This change introduces a prctl that allows the user program to control
which PAC keys are enabled in a particular task. The main reason
why this is useful is to enable a userspace ABI that uses PAC to
sign and authenticate function pointers and other pointers exposed
outside of the function, while still allowing binaries conforming
to the ABI to interoperate with legacy binaries that do not sign or
authenticate pointers.
The idea is that a dynamic loader or early startup code would issue
this prctl very early after establishing that a process may load legacy
binaries, but before executing any PAC instructions.
This change adds a small amount of overhead to kernel entry and exit
due to additional required instruction sequences.
On a DragonBoard 845c (Cortex-A75) with the powersave governor, the
overhead of similar instruction sequences was measured as 4.9ns when
simulating the common case where IA is left enabled, or 43.7ns when
simulating the uncommon case where IA is disabled. These numbers can
be seen as the worst case scenario, since in more realistic scenarios
a better performing governor would be used and a newer chip would be
used that would support PAC unlike Cortex-A75 and would be expected
to be faster than Cortex-A75.
On an Apple M1 under a hypervisor, the overhead of the entry/exit
instruction sequences introduced by this patch was measured as 0.3ns
in the case where IA is left enabled, and 33.0ns in the case where
IA is disabled.
Signed-off-by: Peter Collingbourne <pcc@google.com>
Reviewed-by: Dave Martin <Dave.Martin@arm.com>
Link: https://linux-review.googlesource.com/id/Ibc41a5e6a76b275efbaa126b31119dc197b927a5
Link: https://lore.kernel.org/r/d6609065f8f40397a4124654eb68c9f490b4d477.1616123271.git.pcc@google.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Doing a
prctl(PR_SET_MM, PR_SET_MM_AUXV, addr, 1);
will copy 1 byte from userspace to (quite big) on-stack array
and then stash everything to mm->saved_auxv.
AT_NULL terminator will be inserted at the very end.
/proc/*/auxv handler will find that AT_NULL terminator
and copy original stack contents to userspace.
This devious scheme requires CAP_SYS_RESOURCE.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull Kbuild updates from Masahiro Yamada:
- Fix false-positive build warnings for ARCH=ia64 builds
- Optimize dictionary size for module compression with xz
- Check the compiler and linker versions in Kconfig
- Fix misuse of extra-y
- Support DWARF v5 debug info
- Clamp SUBLEVEL to 255 because stable releases 4.4.x and 4.9.x
exceeded the limit
- Add generic syscall{tbl,hdr}.sh for cleanups across arches
- Minor cleanups of genksyms
- Minor cleanups of Kconfig
* tag 'kbuild-v5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: (38 commits)
initramfs: Remove redundant dependency of RD_ZSTD on BLK_DEV_INITRD
kbuild: remove deprecated 'always' and 'hostprogs-y/m'
kbuild: parse C= and M= before changing the working directory
kbuild: reuse this-makefile to define abs_srctree
kconfig: unify rule of config, menuconfig, nconfig, gconfig, xconfig
kconfig: omit --oldaskconfig option for 'make config'
kconfig: fix 'invalid option' for help option
kconfig: remove dead code in conf_askvalue()
kconfig: clean up nested if-conditionals in check_conf()
kconfig: Remove duplicate call to sym_get_string_value()
Makefile: Remove # characters from compiler string
Makefile: reuse CC_VERSION_TEXT
kbuild: check the minimum linker version in Kconfig
kbuild: remove ld-version macro
scripts: add generic syscallhdr.sh
scripts: add generic syscalltbl.sh
arch: syscalls: remove $(srctree)/ prefix from syscall tables
arch: syscalls: add missing FORCE and fix 'targets' to make if_changed work
gen_compile_commands: prune some directories
kbuild: simplify access to the kernel's version
...
Pull idmapped mounts from Christian Brauner:
"This introduces idmapped mounts which has been in the making for some
time. Simply put, different mounts can expose the same file or
directory with different ownership. This initial implementation comes
with ports for fat, ext4 and with Christoph's port for xfs with more
filesystems being actively worked on by independent people and
maintainers.
Idmapping mounts handle a wide range of long standing use-cases. Here
are just a few:
- Idmapped mounts make it possible to easily share files between
multiple users or multiple machines especially in complex
scenarios. For example, idmapped mounts will be used in the
implementation of portable home directories in
systemd-homed.service(8) where they allow users to move their home
directory to an external storage device and use it on multiple
computers where they are assigned different uids and gids. This
effectively makes it possible to assign random uids and gids at
login time.
