SCHED_FLAG_KEEP_PARAMS can be passed to sched_setattr to specify that
the call must not touch scheduling parameters (nice or priority). This
is particularly handy for uclamp when used in conjunction with
SCHED_FLAG_KEEP_POLICY as that allows to issue a syscall that only
impacts uclamp values.
However, sched_setattr always checks whether the priorities and nice
values passed in sched_attr are valid first, even if those never get
used down the line. This is useless at best since userspace can
trivially bypass this check to set the uclamp values by specifying low
priorities. However, it is cumbersome to do so as there is no single
expression of this that skips both RT and CFS checks at once. As such,
userspace needs to query the task policy first with e.g. sched_getattr
and then set sched_attr.sched_priority accordingly. This is racy and
slower than a single call.
As the priority and nice checks are useless when SCHED_FLAG_KEEP_PARAMS
is specified, simply inherit them in this case to match the policy
inheritance of SCHED_FLAG_KEEP_POLICY.
Reported-by: Wei Wang <wvw@google.com>
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Qais Yousef <qais.yousef@arm.com>
Link: https://lore.kernel.org/r/20210805102154.590709-3-qperret@google.com
Bug: 190237315
(cherry picked from commit f4dddf9
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core)
Signed-off-by: Quentin Perret <qperret@google.com>
Change-Id: Ifdbc9262b82c7f5c0d34952ece07770a53e3f6a5
[panchajanya1999: adapt for k4.14]
Signed-off-by: Panchajanya1999 <panchajanya@azure-dev.live>
Signed-off-by: mcdofrenchfreis <xyzevan@androidist.net>
A more energy efficient update of the IO wait boosting mechanism has
been introduced in:
commit a5a0809 ("cpufreq: schedutil: Make iowait boost more energy
efficient")
where the boost value is expected to be:
- doubled at each successive wakeup from IO
staring from the minimum frequency supported by a CPU
- reset when a CPU is not updated for more then one tick
by either disabling the IO wait boost or resetting its value to the
minimum frequency if this new update requires an IO boost.
This approach is supposed to "ignore" boosting for sporadic wakeups from
IO, while still getting the frequency boosted to the maximum to benefit
long sequence of wakeup from IO operations.
However, these assumptions are not always satisfied.
For example, when an IO boosted CPU enters idle for more the one tick
and then wakes up after an IO wait, since in sugov_set_iowait_boost() we
first check the IOWAIT flag, we keep doubling the iowait boost instead
of restarting from the minimum frequency value.
This misbehavior could happen mainly on non-shared frequency domains,
thus defeating the energy efficiency optimization, but it can also
happen on shared frequency domain systems.
Let fix this issue in sugov_set_iowait_boost() by:
- first check the IO wait boost reset conditions
to eventually reset the boost value
- then applying the correct IO boost value
if required by the caller
Fixes: a5a0809 (cpufreq: schedutil: Make iowait boost more energy
efficient)
Reported-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Pranav Vashi <neobuddy89@gmail.com>
[ Upstream commit b89997aa88f0b07d8a6414c908af75062103b8c9 ]
Being called for each dequeue, util_est reduces the number of its updates
by filtering out when the EWMA signal is different from the task util_avg
by less than 1%. It is a problem for a sudden util_avg ramp-up. Due to the
decay from a previous high util_avg, EWMA might now be close enough to
the new util_avg. No update would then happen while it would leave
ue.enqueued with an out-of-date value.
Taking into consideration the two util_est members, EWMA and enqueued for
the filtering, ensures, for both, an up-to-date value.
This is for now an issue only for the trace probe that might return the
stale value. Functional-wise, it isn't a problem, as the value is always
accessed through max(enqueued, ewma).
This problem has been observed using LISA's UtilConvergence:test_means on
the sd845c board.
No regression observed with Hackbench on sd845c and Perf-bench sched pipe
on hikey/hikey960.
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210225165820.1377125-1-vincent.donnefort@arm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
While calculating untilization of CPU during task placement in fbt(),
current code doesn't take uclamp into account which would lead to
selection of incorrect CPU for the task when uclamp restrictions
are in place for the task.
Change-Id: I8371affe3b37733d222e5c57953e53f91fc19a53
Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org>
Make use of the existing need_idle feature to incorporate upstream latency
sensitive tasks.
Change-Id: Ie1513187d024b93c8b619d9e0a35d84195488696
Signed-off-by: Shaleen Agrawal <shalagra@codeaurora.org>
There is extensive spam in dmesg because of userspace.
