gmane.comp.file-systems.nilfs.user

Re: [PATCH 1/1] nilfs2: add omitted comment for ns_mount_state field of the_nilfs structure

Ryusuke Konishi — 2012-05-23T13:22:48

Thanks.  I will pick it up.

Ryusuke Konishi

--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 1/1] nilfs2: add omitted comment for ns_mount_state field of the_nilfs structure

Vyacheslav Dubeyko — 2012-05-22T14:25:15

Hi,

I found that ns_mount_state field doesn't commented in the_nilfs
structure. I simply add comment for ns_mount_state field.

With the best regards,
Vyacheslav Dubeyko.
--
From: Vyacheslav Dubeyko <slava-yeENwD64cLxBDgjK7y7TUQ< at >public.gmane.org>
Subject: [PATCH 1/1] nilfs2: add omitted comment for ns_mount_state
field of the_nilfs structure

This patch adds omitted comment for ns_mount_state field of the_nilfs
structure.

Signed-off-by: Vyacheslav Dubeyko <slava-yeENwD64cLxBDgjK7y7TUQ< at >public.gmane.org>
---
diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h
index 1a588a3..7885bdb 100644
--- a/fs/nilfs2/the_nilfs.h
+++ b/fs/nilfs2/the_nilfs.h
< at >< at > -53,6 +53,7 < at >< at > enum {
  * < at >ns_sbwtime: previous write time of super block
  * < at >ns_sbwcount: write count of super block
  * < at >ns_sbsize: size of valid data in super block
+ * < at >ns_mount_state: file system state
  * < at >ns_seg_seq: segment sequence counter
  * < at >ns_segnum: index number of the latest full segment.
  * < at >ns_nextnum: index number of the full segment index to be used next
--


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [NILFS2 design doc] draft of the on-disk layout description paragraph

Reinoud Zandijk — 2012-05-20T15:50:46

Hi Vyacheslav,

On Sun, May 20, 2012 at 06:23:29PM +0400, Vyacheslav Dubeyko wrote:

Looks good. The byte code examples i'd leave out; they are confusing. Maybe as
an edditional part?

Cheers,
Reinoud

--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [NILFS2 design doc] draft of the on-disk layout description paragraph

Vyacheslav Dubeyko — 2012-05-21T05:36:26

Hi Reinoud,

On Sun, 2012-05-20 at 17:50 +0200, Reinoud Zandijk wrote:

Ok. I see. I will think about it and try to rework. Thank you.

With the best regards,
Vyacheslav Dubeyko.

--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[NILFS2 design doc] draft of the on-disk layout description paragraph

Vyacheslav Dubeyko — 2012-05-20T14:23:29

Hi all,

I am working on NILFS2 design doc now. Currently, I elaborated draft of the on-disk layout description paragraph. You can download draft of the design document from here: http://dubeyko.com/development/FileSystems/NILFS/nilfs2-design.pdf.

Could you please review the draft of the on-disk layout description paragraph and point out any misunderstandings or misspellings? Thank you in advance.

With the best regards,
Vyacheslav Dubeyko.--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Streaming Service

2012-05-15T19:35:20

Our Streaming Server is powerful and optimized server for live and on-demand audio/ video streaming content delivery. 
Our Streaming Server features with high performance streaming throughput, network & storage I/O and optimized configurations for its high scalability and reliability.
Our server supports latest Windows Media Technologies such as Fast Streaming, Fast Cache, Fast Start, Wireless Streaming as well as feature-rich add-on modules such as sophisticated content management, monitoring, logging reports, billing, authentications, load-balancing & clustering for all needs. 

Pls send us email for further information.Thanks,

Boris 
boris-JYBpPTDdvJEqwVFMiTZqHLZc9IWuHzk0< at >public.gmane.org

If you do not wish to further receive this event message, email "borislamsv2-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org" to unsubscribe this message or remove your email from the list.


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

haalloo,

abi — 2012-05-12T17:06:50

haalloo,
how are you doing,i hope you are fine,my name is miss abi okom i got your
contact and want us to be a good friend,
please try and write back to me so that i will give you my pictures and tell
you more about me,
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Small device

Zeke Rondel — 2012-05-10T18:01:55

Using "-b 1024 -B 512", we would get 512 blocks per segment: (1024*512)
/ 1024

Then, what about to use "-B" to 1024 or to 2048 ? So there were lesser
number of full segments in the device.
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Small device

Ryusuke Konishi — 2012-05-10T15:42:22

Hi,
On Thu, 10 May 2012 00:33:30 +0100, Zeke Rondel wrote:

Basically NILFS is not designed for small devices.  Its most on-disk
data items have 64-bit width.  In addition, it's more likely to suffer
from disk full on such devices due to the copy-on-write nature.