- It is possible to share files from the host with unprivileged
containers without having to change ownership permanently through
chown(2).
- It is possible to idmap a container's rootfs and without having to
mangle every file. For example, Chromebooks use it to share the
user's Download folder with their unprivileged containers in their
Linux subsystem.
- It is possible to share files between containers with
non-overlapping idmappings.
- Filesystem that lack a proper concept of ownership such as fat can
use idmapped mounts to implement discretionary access (DAC)
permission checking.
- They allow users to efficiently changing ownership on a per-mount
basis without having to (recursively) chown(2) all files. In
contrast to chown (2) changing ownership of large sets of files is
instantenous with idmapped mounts. This is especially useful when
ownership of a whole root filesystem of a virtual machine or
container is changed. With idmapped mounts a single syscall
mount_setattr syscall will be sufficient to change the ownership of
all files.
- Idmapped mounts always take the current ownership into account as
idmappings specify what a given uid or gid is supposed to be mapped
to. This contrasts with the chown(2) syscall which cannot by itself
take the current ownership of the files it changes into account. It
simply changes the ownership to the specified uid and gid. This is
especially problematic when recursively chown(2)ing a large set of
files which is commong with the aforementioned portable home
directory and container and vm scenario.
- Idmapped mounts allow to change ownership locally, restricting it
to specific mounts, and temporarily as the ownership changes only
apply as long as the mount exists.
Several userspace projects have either already put up patches and
pull-requests for this feature or will do so should you decide to pull
this:
- systemd: In a wide variety of scenarios but especially right away
in their implementation of portable home directories.
https://systemd.io/HOME_DIRECTORY/
- container runtimes: containerd, runC, LXD:To share data between
host and unprivileged containers, unprivileged and privileged
containers, etc. The pull request for idmapped mounts support in
containerd, the default Kubernetes runtime is already up for quite
a while now: https://github.com/containerd/containerd/pull/4734
- The virtio-fs developers and several users have expressed interest
in using this feature with virtual machines once virtio-fs is
ported.
- ChromeOS: Sharing host-directories with unprivileged containers.
I've tightly synced with all those projects and all of those listed
here have also expressed their need/desire for this feature on the
mailing list. For more info on how people use this there's a bunch of
talks about this too. Here's just two recent ones:
https://www.cncf.io/wp-content/uploads/2020/12/Rootless-Containers-in-Gitpod.pdfhttps://fosdem.org/2021/schedule/event/containers_idmap/
This comes with an extensive xfstests suite covering both ext4 and
xfs:
https://git.kernel.org/brauner/xfstests-dev/h/idmapped_mounts
It covers truncation, creation, opening, xattrs, vfscaps, setid
execution, setgid inheritance and more both with idmapped and
non-idmapped mounts. It already helped to discover an unrelated xfs
setgid inheritance bug which has since been fixed in mainline. It will
be sent for inclusion with the xfstests project should you decide to
merge this.
In order to support per-mount idmappings vfsmounts are marked with
user namespaces. The idmapping of the user namespace will be used to
map the ids of vfs objects when they are accessed through that mount.
By default all vfsmounts are marked with the initial user namespace.
The initial user namespace is used to indicate that a mount is not
idmapped. All operations behave as before and this is verified in the
testsuite.
Based on prior discussions we want to attach the whole user namespace
and not just a dedicated idmapping struct. This allows us to reuse all
the helpers that already exist for dealing with idmappings instead of
introducing a whole new range of helpers. In addition, if we decide in
the future that we are confident enough to enable unprivileged users
to setup idmapped mounts the permission checking can take into account
whether the caller is privileged in the user namespace the mount is
currently marked with.
The user namespace the mount will be marked with can be specified by
passing a file descriptor refering to the user namespace as an
argument to the new mount_setattr() syscall together with the new
MOUNT_ATTR_IDMAP flag. The system call follows the openat2() pattern
of extensibility.
The following conditions must be met in order to create an idmapped
mount:
- The caller must currently have the CAP_SYS_ADMIN capability in the
user namespace the underlying filesystem has been mounted in.
- The underlying filesystem must support idmapped mounts.
- The mount must not already be idmapped. This also implies that the
idmapping of a mount cannot be altered once it has been idmapped.
