Patch series "use per-vma locks for /proc/pid/maps reads", v8.
Reading /proc/pid/maps requires read-locking mmap_lock which prevents any
other task from concurrently modifying the address space. This guarantees
coherent reporting of virtual address ranges, however it can block
important updates from happening. Oftentimes /proc/pid/maps readers are
low priority monitoring tasks and them blocking high priority tasks
results in priority inversion.
Locking the entire address space is required to present fully coherent
picture of the address space, however even current implementation does not
strictly guarantee that by outputting vmas in page-size chunks and
dropping mmap_lock in between each chunk. Address space modifications are
possible while mmap_lock is dropped and userspace reading the content is
expected to deal with possible concurrent address space modifications.
Considering these relaxed rules, holding mmap_lock is not strictly needed
as long as we can guarantee that a concurrently modified vma is reported
either in its original form or after it was modified.
This patchset switches from holding mmap_lock while reading /proc/pid/maps
to taking per-vma locks as we walk the vma tree. This reduces the
contention with tasks modifying the address space because they would have
to contend for the same vma as opposed to the entire address space.
Previous version of this patchset [1] tried to perform /proc/pid/maps
reading under RCU, however its implementation is quite complex and the
results are worse than the new version because it still relied on
mmap_lock speculation which retries if any part of the address space gets
modified. New implementaion is both simpler and results in less
contention. Note that similar approach would not work for /proc/pid/smaps
reading as it also walks the page table and that's not RCU-safe.
Paul McKenney's designed a test [2] to measure mmap/munmap latencies while
concurrently reading /proc/pid/maps. The test has a pair of processes
scanning /proc/PID/maps, and another process unmapping and remapping 4K
pages from a 128MB range of anonymous memory. At the end of each 10
second run, the latency of each mmap() or munmap() operation is measured,
and for each run the maximum and mean latency is printed. The map/unmap
process is started first, its PID is passed to the scanners, and then the
map/unmap process waits until both scanners are running before starting
its timed test. The scanners keep scanning until the specified
/proc/PID/maps file disappears.
The latest results from Paul:
Stock mm-unstable, all of the runs had maximum latencies in excess of 0.5
milliseconds, and with 80% of the runs' latencies exceeding a full
millisecond, and ranging up beyond 4 full milliseconds. In contrast, 99%
of the runs with this patch series applied had maximum latencies of less
than 0.5 milliseconds, with the single outlier at only 0.608 milliseconds.
From a median-performance (as opposed to maximum-latency) viewpoint, this
patch series also looks good, with stock mm weighing in at 11 microseconds
and patch series at 6 microseconds, better than a 2x improvement.
Before the change:
./run-proc-vs-map.sh --nsamples 100 --rawdata -- --busyduration 2
0.011 0.008 0.521
0.011 0.008 0.552
0.011 0.008 0.590
0.011 0.008 0.660
...
0.011 0.015 2.987
0.011 0.015 3.038
0.011 0.016 3.431
0.011 0.016 4.707
After the change:
./run-proc-vs-map.sh --nsamples 100 --rawdata -- --busyduration 2
0.006 0.005 0.026
0.006 0.005 0.029
0.006 0.005 0.034
0.006 0.005 0.035
...
0.006 0.006 0.421
0.006 0.006 0.423
0.006 0.006 0.439
0.006 0.006 0.608
The patchset also adds a number of tests to check for /proc/pid/maps data
coherency. They are designed to detect any unexpected data tearing while
performing some common address space modifications (vma split, resize and
remap). Even before these changes, reading /proc/pid/maps might have
inconsistent data because the file is read page-by-page with mmap_lock
being dropped between the pages. An example of user-visible inconsistency
can be that the same vma is printed twice: once before it was modified and
then after the modifications. For example if vma was extended, it might
be found and reported twice. What is not expected is to see a gap where
there should have been a vma both before and after modification. This
patchset increases the chances of such tearing, therefore it's even more
important now to test for unexpected inconsistencies.