However, I guess you can use NILFS on 64MB SD card by formatting the
device with a smaller block size and a smaller number of blocks per
segment.

My mkfs format parameter for a 64MB sized device is as follows:

$ mkfs -t nilfs2 -b 1024 -B 512 <device>


We may get more disk space overhead for much smaller "blocks per
segment" because every segment of NILFS needs at least one header
block.


Regards,
Ryusuke Konishi
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: fsck.nilfs2: not found

karunakar merugu — 2012-05-10T05:45:16

hi
I am trying to test xfstests on nilfs.During this i got this error
fsck.nilfs2 not found.
Is any one tried xfstests for nilfs and facing same problem ?
thanks,
karunakar.

On Wed, May 9, 2012 at 4:51 PM, Vyacheslav Dubeyko <slava-yeENwD64cLxBDgjK7y7TUQ< at >public.gmane.org> wrote:
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Small device

Zeke Rondel — 2012-05-09T23:33:30

Hi,

I've a small memory card (SD) of 64MB used for keys related to crypto.
stuff so I need a log filesystem that avoids the data can be re-written
or modified. Is NILFS suitable in case of little devices? If so, which
size of blocks per segment would be suitable for it? Thanks in advance!
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: fsck.nilfs2: not found

Vyacheslav Dubeyko — 2012-05-09T11:21:37

Hi,

Sorry, but as I know fsck tool for NILFS filesystem doesn't implemented yet.

With the best regards,
Vyacheslav Dubeyko.

On May 9, 2012, at 1:16 PM, karunakar merugu wrote:


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

fsck.nilfs2: not found

karunakar merugu — 2012-05-09T09:16:00

Hi
I am using Linux karunakar-VirtualBox 3.0.0-12-generic #20-Ubuntu SMP.
I installed nilfs utilities  by
sudo apt-get update
sudo apt-get install nilfs2-tools

Even after this I am getting error like:
fsck from util-linux 2.19.1
fsck: fsck.nilfs2: not found
fsck: Error 2 while executing fsck.nilfs2 for /dev/sdb

can anyone help me in finding the correct package for fsck.nilfs2
utility please.
thanks,
Karunakar.
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/1] mkfs.nilfs2: initialize block info array for DAT file by nilfs_binfo_dat structure

Ryusuke Konishi — 2012-05-06T18:29:29

Thanks for the patch.  But, No, the current mkfs implementation is
correct.

nilfs_binfo_dat is the structure storing summary information only
applied to btree intermediate blocks of the DAT meta-data file.

For data blocks of the DAT file, a 64-bit block offset number (__le64)
is allocated per block.

Since mkfs.nilfs2 only allocates data blocks in the initial format and
never allocates btree node blocks, nilfs_binfo_dat structure will not
appear in the tool.

The current kernel module is using the same binfo format.  If this
patch is applied, compatibility of the disk format will be broken.


Regards,
Ryusuke Konishi


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 1/1] mkfs.nilfs2: initialize block info array for DAT file by nilfs_binfo_dat structure

Vyacheslav Dubeyko — 2012-05-06T12:06:41

Hello,

I think that I have found obsolete code in mkfs.nilfs2. And I prepare a patch which fixes the bug with block info initialization in segment summary for the case of DAT file.

With the best regards,
Vyacheslav Dubeyko.
--
From: Vyacheslav Dubeyko <slava-yeENwD64cLxBDgjK7y7TUQ< at >public.gmane.org>
Subject: [PATCH 1/1] mkfs.nilfs2: initialize block info array for DAT file by nilfs_binfo_dat structure

Segment summary block begins from segment summary header (nilfs_segment_summary structure). It follows array of file information items (nilfs_finfo structure) after it. And after each file information item follows array of block information items (nilfs_binfo union). The block information item can be nilfs_binfo_v structure (information for the block to which a virtual block number is assigned) or nilfs_binfo_dat structure (information for the block which belongs to the DAT file).

Current state of mkfs.nilfs2 code initializes block information items as sizeof(__le64) for the DAT file case. But driver code expects greater structure in size because it operates by bi_level field for the case of DAT file. The patch uses nilfs_binfo_dat structure instead of __le64 during creation of block information items for DAT file case.

