[Devel] linux-next: lockdep whinge in cgroup_rmdir
Seen booting yesterday's linux-next, was not present in 2.6.37-rc7-mmotm1202. Not sure if it's an selinux or cgroup issue, so I'm throwing it at every address I can find for either. This is easily replicatable and happens at every boot, so I can test patches if needed. Am willing to bisect it down if nobody knows right off the bat what the problem is. The 'W' taint is from the already-reported kernel/workqueue.c worker_enter_idle issue. [ 85.100795] systemd[1]: readahead-replay.service: main process exited, code=exited, status=1 [ 85.101530] [ 85.101531] = [ 85.101796] [ INFO: possible recursive locking detected ] [ 85.102002] 2.6.37-next-20110111 #1 [ 85.102009] - [ 85.102009] systemd/1 is trying to acquire lock: [ 85.102009] ((dentry-d_lock)-rlock){+.+...}, at: [8107ca5c] cgroup_rmdir+0x339/0x479 [ 85.102009] [ 85.102009] but task is already holding lock: [ 85.102009] ((dentry-d_lock)-rlock){+.+...}, at: [8107ca54] cgroup_rmdir+0x331/0x479 [ 85.102009] [ 85.102009] other info that might help us debug this: [ 85.102009] 4 locks held by systemd/1: [ 85.102009] #0: (sb-s_type-i_mutex_key#14/1){+.+.+.}, at: [810fea4d] do_rmdir+0x7d/0x121 [ 85.102009] #1: (sb-s_type-i_mutex_key#14){+.+.+.}, at: [810fd4bc] vfs_rmdir+0x4a/0xbe [ 85.102009] #2: (cgroup_mutex){+.+.+.}, at: [8107cb84] cgroup_rmdir+0x461/0x479 [ 85.102009] #3: ((dentry-d_lock)-rlock){+.+...}, at: [8107ca54] cgroup_rmdir+0x331/0x479 [ 85.102009] [ 85.102009] stack backtrace: [ 85.102009] Pid: 1, comm: systemd Tainted: GW 2.6.37-next-20110111 #1 [ 85.102009] Call Trace: [ 85.102009] [81069f22] ? __lock_acquire+0x929/0xd4e [ 85.102009] [8107c6f1] ? cgroup_clear_directory+0xff/0x131 [ 85.102009] [8107c6f1] ? cgroup_clear_directory+0xff/0x131 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [8106a859] ? lock_acquire+0x100/0x126 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [815521ef] ? sub_preempt_count+0x35/0x48 [ 85.102009] [8154e401] ? _raw_spin_lock+0x36/0x45 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [810579cd] ? autoremove_wake_function+0x0/0x34 [ 85.102009] [811e1839] ? selinux_inode_rmdir+0x15/0x17 [ 85.102009] [810fd4eb] ? vfs_rmdir+0x79/0xbe [ 85.102009] [810feaa0] ? do_rmdir+0xd0/0x121 [ 85.102009] [8100256c] ? sysret_check+0x27/0x62 [ 85.102009] [8106ac79] ? trace_hardirqs_on_caller+0x117/0x13b [ 85.102009] [8154e201] ? trace_hardirqs_on_thunk+0x3a/0x3f [ 85.102009] [8110040b] ? sys_rmdir+0x11/0x13 [ 85.102009] [8100253b] ? system_call_fastpath+0x16/0x1b [ 85.268272] systemd[1]: readahead-collect.service: main process exited, code=exited, status=1 Any ideas? pgpzYhMMjNRuO.pgp Description: PGP signature ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] [RFD] reboot / shutdown of a container
Hi all, in the container implementation, we are facing the problem of a process calling the sys_reboot syscall which of course makes the host to poweroff/reboot. If we drop the cap_sys_reboot capability, sys_reboot fails and the container reach a shutdown state but the init process stay there, hence the container becomes stuck waiting indefinitely the process '1' to exit. The current implementation to make the shutdown / reboot of the container to work is we watch, from a process outside of the container, the rootfs/var/run/utmp file and check the runlevel each time the file changes. When the 'reboot' or 'shutdown' level is detected, we wait for a single remaining in the container and then we kill it. That works but this is not efficient in case of a large number of containers as we will have to watch a lot of utmp files. In addition, the /var/run directory must *not* mounted as tmpfs in the distro. Unfortunately, it is the default setup on most of the distros and tends to generalize. That implies, the rootfs init's scripts must be modified for the container when we put in place its rootfs and as /var/run is supposed to be a tmpfs, most of the applications do not cleanup the directory, so we need to add extra services to wipeout the files. More