p));
+ printk(KERN_INFO "%s returned %d and took %lld ns to execute\n",
+ func_name, retval, (long long)delta);
+ }
return 0;
}
--
Thanks,
Gioh Kim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the
and took %lld ns to execute\n,
+ func_name, retval, (long long)delta);
+ }
return 0;
}
--
Thanks,
Gioh Kim
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at http
2014-09-05 오전 7:16, Andrew Morton 쓴 글:
On Thu, 4 Sep 2014 16:29:38 +0900 Gioh Kim wrote:
This patch try to solve problem that a long-lasting page caches of
ext4 superblock and journaling of superblock disturb page migration.
I've been testing CMA feature on my ARM-based platform
and found
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim
Reviewed-by: Jan Kara
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim
Reviewed-by: Jan Kara
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3
want to set page allocation flag for page cache
allocation.
And *_unmovable APIs are for the user wants to allocate page cache from
non-movable area.
Signed-off-by: Gioh Kim
Reviewed-by: Jan Kara
---
fs/buffer.c | 45 -
include/linux
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim (3):
fs.c: support buffer cache allocations with gfp modifiers
ext4: use non-movable memory for the ext4 superblock
jbd/jbd2: use non-movable memory for the jbd superblock
fs/buffer.c | 45
2014-09-05 오전 7:16, Andrew Morton 쓴 글:
On Thu, 4 Sep 2014 16:29:38 +0900 Gioh Kim gioh@lge.com wrote:
This patch try to solve problem that a long-lasting page caches of
ext4 superblock and journaling of superblock disturb page migration.
I've been testing CMA feature on my ARM-based
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim (3):
fs.c: support buffer cache allocations with gfp modifiers
ext4: use non-movable memory for the ext4 superblock
jbd/jbd2: use non-movable memory for the jbd superblock
fs/buffer.c | 45
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim gioh@lge.com
Reviewed-by: Jan Kara j...@suse.cz
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
Reviewed-by: Jan Kara j...@suse.cz
---
fs/ext4/super.c |6 +++---
1 file
want to set page allocation flag for page cache
allocation.
And *_unmovable APIs are for the user wants to allocate page cache from
non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
Reviewed-by: Jan Kara j...@suse.cz
---
fs/buffer.c | 45
2014-09-01 오후 4:53, Jan Kara 쓴 글:
On Fri 29-08-14 13:48:27, Gioh Kim wrote:
What about below?: add gfp for __getblk_slow, change __getblk into __getblk_gfp,
getblk_unmovable and __getblk are, I think, symmetric.
If you say OK, I'm going to send v4 with tabs ;-)
Yes, this looks like what
2014-09-01 오후 4:53, Jan Kara 쓴 글:
On Fri 29-08-14 13:48:27, Gioh Kim wrote:
What about below?: add gfp for __getblk_slow, change __getblk into __getblk_gfp,
getblk_unmovable and __getblk are, I think, symmetric.
If you say OK, I'm going to send v4 with tabs ;-)
Yes, this looks like what
2014-08-28 오후 7:59, Jan Kara 쓴 글:
On Thu 28-08-14 11:31:46, Gioh Kim wrote:
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
New APIs
2014-08-28 오후 10:48, Theodore Ts'o 쓴 글:
On Thu, Aug 28, 2014 at 11:26:31AM +0900, Gioh Kim wrote:
I have 3 patchs:
1. Patch 1/3: introduce a new API that create page cache with allocation flag
2. Patch 2/3: have ext4 use the new API to read superblock
3. Patch 3/3: have jbd/jbd2 use the new
2014-08-28 오후 10:48, Theodore Ts'o 쓴 글:
On Thu, Aug 28, 2014 at 11:26:31AM +0900, Gioh Kim wrote:
I have 3 patchs:
1. Patch 1/3: introduce a new API that create page cache with allocation flag
2. Patch 2/3: have ext4 use the new API to read superblock
3. Patch 3/3: have jbd/jbd2 use the new
2014-08-28 오후 7:59, Jan Kara 쓴 글:
On Thu 28-08-14 11:31:46, Gioh Kim wrote:
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
New APIs
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
want to set page allocation flag for page cache
allocation.
