Jan,
Thanks for taking time to review the patch :-)
Jan Kara wrote:
> Hi Coly,
>
> finally I've found some time to have a look at a new version of your
> patch.
>
>> 5, Performance number
>> On a Core-Duo, 2MB DDM memory, 7200 RPM SATA PC, I built a 50GB
Thanks for the feedback :-)
Mingming Cao wrote:
> On Tue, 2007-11-13 at 22:12 +0800, Coly Li wrote:
>> Basic idea of my dir inode reservation patch can be found here,
>> http://lists.openwall.net/linux-ext4/2007/11/05/3
>>
>> 1, What does dir inode reservation do
>&g
Coly Li wrote:
> Thanks for the feedback :-)
>
> Alex Tomas wrote:
>> hmm. so you trade 265% degradation of creation for 40% improvement of
>> unlink?
>>
> 265% degradation is only for creating 5 empty directories. This is not a
> common case.
> There ar
e so many empty directories, they also can mount
the file system without
dir inode reservation to get better performance.
> thanks, Alex
>
> Coly Li wrote:
>> normal ext4ext4 with dir inode reservation
>> mount options:-o data=writeba
and other friendly guys who
discussed ext4 stuffs
with me in irc. Thanks to make me enjoy open source development.
7, Feedback
For any feedback, criticism, please email to coyli at suse dot de. I am very
happy to hear from you
and improve the dir inode reservation patch.
Signed-off-
Designing and Implementation of Directory Inode Reservation
version 1
Coly Li
Nov 4, 2007
This text explain the designing and implementation of the directory inode
reservation patch to ext4
file system. The original idea was developed by Andreas Dilger and Danial
Phillips when ext3 htree
was
series that
> has a couple of fixes. It also has one additional patch.
>
> http://www.mail-archive.com/linux-ext4@vger.kernel.org/msg03803.html
Thanks for the update :-)
>
> -JRS
--
Coly Li
SuSE PRC Labs
-
To unsubscribe from this list: send the line "unsubscr
Jose R. Santos wrote:
> On Thu, 01 Nov 2007 17:58:42 +0800
> Coly Li <[EMAIL PROTECTED]> wrote:
>
>> Coly Li wrote:
>>> I plan to do some work based on uninit group patch. When I try to patch
Coly Li wrote:
> I plan to do some work based on uninit group patch. When I try to patch the
> patch into kernel, I
sorry
it's typo, I mean into e2fsprogs...
> find the sequence number o
r uninit block groups.
>
> Signed-off-by: Jose R. Santos <[EMAIL PROTECTED]>
> --
>
> tests/m_uninit/expect.1 | 166
> +++
> tests/m_uninit/script |4 +
> 2 files changed, 170 insertions(+), 0 deletions(-)
>
--
Coly Li
S
Theodore Tso wrote:
> On Tue, Oct 30, 2007 at 01:03:38AM +0800, Coly Li wrote:
>> This patch adds directory inode reservation supporting to mke2fs. Ask for
>> review.
>
> So I had completely erased this from my memory, and had to do some
> google searching and some mail
f24112341b0 (maybe my mistake during git-pull). If
there is any error
reported from fs/ext4/inode.c, please fix it before apply this patch.
Thank Mingming's for the feedback.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
Cc: Andreas Dilger <[EMAIL PROTECTED]>
Cc: Mingming Cao <
This patch adds explaining for directory inode reservation option to mke2fs.8.
Ask for review.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
---
misc/mke2fs.8.in |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/misc/mke2fs.8.in b/misc/mke2fs.8.in
index a3dc4a1..9
This patch adds directory inode reservation supporting to mke2fs. Ask for
review.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
---
lib/ext2fs/ext2_fs.h |9 -
misc/mke2fs.c| 27 ---
2 files changed, 32 insertions(+), 4 deletions(-)
diff --git
will post detail text later. Any comments for this patch is great welcome :-)
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
Cc: Andreas Dilger <[EMAIL PROTECTED]>
Cc: Mingming Cao <[EMAIL PROTECTED]>
---
fs/ext4/ialloc.c | 203 ++-
to 2.6.23 ? and after the merge, the
patches will in pached
in the tree again ?
