similar to: [patch]use strnlen to deal with string from user space

I found userspace tool, btrfsctl, uses BTRFS_VOL_NAME_MAX, and it also looks that this one is more proper. Kill BTRFS_PATH_NAME_MAX since no one will use it. Signed-off-by: WANG Cong <wangcong@zeuux.org> Cc: Chris Mason <chris.mason@oracle.com> --- diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index c2aa33e..f229950 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -472,7

The ''char name[BTRFS_PATH_NAME_MAX]'' member of struct btrfs_ioctl_vol_args is passed directly to strlen() after being copied from user. I haven''t verified this, but in theory a userspace program could pass in an unterminated string and cause a kernel crash as strlen walks off the end of the array. This patch terminates the ->name string in all btrfs ioctl functions

From: Ajesh JS <ajesh.js@hp.com> This patch provides a mechanism to modify the permission of a subvolume in a btrfs file system This helps to apply a policy of having read-write subvolume inside a read-only subvolume. One use case is to have the whole root file system as a read-only subvolume and have /etc as a read-write subvolume inside. Signed-off-by: Ajesh JS <ajesh.js@hp.com>

Hi everyone, This patchset is the latest approach I''m using for the Ceph storage daemon to keep track of which data has safely committed to disk. The basic idea is to not use the (problematic) user transaction ioctls at all. Instead, the daemon quiesces its own write requests, initiates an async snapshot, and then continues. The snapshot approach is nice because it provides rollback.

Hello, This patch adds rollback support for the converter, the converter can roll back a conversion if the image file haven't been modified. In addition, I rearrange some codes in convert.c and add a few comments. Regards YZ --- diff -r 12138d4beeb0 convert.c --- a/convert.c Fri Jan 04 11:29:55 2008 -0500 +++ b/convert.c Mon Jan 07 23:35:25 2008 +0800 @@ -33,6 +33,7 @@ #include

hello, Chris In the next several months, I will have a lot of spare time. I will be happy if you assign some work to me. I have learn linux kernel for years but only fix a few small bugs in IP stack. In the past few months, I read some EXT4/VFS codes. I began reading source of btrfs ten days ago and have read it twice now. Regards YZ

Hi Chris, This is the extent of my current queue of Btrfs snapshot/subvol/commit stuff. Most of these were posted several months ago. Can be sent upstream during this merge window? Not having this functionality is becoming a bit of a roadblock for our efforts to keep the Ceph data in a consistent state. These patches are also available from

Joerg Sonnenberger wrote: > On Sat, Feb 06, 2016 at 11:05:14PM -0500, Michael McConville via llvm-dev wrote: > > This addition converts strlen() calls to strnlen() when the result is > > compared to a constant. For example, the following: > > > > strlen(s) < 5 > > > > Becomes: > > > > strnlen(s, 5) < 5 > > > > That way, we

Hi, I recompiled my kernel today, and when the btrfs-0.13 option is enabled (CONFIG_BTR_FS=y) I cannot boot on my reiserfs-3.6 partition any more. The kernel says that is cannot mount /dev/sdc6. I can boot the kernel when I recompile it with the same options and CONFIG_BTR_FS=n, and with the same boot options as well. Everything is amd64 on my system (kernel + programs). Here is a screenshot of

Add a mount option user_subvol_rm_allowed that allows users to delete a (potentially non-empty!) subvol when they would otherwise we allowed to do an rmdir(2). We duplicate the may_delete() checks from the core VFS code to implement identical security checks (minus the directory size check). Signed-off-by: Sage Weil <sage@newdream.net> --- fs/btrfs/ctree.h | 1 + fs/btrfs/ioctl.c |

This addition converts strlen() calls to strnlen() when the result is compared to a constant. For example, the following: strlen(s) < 5 Becomes: strnlen(s, 5) < 5 That way, we don't have to walk through the entire string. There is the added overhead of maintaining a counter when using strnlen(), but I thought I'd start with the general case. It may make sense to only use this

I was playing around with btrfs, and did a mkfs.btrfs on one of my partition. Mistakes, it was my data partition (ext3 based). I have not yet mounted the device, but the mkfs.btrfs continued to completion. Is it possible to salvage my ext3 filesystem? Anyway to reverse the effect of mkfs.btrfs? Thank you for your help/answer. -- Regards, Peter Teoh

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I mentioned a ghost file. Here is the dump (less than 65M): http://dev.mccme.ru/~raskin/btrfs.dump ghost file is: 885dj1l4788pymp3bk2f3dz52ggcvw4v-empty/garbage-1/mwave.h.tmp-31838-1822528541 It has a name, but stat fails (so no inode, I guess). Also I can create another file with the same name in the directory. I cannot remove the file, or

Hi all. I'm working with a short plot (3x3 inches), but the results (via postscript command) are not nice. The lwd command don't affect the lines (that are very large) and the margins don't change using oma, mai, mar, ... Below I put an example. Moreover, save the graphics via postscript command isn't working well (see the attached ps). Thanks by the help, Cezar Freitas.

Dear R People: Hope you're having a nice day. Here is a character vector: > yz [1] "pexp(3.2,rate=1)" > str(yz) chr "pexp(3.2,rate=1)" > And I would like to evaluate that vector. I tried: > eval(as.expression(yz)) [1] "pexp(3.2,rate=1)" > But that doesn't work. Any suggestions would be most welcome. I have a feeling that it's quite

Hi In ext3_get_parent(), quick string (struct qstr) can do the job, in place of declaring a dentry on stack. Following patch does this and saves few bytes on kernel stack. Thanks, Maneesh namei.c | 69 ++++++++++++++++++++++++++++++---------------------------------- 1 files changed, 33 insertions(+), 36 deletions(-) diff -urN linux-2.5.58-base/fs/ext3/namei.c

Hello, in the BSD Authentication system the login script can request environment variables to be set/unset. The call to auth_close() in auth-passwd.c does change the current environment, but those changes are lost for the child environment. It would be really useful to add some kind of mechanism to get those changes into the child environment. I've added two possible solutions. Both

FYI >Mailing-List: contact vulnwatch-help@vulnwatch.org; run by ezmlm >List-Post: <mailto:vulnwatch@vulnwatch.org> >List-Help: <mailto:vulnwatch-help@vulnwatch.org> >List-Unsubscribe: <mailto:vulnwatch-unsubscribe@vulnwatch.org> >List-Subscribe: <mailto:vulnwatch-subscribe@vulnwatch.org> >Delivered-To: mailing list vulnwatch@vulnwatch.org

On 31 January 2018 at 21:34, Freer, Eva B. <freereb at ornl.gov> wrote: > Nithya, > > > > Responding to an earlier question: Before the upgrade, we were at 3.103 on > these servers, but some of the clients were 3.7.6. From below, does this > mean that ?shared-brick-count? needs to be set to 1 for all bricks. > > > > All of the bricks are on separate xfs

On 31 January 2018 at 21:50, Freer, Eva B. <freereb at ornl.gov> wrote: > The values for shared-brick-count are still the same. I did not re-start > the volume after setting the cluster.min-free-inodes to 6%. Do I need to > restart it? > > > That is not necessary. Let me get back to you on this tomorrow. Regards, Nithya > Thanks, > > Eva (865) 574-6894