This kills all "init", "healthd", "logd" messages in kernel logs and makes them
more readable.
Extracted from 59f163ac76.
Change-Id: Ia812d01898ad97747df2be9384e64e4409e5e49a
Signed-off-by: idkwhoiam322 <idkwhoiam322@raphielgang.org>
Signed-off-by: Panchajanya1999 <panchajanya@azure-dev.live>
Constant messages about battery status every second doesn't help battery
or debugging.
Signed-off-by: kdrag0n <dragon@khronodragon.com>
Signed-off-by: clarencelol <clarencekuiek@icloud.com>
DISCLAIMER:
=====================================================================
This patch is intended to go upstream after collecting feedback from
Android community that it resolves the issues reported by various
partners. It is not meant to be merged into android-mainline.
=====================================================================
uclamp_max effectiveness could be easily impacted by small transient
tasks that wake up frequency to do small work then go back to sleep.
If there's a busy task that is capped by uclamp_max to run at a smaller
frequency, due to max-aggregation rule tasks that wake up on the same
cpu will increase the rq->uclamp_max value if they were higher than the
capped task. Given that all tasks by default have a uclamp_max = 1024,
this is the likely case by default.
Note that since the capped task is likely to be a busy and throttled
one, its util, and hence the rq->util, will be very high and as soon as
we lift the capping the requested frequency will be very high.
To address this issue of increasing the resilience of uclamp_max against
these transient tasks that don't really need to run at a higher
frequency, we implement a simple filter mechanism to ignore uclamp_max
for those tasks.
The algorithm looks at the runtime of the task and compares it to
sched_slice(). By default we assume any task that its runtime is 1/4th
of sched_slice() or less is a small transient task that we can ignore
its uclamp_max requirement.
runtime < sched_slice() / divider
We can tweak the divider by
/proc/sys/kernel/sched_util_uclamp_max_filter_divider sysctl. It accepts
values 0-4.
divider = 1 << sched_util_uclamp_max_filter_divider
We add a new task_tick_uclamp() function to verify this condition
periodically and ensure the conditions checked at wake up are still true
- in case this transient task suddenly becomes a busy one.
For EAS, we can't use sched_slice() there to figure out if uclamp_max
will be ignored because the task is not enqueued yet. So we leave it
as-is to figure out the placement based on worst case scenario.
Signed-off-by: Qais Yousef <qais.yousef@arm.com>
Change-Id: Ie3afa93a7d70dab5b7c22e820cc078ffd0e891ef
[yaro: ported to msm-5.4 and remove sysctl parts for now]
Signed-off-by: Yaroslav Furman <yaro330@gmail.com>
task_fits_capacity() drives CPU selection at wakeup time, and is also used
to detect misfit tasks. Right now it does so by comparing task_util_est()
with a CPU's capacity, but doesn't take into account uclamp restrictions.
There's a few interesting uses that can come out of doing this. For
instance, a low uclamp.max value could prevent certain tasks from being
flagged as misfit tasks, so they could merrily remain on low-capacity CPUs.
Similarly, a high uclamp.min value would steer tasks towards high capacity
CPUs at wakeup (and, should that fail, later steered via misfit balancing),
so such "boosted" tasks would favor CPUs of higher capacity.
Introduce uclamp_task_util() and make task_fits_capacity() use it.
[QP: fixed missing dependency on fits_capacity() by using the open coded
alternative]
Bug: 120440300
Tested-By: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Quentin Perret <qperret@google.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191211113851.24241-5-valentin.schneider@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
(cherry picked from commit a7008c07a568278ed2763436404752a98004c7ff)
Signed-off-by: Quentin Perret <qperret@google.com>
Change-Id: Iabde2eda7252c3bcc273e61260a7a12a7de991b1
commit 6f363f5aa845561f7ea496d8b1175e3204470486 upstream.
We found a refcount UAF bug as follows:
refcount_t: addition on 0; use-after-free.
