This is preparation step to report test failed pfn in new tracepoint
to analyze cma allocation failure problem. There is no functional change
in this patch.
Signed-off-by: Joonsoo Kim
---
mm/page_isolation.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/mm
m is hard to find without this
page reference tracepoint facility.
Enabling this feature bloat kernel text 20 KB in my configuration.
textdata bss dec hex filename
120412722223424 1507328 15772024 f0a978 vmlinux_disabled
12064844 2225920 1507328 15798092
. With this facility, we can
easily find reason of CMA allocation failure. There is no functional change
in this patch.
Signed-off-by: Joonsoo Kim
---
arch/mips/mm/gup.c| 2 +-
arch/powerpc/mm/mmu_context_hash64.c | 3 +-
arch/powerpc/mm/pgtable_64.c
2015-11-05 19:32 GMT+09:00 Catalin Marinas :
> On Thu, Nov 05, 2015 at 01:40:14PM +0900, Joonsoo Kim wrote:
>> On Tue, Sep 22, 2015 at 07:59:48PM +0200, Robert Richter wrote:
>> > From: Tirumalesh Chalamarla
>> >
>> > Increase the standard cacheline si
On Thu, Nov 05, 2015 at 09:17:33AM +0100, Vlastimil Babka wrote:
> On 11/05/2015 09:10 AM, Joonsoo Kim wrote:
> > On Wed, Nov 04, 2015 at 04:00:59PM +0100, Vlastimil Babka wrote:
> >> +void __copy_page_owner(struct page *oldpage, struct page *newpage)
> >> +{
>
On Thu, Nov 05, 2015 at 09:15:01AM +0100, Vlastimil Babka wrote:
> On 11/05/2015 09:09 AM, Joonsoo Kim wrote:
> > On Wed, Nov 04, 2015 at 04:00:57PM +0100, Vlastimil Babka wrote:
> >> The information in /sys/kernel/debug/page_owner includes the migratetype
> >> declared
On Wed, Nov 04, 2015 at 04:00:59PM +0100, Vlastimil Babka wrote:
> The page_owner mechanism stores gfp_flags of an allocation and stack trace
> that lead to it. During page migration, the original information is
> essentially replaced by the allocation of free page as the migration target.
> Arguab
On Wed, Nov 04, 2015 at 04:00:57PM +0100, Vlastimil Babka wrote:
> The information in /sys/kernel/debug/page_owner includes the migratetype
> declared during the page allocation via gfp_flags. This is also checked
> against
> the pageblock's migratetype, and reported as Fallback allocation if thes
On Tue, Sep 22, 2015 at 07:59:48PM +0200, Robert Richter wrote:
> From: Tirumalesh Chalamarla
>
> Increase the standard cacheline size to avoid having locks in the same
> cacheline.
>
> Cavium's ThunderX core implements cache lines of 128 byte size. With
> current granulare size of 64 bytes (L1_
On Wed, Nov 04, 2015 at 03:39:10PM +, Catalin Marinas wrote:
> On Wed, Nov 04, 2015 at 09:28:34AM -0600, Christoph Lameter wrote:
> > On Wed, 4 Nov 2015, Catalin Marinas wrote:
> >
> > > BTW, assuming L1_CACHE_BYTES is 512 (I don't ever see this happening but
> > > just in theory), we potentia
On Wed, Oct 14, 2015 at 02:28:31PM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
> > Now, compaction algorithm become powerful. Migration scanner traverses
> > whole zone range. So, old threshold for depleted zone which is designed
> > to imitate
On Wed, Oct 14, 2015 at 02:28:31PM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
> > Now, compaction algorithm become powerful. Migration scanner traverses
> > whole zone range. So, old threshold for depleted zone which is designed
> > to imitate
On Thu, Oct 15, 2015 at 11:05:49AM +0900, Sergey Senozhatsky wrote:
> On (10/14/15 16:38), Joonsoo Kim wrote:
> [..]
> > @@ -352,6 +352,7 @@ static ssize_t comp_algorithm_show(struct device *dev,
> > size_t sz;
> > struct zram *zram = dev_to_zram(dev);
> >
On Wed, Oct 14, 2015 at 01:38:07PM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
>
> [...]