In [3] Lorenzo identified the following possible vma merging/splitting
scenarios:
Merges with changes to existing vmas:
1 Merge both - mapping a vma over another one and between two vmas which
can be merged after this replacement;
2. Merge left full - mapping a vma at the end of an existing one and
completely over its right neighbor;
3. Merge left partial - mapping a vma at the end of an existing one and
partially over its right neighbor;
4. Merge right full - mapping a vma before the start of an existing one
and completely over its left neighbor;
5. Merge right partial - mapping a vma before the start of an existing one
and partially over its left neighbor;
Merges without changes to existing vmas:
6. Merge both - mapping a vma into a gap between two vmas which can be
merged after the insertion;
7. Merge left - mapping a vma at the end of an existing one;
8. Merge right - mapping a vma before the start end of an existing one;
Splits
9. Split with new vma at the lower address;
10. Split with new vma at the higher address;
If such merges or splits happen concurrently with the /proc/maps reading
we might report a vma twice, once before the modification and once after
it is modified:
Case 1 might report overwritten and previous vma along with the final
merged vma;
Case 2 might report previous and the final merged vma;
Case 3 might cause us to retry once we detect the temporary gap caused by
shrinking of the right neighbor;
Case 4 might report overritten and the final merged vma;
Case 5 might cause us to retry once we detect the temporary gap caused by
shrinking of the left neighbor;
Case 6 might report previous vma and the gap along with the final marged
vma;
Case 7 might report previous and the final merged vma;
Case 8 might report the original gap and the final merged vma covering the
gap;
Case 9 might cause us to retry once we detect the temporary gap caused by
shrinking of the original vma at the vma start;
Case 10 might cause us to retry once we detect the temporary gap caused by
shrinking of the original vma at the vma end;
In all these cases the retry mechanism prevents us from reporting possible
temporary gaps.
[1] https://lore.kernel.org/all/20250418174959.1431962-1-surenb@google.com/
[2] https://github.com/paulmckrcu/proc-mmap_sem-test
[3] https://lore.kernel.org/all/e1863f40-39ab-4e5b-984a-c48765ffde1c@lucifer.local/
The /proc/pid/maps file is generated page by page, with the mmap_lock
released between pages. This can lead to inconsistent reads if the
underlying vmas are concurrently modified. For instance, if a vma split
or merge occurs at a page boundary while /proc/pid/maps is being read, the
same vma might be seen twice: once before and once after the change. This
duplication is considered acceptable for userspace handling. However,
observing a "hole" where a vma should be (e.g., due to a vma being
replaced and the space temporarily being empty) is unacceptable.
Implement a test that:
1. Forks a child process which continuously modifies its address
space, specifically targeting a vma at the boundary between two pages.
2. The parent process repeatedly reads the child's /proc/pid/maps.
3. The parent process checks the last vma of the first page and the
first vma of the second page for consistency, looking for the effects
of vma splits or merges.
The test duration is configurable via DURATION environment variable
expressed in seconds. The default test duration is 5 seconds.
Example Command: DURATION=10 ./proc-maps-race
Link: https://lore.kernel.org/all/20250418174959.1431962-1-surenb@google.com/ [1]
Link: https://github.com/paulmckrcu/proc-mmap_sem-test [2]
Link: https://lore.kernel.org/all/e1863f40-39ab-4e5b-984a-c48765ffde1c@lucifer.local/ [3]
Link: https://lkml.kernel.org/r/20250719182854.3166724-1-surenb@google.com
Link: https://lkml.kernel.org/r/20250719182854.3166724-2-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Hildenbrand <david@redhat.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jeongjun Park <aha310510@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Kalesh Singh <kaleshsingh@google.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: "Paul E . McKenney" <paulmck@kernel.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Thomas Weißschuh <linux@weissschuh.net>
Cc: T.J. Mercier <tjmercier@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Ye Bin <yebin10@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
/proc/${pid}/status got Kthread field recently.
Test that userspace program is not reported as kernel thread.
Test that kernel thread is reported as kernel thread.
Use kthreadd with pid 2 for this.
Link: https://lkml.kernel.org/r/818c4c41-8668-4566-97a9-7254abf819ee@p183
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Chunguang Wu <fullspring2018@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Create process without mappings and check
/proc/*/maps
/proc/*/numa_maps
/proc/*/smaps
/proc/*/smaps_rollup
They must be empty (excluding vsyscall page) or full of zeroes.
