And in almost invoking case, order is 0, so return immediately.
You can't make sure it.
Okay.
Let's not invoke it when order 0
Let's not ruin git blame.
Hmm...
When I do git blame, I can't find anything related to this.
I mean if we merge the pointless patch, it could be
2012/7/7 David Rientjes rient...@google.com:
On Sat, 7 Jul 2012, Joonsoo Kim wrote:
__alloc_pages_direct_compact has many arguments so invoking it is very
costly.
And in almost invoking case, order is 0, so return immediately.
If zero cost is very costly, then this might make sense
2012/7/8 Fengguang Wu fengguang...@intel.com:
Hi Vegard,
This warning code is triggered for the attached config:
__lockdep_trace_alloc():
/*
* Oi! Can't be having __GFP_FS allocations with IRQs disabled.
*/
if
2012/7/7 Christoph Lameter c...@linux.com:
On Fri, 6 Jul 2012, JoonSoo Kim wrote:
At CPU2, we don't need lock anymore, because this slab already in partial
list.
For that scenario we could also simply do a trylock there and redo
the loop if we fail. But still what guarantees
2012/7/9 David Rientjes rient...@google.com:
On Mon, 9 Jul 2012, JoonSoo Kim wrote:
diff --git a/mm/slub.c b/mm/slub.c
index 8c691fa..5d41cad 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1324,8 +1324,14 @@ static struct page *allocate_slab(struct
kmem_cache *s, gfp_t flags, int node
2012/7/9 David Rientjes rient...@google.com:
On Sun, 8 Jul 2012, JoonSoo Kim wrote:
__alloc_pages_direct_compact has many arguments so invoking it is very
costly.
And in almost invoking case, order is 0, so return immediately.
If zero cost is very costly, then this might make sense
2012/7/10 Mel Gorman mgor...@suse.de:
You say that invoking the function is very costly. I agree that a function
call with that many parameters is hefty but it is also in the slow path of
the allocator. For order-0 allocations we are about to enter direct reclaim
where I would expect the cost
2012/7/16 Thomas Gleixner t...@linutronix.de:
- static const struct sched_param param = {
- .sched_priority = MAX_RT_PRIO-1
- };
-
- p = per_cpu(ksoftirqd, hotcpu);
- per_cpu(ksoftirqd, hotcpu) = NULL;
-
2012/7/16 Thomas Gleixner t...@linutronix.de:
The following series implements the infrastructure for parking and
unparking kernel threads to avoid the full teardown and fork on cpu
hotplug operations along with management infrastructure for hotplug
and users.
Changes vs. V2:
Use callbacks
static_vm for ARM-specific static mapped area'
https://lkml.org/lkml/2012/11/27/356
But, running properly on x86 without ARM patchset.
Joonsoo Kim (8):
mm, vmalloc: change iterating a vmlist to find_vm_area()
mm, vmalloc: move get_vmalloc_info() to vmalloc.c
mm, vmalloc: protect va-vm
. So move the code to vmalloc.c
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/fs/proc/Makefile b/fs/proc/Makefile
index 99349ef..88092c1 100644
--- a/fs/proc/Makefile
+++ b/fs/proc/Makefile
@@ -5,7 +5,7 @@
obj-y += proc.o
proc-y := nommu.o task_nommu.o
-proc
.
For example, vm_map_ram() allocate area in vmalloc address space,
but it doesn't make a link with vmlist. To provide full information about
vmalloc address space is better idea, so we don't use va-vm and use
vmap_area directly.
This makes get_vmalloc_info() more precise.
Signed-off-by: Joonsoo Kim js1
is sufficient.
So use vmlist_early for full chain of vm_struct and assign a dummy_vm
to vmlist for supporting kexec.
Cc: Eric Biederman ebied...@xmission.com
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index f134950..8a1b959 100644
--- a/mm/vmalloc.c
+++ b/mm
Now, there is no need to maintain vmlist_early after initializing vmalloc.