>
> > This patch solves these problems mentioned in above.
> > Freepage scanner is largely changed to scan zone from zone_start_pfn
> > to zon
On Thu, Oct 15, 2015 at 09:38:41AM +0900, Sergey Senozhatsky wrote:
> Hi,
>
> On (10/14/15 16:38), Joonsoo Kim wrote:
> [..]
> > Creates virtual block devices called /dev/zramX (X = 0, 1, ...).
> > @@ -18,9 +17,8 @@ config ZRAM
> > config ZRAM_LZ4_COM
On Thu, Oct 15, 2015 at 09:29:03AM +0900, Sergey Senozhatsky wrote:
> Hi,
>
> On (10/14/15 16:38), Joonsoo Kim wrote:
> [..]
> > static const char * const backends[] = {
> > "lzo",
> > -#ifdef CONFIG_ZRAM_LZ4_COMPRESS
> > "lz4",
Now, contextless compression API is introduced and it can reduce
memory overhead in some cases. All compression algorithm will
support it.
Signed-off-by: Joonsoo Kim
---
crypto/lz4.c | 77
1 file changed, 67 insertions(+), 10
Now, contextless compression API is introduced and it can reduce
memory overhead in some cases. All compression algorithm will
support it.
Signed-off-by: Joonsoo Kim
---
crypto/Kconfig | 1 +
crypto/lzo.c | 81 --
2 files changed, 69
Now, contextless compression API is introduced and it can reduce
memory overhead in some cases. All compression algorithm will
support it.
Signed-off-by: Joonsoo Kim
---
crypto/deflate.c | 96 ++--
1 file changed, 86 insertions(+), 10
this patch is applied, there is no way to know current
compression algorithm so it's really bad thing. Please let me know proper
solution if someone have better idea.
Signed-off-by: Joonsoo Kim
---
Documentation/blockdev/zram.txt | 29 +++--
drivers/block/zram/Kc
compression backend require zstrm for
decompression or NULL if it does not.
Acked-by: Minchan Kim
Signed-off-by: Sergey Senozhatsky
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c| 3 ++-
drivers/block/zram/zcomp.h| 3 ++-
drivers/block/zram/zram_drv.c | 20
: Sergey Senozhatsky
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c| 27 +--
drivers/block/zram/zcomp.h| 8 ++--
drivers/block/zram/zram_drv.c | 6 +++---
3 files changed, 34 insertions(+), 7 deletions(-)
diff --git a/drivers/block/zram/zcomp.c b
zram in following patches.
Signed-off-by: Joonsoo Kim
---
crypto/Kconfig | 17 +-
crypto/Makefile| 3 +-
crypto/ccompress.c | 95 +
crypto/pcompress.c | 115 ---
crypto/zlib.c
doesn't support zlib yet. Following patch
will change this interface and enable various compression algorithm.
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/Kconfig| 8 +++---
drivers/block/zram/Makefile | 4 +--
drivers/block/zram/zcomp.c| 63 ---
aration, we can save memory in some cases and get the best
performance.
This patchset solves Sergey's concern perfectly and provides possibility
to use various compression algorithm in zram.
Thanks.
Joonsoo Kim (6):
crypto/compress: introduce contextless compression and remove unused
On Fri, Sep 25, 2015 at 01:41:48PM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
> > Purpose of compaction is to make high order page. To achive this purpose,
> > it is the best strategy that compaction migrates contiguous used pages
> > to fragme
On Fri, Sep 25, 2015 at 12:32:02PM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
>
> [...]
>
> >
> > Because we just allow freepage scanner to scan non-movable pageblock
> > in very limited situation, more scanning events happen. But
On Fri, Sep 25, 2015 at 11:33:38AM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
> > Now, we have a way to determine compaction depleted state and compaction
> > activity will be limited according this state and depletion depth so
> > compaction o
Hello, Vlastimil.
First of all, thanks for review!