And *_unmovable APIs are for the user wants to allocate page cache from
non-movable area.
Signed-off-by: Gioh Kim
---
fs/buffer.c | 54 +--
include/linux/buffer_head.h | 14
of superblock
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim (3):
fs/buffer.c: allocate buffer cache with user specific flag
ext4: allocate buffer-cache for superblock in non-movable area
jbd-jbd2-allocate-buffer-cache-for-superblock-inode-.patch
fs/buffer.c
of superblock
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim (3):
fs/buffer.c: allocate buffer cache with user specific flag
ext4: allocate buffer-cache for superblock in non-movable area
jbd-jbd2-allocate-buffer-cache-for-superblock-inode-.patch
fs/buffer.c
want to set page allocation flag for page cache
allocation.
And *_unmovable APIs are for the user wants to allocate page cache from
non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/buffer.c | 54 +--
include/linux
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions
2014-08-21 오전 7:02, Jan Kara 쓴 글:
On Wed 20-08-14 11:38:10, Gioh Kim wrote:
@@ -1381,12 +1383,7 @@ EXPORT_SYMBOL(__find_get_block);
struct buffer_head *
__getblk(struct block_device *bdev, sector_t block, unsigned size)
{
- struct buffer_head *bh = __find_get_block(bdev, block
2014-08-21 오전 7:02, Jan Kara 쓴 글:
On Wed 20-08-14 11:38:10, Gioh Kim wrote:
@@ -1381,12 +1383,7 @@ EXPORT_SYMBOL(__find_get_block);
struct buffer_head *
__getblk(struct block_device *bdev, sector_t block, unsigned size)
{
- struct buffer_head *bh = __find_get_block(bdev, block
@@ -1381,12 +1383,7 @@ EXPORT_SYMBOL(__find_get_block);
struct buffer_head *
__getblk(struct block_device *bdev, sector_t block, unsigned size)
{
- struct buffer_head *bh = __find_get_block(bdev, block, size);
-
- might_sleep();
- if (bh == NULL)
- bh =
2014-08-19 오후 10:03, Jan Kara 쓴 글:
Hello,
On Tue 19-08-14 15:52:38, Gioh Kim wrote:
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
A new
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
if user set flag to zero, buffer cache is allocated from
non-movable area.
Signed-off-by: Gioh Kim
---
fs/buffer.c | 52 +--
include/linux/buffer_head.h | 12 +-
2 files changed, 46 insertions(+), 18 deletions(-)
diff --git a/fs
have 3 patchs:
1. Patch 1/3: introduce a new API that create page cache with allocation flag
2. Patch 2/3: have ext4 use the new API to read superblock
3. Patch 3/3: have jbd/jbd2 use the new API to make journaling of superblock
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim
2014-08-19 오후 10:03, Jan Kara 쓴 글:
Hello,
On Tue 19-08-14 15:52:38, Gioh Kim wrote:
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
A new
@@ -1381,12 +1383,7 @@ EXPORT_SYMBOL(__find_get_block);
struct buffer_head *
__getblk(struct block_device *bdev, sector_t block, unsigned size)
{
- struct buffer_head *bh = __find_get_block(bdev, block, size);
-
- might_sleep();
- if (bh == NULL)
- bh =
have 3 patchs:
1. Patch 1/3: introduce a new API that create page cache with allocation flag
2. Patch 2/3: have ext4 use the new API to read superblock
3. Patch 3/3: have jbd/jbd2 use the new API to make journaling of superblock
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim
if user set flag to zero, buffer cache is allocated from
non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/buffer.c | 52 +--
include/linux/buffer_head.h | 12 +-
2 files changed, 46 insertions(+), 18 deletions(-)
diff
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions
2014-08-18 오후 12:24, Theodore Ts'o 쓴 글:
On Mon, Aug 18, 2014 at 10:15:32AM +0900, Gioh Kim wrote:
My test platform has totally 1GB memory, 256MB for CMA and 768MB for normal.
