, Davidlohr Bueso wrote:
> On Wed, 2013-08-07 at 11:03 +1000, David Gibson wrote:
> > On Tue, Aug 06, 2013 at 05:18:44PM -0700, Davidlohr Bueso wrote:
> > > On Mon, 2013-08-05 at 16:36 +0900, Joonsoo Kim wrote:
> > > > >
On Tue, Aug 06, 2013 at 05:08:04PM -0700, Davidlohr Bueso wrote:
> On Mon, 2013-07-29 at 15:18 +0900, Joonsoo Kim wrote:
> > On Fri, Jul 26, 2013 at 07:27:23AM -0700, Davidlohr Bueso wrote:
> > > This patchset attempts to reduce the amount of contention we i
Hello, Andrew.
2013/8/7 Andrew Morton :
> On Tue, 6 Aug 2013 17:40:40 +0900 Joonsoo Kim wrote:
>
>> If we use a division operation, we can compute a batch count more closed
>> to ideal value. With this value, we can finish our job within
>> MIGRATE_PCPTYPES iteration.
then crashed, I changed the test to be
> just:
>
> perf stat -r 10 ./hackbench 50
>
> And kicked off ktest.pl to do the bisect. It came up with this commit as
> the culprit:
>
> commit 318df36e57c0ca9f2146660d41ff28e8650af423
> Author: Joonsoo Kim
> Date: Wed Jun 19
On Fri, Jul 19, 2013 at 02:24:15PM -0700, Davidlohr Bueso wrote:
> On Fri, 2013-07-19 at 17:14 +1000, David Gibson wrote:
> > On Thu, Jul 18, 2013 at 05:42:35PM +0900, Joonsoo Kim wrote:
> > > On Wed, Jul 17, 2013 at 12:50:25PM -0700, Davidlohr Bueso wrote:
> >
In this time we are holding a hugetlb_lock, so hstate values can't
be changed. If we don't have any usable free huge page in this time,
we don't need to proceede the processing. So move this code up.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e
patch implement it.
Reviewed-by: Wanpeng Li
Reviewed-by: Aneesh Kumar K.V
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 87e73bd..2ea6afd 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -434,25 +434,6 @@ static int is_vma_resv_set(struct vm_area_struct *vma,
unsigned
space.
Reviewed-by: Wanpeng Li
Reviewed-by: Aneesh Kumar K.V
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 8a61638..87e73bd 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -464,6 +464,8 @@ void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
/* Returns true i
if (q == MAP_FAILED) {
fprintf(stderr, "mmap() failed: %s\n", strerror(errno));
}
q[0] = 'c';
This patch solve this problem.
Reviewed-by: Wanpeng Li
Reviewed-by: Aneesh Kumar K.V
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.
If list is empty, list_for_each_entry_safe() doesn't do anything.
So, this check is redundant. Remove it.
Reviewed-by: Aneesh Kumar K.V
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 3ac0a6f..7ca8733 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1017,11 +1
on to a page
in page cache, the problem is disappeared.
Reviewed-by: Wanpeng Li
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 7ca8733..8a61638 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2508,7 +2508,6 @@ static int hugetlb_cow(struct mm_struct *mm, struct
vm
e_mask_to_[alloc|free]" and
fix and clean-up node iteration code to alloc or free.
This makes code more understandable.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 83edd17..3ac0a6f 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -752,33 +752,
We can unify some codes for succeed allocation.
This makes code more readable.
There is no functional difference.
Reviewed-by: Aneesh Kumar K.V
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index d21a33a..83edd17 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1146,12
The name of the mutex written in comment is wrong.
Fix it.
Reviewed-by: Aneesh Kumar K.V
Acked-by: Hillf Danton
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index d87f70b..d21a33a 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -135,9 +135,9 @@ static inline struct
This label is not needed now, because there is no error handling
except returing NULL.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index fc4988c..d87f70b 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -546,11 +546,11 @@ static struct page *dequeue_huge_page_vma(struct
.
Changes from v1.
Split patch 1 into two patches to clear it's purpose.
Remove useless indentation changes in 'clean-up alloc_huge_page()'
Fix new iteration code bug.
Add reviewed-by or acked-by.
Joonsoo Kim (10):
mm, hugetlb: move up the code which check availability of free hug
On Mon, Jul 22, 2013 at 06:11:11PM +0200, Michal Hocko wrote:
> On Mon 22-07-13 17:36:23, Joonsoo Kim wrote:
> > This label is not needed now, because there is no error handling
> > except returing NULL.
> >
> > Signed-off-by: Joonsoo Kim
> >
> > d
On Mon, Jul 22, 2013 at 04:51:50PM +0200, Michal Hocko wrote:
> On Mon 15-07-13 18:52:41, Joonsoo Kim wrote:
> > We can unify some codes for succeed allocation.
> > This makes code more readable.
> > There is no functional difference.
>
> "This patch unifies suc
On Mon, Jul 22, 2013 at 09:21:38PM +0530, Aneesh Kumar K.V wrote:
> Joonsoo Kim writes:
>
> > First 6 patches are almost trivial clean-up patches.