So remove related code and data structure.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 698b1e5..10d19c9 100644
--- a/include/linux/vmalloc.h
+++ b
. So we need smp_[rw]mb for ensuring that proper
values is assigned when we see that VM_UNLIST is removed.
Therefore, this patch not only change a iteration list, but also add a
appropriate smp_[rw]mb to right places.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm/vmalloc.c b/mm
to lock, because vmlist_lock is mutex,
but, vmap_area_lock is spin_lock. It may introduce a spinning overhead
during vread/vwrite() is executing. But, these are debug-oriented
functions, so this overhead is not real problem for common case.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm
...@linutronix.de
Cc: Ingo Molnar mi...@redhat.com
Cc: H. Peter Anvin h...@zytor.com
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/arch/tile/mm/pgtable.c b/arch/tile/mm/pgtable.c
index de0de0c..862782d 100644
--- a/arch/tile/mm/pgtable.c
+++ b/arch/tile/mm/pgtable.c
@@ -592,12 +592,7 @@ void iounmap
that, we should make sure that
when we iterate a vmap_area_list, accessing to va-vm doesn't cause a race
condition. This patch ensure that when iterating a vmap_area_list,
there is no race condition for accessing to vm_struct.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm/vmalloc.c b
2012/11/28 Joonsoo Kim js1...@gmail.com:
In current implementation, we used ARM-specific flag, that is,
VM_ARM_STATIC_MAPPING, for distinguishing ARM specific static mapped area.
The purpose of static mapped area is to re-use static mapped area when
entire physical address range of the ioremap
Hello, Andrew.
2012/12/7 Andrew Morton a...@linux-foundation.org:
On Fri, 7 Dec 2012 01:09:27 +0900
Joonsoo Kim js1...@gmail.com wrote:
This patchset remove vm_struct list management after initializing vmalloc.
Adding and removing an entry to vmlist is linear time complexity, so
2012/12/7 Andrew Morton a...@linux-foundation.org:
On Fri, 7 Dec 2012 01:09:27 +0900
Joonsoo Kim js1...@gmail.com wrote:
I'm not sure that 7/8: makes vmlist only for kexec is fine.
Because it is related to userspace program.
As far as I know, makedumpfile use kexec's output information
Hello, Bob.
2012/12/7 Bob Liu lliu...@gmail.com:
Hi Joonsoo,
On Fri, Dec 7, 2012 at 12:09 AM, Joonsoo Kim js1...@gmail.com wrote:
This patchset remove vm_struct list management after initializing vmalloc.
Adding and removing an entry to vmlist is linear time complexity, so
it is inefficient
Hello, Pekka.
2012/12/7 Pekka Enberg penb...@kernel.org:
On Thu, Dec 6, 2012 at 6:09 PM, Joonsoo Kim js1...@gmail.com wrote:
The purpose of iterating a vmlist is finding vm area with specific
virtual address. find_vm_area() is provided for this purpose
and more efficient, because it uses
Hello, Vivek.
2012/12/7 Vivek Goyal vgo...@redhat.com:
On Fri, Dec 07, 2012 at 10:16:55PM +0900, JoonSoo Kim wrote:
2012/12/7 Andrew Morton a...@linux-foundation.org:
On Fri, 7 Dec 2012 01:09:27 +0900
Joonsoo Kim js1...@gmail.com wrote:
I'm not sure that 7/8: makes vmlist only
cscope O=. SRCARCH=arm SUBARCH=xxx
Signed-off-by: Joonsoo Kim js1...@gmail.com
---
v2: change bash specific '[[]]' to 'case in' statement.
v3: quote the patterns.
diff --git a/scripts/tags.sh b/scripts/tags.sh
index 79fdafb..8fb18d1 100755
--- a/scripts/tags.sh
+++ b/scripts/tags.sh
@@ -48,13 +48,14
the kernel.