WARNING: CPU: 1 PID: 342 at lib/refcount.c:25 refcount_warn_saturate+0xa0/0x148
Workqueue: events cpuset_hotplug_workfn
Call trace:
refcount_warn_saturate+0xa0/0x148
__refcount_add.constprop.0+0x5c/0x80
css_task_iter_advance_css_set+0xd8/0x210
css_task_iter_advance+0xa8/0x120
css_task_iter_next+0x94/0x158
update_tasks_root_domain+0x58/0x98
rebuild_root_domains+0xa0/0x1b0
rebuild_sched_domains_locked+0x144/0x188
cpuset_hotplug_workfn+0x138/0x5a0
process_one_work+0x1e8/0x448
worker_thread+0x228/0x3e0
kthread+0xe0/0xf0
ret_from_fork+0x10/0x20
then a kernel panic will be triggered as below:
Unable to handle kernel paging request at virtual address 00000000c0000010
Call trace:
cgroup_apply_control_disable+0xa4/0x16c
rebind_subsystems+0x224/0x590
cgroup_destroy_root+0x64/0x2e0
css_free_rwork_fn+0x198/0x2a0
process_one_work+0x1d4/0x4bc
worker_thread+0x158/0x410
kthread+0x108/0x13c
ret_from_fork+0x10/0x18
The race that cause this bug can be shown as below:
(hotplug cpu) | (umount cpuset)
mutex_lock(&cpuset_mutex) | mutex_lock(&cgroup_mutex)
cpuset_hotplug_workfn |
rebuild_root_domains | rebind_subsystems
update_tasks_root_domain | spin_lock_irq(&css_set_lock)
css_task_iter_start | list_move_tail(&cset->e_cset_node[ss->id]
while(css_task_iter_next) | &dcgrp->e_csets[ss->id]);
css_task_iter_end | spin_unlock_irq(&css_set_lock)
mutex_unlock(&cpuset_mutex) | mutex_unlock(&cgroup_mutex)
Inside css_task_iter_start/next/end, css_set_lock is hold and then
released, so when iterating task(left side), the css_set may be moved to
another list(right side), then it->cset_head points to the old list head
and it->cset_pos->next points to the head node of new list, which can't
be used as struct css_set.
To fix this issue, switch from all css_sets to only scgrp's css_sets to
patch in-flight iterators to preserve correct iteration, and then
update it->cset_head as well.
Reported-by: Gaosheng Cui <cuigaosheng1@huawei.com>
Link: https://www.spinics.net/lists/cgroups/msg37935.html
Suggested-by: Michal Koutný <mkoutny@suse.com>
Link: https://lore.kernel.org/all/20230526114139.70274-1-xiujianfeng@huaweicloud.com/
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Fixes: 2d8f243a5e ("cgroup: implement cgroup->e_csets[]")
Cc: stable@vger.kernel.org # v3.16+
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 15def34e2635ab7e0e96f1bc32e1b69609f14942 ]
commit e050e3f0a7 ("perf: Fix broken interrupt rate throttling")
introduces a change in throttling threshold judgment. Before this,
compare hwc->interrupts and max_samples_per_tick, then increase
hwc->interrupts by 1, but this commit reverses order of these two
behaviors, causing the semantics of max_samples_per_tick to change.
In literal sense of "max_samples_per_tick", if hwc->interrupts ==
max_samples_per_tick, it should not be throttled, therefore, the judgment
condition should be changed to "hwc->interrupts > max_samples_per_tick".
In fact, this may cause the hardlockup to fail, The minimum value of
max_samples_per_tick may be 1, in this case, the return value of
__perf_event_account_interrupt function is 1.
As a result, nmi_watchdog gets throttled, which would stop PMU (Use x86
architecture as an example, see x86_pmu_handle_irq).
Fixes: e050e3f0a7 ("perf: Fix broken interrupt rate throttling")
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230227023508.102230-1-yangjihong1@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 675751bb20634f981498c7d66161584080cc061e upstream.
If something was written to the buffer just before destruction,
it may be possible (maybe not in a real system, but it did
happen in ARCH=um with time-travel) to destroy the ringbuffer
before the IRQ work ran, leading this KASAN report (or a crash
without KASAN):
BUG: KASAN: slab-use-after-free in irq_work_run_list+0x11a/0x13a
Read of size 8 at addr 000000006d640a48 by task swapper/0
CPU: 0 PID: 0 Comm: swapper Tainted: G W O 6.3.0-rc1 #7
Stack:
60c4f20f 0c203d48 41b58ab3 60f224fc
600477fa 60f35687 60c4f20f 601273dd
00000008 6101eb00 6101eab0 615be548
Call Trace:
[<60047a58>] show_stack+0x25e/0x282
[<60c609e0>] dump_stack_lvl+0x96/0xfd
[<60c50d4c>] print_report+0x1a7/0x5a8
[<603078d3>] kasan_report+0xc1/0xe9
[<60308950>] __asan_report_load8_noabort+0x1b/0x1d
[<60232844>] irq_work_run_list+0x11a/0x13a
[<602328b4>] irq_work_tick+0x24/0x34
[<6017f9dc>] update_process_times+0x162/0x196
[<6019f335>] tick_sched_handle+0x1a4/0x1c3
[<6019fd9e>] tick_sched_timer+0x79/0x10c
[<601812b9>] __hrtimer_run_queues.constprop.0+0x425/0x695
[<60182913>] hrtimer_interrupt+0x16c/0x2c4
[<600486a3>] um_timer+0x164/0x183
[...]