Retroactively this test should've caught embarassing /proc/*/smaps_rollup
oops:
[17752.703567] BUG: kernel NULL pointer dereference, address: 0000000000000000
[17752.703580] #PF: supervisor read access in kernel mode
[17752.703583] #PF: error_code(0x0000) - not-present page
[17752.703587] PGD 0 P4D 0
[17752.703593] Oops: 0000 [#1] PREEMPT SMP PTI
[17752.703598] CPU: 0 PID: 60649 Comm: cat Tainted: G W 5.19.9-100.fc35.x86_64 #1
[17752.703603] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./X99 Extreme6/3.1, BIOS P3.30 08/05/2016
[17752.703607] RIP: 0010:show_smaps_rollup+0x159/0x2e0
Note 1:
ProtectionKey field in /proc/*/smaps is optional,
so check most of its contents, not everything.
Note 2:
due to the nature of this test, child process hardly can signal
its readiness (after unmapping everything!) to parent.
I feel like "sleep(1)" is justified.
If you know how to do it without sleep please tell me.
Note 3:
/proc/*/statm is not tested but can be.
Link: https://lkml.kernel.org/r/Yz3liL6Dn+n2SD8Q@localhost.localdomain
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This new selftest needs an entry in the .gitignore file otherwise git
will try to track the binary.
Link: https://lkml.kernel.org/r/20210601164305.11776-1-dmatlack@google.com
Fixes: 268af17ada ("selftests: proc: test subset=pid")
Signed-off-by: David Matlack <dmatlack@google.com>
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Alexey Gladkov <gladkov.alexey@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The hidepid parameter values are becoming more and more and it becomes
difficult to remember what each new magic number means.
Backward compatibility is preserved since it is possible to specify
numerical value for the hidepid parameter. This does not break the
fsconfig since it is not possible to specify a numerical value through
it. All numeric values are converted to a string. The type
FSCONFIG_SET_BINARY cannot be used to indicate a numerical value.
Selftest has been added to verify this behavior.
Suggested-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Alexey Gladkov <gladkov.alexey@gmail.com>
Reviewed-by: Alexey Dobriyan <adobriyan@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
This patch allows to have multiple procfs instances inside the
same pid namespace. The aim here is lightweight sandboxes, and to allow
that we have to modernize procfs internals.
1) The main aim of this work is to have on embedded systems one
supervisor for apps. Right now we have some lightweight sandbox support,
however if we create pid namespacess we have to manages all the
processes inside too, where our goal is to be able to run a bunch of
apps each one inside its own mount namespace without being able to
notice each other. We only want to use mount namespaces, and we want
procfs to behave more like a real mount point.
2) Linux Security Modules have multiple ptrace paths inside some
subsystems, however inside procfs, the implementation does not guarantee
that the ptrace() check which triggers the security_ptrace_check() hook
will always run. We have the 'hidepid' mount option that can be used to
force the ptrace_may_access() check inside has_pid_permissions() to run.
The problem is that 'hidepid' is per pid namespace and not attached to
the mount point, any remount or modification of 'hidepid' will propagate
to all other procfs mounts.
This also does not allow to support Yama LSM easily in desktop and user
sessions. Yama ptrace scope which restricts ptrace and some other
syscalls to be allowed only on inferiors, can be updated to have a
per-task context, where the context will be inherited during fork(),
clone() and preserved across execve(). If we support multiple private
procfs instances, then we may force the ptrace_may_access() on
/proc/<pids>/ to always run inside that new procfs instances. This will
allow to specifiy on user sessions if we should populate procfs with
pids that the user can ptrace or not.
By using Yama ptrace scope, some restricted users will only be able to see
inferiors inside /proc, they won't even be able to see their other
processes. Some software like Chromium, Firefox's crash handler, Wine
and others are already using Yama to restrict which processes can be
ptracable. With this change this will give the possibility to restrict
/proc/<pids>/ but more importantly this will give desktop users a
generic and usuable way to specifiy which users should see all processes
and which users can not.
Side notes:
* This covers the lack of seccomp where it is not able to parse
arguments, it is easy to install a seccomp filter on direct syscalls
that operate on pids, however /proc/<pid>/ is a Linux ABI using
filesystem syscalls. With this change LSMs should be able to analyze
open/read/write/close...
In the new patch set version I removed the 'newinstance' option
as suggested by Eric W. Biederman.
Selftest has been added to verify new behavior.
Signed-off-by: Alexey Gladkov <gladkov.alexey@gmail.com>
Reviewed-by: Alexey Dobriyan <adobriyan@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
I thought that /proc/sysvipc has the same bug as /proc/net
commit 1fde6f21d9
proc: fix /proc/net/* after setns(2)
However, it doesn't! /proc/sysvipc files do
get_ipc_ns(current->nsproxy->ipc_ns);
in their open() hook and avoid the problem.