Signed-off-by: Joonsoo Kim js1...@gmail.com
---
v2: change bash specific '[[]]' to 'case in' statement.
use COMPILED_SOURCE env var, instead of abusing SUBARCH
v3: change [ $COMPILED_SOURCE=compiled ] to [ -n $COMPILED_SOURCE ]
diff --git a/scripts/tags.sh b/scripts/tags.sh
index
Hello, Andrew.
2012/11/20 Minchan Kim minc...@kernel.org:
Hi Joonsoo,
Sorry for the delay.
On Thu, Nov 15, 2012 at 02:09:04AM +0900, JoonSoo Kim wrote:
Hi, Minchan.
2012/11/14 Minchan Kim minc...@kernel.org:
On Tue, Nov 13, 2012 at 11:12:28PM +0900, JoonSoo Kim wrote:
2012/11/13
and stat inevitably
use vmlist and vmlist_lock. But it is preferable that they are used
as least as possible in outside of vmalloc.c
Changelog
v1-v2:
[2/3]: patch description is improved.
Rebased on v3.7-rc7
Joonsoo Kim (3):
ARM: vmregion: remove vmregion code entirely
ARM: static_vm
Now, there is no user for vmregion.
So remove it.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 8a9c4cb..4e333fa 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -6,7 +6,7 @@ obj-y := dma
-by: Joonsoo Kim js1...@gmail.com
diff --git a/arch/arm/include/asm/mach/static_vm.h
b/arch/arm/include/asm/mach/static_vm.h
new file mode 100644
index 000..1bb6604
--- /dev/null
+++ b/arch/arm/include/asm/mach/static_vm.h
@@ -0,0 +1,45 @@
+/*
+ * arch/arm/include/asm/mach/static_vm.h
.
With it, we don't need to iterate all mapped areas. Instead, we just
iterate static mapped areas. It helps to reduce an overhead of finding
matched area. And architecture dependency on vmalloc layer is removed,
so it will help to maintainability for vmalloc layer.
Signed-off-by: Joonsoo Kim js1
-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm/slub.c b/mm/slub.c
index a0d6984..321afab 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3279,6 +3279,7 @@ static struct kmem_cache *__init
create_kmalloc_cache(const char *name,
if (kmem_cache_open(s, flags))
goto panic
2012/12/26 Joonsoo Kim js1...@gmail.com:
commit cce89f4f6911286500cf7be0363f46c9b0a12ce0('Move kmem_cache
refcounting to common code') moves some refcount manipulation code to
common code. Unfortunately, it also removed refcount assignment for
kmalloc_caches. So, kmalloc_caches's refcount
Hello, Sasha.
2012/12/28 Sasha Levin sasha.le...@oracle.com:
On 12/27/2012 06:04 PM, David Rientjes wrote:
On Thu, 27 Dec 2012, Sasha Levin wrote:
That's exactly what happens with the patch. Note that in the current
upstream
version there are several slab checks scattered all over.
In
into panic_smp_self_stop() which prevent system to restart.
For avoid second panic, skip reboot_fixups in early boot phase.
It makes panic_timeout works in early boot phase.
Cc: Thomas Gleixner t...@linutronix.de
Cc: Ingo Molnar mi...@redhat.com
Cc: H. Peter Anvin h...@zytor.com
Signed-off-by: Joonsoo
Hello, Atsushi.
2012/12/12 Atsushi Kumagai kumagai-atsu...@mxc.nes.nec.co.jp:
Hello,
On Tue, 11 Dec 2012 17:17:05 -0500 (EST)
Dave Anderson ander...@redhat.com wrote:
- Original Message -
On Mon, Dec 10, 2012 at 11:40:47PM +0900, JoonSoo Kim wrote:
[..]
So without
2012/12/7 JoonSoo Kim js1...@gmail.com:
2012/11/28 Joonsoo Kim js1...@gmail.com:
In current implementation, we used ARM-specific flag, that is,
VM_ARM_STATIC_MAPPING, for distinguishing ARM specific static mapped area.
The purpose of static mapped area is to re-use static mapped area when
reduce this overhead.