Allocated by task 411:
save_stack_trace+0x99/0xb5
stack_trace_save+0x81/0x9b
kasan_save_stack+0x2d/0x54
kasan_set_track+0x34/0x3e
kasan_save_alloc_info+0x25/0x28
____kasan_kmalloc+0x8b/0x97
__kasan_kmalloc+0x10/0x12
__kmalloc+0xb2/0xe8
load_elf_phdrs+0xee/0x182
[...]
The buggy address belongs to the object at 000000006d640800
which belongs to the cache kmalloc-1k of size 1024
The buggy address is located 584 bytes inside of
freed 1024-byte region [000000006d640800, 000000006d640c00)
Add the appropriate irq_work_sync() so the work finishes before
the buffers are destroyed.
Prior to the commit in the Fixes tag below, there was only a
single global IRQ work, so this issue didn't exist.
Link: https://lore.kernel.org/linux-trace-kernel/20230427175920.a76159263122.I8295e405c44362a86c995e9c2c37e3e03810aa56@changeid
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Fixes: 15693458c4 ("tracing/ring-buffer: Move poll wake ups into ring buffer code")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ba9182a89626d5f83c2ee4594f55cb9c1e60f0e2 upstream.
After a successful cpuset_can_attach() call which increments the
attach_in_progress flag, either cpuset_cancel_attach() or cpuset_attach()
will be called later. In cpuset_attach(), tasks in cpuset_attach_wq,
if present, will be woken up at the end. That is not the case in
cpuset_cancel_attach(). So missed wakeup is possible if the attach
operation is somehow cancelled. Fix that by doing the wakeup in
cpuset_cancel_attach() as well.
Fixes: e44193d39e ("cpuset: let hotplug propagation work wait for task attaching")
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Cc: stable@vger.kernel.org # v3.11+
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6455b6163d8c680366663cdb8c679514d55fc30c upstream.
When user reads file 'trace_pipe', kernel keeps printing following logs
that warn at "cpu_buffer->reader_page->read > rb_page_size(reader)" in
rb_get_reader_page(). It just looks like there's an infinite loop in
tracing_read_pipe(). This problem occurs several times on arm64 platform
when testing v5.10 and below.
Call trace:
rb_get_reader_page+0x248/0x1300
rb_buffer_peek+0x34/0x160
ring_buffer_peek+0xbc/0x224
peek_next_entry+0x98/0xbc
__find_next_entry+0xc4/0x1c0
trace_find_next_entry_inc+0x30/0x94
tracing_read_pipe+0x198/0x304
vfs_read+0xb4/0x1e0
ksys_read+0x74/0x100
__arm64_sys_read+0x24/0x30
el0_svc_common.constprop.0+0x7c/0x1bc
do_el0_svc+0x2c/0x94
el0_svc+0x20/0x30
el0_sync_handler+0xb0/0xb4
el0_sync+0x160/0x180
Then I dump the vmcore and look into the problematic per_cpu ring_buffer,
I found that tail_page/commit_page/reader_page are on the same page while
reader_page->read is obviously abnormal:
tail_page == commit_page == reader_page == {
.write = 0x100d20,
.read = 0x8f9f4805, // Far greater than 0xd20, obviously abnormal!!!
.entries = 0x10004c,
.real_end = 0x0,
.page = {
.time_stamp = 0x857257416af0,
.commit = 0xd20, // This page hasn't been full filled.
// .data[0...0xd20] seems normal.
}
}
The root cause is most likely the race that reader and writer are on the
same page while reader saw an event that not fully committed by writer.
To fix this, add memory barriers to make sure the reader can see the
content of what is committed. Since commit a0fcaaed0c46 ("ring-buffer: Fix
race between reset page and reading page") has added the read barrier in
rb_get_reader_page(), here we just need to add the write barrier.