On Fri, Sep 25, 2015 at 10:04:49AM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
> > Further compaction attempt is deferred when some of compaction attempts
> > already fails. But, after some numb
On Mon, Sep 21, 2015 at 02:29:18PM +0900, Sergey Senozhatsky wrote:
> On (09/18/15 14:19), Joonsoo Kim wrote:
> > -/* Never return NULL, may sleep */
> > +/* May return NULL, may sleep */
> > struct zcomp_strm *zcomp_decompress_begin(struct zcomp *comp)
> > {
On Mon, Sep 21, 2015 at 04:56:00PM +0900, Sergey Senozhatsky wrote:
> On (09/18/15 14:19), Joonsoo Kim wrote:
> [..]
> > + /*
> > +* Prepare to use crypto decompress_noctx API. One tfm is required
> > +* to initialize crypto algorithm properly and fetch correspon
On Mon, Sep 21, 2015 at 12:51:28PM +0900, Minchan Kim wrote:
> On Fri, Sep 18, 2015 at 02:19:24PM +0900, Joonsoo Kim wrote:
> > Crypto subsystem now supports decompress_noctx API that requires
> > special tfm_noctx. This tfm can be shared by multiple concurrent
> > decompress
On Mon, Sep 21, 2015 at 12:45:12PM +0900, Minchan Kim wrote:
> Hi Joonsoo,
>
> On Fri, Sep 18, 2015 at 02:19:23PM +0900, Joonsoo Kim wrote:
> > Until now, zram uses compression algorithm through direct call
> > to core algorithm function, but, it has drawback that we need t
On Mon, Sep 21, 2015 at 02:19:03PM +0900, Sergey Senozhatsky wrote:
> On (09/18/15 14:19), Joonsoo Kim wrote:
> [..]
> > -static struct zcomp_backend *find_backend(const char *compress)
> > +static const char *find_backend(const char *compress)
> > {
> > int i
On Mon, Sep 21, 2015 at 12:58:12PM +0900, Minchan Kim wrote:
> On Fri, Sep 18, 2015 at 02:19:15PM +0900, Joonsoo Kim wrote:
> > This patchset makes zram to use crypto API in order to support
> > more compression algorithm.
> >
> > The reason we need to support vairou
On Mon, Sep 21, 2015 at 02:38:59PM +0900, Sergey Senozhatsky wrote:
> On (09/18/15 14:19), Joonsoo Kim wrote:
> [..]
> > @@ -61,7 +61,8 @@ static struct crypto_alg alg = {
> > .cra_module = THIS_MODULE,
> > .cra_u = { .compress =
On Tue, Sep 22, 2015 at 08:43:46PM +0800, Herbert Xu wrote:
> On Fri, Sep 18, 2015 at 02:19:16PM +0900, Joonsoo Kim wrote:
> > Until now, tfm object embeds (de)compression context in it and
> > (de)compression in crypto API requires tfm object to use
> > this context
On Mon, Sep 21, 2015 at 03:18:17PM +0900, Sergey Senozhatsky wrote:
> On (09/18/15 14:19), Joonsoo Kim wrote:
> [..]
> > static int __init lzo_mod_init(void)
> > diff --git a/include/linux/crypto.h b/include/linux/crypto.h
> > index e71cb70..31152b1 100644
> >
On Wed, Sep 09, 2015 at 01:39:01PM +0100, Mel Gorman wrote:
> On Tue, Sep 08, 2015 at 05:26:13PM +0900, Joonsoo Kim wrote:
> > 2015-08-24 21:30 GMT+09:00 Mel Gorman :
> > > The primary purpose of watermarks is to ensure that reclaim can always
> > > make forward prog
On Wed, Sep 09, 2015 at 01:22:03PM +0100, Mel Gorman wrote:
> On Tue, Sep 08, 2015 at 03:49:58PM +0900, Joonsoo Kim wrote:
> > 2015-08-24 21:09 GMT+09:00 Mel Gorman :
> > > __GFP_WAIT has been used to identify atomic context in callers that hold
> > > spinlocks or
On Wed, Sep 09, 2015 at 01:32:39PM +0100, Mel Gorman wrote:
> On Tue, Sep 08, 2015 at 05:01:06PM +0900, Joonsoo Kim wrote:
> > 2015-08-24 21:29 GMT+09:00 Mel Gorman :
> > >
> > >
> > > +/*
> > > + * Reserve a pageblock for exclusive use of high-order
lz4's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/lz4.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff
llowing patch support to use
crypto decompress_noctx API and would restore the performance as is.