This patch implement basic infrastructure for numa_zero_pfn.
It is default disabled, because it doesn't provide page coloring and
some architecture use page coloring for zero page.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm/Kconfig b/mm/Kconfig
index a3f8ddd
2012/12/13 Andi Kleen a...@firstfloor.org:
I would expect a processor to fetch the zero page cachelines from the l3
cache from other sockets avoiding memory transactions altogether. The zero
page is likely in use somewhere so no typically no memory accesses should
occur in a system.
It
on v3.7-rc5.
Thanks.
Joonsoo Kim (3):
ARM: vmregion: remove vmregion code entirely
ARM: static_vm: introduce an infrastructure for static mapped area
ARM: mm: use static_vm for managing static mapped areas
arch/arm/include/asm/mach/static_vm.h | 51
arch/arm/mm/Makefile
Now, there is no user for vmregion.
So remove it.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 8a9c4cb..4e333fa 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -6,7 +6,7 @@ obj-y := dma
and vmlist_lock. But it is preferable that they are used outside
of vmalloc.c as least as possible.
Now, I introduce an ARM-specific infrastructure for static mapped area. In
the following patch, we will use this and resolve above mentioned problem.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/arch
.
With it, we don't need to iterate all mapped areas. Instead, we just
iterate static mapped areas. It helps to reduce an overhead of finding
matched area. And architecture dependency on vmalloc layer is removed,
so it will help to maintainability for vmalloc layer.
Signed-off-by: Joonsoo Kim js1
Hi, Minchan.
2012/11/14 Minchan Kim minc...@kernel.org:
On Tue, Nov 13, 2012 at 11:12:28PM +0900, JoonSoo Kim wrote:
2012/11/13 Minchan Kim minc...@kernel.org:
On Tue, Nov 13, 2012 at 09:30:57AM +0900, JoonSoo Kim wrote:
2012/11/3 Minchan Kim minc...@kernel.org:
Hi Joonsoo,
On Sat
, it is possible error situation, because we already
prepare prev vma, rb_link and rb_parent and these are related to original
address.
So add WARN_ON_ONCE for finding that this situtation really happens.
Signed-off-by: Joonsoo Kim js1...@gmail.com
diff --git a/mm/mmap.c b/mm/mmap.c
index 2d94235..36567b7
Hello, Russell.
Thanks for review.
2012/11/15 Russell King - ARM Linux li...@arm.linux.org.uk:
On Thu, Nov 15, 2012 at 01:55:51AM +0900, Joonsoo Kim wrote:
In current implementation, we used ARM-specific flag, that is,
VM_ARM_STATIC_MAPPING, for distinguishing ARM specific static mapped area
, Dec 25, 2012 at 7:30 AM, JoonSoo Kim js1...@gmail.com wrote:
2012/12/26 Joonsoo Kim js1...@gmail.com:
commit cce89f4f6911286500cf7be0363f46c9b0a12ce0('Move kmem_cache
refcounting to common code') moves some refcount manipulation code to
common code. Unfortunately, it also removed
.
After that, we don't need return value of put_cpu_partial().
So remove it.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
---
These are based on v3.8-rc3 and there is no dependency between each other.
If rebase is needed, please notify me.
diff --git a/mm/slub.c b/mm/slub.c
index ba2ca53..abef30e
After boot phase, 'n' always exist.
So add 'likely' macro for helping compiler.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/slub.c b/mm/slub.c
index 830348b..6f82070 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1005,7 +1005,7 @@ static inline void inc_slabs_node(struct
this slab.
These didn't make any error previously, because we normally don't free
objects which comes from kmem_cache's first slab and kmem_cache_node's.
Problem will be solved if we consider a cpu slab in bootstrap().
This patch implement it.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
There is no .gitignore in tools/vm,
so 'git status' always show built binaries.