> >
> > The others are for fixing three bugs.
> > Perhaps, these problems are minor, because this codes are used
> &
On Tue, Jul 23, 2013 at 01:45:50PM +0200, Michal Hocko wrote:
> On Mon 22-07-13 17:36:28, Joonsoo Kim wrote:
> > Currently, we use a page with mapped count 1 in page cache for cow
> > optimization. If we find this condition, we don't allocate a new
> > page and copy cont
On Wed, Jul 24, 2013 at 09:00:41AM +0800, Wanpeng Li wrote:
> On Mon, Jul 22, 2013 at 05:36:24PM +0900, Joonsoo Kim wrote:
> >The name of the mutex written in comment is wrong.
> >Fix it.
> >
> >Reviewed-by: Aneesh Kumar K.V
> >Acked-by: Hillf Danton
> >
(Suggedt by Naoya)
[1] http://lwn.net/Articles/558863/
"[PATCH] mm/hugetlb: per-vma instantiation mutexes"
[2] https://lkml.org/lkml/2013/7/22/96
"[PATCH v2 00/10] mm, hugetlb: clean-up and possible bug fix"
Joonsoo Kim (20):
mm, hugetlb: protect reser
get a SIGBUS signal until there is no
concurrent user, and so, we can ensure that no one get a SIGBUS if there
are enough hugepages.
Signed-off-by: Joonsoo Kim
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index e29e28f..981c539 100644
--- a/include/linux/hugetlb.h
+++ b/include
If we fail with a allocated hugepage, we need some effort to recover
properly. So, it is better not to allocate a hugepage as much as possible.
So move up anon_vma_prepare() which can be failed in OOM situation.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 2372f75
Now, we have prepared to have an infrastructure in order to remove a this
awkward mutex which serialize all faulting tasks, so remove it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 0501fe5..f2c3a51 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2504,9 +2504,7
han how we need, because reserve count already decrease in
dequeue_huge_page_vma(). This patch fix this situation.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 6c8eec2..3f834f1 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -572,6 +572,7 @@ re
Current code include 'Caller expects lock to be held' in every error path.
We can clean-up it as we do error handling in one place.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 7e9a651..8743e5c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2500
same as vma_has_reserves(), so remove vma_has_reserves().
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e6c0c77..22ceb04 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -473,39 +473,6 @@ void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 8dff972..bc666cf 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1110,13 +1110,11 @@ static void vma_commit_reservation(struct hstate *h,
}
static struct page *alloc_huge_page(struct vm_area_struct *vma
Currently, we have two variable to represent whether we can use reserved
page or not, chg and avoid_reserve, respectively. With aggregating these,
we can have more clean code. This makes no functinoal difference.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 22ceb04
Now, alloc_huge_page() only return -ENOSPEC if failed.
So, we don't worry about other return value.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index bc666cf..24de2ca 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2544,7 +2544,6 @@ retry_avoidcopy:
new
Util now, we get a resv_map by two ways according to each mapping type.
This makes code dirty and unreadable. So unfiying it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 869c3e0..e6c0c77 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -421,13 +421,24 @@ void
Just move down outside_reserve check and don't check
vma_need_reservation() when outside_resever is true. It is slightly
optimized implementation.
This makes code more readable.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 24de2ca..2372f75 100644
--- a/mm/huge
To change a protection method for region tracking to find grained one,
we pass the resv_map, instead of list_head, to region manipulation
functions. This doesn't introduce any functional change, and it is just
for preparing a next step.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c
In following patch, I change vma_resv_map() to return resv_map
for all case. This patch prepares it by removing resv_map_put() which
doesn't works properly with following change, because it works only for
HPAGE_RESV_OWNER's resv_map, not for all resv_maps.
Signed-off-by: Joonsoo Kim
re to fine grained lock, and this difference hinder it.
So, before changing it, unify region structure handling.
Signed-off-by: Joonsoo Kim
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index a3f868a..9bf2c4a 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -366,7 +3
Don't use the reserve pool when soft offlining a hugepage.
Check we have free pages outside the reserve pool before we
dequeue the huge page. Otherwise, we can steal other's reserve page.
Reviewed-by: Aneesh Kumar
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/huget
ure, so it can be modified by two processes concurrently.
To solve this, I introduce a lock to resv_map and make region manipulation
function grab a lock before they do actual work. This makes region
tracking safe.
Signed-off-by: Joonsoo Kim
diff --git a/include/linux/hugetlb.h b/include/linux
We don't need to grab a page_table_lock when we try to release a page.
So, defer to grab a page_table_lock.
Reviewed-by: Naoya Horiguchi
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index c017c52..6c8eec2 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2627,10 +26
If we alloc hugepage with avoid_reserve, we don't dequeue reserved one.
So, we should check subpool counter when avoid_reserve.
This patch implement it.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 12b6581..ea1ae0a 100644
--- a/mm/hugetlb.c
+++ b/mm/huge
'reservations' is so long name as a variable and we use 'resv_map'
to represent 'struct resv_map' in other place. To reduce confusion and
unreadability, change it.