- The mount must be a detached/anonymous mount, i.e. it must have
been created by calling open_tree() with the OPEN_TREE_CLONE flag
and it must not already have been visible in the filesystem.
The last two points guarantee easier semantics for userspace and the
kernel and make the implementation significantly simpler.
By default vfsmounts are marked with the initial user namespace and no
behavioral or performance changes are observed.
The manpage with a detailed description can be found here:
1d7b902e28
In order to support idmapped mounts, filesystems need to be changed
and mark themselves with the FS_ALLOW_IDMAP flag in fs_flags. The
patches to convert individual filesystem are not very large or
complicated overall as can be seen from the included fat, ext4, and
xfs ports. Patches for other filesystems are actively worked on and
will be sent out separately. The xfstestsuite can be used to verify
that port has been done correctly.
The mount_setattr() syscall is motivated independent of the idmapped
mounts patches and it's been around since July 2019. One of the most
valuable features of the new mount api is the ability to perform
mounts based on file descriptors only.
Together with the lookup restrictions available in the openat2()
RESOLVE_* flag namespace which we added in v5.6 this is the first time
we are close to hardened and race-free (e.g. symlinks) mounting and
path resolution.
While userspace has started porting to the new mount api to mount
proper filesystems and create new bind-mounts it is currently not
possible to change mount options of an already existing bind mount in
the new mount api since the mount_setattr() syscall is missing.
With the addition of the mount_setattr() syscall we remove this last
restriction and userspace can now fully port to the new mount api,
covering every use-case the old mount api could. We also add the
crucial ability to recursively change mount options for a whole mount
tree, both removing and adding mount options at the same time. This
syscall has been requested multiple times by various people and
projects.
There is a simple tool available at
https://github.com/brauner/mount-idmapped
that allows to create idmapped mounts so people can play with this
patch series. I'll add support for the regular mount binary should you
decide to pull this in the following weeks:
Here's an example to a simple idmapped mount of another user's home
directory:
u1001@f2-vm:/$ sudo ./mount --idmap both:1000:1001:1 /home/ubuntu/ /mnt
u1001@f2-vm:/$ ls -al /home/ubuntu/
total 28
drwxr-xr-x 2 ubuntu ubuntu 4096 Oct 28 22:07 .
drwxr-xr-x 4 root root 4096 Oct 28 04:00 ..
-rw------- 1 ubuntu ubuntu 3154 Oct 28 22:12 .bash_history
-rw-r--r-- 1 ubuntu ubuntu 220 Feb 25 2020 .bash_logout
-rw-r--r-- 1 ubuntu ubuntu 3771 Feb 25 2020 .bashrc
-rw-r--r-- 1 ubuntu ubuntu 807 Feb 25 2020 .profile
-rw-r--r-- 1 ubuntu ubuntu 0 Oct 16 16:11 .sudo_as_admin_successful
-rw------- 1 ubuntu ubuntu 1144 Oct 28 00:43 .viminfo
u1001@f2-vm:/$ ls -al /mnt/
total 28
drwxr-xr-x 2 u1001 u1001 4096 Oct 28 22:07 .
drwxr-xr-x 29 root root 4096 Oct 28 22:01 ..