To ignore this, add .gitignore.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/tools/vm/.gitignore b/tools/vm/.gitignore
new file mode 100644
index 000..44f095f
--- /dev/null
+++ b/tools/vm
On Fri, Aug 02, 2013 at 12:20:40PM +0200, Peter Zijlstra wrote:
On Fri, Aug 02, 2013 at 06:05:51PM +0900, ���ؼ� wrote:
What is with you people; have you never learned to trim emails?
Seriously, I'm going to write a script which tests to too many quoted
lines, too many nested quotes and
On Mon, Aug 05, 2013 at 09:52:28AM +0530, Preeti U Murthy wrote:
On 08/02/2013 04:02 PM, Peter Zijlstra wrote:
On Fri, Aug 02, 2013 at 02:56:14PM +0530, Preeti U Murthy wrote:
You need to iterate over all the groups of the sched domain env-sd and
not just the first group of env-sd like you
+ if (busiest-group_imb) {
+ busiest-sum_weighted_load =
+ min(busiest-sum_weighted_load, sds-sd_avg_load);
Right here we get confused as to why the total load is being compared
against load per task (although you are changing it to load per task above).
Any mapping that doesn't use the reserved pool, not just
MAP_NORESERVE. For example, if a process makes a MAP_PRIVATE mapping,
then fork()s then the mapping is instantiated in the child, that will
not draw from the reserved pool.
Should we ensure them to allocate the last hugepage?
They
get_new_page() sets up result to communicate error codes from the
following checks. While the existing ones (page freed and thp split
failed) don't change rc, somebody else might add a condition whose
error code should be propagated back into *result but miss it.
Please leave
Hello, Michal.
On Fri, Aug 02, 2013 at 11:36:07PM +0200, Michal Hocko wrote:
On Fri 02-08-13 16:47:10, Johannes Weiner wrote:
On Fri, Aug 02, 2013 at 06:27:22PM +0200, Michal Hocko wrote:
On Fri 02-08-13 11:07:56, Joonsoo Kim wrote:
We rarely allocate a page with ALLOC_NO_WATERMARKS
On Mon, Aug 05, 2013 at 05:10:08PM +0900, Joonsoo Kim wrote:
Hello, Michal.
On Fri, Aug 02, 2013 at 11:36:07PM +0200, Michal Hocko wrote:
On Fri 02-08-13 16:47:10, Johannes Weiner wrote:
On Fri, Aug 02, 2013 at 06:27:22PM +0200, Michal Hocko wrote:
On Fri 02-08-13 11:07:56, Joonsoo
, 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 changed, 162
-by: Joonsoo Kim iamjoonsoo@lge.com
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_group
PageSwapCache() is always false when !CONFIG_SWAP, so compiler
properly discard related code. Therefore, we don't need #ifdef explicitly.
Acked-by: Johannes Weiner han...@cmpxchg.org
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/include/linux/swap.h b/include/linux/swap.h
index
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 han...@cmpxchg.org
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/page_alloc.c b/mm
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 iamjoonsoo@lge.com
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 52898dc..c6732d2 100644
--- a/kernel/sched/fair.c
+++ b
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 iamjoonsoo@lge.com
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
Remove one division operation in find_buiest_queue().
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
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
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 iamjoonsoo@lge.com
diff --git
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 iamjoonsoo@lge.com
diff --git
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 iamjoonsoo@lge.com
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 iamjoonsoo@lge.com
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
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 iamjoonsoo@lge.com
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
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 helpful to cache usage
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;
- avc =
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(!avc)) {
-
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 holding a lock. Then, iterate
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:
Any mapping that doesn't use the reserved pool
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 impose
on the hugetlb_instantiation_mutex
Hello, Andrew.
2013/8/7 Andrew Morton a...@linux-foundation.org:
On Tue, 6 Aug 2013 17:40:40 +0900 Joonsoo Kim iamjoonsoo@lge.com 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
. This patch implement it.
Reviewed-by: Wanpeng Li liw...@linux.vnet.ibm.com
Reviewed-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index ca15854..4b1b043 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e2bfbf7..fc4988c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -539,10 +539,6 @@ static struct page *dequeue_huge_page_vma(struct hstate *h
, this patch solve the problem.