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/Kconfig | 8 +++
drivers/block/zram/Makefile| 4 +---
drivers/block/zram/zcomp.c | 54 +++---
842's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/842.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c| 27 +--
drivers/block/zram/zcomp.h| 8 ++--
drivers/block/zram/zram_drv.c | 6 +++---
3 files changed, 34 insertions(+), 7 deletions(-)
diff --git a/drivers/block/zram/zcomp.c b/drivers/block/zram
ple tfm
if decompression doesn't require any context to operate.
Signed-off-by: Joonsoo Kim
---
crypto/842.c | 3 ++-
crypto/compress.c | 36
crypto/crypto_null.c | 3 ++-
crypto/deflate.c | 3 ++-
crypto/lz4.c | 3 ++-
lz4hc's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/lz4hc.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-
x
Read10411701.88 6426911.62 9423894.38
Re-read 10017386.62 6428218.88 11000063.50
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c | 24 +++-
drivers/block/zram/zcomp.h | 1 +
2 files changed, 24 insertions(+), 1 deletion(-)
diff --git
compression backend require zstrm for
decompression or NULL if it does not.
Signed-off-by: Sergey Senozhatsky
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c| 3 ++-
drivers/block/zram/zcomp.h| 3 ++-
drivers/block/zram/zram_drv.c | 20
3 files changed, 20
lzo's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/lzo.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff
/sec
zram-base vs zram-crypto vs zram-crypto-noctx
Read10411701.88 6426911.62 9423894.38
Re-read 10017386.62 6428218.88 1163.50
Thanks.
Joonsoo Kim (7):
crypto: introduce decompression API that can be called via sharable
tfm object
crypto/lzo: s
blem of kmalloc_size(INDEX_NODE + 1) and removes
the BUG by adding 'size >= 256' check to guarantee that all necessary
small sized slabs are initialized regardless sequence of slab size in
mapping table.
Fixes: e33660165c90 ("slab: Use common kmalloc_index/kmalloc_size...")
Rep
2015-08-24 21:30 GMT+09:00 Mel Gorman :
> The primary purpose of watermarks is to ensure that reclaim can always
> make forward progress in PF_MEMALLOC context (kswapd and direct reclaim).
> These assume that order-0 allocations are all that is necessary for
> forward progress.
>
> High-order water
2015-08-24 21:29 GMT+09:00 Mel Gorman :
> High-order watermark checking exists for two reasons -- kswapd high-order
> awareness and protection for high-order atomic requests. Historically the
> kernel depended on MIGRATE_RESERVE to preserve min_free_kbytes as high-order
> free pages for as long as
2015-08-24 21:09 GMT+09:00 Mel Gorman :
> __GFP_WAIT has been used to identify atomic context in callers that hold
> spinlocks or are in interrupts. They are expected to be high priority and
> have access one of two watermarks lower than "min" which can be referred
> to as the "atomic reserve". __G
On Thu, Aug 27, 2015 at 05:24:04PM +0200, Vlastimil Babka wrote:
> Compaction returns prematurely with COMPACT_PARTIAL when contended or has
> fatal
> signal pending. This is ok for the callers, but might be misleading in the
> traces, as the usual reason to return COMPACT_PARTIAL is that we think
On Mon, Aug 24, 2015 at 11:07:12AM +0200, Vlastimil Babka wrote:
> On 08/24/2015 04:19 AM, Joonsoo Kim wrote:
> >Cached pfn is used to determine the start position of scanner
> >at next compaction run. Current cached pfn points the skipped pageblock
> >so we uselessly checks
On Tue, Sep 01, 2015 at 11:12:39AM +0900, Sergey Senozhatsky wrote:
> On (09/01/15 11:06), Minchan Kim wrote:
> > Thanks.
> >
> > I want to include this patchset in Joonsoo's crypto support patch
> > if you don't mind.