problems arise when we do an upgrade of the distro inside the container, because all the setup we made at creation time will be lost. The upgrade overwrite the scripts, the fstab and so on. We did what was possible to solve the problem from userspace but we reach always a limit because there are different implementations of the 'init' process and the init's scripts differ from a distro to another and the same with the versions. We think this problem can only be solved from the kernel. The idea was to send a signal SIGPWR to the parent of the pid '1' of the pid namespace when the sys_reboot is called. Of course that won't occur for the init pid namespace. Does it make sense ? Any idea is very welcome :) -- Daniel ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] Re: [PATCH] Teach cifs about network namespaces (take 3)
On Thu, 13 Jan 2011 12:55:04 -0600 Rob Landley rland...@parallels.com wrote: From: Rob Landley rland...@parallels.com Teach cifs about network namespaces, so mounting uses adresses/routing visible from the container rather than from init context. Signed-off-by: Rob Landley rland...@parallels.com --- Now using net_eq(), with the initialization moved up so the error path doesn't dereference a null on the put. fs/cifs/cifsglob.h | 33 + fs/cifs/connect.c | 12 ++-- 2 files changed, 43 insertions(+), 2 deletions(-) Looks good to me: Reviewed-by: Jeff Layton jlay...@redhat.com ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] Re: [RFD] reboot / shutdown of a container
On 01/13/2011 10:50 PM, Bruno Prémont wrote: On Thu, 13 January 2011 Daniel Lezcanodaniel.lezc...@free.fr wrote: On 01/13/2011 09:09 PM, Bruno Prémont wrote: On Thu, 13 January 2011 Daniel Lezcanodaniel.lezc...@free.fr wrote: in the container implementation, we are facing the problem of a process calling the sys_reboot syscall which of course makes the host to poweroff/reboot. If we drop the cap_sys_reboot capability, sys_reboot fails and the container reach a shutdown state but the init process stay there, hence the container becomes stuck waiting indefinitely the process '1' to exit. The current implementation to make the shutdown / reboot of the container to work is we watch, from a process outside of the container, therootfs/var/run/utmp file and check the runlevel each time the file changes. When the 'reboot' or 'shutdown' level is detected, we wait for a single remaining in the container and then we kill it. That works but this is not efficient in case of a large number of containers as we will have to watch a lot of utmp files. In addition, the /var/run directory must *not* mounted as tmpfs in the distro. Unfortunately, it is the default setup on most of the distros and tends to generalize. That implies, the rootfs init's scripts must be modified for the container when we put in place its rootfs and as /var/run is supposed to be a tmpfs, most of the applications do not cleanup the directory, so we need to add extra services to wipeout the files. More problems arise when we do an upgrade of the distro inside the container, because all the setup we made at creation time will be lost. The upgrade overwrite the scripts, the fstab and so on. We did what was possible to solve the problem from userspace but we reach always a limit because there are different implementations of the 'init' process and the init's scripts differ from a distro to another and the same with the versions. We think this problem can only be solved from the kernel. The idea was to send a signal SIGPWR to the parent of the pid '1' of the pid namespace when the sys_reboot is called. Of course that won't occur for the init pid namespace. Wouldn't sending SIGKILL to the pid '1' process of the originating PID namespace be sufficient (that would trigger a SIGCHLD for the parent process in the outer PID namespace. This is already the case. The question is : when do we send this signal ? We have to wait for the container system shutdown before killing it. I meant that sys_reboot() would kill the namespace's init if it's