similar to: One more patch change

Jan, since i can not cc you mail address when filing a bug on oss bugzilla,just paste the bug description here. Testing env, Kernel version:latest linus's mainline 2.6.27 Testing Nodes:test5/6 Description: Jan,i used your provided quota tools,now it works for me:),currently,i'm doing a sanity check for the quota on ocfs2. Found a strange thing when comparing with the ext3,the quota

Hi, I'm sending a patch for proper quota support when disabling sparse feature. The second patch fixes a minor problem in tunefs.ocfs2 when disabling the sparse feature. In a few days I plan to resend the whole "quota support" series with all the changes people request included... Honza

On 5/23/23 5:33 PM, Lu?s Henriques wrote: > It's trivial to trigger a use-after-free bug in the ocfs2 quotas code using > fstest generic/452. After a read-only remount, quotas are suspended and > ocfs2_mem_dqinfo is freed through ->ocfs2_local_free_info(). When unmounting > the filesystem, an UAF access to the oinfo will eventually cause a crash. > > BUG: KASAN:

Hi, On our 2-nodes cluster (ubuntu server 12.04 LTS, 64bits 3.2.0-24-generic kernel, ocfs2 1.6.3-4ubuntu1 package), things seem to run nicely. But /var/log/kern.log on both nodes is constantly showing the errors below : Jun 25 10:46:46 serv-fs-adm1 kernel: [ 4910.713981] __quota_error: 2 callbacks suppressed Jun 25 10:46:46 serv-fs-adm1 kernel: [ 4910.713986] Quota error (device dm-6):

Since these include files are used only by implementation of quota formats, there's no need to have them in include/linux/. Signed-off-by: Jan Kara <jack at suse.cz> --- fs/quota_v1.c | 3 +- fs/quota_v2.c | 7 ++-- fs/quotaio_v1.h | 33 ++++++++++++++++++ fs/quotaio_v2.h | 79 ++++++++++++++++++++++++++++++++++++++++++++

Hi, the first four patches in this series fix locking problems in OCFS2 quota code (three of them can lead to potential deadlocks). The fifth patch reorders ip_alloc_sem for directories to be acquired before localalloc locks. Mark would you please merge these? The last two patches implement lockdep annotations for OCFS2 cluster locks. We annotate all the cluster locks except for special ones

hi Jan, I've attached Yu-Chan's quota.h -- does this look familiar to you? It is clearly not the quota.h from the XFS CVS as I had originally thought. It may be the one from the XFS 1.1 CDs, which is a Redhat-based, XFS-enabled distro, so would have an earlier version of your VFS quota patches (via -ac kernel). Either way, it doesn't really matter. We need to fix Samba, and a

Hi, All, When mount ocfs2 with option usrquota and grpquota, we will see below message: (mount.ocfs2,9434,1):ocfs2_global_read_info:403 ERROR: status = 24 (mount.ocfs2,9434,3):ocfs2_global_read_info:403 ERROR: status = 24 But actually this should not be a error. In ocfs2_global_read_info, if read_quota() succeed, status is the size of the struct ocfs2_global_disk_dqinfo, only when it is not

Hello, the following series of patches fixes some issues with OCFS2 quotas. The first patch modifies VFS quota locking, the next patch uses the fact to simplify OCFS2 quota locking and solves a few deadlock issues. The third and the fourth patches fix another possible deadlocks in OCFS2 quota code and the last patch is a minor cleanup. Honza

Hello Sunil, attached is a patch which implements full quota support for ocfs2 tools. It's based against ocfs2-tools git's master branch. Would you please consider inclusion of the patch? Thanks. Honza -- Jan Kara <jack at suse.cz> SUSE Labs, CR -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Quota-support-for-ocfs2-tools.patch

Hi, I'm sending a series of patches implementing quota support into ocfs2-tools. It's the same as the original huge patch I've sent but now it's split as Joel asked. I've also realized that when disabling SPARSE feature, we should update quota information. That piece of code is missing, I'll implement it soon. Comments welcome. Honza

Currently, default behavior of O_DIRECT writes was allowing concurrent writing among nodes, no cluster coherency guaranteed (no EX locks was taken), it hurts buffered reads on other nodes by reading stale data from cache. The new mount option introduce a chance to choose two different behaviors for O_DIRECT writes: * coherency=full, as the default value, will disallow concurrent

Hi, I'm resending this patch series. It's rediffed against linux-next branch of Joel's git tree. The first four patches are obvious fixes of deadlocks in quota code and should go in as soon as possible. The other three patches implement lockdep support for OCFS2 cluster locks. So you can have a look whether the code make sence to you and possibly merge them. They should be NOP when

Hello, the following patch series implements quotas for OCFS2. The patch series is based on: git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2.git linux-next I've adressed Joel's comments, also node recovery is now fully working and I've fixed a few issues I found during my testing. So I'm currently not aware of any bugs. Please review, test, comment. Thanks.

Hello, attached is a patch which implements necessary support for mkfs so that quota files can be created. Honza -- Jan Kara <jack at suse.cz> SUSE Labs, CR -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Implement-quota-support-to-mkfs-and-add-a-few-auxili.patch Type: text/x-patch Size: 16966 bytes Desc: not available Url :

Dear Robert, Before sending 'grep -r /home /etc' data, I want tell you what happned this morning. In order to solve the /home/home problem, 'umount /home' had been done, system had been running in a normal file system. But suddenly /home has been lost. Key information of that time is as in the **** lines. And I found in the //// lines messages log. How do you think the reason

On Sun 18-06-23 23:32:49, Bean Huo wrote: > From: Bean Huo <beanhuo at micron.com> > > Remove unnecessary check on the return value of block_commit_write(). > > Signed-off-by: Bean Huo <beanhuo at micron.com> Looks good to me. Feel free to add: Reviewed-by: Jan Kara <jack at suse.cz> Honza > --- > fs/ocfs2/file.c | 7 +------ > 1 file

Dear Robert, Thank you. The state after reboot are as follows. # LANG=C lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT sr0 11:0 1 1024M 0 rom vda 252:0 0 15G 0 disk |-vda1 252:1 0 500M 0 part /boot `-vda2 252:2 0 14.5G 0 part |-VolGroup-lv_root (dm-0) 253:0 0

Hi, I've been playing lately with lockdep annotations of OCFS2. I seem to have most of the false positives sorted out and currently I hit the report below. I've analyzed that ocfs2_extend_dir() does first lock local alloc inode in ocfs2_reserve_clusters() and then acquires ip_alloc_sem from the directory. The usual ordering e.g. in ocfs2_write_begin() is to first acquire ip_alloc_sem

Hi, I've looked at how OCFS2 call jbd2_journal_begin_ordered_truncate() (because I've been adding some comments about how is should be used) and noticed that OCFS2 has a potential race in truncate vs transaction commit leading to stale data in file. In particular: There is a race if someone writes new data and we start committing the transaction after