Hello,
Sorry for long delay.
I will work crypto support with this patchset.
S
d and this threshold is also adjusted to that change.
In this patch, only state definition is implemented. There is no action
for this new state so no functional change. But, following patch will
add some handling for this new state.
Signed-off-by: Joonsoo Kim
---
include/linux/mmzone.h | 3 ++
will be mentioned
on following patch. Please refer it.
Signed-off-by: Joonsoo Kim
---
include/linux/mmzone.h | 3 ---
include/linux/pageblock-flags.h | 37 +++--
mm/compaction.c | 25 -
mm/page_al
le pageblock
in very limited situation, more scanning events happen. But, allowing
in very limited situation results in a very important benefit that
memory isn't fragmented more than before. Fragmentation effect is
measured on following patch so please refer it.
Signed-off-by: Joonsoo Kim
---
in
sting depletion check with this new algorithm.
Signed-off-by: Joonsoo Kim
---
mm/compaction.c | 145 ++--
1 file changed, 77 insertions(+), 68 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index a259608..ca4d6d1 100644
--- a/mm
han compaction order.
Signed-off-by: Joonsoo Kim
---
mm/compaction.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/mm/compaction.c b/mm/compaction.c
index ca4d6d1..e61ee77 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -455,6 +455,7 @@ static unsigned long isolate_fr
45
compact_migrate_scanned 69714502 2307433
compact_free_scanned20243121 2325295
This looks like reasonable trade-off.
Signed-off-by: Joonsoo Kim
---
mm/compaction.c | 19 ---
1 file changed, 12 insertions(+), 7 deletions(-)
diff --git a/mm/compaction.c b/mm/
activity limit.
Various functions are renamed and tracepoint outputs are changed due to
this removing.
Signed-off-by: Joonsoo Kim
---
include/linux/compaction.h| 14 +--
include/linux/mmzone.h| 3 +-
include/trace/events/compaction.h | 30 ++---
mm/compaction.c
ring.
Signed-off-by: Joonsoo Kim
---
mm/compaction.c | 41 +
mm/internal.h | 1 +
2 files changed, 42 insertions(+)
diff --git a/mm/compaction.c b/mm/compaction.c
index de96e9d..c6b8277 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -130,6 +13
k, we don't need to do this check.
Signed-off-by: Joonsoo Kim
---
mm/compaction.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
index 6ef2fdf..c2d3d6a 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -261,10 +261
ll
decreases but just a little.
In conclusion, this series solves three limitations of current compaction
algorithm successfully. :)
This patchset is based on linux-next-20150515 and not targeted to merge.
After merge window is finished, I will rebase it to recent kernel and
send v3.
Feel free to com
() before entering actual isolation. This patch
implements it and fix the problem.
Signed-off-by: Joonsoo Kim
---
mm/compaction.c | 30 +++---
mm/internal.h | 1 +
2 files changed, 16 insertions(+), 15 deletions(-)
diff --git a/mm/compaction.c b/mm/compaction.c
inde
On Thu, Aug 20, 2015 at 03:50:35PM +0800, Herbert Xu wrote:
> On Thu, Aug 20, 2015 at 04:52:17PM +0900, Joonsoo Kim wrote:
> >
> > Hmm... I guess there is no problem. crypto_alg object fetched by
> > crypto_get_comp() introduced in this patch could be hardware device
> &
On Thu, Aug 20, 2015 at 02:47:28PM +0800, Herbert Xu wrote:
> On Thu, Aug 20, 2015 at 03:34:57PM +0900, Joonsoo Kim wrote:
> > Until now, tfm object embeds (de)compression context in it and
> > (de)compression in crypto API requires tfm object to use
> > this context
lzo's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/lzo.c | 18 +-
1 file changed, 17 insertions(+), 1 de
2 9423894.38
Re-read 10017386.62 6428218.88 11000063.50
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c| 28 +++-
drivers/block/zram/zcomp.h| 9 -
drivers/block/zram/zram_drv.c | 9 +
3 files changed, 40 insertions(+), 6
842's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/842.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff
zram regards zstrm's buffer as compression destination buffer, but,
it is not intuitive and there is no document about it. Providing
destination buffer to zcomp_compress() directly seems more intuitive
interface to me so this patch changes zcomp_compress interface.