not called from boot namespace. See below (as far as I remember the PID namespace is killed when its 'init' exits, if this is not the case all other processes in the given namespace would have to be killed as well) Yes, absolutely but this is not the point, reaping the container is not a problem. What we are trying to achieve is to shutdown properly the container from inside (from outside will be possible too with the setns syscall). Assuming the process '1234' creates a new process in a new namespace set and wait for it. The new process '1' will exec /sbin/init and the system will boot up. But, when the system is shutdown or rebooted, after the down scripts are executed the kill -15 -1 will be invoked, killing all the processes expect the process '1' and the caller. This one will then call 'sys_reboot' and exit. Hence we still have the init process idle and its parent '1234' waiting for it to die. This call to sys_reboot() would kill new process '1' instead of trying to operate on the HW box. This also has the advantage that a container would not require an informed parent monitoring it from outside (though it would not be restarted even if requested without such informed outside parent). Oh, ok. Sorry I misunderstood. Yes, that could be better than crossing the namespace boundaries. If we are able to receive the information in the process '1234' : the sys_reboot was called in the child pid namespace, we can take then kill our child pid. If this information is raised via a signal sent by the kernel with the proper information in the siginfo_t (eg. si_code contains LINUX_REBOOT_CMD_RESTART, LINUX_REBOOT_CMD_HALT, ... ), the solution will be generic for all the shutdown/reboot of any kind of container and init version. Could this be passed for a SIGCHLD? (when namespace is reaped, and received by 1234 from above example assuming sys_reboot() kills the new process '1') Yes, that sounds a good idea. Looks like yes, but with the need to define new values for si_code (reusing LINUX_REBOOT_CMD_* would certainly clash, no matter which signal is choosen). CLD_REBOOT_CMD_RESTART CLD_REBOOT_CMD_HALT CLD_REBOOT_CMD_POWER_OFF CLD_REBOOT_CMD_RESTART2 (what about the cmd buffer, shall we ignore it ?) CLD_REBOOT_CMD_KEXEC (?) CLD_REBOOT_CMD_SW_SUSPEND (useful for the future checkpoint/restart) LINUX_REBOOT_CMD_CAD_ON and LINUX_REBOOT_CMD_CAD_OFF
[Devel] Re: [RFD] reboot / shutdown of a container
On 01/13/2011 09:09 PM, Bruno Prémont wrote: On Thu, 13 January 2011 Daniel Lezcanodaniel.lezc...@free.fr wrote: in the container implementation, we are facing the problem of a process calling the sys_reboot syscall which of course makes the host to poweroff/reboot. If we drop the cap_sys_reboot capability, sys_reboot fails and the container reach a shutdown state but the init process stay there, hence the container becomes stuck waiting indefinitely the process '1' to exit. The current implementation to make the shutdown / reboot of the container to work is we watch, from a process outside of the container, therootfs/var/run/utmp file and check the runlevel each time the file changes. When the 'reboot' or 'shutdown' level is detected, we wait for a single remaining in the container and then we kill it. That works but this is not efficient in case of a large number of containers as we will have to watch a lot of utmp files. In addition, the /var/run directory must *not* mounted as tmpfs in the distro. Unfortunately, it is the default setup on most of the distros and tends to generalize. That implies, the rootfs init's scripts must be modified for the container when we put in place its rootfs and as /var/run is supposed to be a tmpfs, most of the applications do not cleanup the directory, so we need to add extra services to wipeout the files. More problems arise when we do an upgrade of the distro inside the container, because all the setup we made at creation time will be lost. The upgrade overwrite the scripts, the fstab and so on. We did what was possible to solve the problem from userspace