There are several storage drivers like dm-multipath, iscsi, tcmu-runner,
amd nbd that have userspace components that can run in the IO path. For
example, iscsi and nbd's userspace deamons may need to recreate a socket
and/or send IO on it, and dm-multipath's daemon multipathd may need to
send SG IO or read/write IO to figure out the state of paths and re-set
them up.
In the kernel these drivers have access to GFP_NOIO/GFP_NOFS and the
memalloc_*_save/restore functions to control the allocation behavior,
but for userspace we would end up hitting an allocation that ended up
writing data back to the same device we are trying to allocate for.
The device is then in a state of deadlock, because to execute IO the
device needs to allocate memory, but to allocate memory the memory
layers want execute IO to the device.
Here is an example with nbd using a local userspace daemon that performs
network IO to a remote server. We are using XFS on top of the nbd device,
but it can happen with any FS or other modules layered on top of the nbd
device that can write out data to free memory. Here a nbd daemon helper
thread, msgr-worker-1, is performing a write/sendmsg on a socket to execute
a request. This kicks off a reclaim operation which results in a WRITE to
the nbd device and the nbd thread calling back into the mm layer.
[ 1626.609191] msgr-worker-1 D 0 1026 1 0x00004000
[ 1626.609193] Call Trace:
[ 1626.609195] ? __schedule+0x29b/0x630
[ 1626.609197] ? wait_for_completion+0xe0/0x170
[ 1626.609198] schedule+0x30/0xb0
[ 1626.609200] schedule_timeout+0x1f6/0x2f0
[ 1626.609202] ? blk_finish_plug+0x21/0x2e
[ 1626.609204] ? _xfs_buf_ioapply+0x2e6/0x410
[ 1626.609206] ? wait_for_completion+0xe0/0x170
[ 1626.609208] wait_for_completion+0x108/0x170
[ 1626.609210] ? wake_up_q+0x70/0x70
[ 1626.609212] ? __xfs_buf_submit+0x12e/0x250
[ 1626.609214] ? xfs_bwrite+0x25/0x60
[ 1626.609215] xfs_buf_iowait+0x22/0xf0
[ 1626.609218] __xfs_buf_submit+0x12e/0x250
[ 1626.609220] xfs_bwrite+0x25/0x60
[ 1626.609222] xfs_reclaim_inode+0x2e8/0x310
[ 1626.609224] xfs_reclaim_inodes_ag+0x1b6/0x300
[ 1626.609227] xfs_reclaim_inodes_nr+0x31/0x40
[ 1626.609228] super_cache_scan+0x152/0x1a0
[ 1626.609231] do_shrink_slab+0x12c/0x2d0
[ 1626.609233] shrink_slab+0x9c/0x2a0
[ 1626.609235] shrink_node+0xd7/0x470
[ 1626.609237] do_try_to_free_pages+0xbf/0x380
[ 1626.609240] try_to_free_pages+0xd9/0x1f0
[ 1626.609245] __alloc_pages_slowpath+0x3a4/0xd30
[ 1626.609251] ? ___slab_alloc+0x238/0x560
[ 1626.609254] __alloc_pages_nodemask+0x30c/0x350
[ 1626.609259] skb_page_frag_refill+0x97/0xd0
[ 1626.609274] sk_page_frag_refill+0x1d/0x80
[ 1626.609279] tcp_sendmsg_locked+0x2bb/0xdd0
[ 1626.609304] tcp_sendmsg+0x27/0x40
[ 1626.609307] sock_sendmsg+0x54/0x60
[ 1626.609308] ___sys_sendmsg+0x29f/0x320
[ 1626.609313] ? sock_poll+0x66/0xb0
[ 1626.609318] ? ep_item_poll.isra.15+0x40/0xc0
[ 1626.609320] ? ep_send_events_proc+0xe6/0x230
[ 1626.609322] ? hrtimer_try_to_cancel+0x54/0xf0
[ 1626.609324] ? ep_read_events_proc+0xc0/0xc0
[ 1626.609326] ? _raw_write_unlock_irq+0xa/0x20
[ 1626.609327] ? ep_scan_ready_list.constprop.19+0x218/0x230
[ 1626.609329] ? __hrtimer_init+0xb0/0xb0
[ 1626.609331] ? _raw_spin_unlock_irq+0xa/0x20
[ 1626.609334] ? ep_poll+0x26c/0x4a0
[ 1626.609337] ? tcp_tsq_write.part.54+0xa0/0xa0
[ 1626.609339] ? release_sock+0x43/0x90
[ 1626.609341] ? _raw_spin_unlock_bh+0xa/0x20
[ 1626.609342] __sys_sendmsg+0x47/0x80
[ 1626.609347] do_syscall_64+0x5f/0x1c0
[ 1626.609349] ? prepare_exit_to_usermode+0x75/0xa0
[ 1626.609351] entry_SYSCALL_64_after_hwframe+0x44/0xa9
This patch adds a new prctl command that daemons can use after they have
done their initial setup, and before they start to do allocations that
are in the IO path. It sets the PF_MEMALLOC_NOIO and PF_LESS_THROTTLE
flags so both userspace block and FS threads can use it to avoid the
allocation recursion and try to prevent from being throttled while
writing out data to free up memory.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Tested-by: Masato Suzuki <masato.suzuki@wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Link: https://lore.kernel.org/r/20191112001900.9206-1-mchristi@redhat.com
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Many user space API headers are missing licensing information, which
makes it hard for compliance tools to determine the correct license.