Should I generate my patch based on current version, or wait for some day ?
Indeed, I try to add my
patch to current version, but some patches missing from ext4 tree.
Thanks for information :-)
--
Coly Li
SuSE PRC
Theodore Tso wrote:
> On Mon, Oct 15, 2007 at 10:19:11PM +0800, Coly Li wrote:
>> -E option is missed from usage information of mke2fs. This patch adds -E
>> option to usage
>> information, and names -E|-R as extended-option.
>> Now the usage information can be
Theodore Tso wrote:
> On Mon, Oct 15, 2007 at 10:19:11PM +0800, Coly Li wrote:
>> -E option is missed from usage information of mke2fs. This patch adds -E
>> option to usage
>> information, and names -E|-R as extended-option.
>> Now the usage information can be
This patch extends bg_itable_unused of ext4 group descriptor from 16bit into
32bit. In order to add
bg_itable_unused_hi into struct ext4_group_desc, some extra fields which are
already introduced into
e2fsprogs are also added in for consistency.
Signed-off-by: Coly Li <[EMAIL PROTECTED]&
Jose R. Santos wrote:
> On Mon, 15 Oct 2007 20:46:32 +0800
> Coly Li <[EMAIL PROTECTED]> wrote:
>
>> Thanks for the replying :-)
>>
>> Jose R. Santos wrote:
>>> On Mon, 15 Oct 2007 13:36:05 +0800
>>> Coly Li <[EMAIL PROTECTED]> wrote:
-E option is missed from usage information of mke2fs. This patch adds -E option
to usage
information, and names -E|-R as extended-option.
Now the usage information can be consistent with mke2fs.8.in.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
---
misc/mke2fs.c |2 +-
1 files chan
Thanks for the replying :-)
Jose R. Santos wrote:
> On Mon, 15 Oct 2007 13:36:05 +0800
> Coly Li <[EMAIL PROTECTED]> wrote:
>
>> Now in my mind there are several words for ext4 patches, most frequently one
>> are "patch queue".
>>
>> I s
--
we need to extend
this field into 32bit for ext4.
In latest e2fsprogs git, it seems only adding a bg_itable_unused_hi in struct
ext4_group_desc is OK
(do not interfere current e2fsprogs and kernel code).
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
---
diff --git a/lib/ext2fs/ext2_fs.h
nel/people/tytso/ext4-patches/ ?
Thanks for clarifying :-)
--
Coly Li
SuSE PRC Labs
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
This patch remove masix support from mke2fs.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
diff --git a/misc/mke2fs.c b/misc/mke2fs.c
index 4a6cace..e317d48 100644
--- a/misc/mke2fs.c
+++ b/misc/mke2fs.c
@@ -740,8 +740,6 @@ static int set_os(struct ext2_super_block *sb, char *os)
This patch remove masix supporting from debugfs.
Singed-off-by: Coly Li <[EMAIL PROTECTED]>
--
diff --git a/debugfs/debugfs.c b/debugfs/debugfs.c
index 190c4b7..37dbeba 100644
--- a/debugfs/debugfs.c
+++ b/debugfs/debugfs.c
@@ -569,10 +569,6 @@ void internal_dump_inode(FILE *out, cons
This patch remove masix support from lib/ext2fs.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
diff --git a/lib/ext2fs/ext2_fs.h b/lib/ext2fs/ext2_fs.h
index a316665..3585baf 100644
--- a/lib/ext2fs/ext2_fs.h
+++ b/lib/ext2fs/ext2_fs.h
@@ -315,9 +315,6 @@ struct ext2_inode {
This patch remove masix support from e2fsck.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
diff --git a/e2fsck/pass1.c b/e2fsck/pass1.c
index ceb9c7f..d072e73 100644
--- a/e2fsck/pass1.c
+++ b/e2fsck/pass1.c
@@ -834,10 +834,6 @@ void e2fsck_pass1(e2fsck_t ctx)
frag =
This patch remove masix union from data structure in ext2ed-design.sgml.