but we reach always a limit because there are different implementations of the 'init' process and the init's scripts differ from a distro to another and the same with the versions. We think this problem can only be solved from the kernel. The idea was to send a signal SIGPWR to the parent of the pid '1' of the pid namespace when the sys_reboot is called. Of course that won't occur for the init pid namespace. Wouldn't sending SIGKILL to the pid '1' process of the originating PID namespace be sufficient (that would trigger a SIGCHLD for the parent process in the outer PID namespace. This is already the case. The question is : when do we send this signal ? We have to wait for the container system shutdown before killing it. (as far as I remember the PID namespace is killed when its 'init' exits, if this is not the case all other processes in the given namespace would have to be killed as well) Yes, absolutely but this is not the point, reaping the container is not a problem. What we are trying to achieve is to shutdown properly the container from inside (from outside will be possible too with the setns syscall). Assuming the process '1234' creates a new process in a new namespace set and wait for it. The new process '1' will exec /sbin/init and the system will boot up. But, when the system is shutdown or rebooted, after the down scripts are executed the kill -15 -1 will be invoked, killing all the processes expect the process '1' and the caller. This one will then call 'sys_reboot' and exit. Hence we still have the init process idle and its parent '1234' waiting for it to die. If we are able to receive the information in the process '1234' : the sys_reboot was called in the child pid namespace, we can take then kill our child pid. If this information is raised via a signal sent by the kernel with the proper information in the siginfo_t (eg. si_code contains LINUX_REBOOT_CMD_RESTART, LINUX_REBOOT_CMD_HALT, ... ), the solution will be generic for all the shutdown/reboot of any kind of container and init version. Only issue is how to differentiate the various reboot() modes (restart, power-off/halt) from outside, though that one also exists with the SIGPWR signal. javascript:void(0); ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] Re: linux-next: lockdep whinge in cgroup_rmdir
Nick Piggin wrote: On Fri, Jan 14, 2011 at 2:34 AM, valdis.kletni...@vt.edu wrote: Seen booting yesterday's linux-next, was not present in 2.6.37-rc7-mmotm1202. Not sure if it's an selinux or cgroup issue, so I'm throwing it at every address I can find for either. This is easily replicatable and happens at every boot, so I can test patches if needed. Am willing to bisect it down if nobody knows right off the bat what the problem is. The 'W' taint is from the already-reported kernel/workqueue.c worker_enter_idle issue. [ 85.100795] systemd[1]: readahead-replay.service: main process exited, code=exited, status=1 [ 85.101530] [ 85.101531] = [ 85.101796] [ INFO: possible recursive locking detected ] [ 85.102002] 2.6.37-next-20110111 #1 [ 85.102009] - [ 85.102009] systemd/1 is trying to acquire lock: [ 85.102009] ((dentry-d_lock)-rlock){+.+...}, at: [8107ca5c] cgroup_rmdir+0x339/0x479 [ 85.102009] [ 85.102009] but task is already holding lock: [ 85.102009] ((dentry-d_lock)-rlock){+.+...}, at: [8107ca54] cgroup_rmdir+0x331/0x479 [ 85.102009] [ 85.102009] other info that might help us debug this: [ 85.102009] 4 locks held by systemd/1: [ 85.102009] #0: (sb-s_type-i_mutex_key#14/1){+.+.+.}, at: [810fea4d] do_rmdir+0x7d/0x121 [ 85.102009] #1: (sb-s_type-i_mutex_key#14){+.+.+.}, at: [810fd4bc] vfs_rmdir+0x4a/0xbe [ 85.102009] #2: (cgroup_mutex){+.+.+.}, at: [8107cb84] cgroup_rmdir+0x461/0x479 [ 85.102009] #3: ((dentry-d_lock)-rlock){+.+...