I applied Joonsoo's patch: https://lkml.org/lkml/2014/5/28/64, so that
3/4 of allocation take place in normal area
2014-08-15 오전 6:22, Andrew Morton 쓴 글:
On Thu, 14 Aug 2014 14:15:40 +0900 Gioh Kim wrote:
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration
2014-08-17 오전 3:52, Jan Kara 쓴 글:
On Thu 14-08-14 14:26:10, Andrew Morton wrote:
On Thu, 14 Aug 2014 14:12:17 +0900 Gioh Kim wrote:
This patch try to solve problem that a long-lasting page caches of
ext4 superblock and journaling of superblock disturb page migration.
I've been testing CMA
2014-08-17 오전 3:52, Jan Kara 쓴 글:
On Thu 14-08-14 14:26:10, Andrew Morton wrote:
On Thu, 14 Aug 2014 14:12:17 +0900 Gioh Kim gioh@lge.com wrote:
This patch try to solve problem that a long-lasting page caches of
ext4 superblock and journaling of superblock disturb page migration.
I've
2014-08-15 오전 6:22, Andrew Morton 쓴 글:
On Thu, 14 Aug 2014 14:15:40 +0900 Gioh Kim gioh@lge.com wrote:
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page
2014-08-18 오후 12:24, Theodore Ts'o 쓴 글:
On Mon, Aug 18, 2014 at 10:15:32AM +0900, Gioh Kim wrote:
My test platform has totally 1GB memory, 256MB for CMA and 768MB for normal.
I applied Joonsoo's patch: https://lkml.org/lkml/2014/5/28/64, so that
3/4 of allocation take place in normal area
2014-08-14 오후 2:12, Gioh Kim 쓴 글:
> Hello,
>
> This patch try to solve problem that a long-lasting page caches of
> ext4 superblock and journaling of superblock disturb page migration.
>
> I've been testing CMA feature on my ARM-based platform
> and found that t
I'm sorry for a typo at the title.
It is 1/3, not 1/2.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at
I'm sorry for a typo at the tile.
It is 0/3, not 0/2.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
A new API should be introduced to allocate buffer cache from
non-movable area.
Signed-off-by: Gioh
the new API to read superblock
3. Patch 3/3: have jbd/jbd2 use the new API to make journaling of superblock
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim (3):
fs/buffer.c: allocate buffer cache from non-movable area
ext4: allocate buffer-cache for superblock in non
the new API to read superblock
3. Patch 3/3: have jbd/jbd2 use the new API to make journaling of superblock
This patchset is based on linux-next-20140814.
Thanks a lot.
Gioh Kim (3):
fs/buffer.c: allocate buffer cache from non-movable area
ext4: allocate buffer-cache for superblock in non
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
A new API should be introduced to allocate buffer cache from
non-movable area.
Signed-off-by: Gioh
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions
for a long time
and it can distrub page migration.
This patch make the buffer-cache be allocated from non-movable area
not to distrub page migration.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/jbd/journal.c |2 +-
fs/jbd2/journal.c |2 +-
2 files changed, 2 insertions(+), 2 deletions
I'm sorry for a typo at the tile.
It is 0/3, not 0/2.
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
I'm sorry for a typo at the title.
It is 1/3, not 1/2.
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
2014-08-14 오후 2:12, Gioh Kim 쓴 글:
Hello,
This patch try to solve problem that a long-lasting page caches of
ext4 superblock and journaling of superblock disturb page migration.
I've been testing CMA feature on my ARM-based platform
and found that two page caches cannot be migrated
2014-08-02 오전 1:04, Peter Zijlstra 쓴 글:
On Fri, Aug 01, 2014 at 05:24:59PM +0200, Jan Kara wrote:
OK, makes sense. But then if there's heavy IO going on, anything that has
IO pending on it is pinned and IO completion can easily take something
close to a second or more. So meeting
2014-08-02 오전 1:04, Peter Zijlstra 쓴 글:
On Fri, Aug 01, 2014 at 05:24:59PM +0200, Jan Kara wrote:
OK, makes sense. But then if there's heavy IO going on, anything that has
IO pending on it is pinned and IO completion can easily take something
close to a second or more. So meeting
2014-08-01 오전 9:07, Gioh Kim 쓴 글:
2014-07-31 오후 9:21, Jan Kara 쓴 글:
On Thu 31-07-14 09:37:15, Gioh Kim wrote:
2014-07-31 오전 9:03, Jan Kara 쓴 글:
On Thu 31-07-14 08:54:40, Gioh Kim wrote:
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38
2014-08-01 오전 7:57, Andrew Morton 쓴 글:
On Thu, 31 Jul 2014 11:22:35 +0900 Gioh Kim wrote:
The previous PATCH inserts invalidate_bh_lrus() only into CMA code.