-rw------- 1 u1001 u1001 3154 Oct 28 22:12 .bash_history
-rw-r--r-- 1 u1001 u1001 220 Feb 25 2020 .bash_logout
-rw-r--r-- 1 u1001 u1001 3771 Feb 25 2020 .bashrc
-rw-r--r-- 1 u1001 u1001 807 Feb 25 2020 .profile
-rw-r--r-- 1 u1001 u1001 0 Oct 16 16:11 .sudo_as_admin_successful
-rw------- 1 u1001 u1001 1144 Oct 28 00:43 .viminfo
u1001@f2-vm:/$ touch /mnt/my-file
u1001@f2-vm:/$ setfacl -m u:1001:rwx /mnt/my-file
u1001@f2-vm:/$ sudo setcap -n 1001 cap_net_raw+ep /mnt/my-file
u1001@f2-vm:/$ ls -al /mnt/my-file
-rw-rwxr--+ 1 u1001 u1001 0 Oct 28 22:14 /mnt/my-file
u1001@f2-vm:/$ ls -al /home/ubuntu/my-file
-rw-rwxr--+ 1 ubuntu ubuntu 0 Oct 28 22:14 /home/ubuntu/my-file
u1001@f2-vm:/$ getfacl /mnt/my-file
getfacl: Removing leading '/' from absolute path names
# file: mnt/my-file
# owner: u1001
# group: u1001
user::rw-
user:u1001:rwx
group::rw-
mask::rwx
other::r--
u1001@f2-vm:/$ getfacl /home/ubuntu/my-file
getfacl: Removing leading '/' from absolute path names
# file: home/ubuntu/my-file
# owner: ubuntu
# group: ubuntu
user::rw-
user:ubuntu:rwx
group::rw-
mask::rwx
other::r--"
* tag 'idmapped-mounts-v5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux: (41 commits)
xfs: remove the possibly unused mp variable in xfs_file_compat_ioctl
xfs: support idmapped mounts
ext4: support idmapped mounts
fat: handle idmapped mounts
tests: add mount_setattr() selftests
fs: introduce MOUNT_ATTR_IDMAP
fs: add mount_setattr()
fs: add attr_flags_to_mnt_flags helper
fs: split out functions to hold writers
namespace: only take read lock in do_reconfigure_mnt()
mount: make {lock,unlock}_mount_hash() static
namespace: take lock_mount_hash() directly when changing flags
nfs: do not export idmapped mounts
overlayfs: do not mount on top of idmapped mounts
ecryptfs: do not mount on top of idmapped mounts
ima: handle idmapped mounts
apparmor: handle idmapped mounts
fs: make helpers idmap mount aware
exec: handle idmapped mounts
would_dump: handle idmapped mounts
...
Instead of storing the version in a single integer and having various
kernel (and userspace) code how it's constructed, export individual
(major, patchlevel, sublevel) components and simplify kernel code that
uses it.
This should also make it easier on userspace.
Signed-off-by: Sasha Levin <sashal@kernel.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Introduce a mechanism to quickly disable/enable syscall handling for a
specific process and redirect to userspace via SIGSYS. This is useful
for processes with parts that require syscall redirection and parts that
don't, but who need to perform this boundary crossing really fast,
without paying the cost of a system call to reconfigure syscall handling
on each boundary transition. This is particularly important for Windows
games running over Wine.
The proposed interface looks like this:
prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <off>, <length>, [selector])
The range [<offset>,<offset>+<length>) is a part of the process memory
map that is allowed to by-pass the redirection code and dispatch
syscalls directly, such that in fast paths a process doesn't need to
disable the trap nor the kernel has to check the selector. This is
essential to return from SIGSYS to a blocked area without triggering
another SIGSYS from rt_sigreturn.
selector is an optional pointer to a char-sized userspace memory region
that has a key switch for the mechanism. This key switch is set to
either PR_SYS_DISPATCH_ON, PR_SYS_DISPATCH_OFF to enable and disable the
redirection without calling the kernel.
The feature is meant to be set per-thread and it is disabled on
fork/clone/execv.
Internally, this doesn't add overhead to the syscall hot path, and it
requires very little per-architecture support. I avoided using seccomp,
even though it duplicates some functionality, due to previous feedback
that maybe it shouldn't mix with seccomp since it is not a security
mechanism. And obviously, this should never be considered a security
mechanism, since any part of the program can by-pass it by using the
syscall dispatcher.
For the sysinfo benchmark, which measures the overhead added to
executing a native syscall that doesn't require interception, the
overhead using only the direct dispatcher region to issue syscalls is
pretty much irrelevant. The overhead of using the selector goes around
40ns for a native (unredirected) syscall in my system, and it is (as
expected) dominated by the supervisor-mode user-address access. In
fact, with SMAP off, the overhead is consistently less than 5ns on my
test box.
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20201127193238.821364-4-krisman@collabora.com
tid_addr is not a "pointer to (pointer to int in userspace)"; it is in
fact a "pointer to (pointer to int in userspace) in userspace". So
sparse rightfully complains about passing a kernel pointer to
put_user().
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull SafeSetID updates from Micah Morton:
"The changes are mostly contained to within the SafeSetID LSM, with the
exception of a few 1-line changes to change some ns_capable() calls to
ns_capable_setid() -- causing a flag (CAP_OPT_INSETID) to be set that
is examined by SafeSetID code and nothing else in the kernel.