Signed-off-by: Vyacheslav Dubeyko <slava-yeENwD64cLxBDgjK7y7TUQ< at >public.gmane.org>
---
diff -up nilfs-utils/sbin/mkfs/mkfs.c{.orig,} 
--- nilfs-utils/sbin/mkfs/mkfs.c.orig2012-05-04 12:30:58.308461058 +0400
+++ nilfs-utils/sbin/mkfs/mkfs.c2012-05-04 12:32:39.593861721 +0400
< at >< at > -444,7 +444,7 < at >< at > static void increment_segsum_size(struct
   unsigned nblocks_in_file, int dat_flag)
 {
 unsigned binfo_size = dat_flag ? 
-sizeof(__le64) /* offset */ : sizeof(struct nilfs_binfo_v);
+sizeof(struct nilfs_binfo_dat) : sizeof(struct nilfs_binfo_v);

 si->sumbytes = __increment_segsum_size(si->sumbytes,
        sizeof(struct nilfs_finfo), 1);
< at >< at > -1240,13 +1240,13 < at >< at > static void update_blocknr(struct nilfs_
 finfo->fi_cno = cpu_to_le64(1);

 if (fi->ino == NILFS_DAT_INO) {
-__le64 *pblkoff;
+struct nilfs_binfo_dat *pbinfo_dat;

 fi->raw_inode->i_bmap[0] = 0;
 for (i = 0; i < fi->nblocks; i++) {
-pblkoff = map_segsum_info(start, sum_offset,
-  sizeof(*pblkoff));
-*pblkoff = cpu_to_le64(i);
+pbinfo_dat = map_segsum_info(start, sum_offset,
+  sizeof(*pbinfo_dat));
+pbinfo_dat->bi_blkoff = cpu_to_le64(i);
 fi->raw_inode->i_bmap[i + 1] =
 cpu_to_le64(fi->start + i);
 }
--

--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH] nilfs2: flush disk caches in syncing

Ryusuke Konishi — 2012-05-05T15:30:19

Hi Andrew,

Could you please queue the following patch in the -mm tree ?

I hope this will be sent to upstream at the next merge window through
your tree. (probably only this single change).

Thanks,
Ryusuke Konishi
--
From: Ryusuke Konishi <konishi.ryusuke-Zyj7fXuS5i5L9jVzuh4AOg< at >public.gmane.org>

There are two cases that the cache flush is needed to avoid data loss
against unexpected hang or power failure.  One is sync file function
(i.e. nilfs_sync_file) and another is checkpointing ioctl.

This issues a cache flush request to device for such cases if barrier
mount option is enabled, and makes sure data really is on persistent
storage on their completion.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke-Zyj7fXuS5i5L9jVzuh4AOg< at >public.gmane.org>
---
 fs/nilfs2/file.c  |   24 ++++++++++++++----------
 fs/nilfs2/ioctl.c |    8 +++++++-
 2 files changed, 21 insertions(+), 11 deletions(-)

diff --git a/fs/nilfs2/file.c b/fs/nilfs2/file.c
index 2660152..62cebc8 100644
--- a/fs/nilfs2/file.c
+++ b/fs/nilfs2/file.c
< at >< at > -37,6 +37,7 < at >< at > int nilfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
  * This function should be implemented when the writeback function
  * will be implemented.
  */
+struct the_nilfs *nilfs;
 struct inode *inode = file->f_mapping->host;
 int err;
 
< at >< at > -45,18 +46,21 < at >< at > int nilfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 return err;
 mutex_lock(&inode->i_mutex);
 
-if (!nilfs_inode_dirty(inode)) {
-mutex_unlock(&inode->i_mutex);
-return 0;
+if (nilfs_inode_dirty(inode)) {
+if (datasync)
+err = nilfs_construct_dsync_segment(inode->i_sb, inode,
+    0, LLONG_MAX);
+else
+err = nilfs_construct_segment(inode->i_sb);
 }
-
-if (datasync)
-err = nilfs_construct_dsync_segment(inode->i_sb, inode, 0,
-    LLONG_MAX);
-else
-err = nilfs_construct_segment(inode->i_sb);
-
 mutex_unlock(&inode->i_mutex);
+
+nilfs = inode->i_sb->s_fs_info;
+if (!err && nilfs_test_opt(nilfs, BARRIER)) {
+err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+if (err != -EIO)
+err = 0;
+}
 return err;
 }
 
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index 2a70fce..06658ca 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
< at >< at > -692,8 +692,14 < at >< at > static int nilfs_ioctl_sync(struct inode *inode, struct file *filp,
 if (ret < 0)
 return ret;
 
+nilfs = inode->i_sb->s_fs_info;
+if (nilfs_test_opt(nilfs, BARRIER)) {
+ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+if (ret == -EIO)
+return ret;
+}
+
 if (argp != NULL) {
-nilfs = inode->i_sb->s_fs_info;
 down_read(&nilfs->ns_segctor_sem);
 cno = nilfs->ns_cno - 1;
 up_read(&nilfs->ns_segctor_sem);

Re: Nilfs2 and Grub2

Jiro SEKIBA — 2012-05-04T05:13:46

Hi,

I'm original author of the grub2 NILFS2 module.