Reviewed-by: Wanpeng Li liw...@linux.vnet.ibm.com
Reviewed-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 4b1b043..b3b8252 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
. If this page is not
AnonPage, we don't do optimization. This makes this optimization turning
off for a page cache.
Acked-by: Michal Hocko mho...@suse.cz
Reviewed-by: Wanpeng Li liw...@linux.vnet.ibm.com
Reviewed-by: Naoya Horiguchi n-horigu...@ah.jp.nec.com
Signed-off-by: Joonsoo Kim iamjoonsoo
.
With this change, above test generate a SIGBUG which is correct,
because all free pages are reserved and non reserved shared mapping
can't get a free page.
Reviewed-by: Wanpeng Li liw...@linux.vnet.ibm.com
Reviewed-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim iamjoonsoo
.
Remove useless indentation changes in 'clean-up alloc_huge_page()'
Fix new iteration code bug.
Add reviewed-by or acked-by.
Joonsoo Kim (9):
mm, hugetlb: move up the code which check availability of free huge
page
mm, hugetlb: trivial commenting fix
mm, hugetlb: clean-up alloc_huge_page
[alloc|free] and
fix and clean-up node iteration code to alloc or free.
This makes code more understandable.
Reviewed-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 31d78c5..87d7637 100644
--- a/mm
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 iamjoonsoo@lge.com
diff --git a/include/linux/hugetlb.h b
'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.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index d971233..12b6581 100644
--- a/mm
This patch unifies successful allocation paths to make the code more
readable. There are no functional changes.
Acked-by: Michal Hocko mho...@suse.cz
Reviewed-by: Wanpeng Li liw...@linux.vnet.ibm.com
Reviewed-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim
If list is empty, list_for_each_entry_safe() doesn't do anything.
So, this check is redundant. Remove it.
Acked-by: Michal Hocko mho...@suse.cz
Reviewed-by: Wanpeng Li liw...@linux.vnet.ibm.com
Reviewed-by: Aneesh Kumar K.V aneesh.ku...@linux.vnet.ibm.com
Signed-off-by: Joonsoo Kim iamjoonsoo
] 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 (18):
mm, hugetlb: protect reserved pages when softofflining requests the
pages
mm, hugetlb: change variable name reservations
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 iamjoonsoo@lge.com
diff
need, because reserve count already decrease in
dequeue_huge_page_vma(). This patch fix this situation.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index bb8a45f..6a9ec69 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -649,6 +649,34 @@ struct hstate
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 iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 909075b..4fab047 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
. So move up anon_vma_prepare() which can be failed
in OOM situation.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 683fd38..bb8a45f 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2536,6 +2536,15 @@ retry_avoidcopy:
/* Drop
. use_reserve
represent that this user is legimate one who are ensured to have enough
reserved pages. This prevent these thread not to get a SIGBUS signal and
make these thread retrying fault handling.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 6a9ec69
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 iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 255bd9e..683fd38 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
alloc_huge_page_node() use dequeue_huge_page_node() without
any validation check, so it can steal reserved page unconditionally.
To fix it, check the number of free_huge_page in alloc_huge_page_node().
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
Now, alloc_huge_page() only return -ENOSPEC if failed.
So, we don't worry about other return value.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 94173e0..35ccdad 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2562,7 +2562,6 @@ retry_avoidcopy
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.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 35ccdad..255bd9e 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2630,10 +2630,11
Just move down outsider_reserve check.
This makes code more readable.
There is no functional change.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 5f31ca5..94173e0 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2530,20 +2530,6
handling and remove
a hugetlb_instantiation_mutex.
Signed-off-by: Joonsoo Kim iamjoonsoo@lge.com
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a66226e..5f31ca5 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1123,12 +1123,12 @@ static void vma_commit_reservation(struct hstate *h,
}
static
301 - 400 of 4538 matches
Mail list logo