}, at: [8107ca54] cgroup_rmdir+0x331/0x479 [ 85.102009] [ 85.102009] stack backtrace: [ 85.102009] Pid: 1, comm: systemd Tainted: GW 2.6.37-next-20110111 #1 [ 85.102009] Call Trace: [ 85.102009] [81069f22] ? __lock_acquire+0x929/0xd4e [ 85.102009] [8107c6f1] ? cgroup_clear_directory+0xff/0x131 [ 85.102009] [8107c6f1] ? cgroup_clear_directory+0xff/0x131 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [8106a859] ? lock_acquire+0x100/0x126 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [815521ef] ? sub_preempt_count+0x35/0x48 [ 85.102009] [8154e401] ? _raw_spin_lock+0x36/0x45 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [8107ca5c] ? cgroup_rmdir+0x339/0x479 [ 85.102009] [810579cd] ? autoremove_wake_function+0x0/0x34 [ 85.102009] [811e1839] ? selinux_inode_rmdir+0x15/0x17 [ 85.102009] [810fd4eb] ? vfs_rmdir+0x79/0xbe [ 85.102009] [810feaa0] ? do_rmdir+0xd0/0x121 [ 85.102009] [8100256c] ? sysret_check+0x27/0x62 [ 85.102009] [8106ac79] ? trace_hardirqs_on_caller+0x117/0x13b [ 85.102009] [8154e201] ? trace_hardirqs_on_thunk+0x3a/0x3f [ 85.102009] [8110040b] ? sys_rmdir+0x11/0x13 [ 85.102009] [8100253b] ? system_call_fastpath+0x16/0x1b [ 85.268272] systemd[1]: readahead-collect.service: main process exited, code=exited, status=1 Any ideas? It looks like it is just a missing parent-child lock order annotation, but mainline cgroupfs code looks to be OK there. What does cgroup_clear_directory() look like in mmotm? It's not from cgroup_clear_directory().. This should fix it: = From: Li Zefan l...@cn.fujitsu.com Date: Fri, 14 Jan 2011 11:34:34 +0800 Subject: [PATCH] cgroups: Fix a lockdep warning at cgroup removal Commit 2fd6b7f5 (fs: dcache scale subdirs) forgot to annotate a dentry lock, which caused a lockdep warning. Reported-by: Valdis Kletnieks valdis.kletni...@vt.edu Signed-off-by: Li Zefan l...@cn.fujitsu.com --- kernel/cgroup.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 5c5f4cc..db983e2 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -910,7 +910,7 @@ static void cgroup_d_remove_dir(struct dentry *dentry) parent = dentry-d_parent; spin_lock(parent-d_lock); - spin_lock(dentry-d_lock); + spin_lock_nested(dentry-d_lock, DENTRY_D_LOCK_NESTED); list_del_init(dentry-d_u.d_child); spin_unlock(dentry-d_lock); spin_unlock(parent-d_lock); -- 1.7.3.1 ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] kernel BUG at fs/dcache.c:1363 (from cgroup)
Just mount the cgroupfs: # mount -t cgroup -o cpuset xxx /mnt (oops!!) The bug is caused by: = commit 0df6a63f8735a7c8a877878bc215d4312e41ef81 Author: Al Viro v...@zeniv.linux.org.uk Date: Tue Dec 21 13:29:29 2010 -0500 switch cgroup switching it to s_d_op allows to kill the cgroup_lookup() kludge. Signed-off-by: Al Viro v...@zeniv.linux.org.uk = This line: + sb-s_d_op = cgroup_dops; will cause the dentry op be set twice, and thus trigger the bomb: struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) { ... if (parent) { ... d_set_d_op(dentry, dentry-d_sb-s_d_op); ... } ... } static struct dentry *d_alloc_and_lookup(struct dentry *parent, struct qstr *name, struct nameidata *nd) { ... dentry = d_alloc(parent, name); ... old = inode-i_op-lookup(inode, dentry, nd); ... } simple_lookup() will call d_set_d_op()... == [ 90.740906] kernel BUG at fs/dcache.c:1360! .. [ 90.810321] Call Trace: [ 90.814166] [c04f97ad] simple_lookup+0x26/0x3c [ 90.818015] [c04e86ce] d_alloc_and_lookup+0x36/0x54 [ 90.818021] [c04e8aa8] __lookup_hash+0x6a/0x71 [ 90.818026] [c04e8f33] lookup_one_len+0x81/0x90 [ 90.818034] [c0473083] cgroup_add_file+0x8e/0x132 [ 90.818041] [c0473152] cgroup_add_files+0x2b/0x3d [ 90.818047] [c0473188] cgroup_populate_dir+0x24/0xdb [ 90.818053] [c047360b] cgroup_mount+0x3cc/0x431 [ 90.818061] [c04e238d] vfs_kern_mount+0x57/0x109 [ 90.818066] [c047323f] ? cgroup_mount+0x0/0x431 [ 90.818072] [c04e248e] do_kern_mount+0x38/0xba [ 90.818077] [c04f6706] do_mount+0x5e4/0x60f [ 90.818082] [c04f6094] ? copy_mount_options+0x78/0xd7 [ 90.818087] [c04f68de] sys_mount+0x66/0x94 [ 90.818093] [c040329f] sysenter_do_call+0x12/0x38 ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] Re: kernel BUG at fs/dcache.c:1363 (from cgroup)