The changes to SafeSetID internally allow for setting up GID
transition security policies, as already existed for UIDs"
* tag 'safesetid-5.10' of git://github.com/micah-morton/linux:
LSM: SafeSetID: Fix warnings reported by test bot
LSM: SafeSetID: Add GID security policy handling
LSM: Signal to SafeSetID when setting group IDs
For SafeSetID to properly gate set*gid() calls, it needs to know whether
ns_capable() is being called from within a sys_set*gid() function or is
being called from elsewhere in the kernel. This allows SafeSetID to deny
CAP_SETGID to restricted groups when they are attempting to use the
capability for code paths other than updating GIDs (e.g. setting up
userns GID mappings). This is the identical approach to what is
currently done for CAP_SETUID.
NOTE: We also add signaling to SafeSetID from the setgroups() syscall,
as we have future plans to restrict a process' ability to set
supplementary groups in addition to what is added in this series for
restricting setting of the primary group.
Signed-off-by: Thomas Cedeno <thomascedeno@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>
Originally, only a local CAP_SYS_ADMIN could change the exe link,
making it difficult for doing checkpoint/restore without CAP_SYS_ADMIN.
This commit adds CAP_CHECKPOINT_RESTORE in addition to CAP_SYS_ADMIN
for permitting changing the exe link.
The following describes the history of the /proc/self/exe permission
checks as it may be difficult to understand what decisions lead to this
point.
* [1] May 2012: This commit introduces the ability of changing
/proc/self/exe if the user is CAP_SYS_RESOURCE capable.
In the related discussion [2], no clear thread model is presented for
what could happen if the /proc/self/exe changes multiple times, or why
would the admin be at the mercy of userspace.
* [3] Oct 2014: This commit introduces a new API to change
/proc/self/exe. The permission no longer checks for CAP_SYS_RESOURCE,
but instead checks if the current user is root (uid=0) in its local
namespace. In the related discussion [4] it is said that "Controlling
exe_fd without privileges may turn out to be dangerous. At least
things like tomoyo examine it for making policy decisions (see
tomoyo_manager())."
* [5] Dec 2016: This commit removes the restriction to change
/proc/self/exe at most once. The related discussion [6] informs that
the audit subsystem relies on the exe symlink, presumably
audit_log_d_path_exe() in kernel/audit.c.
* [7] May 2017: This commit changed the check from uid==0 to local
CAP_SYS_ADMIN. No discussion.
* [8] July 2020: A PoC to spoof any program's /proc/self/exe via ptrace
is demonstrated
Overall, the concrete points that were made to retain capability checks
around changing the exe symlink is that tomoyo_manager() and
audit_log_d_path_exe() uses the exe_file path.
Christian Brauner said that relying on /proc/<pid>/exe being immutable (or
guarded by caps) in a sake of security is a bit misleading. It can only
be used as a hint without any guarantees of what code is being executed
once execve() returns to userspace. Christian suggested that in the
future, we could call audit_log() or similar to inform the admin of all
exe link changes, instead of attempting to provide security guarantees
via permission checks. However, this proposed change requires the
understanding of the security implications in the tomoyo/audit subsystems.
[1] b32dfe3771 ("c/r: prctl: add ability to set new mm_struct::exe_file")
[2] https://lore.kernel.org/patchwork/patch/292515/
[3] f606b77f1a ("prctl: PR_SET_MM -- introduce PR_SET_MM_MAP operation")
[4] https://lore.kernel.org/patchwork/patch/479359/
[5] 3fb4afd9a5 ("prctl: remove one-shot limitation for changing exe link")
[6] https://lore.kernel.org/patchwork/patch/697304/
[7] 4d28df6152 ("prctl: Allow local CAP_SYS_ADMIN changing exe_file")
[8] https://github.com/nviennot/run_as_exe
Signed-off-by: Nicolas Viennot <Nicolas.Viennot@twosigma.com>
Signed-off-by: Adrian Reber <areber@redhat.com>
Link: https://lore.kernel.org/r/20200719100418.2112740-6-areber@redhat.com
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Pull SafeSetID update from Micah Morton:
"Add additional LSM hooks for SafeSetID
SafeSetID is capable of making allow/deny decisions for set*uid calls
on a system, and we want to add similar functionality for set*gid
calls.
The work to do that is not yet complete, so probably won't make it in
for v5.8, but we are looking to get this simple patch in for v5.8
since we have it ready.