Signed-off-by: Coly Li <[EMAIL PROTECTED]>
diff --git a/ext2ed/doc/ext2ed-design.sgml b/ext2ed/doc/ext2ed-design.sgml
index 3eaca69..ad2df96 100644
--- a/ext2ed/doc/ext2ed-design.sgml
+++ b/ext2ed/doc/ext2ed-desig
Theodore Tso wrote:
> On Mon, Aug 13, 2007 at 12:33:26AM +0800, Coly Li wrote:
>> Theodore,
>>
>> Another question of mine is for the .po files: Once I modify a string from
>> _(""), should I mofify
>> all the corresponded strings from all .po files ?
t; is hurd specific only. What is your comment ?
This time, I use MIME/PGP in my signature, and tested by sending the patch to
another account. Hope
the format is ok this time :-)
Signed-Off-By: Coly Li <[EMAIL PROTECTED]>
diff --git a/debugfs/debugfs.c b/debugfs/debugfs.c
index 190c4b7
Theodore,
Another question of mine is for the .po files: Once I modify a string from
_(""), should I mofify
all the corresponded strings from all .po files ?
Theodore Tso wrote:
> On Sat, Aug 11, 2007 at 01:13:39AM +0800, Coly Li wrote:
>> -BEGIN PGP SIGNED MESSAGE-
&
I should modify the names of these fields, and keep the checking routines.
Andreas Dilger wrote:
> On Aug 11, 2007 01:13 +0800, Coly Li wrote:
>> - --- a/debugfs/set_fields.c
>> +++ b/debugfs/set_fields.c
>> @@ -65,9 +65,7 @@ static struct field_
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Andreas Dilger wrote:
> On Aug 10, 2007 11:37 +0800, Coly Li wrote:
>> - --- a/e2fsck/problem.h
>> +++ b/e2fsck/problem.h
>> @@ -558,8 +558,8 @@ struct problem_context {
>> +/* Duplicate directory entry found */
>>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
- From last patch, I found some strings contain '@' are treated as email
address by my email client.
Therefore I resend this post again with modified format. Hope it is OK this
time :-)
Signed-Off-By: Coly Li <[EMAIL PROTECTED]&g
linux-ext4 removing fragment supporting from e2fsprogs.
I tested both patch.
Signed-Off-By: Coly Li <[EMAIL PROTECTED]>
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 427f830..b8b538d 100644
- --- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -576,11 +576,6 @@ got:
/* dirsyn
Valerie,
How about this:
git clone git://git.kernel.org/pub/scm/fs/ext2/e2fsprogs.git
It works for me.
Valerie Clement wrote:
> Theodore Ts'o wrote:
>> People may have noticed that e2fsprogs 1.40 and roughly a week later
>> e2fsprogs 1.40.1 have been released, while there hasn't been any
>> act
ll related sentences to fragment supporting in documents should
be removed too. Isn't it ?
Signed-Off-By: Coly Li <[EMAIL PROTECTED]>
diff --git a/debugfs/debugfs.c b/debugfs/debugfs.c
index 190c4b7..e94302d 100644
- --- a/debugfs/debugfs.c
+++ b/debugfs/debugfs.c
@@ -521,7 +521,6 @
clear, this is a perfect explaining for Signed-off-by :-)
Thanks a lot.
Coly
Theodore Tso wrote:
> On Sat, Aug 04, 2007 at 02:07:30PM +0800, Coly Li wrote:
>> Do you mean "Signed-off-by Coly Li <[EMAIL PROTECTED]>" ?
>> I just though this is used for others to s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
>Do you mean "Signed-off-by Coly Li <[EMAIL PROTECTED]>" ?
>I just though this is used for others to sign my patch :-) Should I add this
>to the patch this time ?
Oh, I find this patch is applied already. Thanks for notin
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Theodore,
Do you mean "Signed-off-by Coly Li <[EMAIL PROTECTED]>" ?
I just though this is used for others to sign my patch :-) Should I add this to
the patch this time ?