On Fri, Jan 14, 2011 at 12:56:17PM +0800, Li Zefan wrote: Just mount the cgroupfs: # mount -t cgroup -o cpuset xxx /mnt (oops!!) The bug is caused by: = commit 0df6a63f8735a7c8a877878bc215d4312e41ef81 Author: Al Viro v...@zeniv.linux.org.uk Date: Tue Dec 21 13:29:29 2010 -0500 switch cgroup switching it to s_d_op allows to kill the cgroup_lookup() kludge. Signed-off-by: Al Viro v...@zeniv.linux.org.uk = This line: + sb-s_d_op = cgroup_dops; Oh, crap... Right, it's using simple_lookup(). Let me check if anything else might be stepping on that. Umm... There's a very strange codepath in btrfs that also might. Interesting. Fix for cgroup, AFAICS, should be this: Signed-off-by: Al Viro v...@zeniv.linux.org.uk --- diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 5c5f4cc..ffb7bba 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -764,6 +764,7 @@ EXPORT_SYMBOL_GPL(cgroup_unlock); */ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode); +static struct dentry *cgroup_lookup(struct inode *, struct dentry *, struct nameidata *); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); static int cgroup_populate_dir(struct cgroup *cgrp); static const struct inode_operations cgroup_dir_inode_operations; @@ -860,6 +861,11 @@ static void cgroup_diput(struct dentry *dentry, struct inode *inode) iput(inode); } +static int cgroup_delete(const struct dentry *d) +{ + return 1; +} + static void remove_dir(struct dentry *d) { struct dentry *parent = dget(d-d_parent); @@ -1451,6 +1457,7 @@ static int cgroup_get_rootdir(struct super_block *sb) { static const struct dentry_operations cgroup_dops = { .d_iput = cgroup_diput, + .d_delete = cgroup_delete, }; struct inode *inode = @@ -2195,12 +2202,20 @@ static const struct file_operations cgroup_file_operations = { }; static const struct inode_operations cgroup_dir_inode_operations = { - .lookup = simple_lookup, + .lookup = cgroup_lookup, .mkdir = cgroup_mkdir, .rmdir = cgroup_rmdir, .rename = cgroup_rename, }; +static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +{ + if (dentry-d_name.len NAME_MAX) + return ERR_PTR(-ENAMETOOLONG); + d_add(dentry, NULL); + return NULL; +} + /* * Check if a file is a control file */ ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] Re: kernel BUG at fs/dcache.c:1363 (from cgroup)
Al Viro wrote: On Fri, Jan 14, 2011 at 12:56:17PM +0800, Li Zefan wrote: Just mount the cgroupfs: # mount -t cgroup -o cpuset xxx /mnt (oops!!) The bug is caused by: = commit 0df6a63f8735a7c8a877878bc215d4312e41ef81 Author: Al Viro v...@zeniv.linux.org.uk Date: Tue Dec 21 13:29:29 2010 -0500 switch cgroup switching it to s_d_op allows to kill the cgroup_lookup() kludge. Signed-off-by: Al Viro v...@zeniv.linux.org.uk = This line: + sb-s_d_op = cgroup_dops; Oh, crap... Right, it's using simple_lookup(). Let me check if anything else might be stepping on that. Umm... There's a very strange codepath in btrfs that also might. Interesting. Fix for cgroup, AFAICS, should be this: patch tested. thanks! ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel
[Devel] Re: linux-cr: v23-rc1 pushed
Oren Laadan [or...@cs.columbia.edu] wrote: | Folks, | | I just pushed out a new v23-rc1 branch of linux-cr. This one is | rebased to 2.6.37, and contains nearly all the patches pulled | on v22-dev. I only gave it a brief test drive... feel free to | throw all your ammo it. Oren, We need the file_tty() helper to get the tty object from the file pointer (otherwise we will be off by 4 bytes and fail tty_paranoia_check() in tty_file_checkpoint()). Thanks, Sukadev diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index c89f055..6aa458e 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -2781,7 +2784,7 @@ static int tty_file_checkpoint(struct ckpt_ctx *ctx, struc int master_objref, slave_objref; int ret; - tty = (struct tty_struct *)file-private_data; + tty = file_tty(file); inode = file-f_path.dentry-d_inode; if (tty_paranoia_check(tty, inode, tty_file_checkpoint)) return -EIO; ___ Containers mailing list contain...@lists.linux-foundation.org https://lists.linux-foundation.org/mailman/listinfo/containers ___ Devel mailing list Devel@openvz.org https://openvz.org/mailman/listinfo/devel