HOTPLUG needs also dropping bh of lru.
So v2 inserts invalidate_bh_lrus() into both of CMA and HOTPLUG
2014-07-31 오후 9:21, Jan Kara 쓴 글:
On Thu 31-07-14 09:37:15, Gioh Kim wrote:
2014-07-31 오전 9:03, Jan Kara 쓴 글:
On Thu 31-07-14 08:54:40, Gioh Kim wrote:
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05
2014-07-31 오후 9:21, Jan Kara 쓴 글:
On Thu 31-07-14 09:37:15, Gioh Kim wrote:
2014-07-31 오전 9:03, Jan Kara 쓴 글:
On Thu 31-07-14 08:54:40, Gioh Kim wrote:
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05
2014-08-01 오전 7:57, Andrew Morton 쓴 글:
On Thu, 31 Jul 2014 11:22:35 +0900 Gioh Kim gioh@lge.com wrote:
The previous PATCH inserts invalidate_bh_lrus() only into CMA code.
HOTPLUG needs also dropping bh of lru.
So v2 inserts invalidate_bh_lrus() into both of CMA and HOTPLUG
2014-08-01 오전 9:07, Gioh Kim 쓴 글:
2014-07-31 오후 9:21, Jan Kara 쓴 글:
On Thu 31-07-14 09:37:15, Gioh Kim wrote:
2014-07-31 오전 9:03, Jan Kara 쓴 글:
On Thu 31-07-14 08:54:40, Gioh Kim wrote:
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38
ped.
The reference count of bh is increased when it is installed
into lru so that the bh of lru must be freed before migrating the page.
This frees every bh of lru. We could free only bh of migrating page.
But searching lru sometimes costs more than invalidating entire lru.
Signed-off-by: Gioh Kim
Ac
2014-07-31 오전 9:03, Jan Kara 쓴 글:
On Thu 31-07-14 08:54:40, Gioh Kim wrote:
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05, Peter Zijlstra wrote:
On Tue, Jul 22, 2014 at 02:18:47PM +0900, Gioh Kim
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05, Peter Zijlstra wrote:
On Tue, Jul 22, 2014 at 02:18:47PM +0900, Gioh Kim wrote:
Hello,
This patch try to solve problem that a long-lasting page cache
2014-07-30 오후 7:19, Peter Zijlstra 쓴 글:
On Wed, Jul 30, 2014 at 12:11:43PM +0200, Jan Kara wrote:
sb_bread allocates page from movable area but it is not movable until the
reference counter of the buffer-head becomes zero.
There is no lock for the buffer but the reference counter acts like
2014-07-22 오전 10:04, Gioh Kim 쓴 글:
2014-07-22 오전 9:15, Minchan Kim 쓴 글:
Hello Mel,
On Mon, Jul 21, 2014 at 02:01:46PM +0100, Mel Gorman wrote:
On Mon, Jul 21, 2014 at 04:36:51PM +0900, Minchan Kim wrote:
I'm not reviewing this in detail at all, didn't even look at the patch
but two
2014-07-27 오전 10:01, Theodore Ts'o 쓴 글:
Gioh,
As follow up, if you want some further discussions about why these
patches should be accepted, it would be good to get some hard data
about why the keeping the ext4 superblock pinned is causing such a
problem for page migation. Can you give us
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05, Peter Zijlstra wrote:
On Tue, Jul 22, 2014 at 02:18:47PM +0900, Gioh Kim wrote:
Hello,
This patch try to solve problem that a long-lasting page cache of
ext4 superblock disturbs page migration.