Thanks
Theodore Tso wrote:
> On Wed, Aug 01, 2007 at 03:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Eric Sandeen wrote:
> Andreas Dilger wrote:
>> On Aug 02, 2007 17:55 +0800, Coly Li wrote:
>>> These days, when I read the source code of mke2fs, I found an option -f
>>> to set the fragment size.
>>>
>>>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I read from manual and source code, it seems fragment size should be
same to filesystem block size. Therefore, can I say current ext3/4
filesystem still do not support fragment ?
Thanks.
Coly
Andreas Dilger wrote:
> On Aug 02, 2007 17:55 +0
what is the function or purpose of fragment of ext
file system ?
Thanks in advance :-)
Coly
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org
iD8DBQFGsan5uTp8cyZ5lTERAvNJAJ0RVII/NFrq8V37AB+tNZ7r291hggCgltef
This patch remove 3 compiling warning of current e2fsprogs git.
Best regards.
Coly
diff --git a/debugfs/set_fields.c b/debugfs/set_fields.c
index b50ff7c..ce0301f 100644
--- a/debugfs/set_fields.c
+++ b/debugfs/set_fields.c
@@ -16,6 +16,7 @@
#include
#include
#include
+#include
Hi, once we decide to do this, how about storing inode inside the
directory ?
IMHO, the latter one is more attractive :-)
Coly
在 2007-07-10二的 12:12 -0500，Jose R. Santos写道：
> Hi folks,
>
> As I play with the allocation of the metadata for the FLEX_BG feature,
> it seems that we c
Badari,
How about the performance improvement, is there any benchmark result ?
Coly
在 2007-07-05四的 11:11 -0700，Badari Pulavarty写道：
> This is a patch that speeds up statfs. It is very simple -
> the "overhead" calculation, which takes a huge amount of time
> for large
Andrew,
Thanks for your information :-)
Coly
在 2007-05-25五的 16:33 +1000，andrew hendry写道：
> select your whole mail and use pre-format in evolution.
> or change it to pre-format and do insert->text file.
> this should send it with tabs intact, the confusing bit i think is if
> you
Hi,
I tested again, it seems Evolution removes the Tabs with blanks.
How to resolve this issue on Evolution ? I am trying :-)
Coly
在 2007-05-25五的 07:52 +0200，Jan Engelhardt写道：
> On May 25 2007 09:30, WANG Cong wrote:
> >>>
> >>>Yes, I found all TABs gone when I rec
Dave,
Yes, I found all TABs gone when I received the mail. When I post next
version of the patch, I will test to send to me first :-)
Thanks for your information.
Coly
在 2007-05-24四的 08:20 -0500，Dave Kleikamp写道：
> On Thu, 2007-05-24 at 02:06 +0800, coly wrote:
> > The patch is genera
Theodore,
I can see all the files, but how can I download the files by cvs ? I
tried pserver, but it seems does not work ..
could you give me some hints ?
Thanks
Coly
在 2007-05-23三的 08:17 -0400，Theodore Tso写道：
> On Wed, May 23, 2007 at 05:00:43PM +0800, coly wrote:
> >
> > T
I/O for magic inode.
4, Modify mke2fs to support new on-disk layout for inode reservation.
Coly
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Current patch avoids inodes from different directories mixed together in
the inode table. Therefore the benchmakr emulate a situation that mixes
inodes of different sub-directories together. and record the time on
removing them all.
In the first part, reserving 16 inodes for each new created direc
This patch only makes mke2fs support on-disk layout for inode
reservation. Just for experiment. e2fsck and other utils can not work
with magic inode yet.
diff -u -r
e2fsprogs-1.39-tyt1/debugfs/debugfs.c ../e2fsprogs/debugfs/debugfs.c
--- e2fsprogs-1.39-tyt1/debugfs/debugfs.c 2006-10-07 11:42:54.
The patch is generated based on 2.6.20-ext4-2 branch. you can find the
benchmark from other email.
DO NOT waste time on reading the patch :-) I post this patch here is to
show that I really spent time on it and the patch can work (even not
well).
diff --git a/Makefile b/Makefile
index 7e2750f..2
several months, I will try best to write a usable patch :-)
Coly
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Theodore,
Thanks for your kindly replying.