Link: https://lore.kernel.org/linux-trace-kernel/20230325021247.2923907-1-zhengyejian1@huawei.com
Cc: stable@vger.kernel.org
Fixes: 77ae365eca ("ring-buffer: make lockless")
Suggested-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 24d3ae2f37d8bc3c14b31d353c5d27baf582b6a6 ]
The same task check in perf_event_set_output has some potential issues
for some usages.
For the current perf code, there is a problem if using of
perf_event_open() to have multiple samples getting into the same mmap’d
memory when they are both attached to the same process.
https://lore.kernel.org/all/92645262-D319-4068-9C44-2409EF44888E@gmail.com/
Because the event->ctx is not ready when the perf_event_set_output() is
invoked in the perf_event_open().
Besides the above issue, before the commit bd2756811766 ("perf: Rewrite
core context handling"), perf record can errors out when sampling with
a hardware event and a software event as below.
$ perf record -e cycles,dummy --per-thread ls
failed to mmap with 22 (Invalid argument)
That's because that prior to the commit a hardware event and a software
event are from different task context.
The problem should be a long time issue since commit c3f00c7027
("perk: Separate find_get_context() from event initialization").
The task struct is stored in the event->hw.target for each per-thread
event. It is a more reliable way to determine whether two events are
attached to the same task.
The event->hw.target was also introduced several years ago by the
commit 50f16a8bf9 ("perf: Remove type specific target pointers"). It
can not only be used to fix the issue with the current code, but also
back port to fix the issues with an older kernel.
Note: The event->hw.target was introduced later than commit
c3f00c7027. The patch may cannot be applied between the commit
c3f00c7027 and commit 50f16a8bf9. Anybody that wants to back-port
this at that period may have to find other solutions.
Fixes: c3f00c7027 ("perf: Separate find_get_context() from event initialization")
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Zhengjun Xing <zhengjun.xing@linux.intel.com>
Link: https://lkml.kernel.org/r/20230322202449.512091-1-kan.liang@linux.intel.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6015b1aca1a233379625385feb01dd014aca60b5 ]
The getaffinity() system call uses 'cpumask_size()' to decide how big
the CPU mask is - so far so good. It is indeed the allocation size of a
cpumask.
But the code also assumes that the whole allocation is initialized
without actually doing so itself. That's wrong, because we might have
fixed-size allocations (making copying and clearing more efficient), but
not all of it is then necessarily used if 'nr_cpu_ids' is smaller.
Having checked other users of 'cpumask_size()', they all seem to be ok,
either using it purely for the allocation size, or explicitly zeroing
the cpumask before using the size in bytes to copy it.
See for example the ublk_ctrl_get_queue_affinity() function that uses
the proper 'zalloc_cpumask_var()' to make sure that the whole mask is
cleared, whether the storage is on the stack or if it was an external
allocation.
Fix this by just zeroing the allocation before using it. Do the same
for the compat version of sched_getaffinity(), which had the same logic.
Also, for consistency, make sched_getaffinity() use 'cpumask_bits()' to
access the bits. For a cpumask_var_t, it ends up being a pointer to the
same data either way, but it's just a good idea to treat it like you
would a 'cpumask_t'. The compat case already did that.
Reported-by: Ryan Roberts <ryan.roberts@arm.com>
Link: https://lore.kernel.org/lkml/7d026744-6bd6-6827-0471-b5e8eae0be3f@arm.com/
Cc: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit a53ce18cacb477dd0513c607f187d16f0fa96f71 upstream.
Commit 829c1651e9c4 ("sched/fair: sanitize vruntime of entity being placed")
fixes an overflowing bug, but ignore a case that se->exec_start is reset
after a migration.
For fixing this case, we delay the reset of se->exec_start after
placing the entity which se->exec_start to detect long sleeping task.
In order to take into account a possible divergence between the clock_task
of 2 rqs, we increase the threshold to around 104 days.
Fixes: 829c1651e9c4 ("sched/fair: sanitize vruntime of entity being placed")
Originally-by: Zhang Qiao <zhangqiao22@huawei.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Zhang Qiao <zhangqiao22@huawei.com>
Link: https://lore.kernel.org/r/20230317160810.107988-1-vincent.guittot@linaro.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 829c1651e9c4a6f78398d3e67651cef9bb6b42cc upstream.
When a scheduling entity is placed onto cfs_rq, its vruntime is pulled
to the base level (around cfs_rq->min_vruntime), so that the entity
doesn't gain extra boost when placed backwards.