I've been testing CMA feature on my
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05, Peter Zijlstra wrote:
On Tue, Jul 22, 2014 at 02:18:47PM +0900, Gioh Kim wrote:
Hello,
This patch try to solve problem that a long-lasting page cache of
ext4 superblock disturbs page migration.
I've been testing CMA feature on my
2014-07-27 오전 10:01, Theodore Ts'o 쓴 글:
Gioh,
As follow up, if you want some further discussions about why these
patches should be accepted, it would be good to get some hard data
about why the keeping the ext4 superblock pinned is causing such a
problem for page migation. Can you give us
2014-07-22 오전 10:04, Gioh Kim 쓴 글:
2014-07-22 오전 9:15, Minchan Kim 쓴 글:
Hello Mel,
On Mon, Jul 21, 2014 at 02:01:46PM +0100, Mel Gorman wrote:
On Mon, Jul 21, 2014 at 04:36:51PM +0900, Minchan Kim wrote:
I'm not reviewing this in detail at all, didn't even look at the patch
but two
2014-07-30 오후 7:19, Peter Zijlstra 쓴 글:
On Wed, Jul 30, 2014 at 12:11:43PM +0200, Jan Kara wrote:
sb_bread allocates page from movable area but it is not movable until the
reference counter of the buffer-head becomes zero.
There is no lock for the buffer but the reference counter acts like
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05, Peter Zijlstra wrote:
On Tue, Jul 22, 2014 at 02:18:47PM +0900, Gioh Kim wrote:
Hello,
This patch try to solve problem that a long-lasting page cache
2014-07-31 오전 9:03, Jan Kara 쓴 글:
On Thu 31-07-14 08:54:40, Gioh Kim wrote:
2014-07-30 오후 7:11, Jan Kara 쓴 글:
On Wed 30-07-14 16:44:24, Gioh Kim wrote:
2014-07-22 오후 6:38, Jan Kara 쓴 글:
On Tue 22-07-14 09:30:05, Peter Zijlstra wrote:
On Tue, Jul 22, 2014 at 02:18:47PM +0900, Gioh Kim
.
The reference count of bh is increased when it is installed
into lru so that the bh of lru must be freed before migrating the page.
This frees every bh of lru. We could free only bh of migrating page.
But searching lru sometimes costs more than invalidating entire lru.
Signed-off-by: Gioh Kim gioh
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
A new API should be introduced to allocate buffer cache from
non-movable area.
Signed-off-by: Gioh
for ext4 and others that want to hold page cache for a long time.
I have 2 patchs:
1. Patch 1/2: introduce a new API that create page cache from non-movable area
2. Patch 2/2: have ext4 use the new API to read superblock
This patchset is based on linux-next-20140717.
Thanks a lot.
Gioh Kim (2):
fs
2014-07-22 오전 9:15, Minchan Kim 쓴 글:
Hello Mel,
On Mon, Jul 21, 2014 at 02:01:46PM +0100, Mel Gorman wrote:
On Mon, Jul 21, 2014 at 04:36:51PM +0900, Minchan Kim wrote:
I'm not reviewing this in detail at all, didn't even look at the patch
but two things popped out at me during the
2014-07-21 오전 11:50, Minchan Kim 쓴 글:
Hi Gioh,
On Fri, Jul 18, 2014 at 03:45:36PM +0900, Gioh Kim wrote:
Hi,
For page migration of CMA, buffer-heads of lru should be dropped.
Please refer to https://lkml.org/lkml/2014/7/4/101 for the history.
Just nit:
Please write *problem
2014-07-21 오전 11:50, Minchan Kim 쓴 글:
Hi Gioh,
On Fri, Jul 18, 2014 at 03:45:36PM +0900, Gioh Kim wrote:
Hi,
For page migration of CMA, buffer-heads of lru should be dropped.
Please refer to https://lkml.org/lkml/2014/7/4/101 for the history.