Signed-off-by: Joonso
lz4hc's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/lz4hc.c | 17 -
1 file changed, 16 insertions(+), 1 de
tch support to use
crypto noctx API and would restore the performance as is.
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/Kconfig | 13 +--
drivers/block/zram/Makefile| 4 +-
drivers/block/zram/zcomp.c | 84 +-
lz4's decompression doesn't requires any scratch buffer so
it doesn't need tfm context. Hence, it can support
crypto compression noctx API and this patch implements it.
Signed-off-by: Joonsoo Kim
---
crypto/lz4.c | 17 -
1 file changed, 16 insertions(+), 1 de
ssion function via crypto_alg,
instead of tfm object. crypto_alg is required to get appropriate
(de)compression function pointer. This can reduce overhead of
maintaining multiple tfm if (de)compression doesn't require
any context to operate.
Signed-off-by: Joonsoo Kim
---
crypto/842.c
/sec
zram-base vs zram-crypto vs zram-crypto-noctx
Read10411701.88 6426911.62 9423894.38
Re-read 10017386.62 6428218.88 1163.50
Thanks.
Joonsoo Kim (8):
crypto: support (de)compression API that doesn't require tfm object
crypto/l
On Thu, Aug 13, 2015 at 03:29:18PM +0800, Herbert Xu wrote:
> On Thu, Aug 13, 2015 at 04:30:31PM +0900, Joonsoo Kim wrote:
> >
> > How about introducing new functions to search supported algorithm in
> > kernel-side? As crypto API is used in more places, this interface
> &
On Thu, Aug 13, 2015 at 03:32:33PM +0800, Herbert Xu wrote:
> On Thu, Aug 13, 2015 at 04:19:41PM +0900, Joonsoo Kim wrote:
> >
> > If that optimization is really needed for the case that doesn't need
> > tfm except fetching function pointer, we can implement sharable t
On Thu, Aug 13, 2015 at 02:38:23PM +0800, Herbert Xu wrote:
> On Thu, Aug 13, 2015 at 03:37:55PM +0900, Joonsoo Kim wrote:
> >
> > Is there any way to access netlink interface and get the output from
> > kernel-side? I'd like to show information through
> >
On Thu, Aug 13, 2015 at 12:51:13PM +0900, Sergey Senozhatsky wrote:
> On (08/13/15 11:24), Joonsoo Kim wrote:
> > Until now, zram uses compression algorithm through direct call
> > to core algorithm function, but, it has drawback that we need to add
> > compression algorith
On Thu, Aug 13, 2015 at 11:19:54AM +0800, Herbert Xu wrote:
> On Thu, Aug 13, 2015 at 11:24:13AM +0900, Joonsoo Kim wrote:
> > Until now, zram uses compression algorithm through direct call
> > to core algorithm function, but, it has drawback that we need to add
> > compressi
hm in cryto subsystem. To implement
it on zram-side, crypto_alg_list and rwsem should be exported so
this patch do it.
Signed-off-by: Joonsoo Kim
---
crypto/internal.h | 2 --
include/linux/crypto.h | 4
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/crypto/internal.h b/c
patch will help
that situation.
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/Kconfig | 13 +--
drivers/block/zram/Makefile| 4 +-
drivers/block/zram/zcomp.c | 87 ++
drivers/block/zram/zcomp.h | 34 +
drive
On Mon, Aug 10, 2015 at 11:16:59AM +0900, Sergey Senozhatsky wrote:
> Hello Joonsoo,
>
> On (08/10/15 09:32), Joonsoo Kim wrote:
> > > on the other hand... it's actually
> > >
> > > wait_event() is
> > >
> > > if (condition)
> &
On Tue, Aug 11, 2015 at 08:26:33AM +0900, Minchan Kim wrote:
> Hi Joonsoo,
>
> On Mon, Aug 10, 2015 at 09:32:30AM +0900, Joonsoo Kim wrote:
> > On Fri, Aug 07, 2015 at 06:58:16PM +0900, Sergey Senozhatsky wrote:
> > > On (08/07/15 18:14), Sergey Senozhatsky wrote:
>
On Fri, Aug 07, 2015 at 05:05:01PM +0200, Michal Hocko wrote:
> On Fri 07-08-15 11:10:03, Joonsoo Kim wrote:
> [...]