Reviewed-by: Aneesh Kumar
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index
is_vma_resv_set(vma, HPAGE_RESV_OWNER) implys that this mapping is
for private. So we don't need to check whether this mapping is for
shared or not.
This patch is just for clean-up.
Signed-off-by: Joonsoo Kim
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index ea1ae0a..c017c52 100644
---
> I once attempted an approach involving an atomic counter of the number
> of "in flight" hugepages, only retrying when it's non zero. Working
> out a safe ordering for all the updates to get all the cases right
> made my brain melt though, and I never got it working.
I sent v2 few seconds before
> > @@ -202,15 +199,27 @@ static long region_chg(struct resv_map *resv, long f,
> > long t)
> > * Subtle, allocate a new region at the position but make it zero
> > * size such that we can guarantee to record the reservation. */
> > if (&rg->link == head || t < rg->from) {
> > -
.
Signed-off-by: Joonsoo Kim
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index c7e3759..b7711be 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -822,16 +822,16 @@ static void dma_cache_maint_page(struct page *page,
unsigned long offset
With SMP and enabling kmap_high_get(), it makes users of kmap_atomic()
sequential ordered, because kmap_high_get() use global kmap_lock().
It is not welcome situation, so turn off this optimization for SMP.
Cc: Nicolas Pitre
Signed-off-by: Joonsoo Kim
diff --git a/arch/arm/include/asm
de to count these pages for this task's reclaimed_slab.
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: Matt Mackall
Signed-off-by: Joonsoo Kim
diff --git a/mm/slub.c b/mm/slub.c
index 4aec537..16fd2d5 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1409,8 +1409,6 @@ static void __free_slab(struct
de to count these pages for this task's reclaimed_slab.
Cc: Christoph Lameter
Cc: Pekka Enberg
Cc: Matt Mackall
Signed-off-by: Joonsoo Kim
diff --git a/mm/slab.c b/mm/slab.c
index 856e4a1..4d94bcb 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -1934,8 +1934,6 @@ static void kmem_freepa
In shrink_(in)active_list(), we can fail to put into lru, and these pages
are reclaimed accidentally. Currently, these pages are not counted
for sc->nr_reclaimed, but with this information, we can stop to reclaim
earlier, so can reduce overhead of reclaim.
Signed-off-by: Joonsoo Kim
diff --
Hello, Mel.
Sorry for too late question.
On Sun, Mar 17, 2013 at 01:04:14PM +, Mel Gorman wrote:
> If kswaps fails to make progress but continues to shrink slab then it'll
> either discard all of slab or consume CPU uselessly scanning shrinkers.
> This patch causes kswapd to only call the shri
Hello, Mel.
On Tue, Apr 09, 2013 at 12:13:59PM +0100, Mel Gorman wrote:
> On Tue, Apr 09, 2013 at 03:53:25PM +0900, Joonsoo Kim wrote:
> > Hello, Mel.
> > Sorry for too late question.
> >
>
> No need to apologise at all.
>
> > On Sun, Mar 17, 2013 a
Hello, Dave.
On Wed, Apr 10, 2013 at 11:07:34AM +1000, Dave Chinner wrote:
> On Tue, Apr 09, 2013 at 12:13:59PM +0100, Mel Gorman wrote:
> > On Tue, Apr 09, 2013 at 03:53:25PM +0900, Joonsoo Kim wrote:
> >
> > > I think that outside of zone loop is better
Hello, Christoph.
On Tue, Apr 09, 2013 at 02:28:06PM +, Christoph Lameter wrote:
> On Tue, 9 Apr 2013, Joonsoo Kim wrote:
>
> > Currently, freed pages via rcu is not counted for reclaimed_slab, because
> > it is freed in rcu context, not current task context. But, this fre
Hello, Minchan.
On Tue, Apr 09, 2013 at 02:55:14PM +0900, Minchan Kim wrote:
> Hello Joonsoo,
>
> On Tue, Apr 09, 2013 at 10:21:16AM +0900, Joonsoo Kim wrote:
> > In shrink_(in)active_list(), we can fail to put into lru, and these pages
> > are reclaimed accidentally. Curre
On Wed, Apr 10, 2013 at 09:31:10AM +0300, Pekka Enberg wrote:
> On Mon, Apr 8, 2013 at 3:32 PM, Steven Rostedt wrote:
> >> > Index: linux/mm/slub.c
> >> > ===
> >> > --- linux.orig/mm/slub.c2013-03-28 12:14:26.958358688 -0500
> >>
2013/4/10 Christoph Lameter :
> On Wed, 10 Apr 2013, Joonsoo Kim wrote:
>
>> Hello, Christoph.
>>
>> On Tue, Apr 09, 2013 at 02:28:06PM +, Christoph Lameter wrote:
>> > On Tue, 9 Apr 2013, Joonsoo Kim wrote:
>> >
>> > > Currently, freed pa
When we fail to mutex_trylock(), we release the pool spin_lock and do
mutex_lock(). After that, we should regrab the pool spin_lock, but,
regrabbing is missed in current code. So correct it.
Cc: Lai Jiangshan
Signed-off-by: Joonsoo Kim
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index
401 - 500 of 2325 matches
Mail list logo