Keep the test, maybe /proc/sysvipc will become broken someday :-\
Link: http://lkml.kernel.org/r/20190706180146.GA21015@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Start testing VM related fiels found in per-process files.
Do it by jiting small executable which brings its address space to
precisely known state, then comparing /proc/*/maps, smaps, smaps_rollup,
and statm files to expected values.
Currently only x86_64 is supported.
[adobriyan@gmail.com: exit correctly in /proc/*/maps test]
Link: http://lkml.kernel.org/r/20190206073659.GB15311@avx2
Link: http://lkml.kernel.org/r/20190203165806.GA14568@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Same story: I have WIP patch to make it faster, so better have a test
as well.
Link: http://lkml.kernel.org/r/20180627195209.GC18113@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuahkh@osg.samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are plans to change how /proc/self result is calculated,
for that a test is necessary.
Use direct system call because of this whole getpid caching story.
Link: http://lkml.kernel.org/r/20180627195103.GB18113@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuahkh@osg.samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Test lookup in /proc/self/fd.
"map_files" lookup story showed that lookup is not that simple.
* Test that all those symlinks open the same file.
Check with (st_dev, st_info).
* Test that kernel threads do not have anything in their /proc/*/fd/
directory.
Now this is where things get interesting.
First, kernel threads aren't pinned by /proc/self or equivalent,
thus some "atomicity" is required.
Second, ->comm can contain whitespace and ')'.
No, they are not escaped.
Third, the only reliable way to check if process is kernel thread
appears to be field #9 in /proc/*/stat.
This field is struct task_struct::flags in decimal!
Check is done by testing PF_KTHREAD flags like we do in kernel.
PF_KTREAD value is a part of userspace ABI !!!
Other methods for determining kernel threadness are not reliable:
* RSS can be 0 if everything is swapped, even while reading
from /proc/self.
* ->total_vm CAN BE ZERO if process is finishing
munmap(NULL, whole address space);
* /proc/*/maps and similar files can be empty because unmapping
everything works. Read returning 0 can't distinguish between
kernel thread and such suicide process.
Link: http://lkml.kernel.org/r/20180505000414.GA15090@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The only tests I could come up with for /proc/uptime are:
- test that values increase monotonically for 1 second,
- bounce around CPUs and test the same thing.
Avoid glibc like plague for affinity given patches like this:
https://marc.info/?l=linux-kernel&m=152130031912594&w=4
Link: http://lkml.kernel.org/r/20180317165235.GB3445@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Perform reads with nearly everything in /proc, and some writing as well.
Hopefully memleak checkers and KASAN will find something.
[adobriyan@gmail.com: /proc/kmsg can and will block if read under root]
Link: http://lkml.kernel.org/r/20180316232147.GA20146@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
[adobriyan@gmail.com: /proc/sysrq-trigger lives on the ground floor]
Link: http://lkml.kernel.org/r/20180317164911.GA3445@avx2
Link: http://lkml.kernel.org/r/20180315201251.GA12396@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Test fork counter formerly known as ->last_pid, the only part of
/proc/loadavg which can be tested.
Testing in init pid namespace is not reliable because of background
activity.
Link: http://lkml.kernel.org/r/20180311152241.GA26247@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
I totally forgot that _parse_integer() accepts arbitrary amount of
leading zeroes leading to the following lookups:
OK
# readlink /proc/1/map_files/56427ecba000-56427eddc000
/lib/systemd/systemd
bogus
# readlink /proc/1/map_files/00000000000056427ecba000-56427eddc000
/lib/systemd/systemd
# readlink /proc/1/map_files/56427ecba000-00000000000056427eddc000
/lib/systemd/systemd
Link: http://lkml.kernel.org/r/20180303215130.GA23480@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Reviewed-by: Cyrill Gorcunov <gorcunov@gmail.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Pavel Emelyanov <xemul@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Read from /proc/self/syscall should yield read system call and correct
args in the output as current is reading /proc/self/syscall.
Link: http://lkml.kernel.org/r/20180226212145.GB742@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch starts testing /proc. Many more tests to come (I promise).
Read from /proc/self/wchan should always return "0" as current is in
TASK_RUNNING state while reading /proc/self/wchan.
Link: http://lkml.kernel.org/r/20180226212006.GA742@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