Coly
在 2007-05-23三的 08:17 -0400，Theodore Tso写道：
> Mercurial
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
guess maybe I should generate the patch
for e2fsprogs based on e2fsprogs-TEST branch, a.k.a 1.40-WIP-0407. Isn't
it ?
This information has to be confirmed for me to generate the patch.
Thanks for anyone replys me :-)
Best regards.
Coly
-
To unsubscribe from this list: send the line "u
Badari:
Thanks, maybe Ted's tree is a good choice :-)
Best regards.
Coly
在 2007-05-02三的 16:16 -0700，Badari Pulavarty写道：
> On Thu, 2007-05-03 at 02:13 +0800, coly wrote:
> > Hi, list.
> >
> > I find these days 2.6.21 released. If I will submit patches for ext4dev,
Hi, list.
I find these days 2.6.21 released. If I will submit patches for ext4dev,
should I generate the patch against 2.6.21 ?
Best regards.
Coly
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to [EMAIL PROTECTED]
More majordomo inf
Theodore:
Thanks for your explaining. I ignored this detail before, it is more
clear to me now.
Best regards.
Coly
在 2007-04-09一的 11:26 -0400，Theodore Tso写道：
> On Mon, Apr 09, 2007 at 10:33:13AM +0800, coly wrote:
> > Theodore:
> >
> > Thanks for your replying.
> >
inode
will be 256, or more bytes.
* If on disk inode size is 128 bytes, only first 128 bytes of struct
ext4_inode take effects.
Best regards.
Coly
在 2007-04-08日的 13:52 -0400，Theodore Tso写道：
> On Sun, Apr 08, 2007 at 02:39:03PM +0800, coly wrote:
> > Hi, list:
> >
> >
inode table (with
152 bytes inode size), I can not read proper data from the bh->b_data.
Once I use 128 bytes inode size, I can read what I want from the
bh->b_data.
I believe there is something I missed, can anybody give me some points ?
Thanks in advance.
Coly
-
To unsubscribe from thi
Hi, list,
I am working on the directory inode reservation feature now. Here is the
detailed description of my understand of the designing, and current
implementations.
Please give me your comments on this idea. Thanks for your help in
advance.
Best regards.
Coly Li
gic inode can ignore and remove the magic inodes. This
can only happen when fsck is performced and filesystem code can rebuild
the magic inode if it can not be found (this will take a some time for
reading inode table when mount).
Best regards.
Coly
P.S here is the result of my benchmark:
I crea
Ted:
Thank you ! It is clear to me :-)
Coly
在 2007-01-27六的 16:53 -0500，Theodore Tso写道：
> On Sat, Jan 27, 2007 at 09:26:27PM +0800, coly wrote:
> >
> > It seems the patch file is based on stable kernel tree, not -mm tree.
> > Isn't it ?
>
> Yes, it's
Theodore:
It seems the patch file is based on stable kernel tree, not -mm tree.
Isn't it ?
Thanks.
Coly
在 2007-01-26五的 11:51 -0500，Theodore Tso写道：
> On Fri, Jan 26, 2007 at 07:56:59AM -0800, Mingming Cao wrote:
> > >
> > >I can not find the ext4 git tree (on kernel.
Cao:
Thanks for your reply. I have download -mm tree and patched it into
2.6.20-rc4. I can learn the source code fist :-)
Best regards.
Coly
在 2007-01-26五的 07:56 -0800，Mingming Cao写道：
> coly wrote:
>
> > Hi friends:
> >
> > I can not find the ext4 git tree (on kernel
Hi friends:
I can not find the ext4 git tree (on kernel.org) which Te mentioned on
the talk. Can anybody give me a point ? I am new to ext4.
Coly
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to [EMAIL PROTECTED]
More majordomo inf
Hi, team:
This is a very cool feature. Is there any planing or schedule on it ?
Coly
在 2007-01-25四的 18:28 +0100，Vitez Gabor写道：
> Hi,
>
> It would be really great if ext4 supported an xfs_freeze like operation.
> On xfs it spared me a lot of headaches when I was playing with unstab
70 matches
Mail list logo