Just nit:
Please write *problem
2014-07-22 오전 9:15, Minchan Kim 쓴 글:
Hello Mel,
On Mon, Jul 21, 2014 at 02:01:46PM +0100, Mel Gorman wrote:
On Mon, Jul 21, 2014 at 04:36:51PM +0900, Minchan Kim wrote:
I'm not reviewing this in detail at all, didn't even look at the patch
but two things popped out at me during the
for ext4 and others that want to hold page cache for a long time.
I have 2 patchs:
1. Patch 1/2: introduce a new API that create page cache from non-movable area
2. Patch 2/2: have ext4 use the new API to read superblock
This patchset is based on linux-next-20140717.
Thanks a lot.
Gioh Kim (2):
fs
A buffer cache is allocated from movable area
because it is referred for a while and released soon.
But some filesystems are taking buffer cache for a long time
and it can disturb page migration.
A new API should be introduced to allocate buffer cache from
non-movable area.
Signed-off-by: Gioh
A buffer-cache for superblock is disturbing page migration,
because the buffer-cache is not released until unmount.
The buffer-cache must be allocated from non-movable area.
Signed-off-by: Gioh Kim gioh@lge.com
---
fs/ext4/super.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions
I removed checking migratetype of v1: https://lkml.org/lkml/2014/7/18/82.
Thanks a lot.
8< --
>From 95addf7f1644cde3a0944629ff5c6a8d6f43e5d6 Mon Sep 17 00:00:00 2001
From: Gioh Kim
Date: Fri, 18 Jul 2014 13:40:01 +0900
Subject:
I removed checking migratetype of v1: https://lkml.org/lkml/2014/7/18/82.
Thanks a lot.
8 --
From 95addf7f1644cde3a0944629ff5c6a8d6f43e5d6 Mon Sep 17 00:00:00 2001
From: Gioh Kim gioh@lge.com
Date: Fri, 18 Jul 2014 13:40:01 +0900
2014-07-18 오후 4:50, Marek Szyprowski 쓴 글:
Hello,
On 2014-07-18 08:45, Gioh Kim wrote:
For page migration of CMA, buffer-heads of lru should be dropped.
Please refer to https://lkml.org/lkml/2014/7/4/101 for the history.
I have two solution to drop bhs.
One is invalidating entire lru
wing patch.
Laura, can I have you name on Acked-by line?
Please let me represent my thanks.
Thanks for any feedback.
--- 8< --
>From 33c894b1bab9bc26486716f0c62c452d3a04d35d Mon Sep 17 00:00:00 2001
From: Gioh Kim
Date: Fri, 18
have you name on Acked-by line?
Please let me represent my thanks.
Thanks for any feedback.
--- 8 --
From 33c894b1bab9bc26486716f0c62c452d3a04d35d Mon Sep 17 00:00:00 2001
From: Gioh Kim gioh@lge.com
Date: Fri, 18 Jul 2014 13:40:01
2014-07-18 오후 4:50, Marek Szyprowski 쓴 글:
Hello,
On 2014-07-18 08:45, Gioh Kim wrote:
For page migration of CMA, buffer-heads of lru should be dropped.
Please refer to https://lkml.org/lkml/2014/7/4/101 for the history.
I have two solution to drop bhs.
One is invalidating entire lru
2014-07-15 오전 5:37, Andrew Morton 쓴 글:
On Mon, 14 Jul 2014 16:02:25 +0900 Joonsoo Kim wrote:
On Tue, Jul 08, 2014 at 06:46:31PM +0200, Michal Nazarewicz wrote:
On Mon, Jul 07 2014, Andrew Morton wrote:
What I proposed is that CMA call invalidate_bh_lrus() right at the
outset. Something
2014-07-15 오전 5:37, Andrew Morton 쓴 글:
On Mon, 14 Jul 2014 16:02:25 +0900 Joonsoo Kim iamjoonsoo@lge.com wrote:
On Tue, Jul 08, 2014 at 06:46:31PM +0200, Michal Nazarewicz wrote:
On Mon, Jul 07 2014, Andrew Morton a...@linux-foundation.org wrote:
What I proposed is that CMA call
301 - 400 of 492 matches
Mail list logo