> > diff --git a/mm/slub.c b/mm/slub.c
> > index 257283f..52b9025 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -1364,6 +1364,8 @@ stat
On Fri, Aug 07, 2015 at 11:49:04PM +0900, Minchan Kim wrote:
> Hi Joonsoo,
>
> On Fri, Aug 07, 2015 at 05:03:29PM +0900, Joonsoo Kim wrote:
> > Currently, when we enter the wait state due to lack of idle stream,
> > we check idle_strm list without holding the lock in expand
On Fri, Aug 07, 2015 at 06:58:16PM +0900, Sergey Senozhatsky wrote:
> On (08/07/15 18:14), Sergey Senozhatsky wrote:
> > hm... I need to think about it more.
> >
> > we do wake_up every time we put stream back to the list
> >
> > zcomp_strm_multi_release():
> >
> > spin_lock(&zs->strm_lo
e any error related to this race, it should be fixed.
To fix it, we should check idle_strm with holding the lock and
this patch implements it.
Signed-off-by: Joonsoo Kim
---
drivers/block/zram/zcomp.c | 15 +--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/
er occupies 150 MB memory.
Workload: hackbench -g 20 -l 1000
Average result by 10 runs (Base va Patched)
elapsed_time(s): 4.3468 vs 2.9838
compact_stall: 461.7 vs 73.6
pgmigrate_success: 28315.9 vs 7256.1
Signed-off-by: Joonsoo Kim
---
mm/slub.c | 2 ++
1 file changed, 2 insertions(+)
diff
On Tue, Aug 04, 2015 at 03:15:26PM +0200, Michal Hocko wrote:
> On Fri 31-07-15 10:09:50, Joonsoo Kim wrote:
> > Almost description is copied from commit fb05e7a89f50
> > ("net: don't wait for order-3 page allocation").
> >
> > I saw excessive direct m
On Fri, Jul 31, 2015 at 08:19:07AM +0100, Mel Gorman wrote:
> On Fri, Jul 31, 2015 at 03:08:38PM +0900, Joonsoo Kim wrote:
> > On Mon, Jul 20, 2015 at 09:00:19AM +0100, Mel Gorman wrote:
> > > From: Mel Gorman
> > >
> > > The primary purpose of watermarks
On Fri, Jul 31, 2015 at 09:25:13AM +0200, Vlastimil Babka wrote:
> On 07/31/2015 09:11 AM, Mel Gorman wrote:
> >On Fri, Jul 31, 2015 at 02:54:07PM +0900, Joonsoo Kim wrote:
> >>Hello, Mel.
> >>
> >>On Mon, Jul 20, 2015 at 09:00:18AM +0100, Mel Gorman wrote:
&g
On Fri, Jul 31, 2015 at 08:11:13AM +0100, Mel Gorman wrote:
> On Fri, Jul 31, 2015 at 02:54:07PM +0900, Joonsoo Kim wrote:
> > Hello, Mel.
> >
> > On Mon, Jul 20, 2015 at 09:00:18AM +0100, Mel Gorman wrote:
> > > From: Mel Gorman
> > >
> > > High
On Mon, Jul 20, 2015 at 09:00:09AM +0100, Mel Gorman wrote:
> From: Mel Gorman
>
> This series started with the idea to move LRU lists to pgdat but this
> part was more important to start with. It was written against 4.2-rc1 but
> applies to 4.2-rc3.
>
> The zonelist cache has been around for a
On Mon, Jul 20, 2015 at 09:00:19AM +0100, Mel Gorman wrote:
> From: Mel Gorman
>
> The primary purpose of watermarks is to ensure that reclaim can always
> make forward progress in PF_MEMALLOC context (kswapd and direct reclaim).
> These assume that order-0 allocations are all that is necessary f
701 - 800 of 2325 matches
Mail list logo