By default are files without license information under the default
license of the kernel, which is GPLV2. Marking them GPLV2 would exclude
them from being included in non GPLV2 code, which is obviously not
intended. The user space API headers fall under the syscall exception
which is in the kernels COPYING file:
NOTE! This copyright does *not* cover user programs that use kernel
services by normal system calls - this is merely considered normal use
of the kernel, and does *not* fall under the heading of "derived work".
otherwise syscall usage would not be possible.
Update the files which contain no license information with an SPDX
license identifier. The chosen identifier is 'GPL-2.0 WITH
Linux-syscall-note' which is the officially assigned identifier for the
Linux syscall exception. SPDX license identifiers are a legally binding
shorthand, which can be used instead of the full boiler plate text.
This patch is based on work done by Thomas Gleixner and Kate Stewart and
Philippe Ombredanne. See the previous patch in this series for the
methodology of how this patch was researched.
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Root in a non-initial user ns cannot be trusted to write a traditional
security.capability xattr. If it were allowed to do so, then any
unprivileged user on the host could map his own uid to root in a private
namespace, write the xattr, and execute the file with privilege on the
host.
However supporting file capabilities in a user namespace is very
desirable. Not doing so means that any programs designed to run with
limited privilege must continue to support other methods of gaining and
dropping privilege. For instance a program installer must detect
whether file capabilities can be assigned, and assign them if so but set
setuid-root otherwise. The program in turn must know how to drop
partial capabilities, and do so only if setuid-root.
This patch introduces v3 of the security.capability xattr. It builds a
vfs_ns_cap_data struct by appending a uid_t rootid to struct
vfs_cap_data. This is the absolute uid_t (that is, the uid_t in user
namespace which mounted the filesystem, usually init_user_ns) of the
root id in whose namespaces the file capabilities may take effect.
When a task asks to write a v2 security.capability xattr, if it is
privileged with respect to the userns which mounted the filesystem, then
nothing should change. Otherwise, the kernel will transparently rewrite
the xattr as a v3 with the appropriate rootid. This is done during the
execution of setxattr() to catch user-space-initiated capability writes.
Subsequently, any task executing the file which has the noted kuid as
its root uid, or which is in a descendent user_ns of such a user_ns,
will run the file with capabilities.
Similarly when asking to read file capabilities, a v3 capability will
be presented as v2 if it applies to the caller's namespace.
If a task writes a v3 security.capability, then it can provide a uid for
the xattr so long as the uid is valid in its own user namespace, and it
is privileged with CAP_SETFCAP over its namespace. The kernel will
translate that rootid to an absolute uid, and write that to disk. After
this, a task in the writer's namespace will not be able to use those
capabilities (unless rootid was 0), but a task in a namespace where the
given uid is root will.
Only a single security.capability xattr may exist at a time for a given
file. A task may overwrite an existing xattr so long as it is
privileged over the inode. Note this is a departure from previous
semantics, which required privilege to remove a security.capability
xattr. This check can be re-added if deemed useful.
This allows a simple setxattr to work, allows tar/untar to work, and
allows us to tar in one namespace and untar in another while preserving
the capability, without risking leaking privilege into a parent
namespace.
Example using tar:
$ cp /bin/sleep sleepx
$ mkdir b1 b2
$ lxc-usernsexec -m b:0:100000:1 -m b:1:$(id -u):1 -- chown 0:0 b1
$ lxc-usernsexec -m b:0:100001:1 -m b:1:$(id -u):1 -- chown 0:0 b2
$ lxc-usernsexec -m b:0:100000:1000 -- tar --xattrs-include=security.capability --xattrs -cf b1/sleepx.tar sleepx
$ lxc-usernsexec -m b:0:100001:1000 -- tar --xattrs-include=security.capability --xattrs -C b2 -xf b1/sleepx.tar
$ lxc-usernsexec -m b:0:100001:1000 -- getcap b2/sleepx
b2/sleepx = cap_sys_admin+ep
# /opt/ltp/testcases/bin/getv3xattr b2/sleepx
v3 xattr, rootid is 100001
A patch to linux-test-project adding a new set of tests for this
functionality is in the nsfscaps branch at github.com/hallyn/ltp
Changelog:
Nov 02 2016: fix invalid check at refuse_fcap_overwrite()
Nov 07 2016: convert rootid from and to fs user_ns
(From ebiederm: mar 28 2017)
commoncap.c: fix typos - s/v4/v3
get_vfs_caps_from_disk: clarify the fs_ns root access check
nsfscaps: change the code split for cap_inode_setxattr()
Apr 09 2017:
don't return v3 cap for caps owned by current root.
return a v2 cap for a true v2 cap in non-init ns
Apr 18 2017:
. Change the flow of fscap writing to support s_user_ns writing.
. Remove refuse_fcap_overwrite(). The value of the previous
xattr doesn't matter.
Apr 24 2017:
. incorporate Eric's incremental diff
. move cap_convert_nscap to setxattr and simplify its usage
May 8, 2017:
. fix leaking dentry refcount in cap_inode_getsecurity
Signed-off-by: Serge Hallyn <serge@hallyn.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Since when we got rid of usbfs, the /proc/bus/usb is now
elsewhere. Fix references for it.
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Acked-by: Serge Hallyn <serge@hallyn.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Register a netlink per-protocol bind fuction for audit to check userspace
process capabilities before allowing a multicast group connection.
Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Michael Kerrisk <mtk.manpages@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Dave Jones <davej@redhat.com>