Yes, it lacks roll back/forward or recovery to keep the boot loader simple.
It also prevent from unintentional boot, like old buggy kernel is used.

Aside from the way it works, the code itself is capable to boot from
any checkpoints.  It happens to choose the latest one.  The only thing
we need is the way to _choose_ the checkpoint from grub interface :(.

thanks

regards

Server Rental Service in Hong Kong

2012-05-03T18:25:10

Dear All,

We have our own datacenter in Hong Kong & provide email/application/web rental service to clients.We are APNIC member & provide clean IP to clients.

Dell? PowerEdge? EnterpriseRack Mount Server
-Intel(R) Xeon(R) E3-1240 Processor (3.3GHz, 8M Cache, Turbo, 4C/8T, 80W)
-8GB RAM, 2x4GB, 1333MHz, DDR-3, Dual Ranked UDIMMs
-500GB, 3.5", 6Gbps SAS x 2
-Raid 1 Mirroring Protection
-Remote KVM (iDRAC6 Enterprise)

Dell(TM) PowerEdge(TM) R410 Rack Mount Server
-Intel(R) Quad Core E5606 Xeon(R) CPU, 2.13GHz, 4M Cache, 4.86 GT/s QPI
-4GB Memory (2x2GB), 1333MHz Dual Ranked RDIMMs Fully-Buffered
-500GB 7.2K RPM SATAII 3.5" Hard Drive x 2
-iDRAC6 Enterprise or Express (Remote KVM Management)

Every Dedicated Server Hosting Solution Also Includes: 
 
Software Specification 
- CentOS / Fedora / Debian / FreeBSD / Ubuntu / Redhat Linux 
- Full root-level access 
- Data Center Facilities 
- Shared Local & International Bandwidth 
- 2 IP Addresses Allocation 
- Un-interruptible Power Supply (UPS) backed up by private diesel generator 
- FM200¡§based fire suppression system 
- 24x7 CRAC Air Conditioning and Humidity Control 
- 24x7 Security Control 
- 24x7 Remote Hand Service 

Pls send us email for further information.Thanks,

Boris 
boris-JYBpPTDdvJEqwVFMiTZqHLZc9IWuHzk0< at >public.gmane.org

If you do not wish to further receive this event message, email "borislamsv2-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org" to unsubscribe this message or remove your email from the list.


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: NILFS development contribution

Ryusuke Konishi — 2012-05-03T06:59:38

Hi,
On Wed, 25 Apr 2012 11:18:41 +0530, santhosh kumars wrote:

I think one good start is to use "xfstest", an test suites for xfs and
other filesystems.

Thanks,
Ryusuke Konishi
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Nilfs2 and Grub2

Ryusuke Konishi — 2012-05-03T06:49:04

Hi,
On Tue, 1 May 2012 03:38:57 -0700 (PDT), Sven-Göran_Bergh wrote:

Thank you for reporting the issue in the nilfs2 boot loader of grub2.

Your analysis seems right to me.

The nilfs2 grub code looks to be lacking roll forwarding and crc
verification (the latter both for super blocks and log blocks).

It is incomplete, or rather I should say it took a good trade-off
between simplicity and reliability as an early practicable
implementation.

Actually it works without trouble in most cases since most checkpoints
coherently include files needed to boot up system even though that
omission may lead the boot loader to read not-updated files after
unclean shutdown.

In my opinion, nilfs2 grub code does not have to do recovery, but it
should scan the latest checkpoint in order to minimize troubles.
However, I don't know whether it's possible or not in the boot loader.

Of course, there is a possibility that simply an unknown bug caused
your trouble.  In that case, inserting debug print functions in the
nilfs2 grub code and booting up with it, would be helpful to narrow
down the cause of the problem.


Thanks,
Ryusuke Konishi
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: man nilfs-clean

Ryusuke Konishi — 2012-05-03T04:13:29

Thanks, I will apply it.


That's OK.

One thing I will ask you to do, is to include a "Singed-off-by" line
(sob line for short) in your patch, which certifies that you wrote it
or otherwise have the right to pass it on as an open-source patch.
even if it's a typo fix ;)

The sob line is inserted before the source code diff in the following
form:

Signed-off-by: Tiago Caxias <your e-mail address>

This is automatically done if you used "git commit" command with an -s
option; usually you can make a patch from the local change by "git
commit -a -s".

Then you can confirm or output the patch by "git show" or "git
format-patch HEAD^.."

I think that it is good to learn the function of git little by little
starting from hand working since git has various functions.


Thanks,
Ryusuke Konishi

--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA< at >public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html