hey everybody.
1. Deduplication
"When given a list of files it will hash their contents on a block by
block basis" - are those static blocks or is the length of a block
defined by its content? (that would be more resilient regarding inserts
of data and the shift of the following data caused by
On 09/25/2015 06:31 PM, David Sterba wrote:
On Fri, Sep 25, 2015 at 02:43:01PM +0800, Anand Jain wrote:
btrfs_error() and btrfs_std_error() does the same thing
and calls _btrfs_std_error(), so consolidate them together.
And the main motivation is that btrfs_error() is closely
named with btrfs_
On 10/02/2015 12:17 AM, David Sterba wrote:
On Thu, Oct 01, 2015 at 11:00:39PM +0800, Anand Jain wrote:
Its been challenging to realign the patch sequence at my end as well,
I am trying. I think it will take a bit longer than expected.
I'd suggest to send several small patchsets instead o
在 2015年10月02日 15:41, Anand Jain 写道:
On 09/25/2015 06:31 PM, David Sterba wrote:
On Fri, Sep 25, 2015 at 02:43:01PM +0800, Anand Jain wrote:
btrfs_error() and btrfs_std_error() does the same thing
and calls _btrfs_std_error(), so consolidate them together.
And the main motivation is that btr
Hi Dave, I updated the patch and moved it to btrfs.
But I still has some question about the fallocate behavior.
Just as the new btrfs test case, I changed the fallocate range, not to
cover the last part, to make the problem more obvious:
Btrfs will truncate beyond EOF even that's *not covered
I'm not sure if it was my last email with attachement delivered.
My last email is below. Attachement is uploaded here:
http://franta.naobzoru.cz/files/btrfs-failed.txt
Thank you.
Frantisek
2015-09-27 11:51 GMT+02:00 Vackář František :
> So I copy with dd all my 1.7 TB partition with broken btrf
On Fri, Oct 02, 2015 at 09:06:17AM +0200, - wrote:
> hey everybody.
>
> 1. Deduplication
> "When given a list of files it will hash their contents on a block
> by block basis" - are those static blocks or is the length of a
> block defined by its content? (that would be more resilient
> regarding
It only needs to log an error, no need to mark FS_ERROR bit.
As it may just be an invalid parameter.
This ans-ed my question. Then there are two bugs. one which David
pointed out. two In original code (before this patch), it should
have been calling btrfs_err() instead of btrfs_error().
T
So... I find a solution! :)
I tried to play with mount options. Finally I sucessfully mounted
corrupted btrfs with
mount -t btrfs -oro,degraded,recovery /dev/sdb2 /mnt/hdd-root/
Thank you for help. Now I copy all data to external drive and make new
filestystem.
Frantisek
2015-10-02 11:05 GM
From: Filipe Manana
If we have a file that shares an extent with other files, when processing
the extent item relative to a shared extent, we blindly issue a clone
operation that will target a length matching the length in the extent item
and uses as a source some other file the receiver already
From: Filipe Manana
Test that a send operation works correctly with reflinked files (cloned
extents which multiple files point to).
The btrfs issue was fixed by the linux kernel patch titled:
"Btrfs: send, fix file corruption due to incorrect cloning operations"
Signed-off-by: Filipe Manana
--
This patch will log return value of add/del_qgroup_relation()
and pass the err code of btrfs_run_qgroups to the btrfs_std_error().
Signed-off-by: Anand Jain
---
fs/btrfs/ioctl.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index f7
> btrfs_std_error() naming is quite confusing for me.
>
> Std_error() means more like stderr, for my first glance, I'd think
> it's just a new printk() warpper, until I checked the code.
> It does more than printk, but also set FS_STATE_ERROR bit and set fs
> to readonly.
> I'd like it to be so
This patch will log return value of add/del_qgroup_relation()
and pass the err code of btrfs_run_qgroups to the btrfs_std_error().
Signed-off-by: Anand Jain
---
v2: fix the forgotten git commit amend, to take in compile fail, sorry
fs/btrfs/ioctl.c | 7 ++-
1 file changed, 6 insertions(+),
On 2015-09-29 23:50, Omar Sandoval wrote:
Hi,
Here's one more reroll of the free space B-tree patches, a more scalable
alternative to the free space cache. Minimal changes this time around, I
mainly wanted to resend this after Holger and I cleared up his bug
report here: http://www.spinics.net/l
On 2015-10-02 00:21, Russell Coker wrote:
On Sat, 26 Sep 2015 12:20:41 AM Austin S Hemmelgarn wrote:
FYI:
Linux pagecache use LRU cache algo, and in general case it's working good
enough
I'd argue that 'general usage' should be better defined in this
statement. Obviously, ZFS's ARC implementa
On Thursday 01 Oct 2015 10:39:29 Josef Bacik wrote:
> On 09/30/2015 06:28 AM, Chandan Rajendra wrote:
> > Checksums are applicable to sectorsize units. The current code uses
> > bio->bv_len units to compute and look up checksums. This works on machines
> > where sectorsize == PAGE_SIZE. This patch
On Thursday 01 Oct 2015 10:50:30 Josef Bacik wrote:
> On 09/30/2015 06:28 AM, Chandan Rajendra wrote:
> > When reading the page from the disk, we can race with Direct I/O which can
> > get the page lock (before prepare_uptodate_page() gets it) and can go
> > ahead and invalidate the page. Hence if
Hi Omar,
[auto build test results on v4.3-rc3 -- if it's inappropriate base, please
ignore]
config: mips-ip27_defconfig (attached as .config)
reproduce:
wget
https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross
-O ~/bin/make.cross
chmod +x ~/bin/
load_device_info queries the FS_INFO ioctl and this may fail with EPERM
on older kernels. The check did not verify the ioctl return value and
incorrectly returned EPERM if it was previously stored in errno.