We are planning on the rest of the work for extending the SafeSetID
LSM being merged during the v5.9 merge window"
* tag 'LSM-add-setgid-hook-5.8-author-fix' of git://github.com/micah-morton/linux:
security: Add LSM hooks to set*gid syscalls
The SafeSetID LSM uses the security_task_fix_setuid hook to filter
set*uid() syscalls according to its configured security policy. In
preparation for adding analagous support in the LSM for set*gid()
syscalls, we add the requisite hook here. Tested by putting print
statements in the security_task_fix_setgid hook and seeing them get hit
during kernel boot.
Signed-off-by: Thomas Cedeno <thomascedeno@google.com>
Signed-off-by: Micah Morton <mortonm@chromium.org>
Merge updates from Andrew Morton:
"A few little subsystems and a start of a lot of MM patches.
Subsystems affected by this patch series: squashfs, ocfs2, parisc,
vfs. With mm subsystems: slab-generic, slub, debug, pagecache, gup,
swap, memcg, pagemap, memory-failure, vmalloc, kasan"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (128 commits)
kasan: move kasan_report() into report.c
mm/mm_init.c: report kasan-tag information stored in page->flags
ubsan: entirely disable alignment checks under UBSAN_TRAP
kasan: fix clang compilation warning due to stack protector
x86/mm: remove vmalloc faulting
mm: remove vmalloc_sync_(un)mappings()
x86/mm/32: implement arch_sync_kernel_mappings()
x86/mm/64: implement arch_sync_kernel_mappings()
mm/ioremap: track which page-table levels were modified
mm/vmalloc: track which page-table levels were modified
mm: add functions to track page directory modifications
s390: use __vmalloc_node in stack_alloc
powerpc: use __vmalloc_node in alloc_vm_stack
arm64: use __vmalloc_node in arch_alloc_vmap_stack
mm: remove vmalloc_user_node_flags
mm: switch the test_vmalloc module to use __vmalloc_node
mm: remove __vmalloc_node_flags_caller
mm: remove both instances of __vmalloc_node_flags
mm: remove the prot argument to __vmalloc_node
mm: remove the pgprot argument to __vmalloc
...
PF_LESS_THROTTLE exists for loop-back nfsd (and a similar need in the
loop block driver and callers of prctl(PR_SET_IO_FLUSHER)), where a
daemon needs to write to one bdi (the final bdi) in order to free up
writes queued to another bdi (the client bdi).
The daemon sets PF_LESS_THROTTLE and gets a larger allowance of dirty
pages, so that it can still dirty pages after other processses have been
throttled. The purpose of this is to avoid deadlock that happen when
the PF_LESS_THROTTLE process must write for any dirty pages to be freed,
but it is being thottled and cannot write.
This approach was designed when all threads were blocked equally,
independently on which device they were writing to, or how fast it was.
Since that time the writeback algorithm has changed substantially with
different threads getting different allowances based on non-trivial
heuristics. This means the simple "add 25%" heuristic is no longer
reliable.
The important issue is not that the daemon needs a *larger* dirty page
allowance, but that it needs a *private* dirty page allowance, so that
dirty pages for the "client" bdi that it is helping to clear (the bdi
for an NFS filesystem or loop block device etc) do not affect the
throttling of the daemon writing to the "final" bdi.
This patch changes the heuristic so that the task is not throttled when
the bdi it is writing to has a dirty page count below below (or equal
to) the free-run threshold for that bdi. This ensures it will always be
able to have some pages in flight, and so will not deadlock.
In a steady-state, it is expected that PF_LOCAL_THROTTLE tasks might
still be throttled by global threshold, but that is acceptable as it is
only the deadlock state that is interesting for this flag.
This approach of "only throttle when target bdi is busy" is consistent
with the other use of PF_LESS_THROTTLE in current_may_throttle(), were
it causes attention to be focussed only on the target bdi.
So this patch
- renames PF_LESS_THROTTLE to PF_LOCAL_THROTTLE,
- removes the 25% bonus that that flag gives, and
- If PF_LOCAL_THROTTLE is set, don't delay at all unless the
global and the local free-run thresholds are exceeded.
Note that previously realtime threads were treated the same as
PF_LESS_THROTTLE threads. This patch does *not* change the behvaiour
for real-time threads, so it is now different from the behaviour of nfsd
and loop tasks. I don't know what is wanted for realtime.
[akpm@linux-foundation.org: coding style fixes]
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Acked-by: Chuck Lever <chuck.lever@oracle.com> [nfsd]
Cc: Christoph Hellwig <hch@lst.de>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Link: http://lkml.kernel.org/r/87ftbf7gs3.fsf@notabene.neil.brown.name
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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>