This fixes 'btrfs fi usage' that will print the overall summary for all
users (provided tha
If there are different devices mounted to the same directory we can run
into double free issue in the scanning code and this can lead to a
crash. The dev_info_arg buffer allocation get_fs_info might be skipped,
eg. if the FS_INFO ioctl fails due to EPERM in older kernels. Reset the
pointer before e
On Thursday 01 Oct 2015 10:57:52 Josef Bacik wrote:
> On 09/30/2015 06:28 AM, Chandan Rajendra wrote:
> > When extending a file by either "truncate up" or by writing beyond i_size,
> > the page which had i_size needs to be marked "read only" so that future
> > writes to the page via mmap interface
Old implementation used tabs "\t", and tried to work around problems
by guessing amount of tabs needed (e.g. "\t\t" after top level", with
buggy output as soon as empty uuids are printed). This will never work
correctly, as tab width is a user-defined setting in the terminal.
Keep it simple and do
Signed-off-by: Axel Burri
---
btrfs-list.c | 14 ++
btrfs-list.h | 1 +
2 files changed, 15 insertions(+)
diff --git a/btrfs-list.c b/btrfs-list.c
index 7529e11..ff337f9 100644
--- a/btrfs-list.c
+++ b/btrfs-list.c
@@ -80,6 +80,11 @@ static struct {
.need_print =
Affects time format of "otime". Supporting unix time and ISO8601 makes
the output of "subvolume list" machine-readable.
Default is "short", keeping default output as it was before (I suggest
changing the default to "iso", as with "short" is it not clear to the
user if localtime or gmtime is printe
Improvements to "subvolume list" command, with the goal to improve
machine-readability. While the first three patches simply add new
options (-A and --time-format), the last patch [PATCH 4/4] is more
intrusive and changes the output format of the '-t' option. Consider
it "my view of how tabular out
Signed-off-by: Axel Burri
---
Documentation/btrfs-subvolume.asciidoc | 2 ++
cmds-subvolume.c | 6 +-
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/Documentation/btrfs-subvolume.asciidoc
b/Documentation/btrfs-subvolume.asciidoc
index c187fd8..afbec83 100
From: Filipe Manana
Josef ran into a deadlock while a transaction handle was finalizing the
creation of its block groups, which produced the following trace:
[260445.593112] fio D 88022a9df468 0 8924 4518
0x0084
[260445.593119] 88022a9df468 81c134c0 f
On 10/02/2015 01:43 PM, fdman...@kernel.org wrote:
From: Filipe Manana
Josef ran into a deadlock while a transaction handle was finalizing the
creation of its block groups, which produced the following trace:
[260445.593112] fio D 88022a9df468 0 8924 4518
0x0084
Hello,
Is there a way (and appropriate user space tool) to force compression of
existing file? I mean there was directory with some content and then I
set compression attribute (property) to it. How to force data recompression?
--
To unsubscribe from this list: send the line "unsubscribe linux-bt
On Fri, Oct 02, 2015 at 11:41:00PM +0300, Matwey V. Kornilov wrote:
> Hello,
>
> Is there a way (and appropriate user space tool) to force compression of
> existing file? I mean there was directory with some content and then I
> set compression attribute (property) to it. How to force data recompr
Hi
I am interested in combining BtrFS RAID-5 with erasure-coded replication
for HDFS. We have an implementation of Reed-Solomon replication for our
HDFS distribution called HopsFS (www.hops.io).
Some of the nice features of HDFS that make it suitable are:
* not many small files
* not excessive
2015-10-03 0:07 GMT+02:00 Jim Dowling :
> Hi
Hi, I'm not a btrfs developer but we run HDFS on top of btrfs (mainly
due to other use-cases that profit from checksumming data)
> I am interested in combining BtrFS RAID-5 with erasure-coded replication for
> HDFS. We have an implementation of Reed-So
tree: https://git.kernel.org/pub/scm/linux/kernel/git/josef/btrfs-next.git
allocator-fixes
head: 5be061b1329c6f3b1926496d44560f924c582deb
commit: af8df18b295f2e6ff26d27386a5183c545044e48 [14/15] Btrfs: keep track of
largest extent in bitmaps
config: i386-randconfig-n0-201539 (attached as .con
34 matches
Mail list logo