However, if the entity being placed wasn't executed for a long time, its
vruntime may get too far behind (e.g. while cfs_rq was executing a
low-weight hog), which can inverse the vruntime comparison due to s64
overflow. This results in the entity being placed with its original
vruntime way forwards, so that it will effectively never get to the cpu.
To prevent that, ignore the vruntime of the entity being placed if it
didn't execute for much longer than the characteristic sheduler time
scale.
[rkagan: formatted, adjusted commit log, comments, cutoff value]
Signed-off-by: Zhang Qiao <zhangqiao22@huawei.com>
Co-developed-by: Roman Kagan <rkagan@amazon.de>
Signed-off-by: Roman Kagan <rkagan@amazon.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230130122216.3555094-1-rkagan@amazon.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 10ec8ca8ec1a2f04c4ed90897225231c58c124a7 ]
We've seen recent AWS EKS (Kubernetes) user reports like the following:
After upgrading EKS nodes from v20230203 to v20230217 on our 1.24 EKS
clusters after a few days a number of the nodes have containers stuck
in ContainerCreating state or liveness/readiness probes reporting the
following error:
Readiness probe errored: rpc error: code = Unknown desc = failed to
exec in container: failed to start exec "4a11039f730203ffc003b7[...]":
OCI runtime exec failed: exec failed: unable to start container process:
unable to init seccomp: error loading seccomp filter into kernel:
error loading seccomp filter: errno 524: unknown
However, we had not been seeing this issue on previous AMIs and it only
started to occur on v20230217 (following the upgrade from kernel 5.4 to
5.10) with no other changes to the underlying cluster or workloads.
We tried the suggestions from that issue (sysctl net.core.bpf_jit_limit=452534528)
which helped to immediately allow containers to be created and probes to
execute but after approximately a day the issue returned and the value
returned by cat /proc/vmallocinfo | grep bpf_jit | awk '{s+=$2} END {print s}'
was steadily increasing.
I tested bpf tree to observe bpf_jit_charge_modmem, bpf_jit_uncharge_modmem
their sizes passed in as well as bpf_jit_current under tcpdump BPF filter,
seccomp BPF and native (e)BPF programs, and the behavior all looks sane
and expected, that is nothing "leaking" from an upstream perspective.
The bpf_jit_limit knob was originally added in order to avoid a situation
where unprivileged applications loading BPF programs (e.g. seccomp BPF
policies) consuming all the module memory space via BPF JIT such that loading
of kernel modules would be prevented. The default limit was defined back in
2018 and while good enough back then, we are generally seeing far more BPF
consumers today.
Adjust the limit for the BPF JIT pool from originally 1/4 to now 1/2 of the
module memory space to better reflect today's needs and avoid more users
running into potentially hard to debug issues.
Fixes: fdadd04931c2 ("bpf: fix bpf_jit_limit knob for PAGE_SIZE >= 64K")
Reported-by: Stephen Haynes <sh@synk.net>
Reported-by: Lefteris Alexakis <lefteris.alexakis@kpn.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://github.com/awslabs/amazon-eks-ami/issues/1179
Link: https://github.com/awslabs/amazon-eks-ami/issues/1219
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20230320143725.8394-1-daniel@iogearbox.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit ee92fa443358f4fc0017c1d0d325c27b37802504 upstream.
KASAN reported follow problem:
BUG: KASAN: use-after-free in lookup_rec
Read of size 8 at addr ffff000199270ff0 by task modprobe
CPU: 2 Comm: modprobe
Call trace:
kasan_report
__asan_load8
lookup_rec
ftrace_location
arch_check_ftrace_location
check_kprobe_address_safe
register_kprobe
When checking pg->records[pg->index - 1].ip in lookup_rec(), it can get a
pg which is newly added to ftrace_pages_start in ftrace_process_locs().
Before the first pg->index++, index is 0 and accessing pg->records[-1].ip
will cause this problem.
Don't check the ip when pg->index is 0.
Link: https://lore.kernel.org/linux-trace-kernel/20230309080230.36064-1-chenzhongjin@huawei.com
Cc: stable@vger.kernel.org
Fixes: 9644302e33 ("ftrace: Speed up search by skipping pages by address")
Suggested-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Chen Zhongjin <chenzhongjin@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 3e4272b9954094907f16861199728f14002fcaf6 ]
In a previous commit 7433632c9ff6, buffer, buffer->buffers and
buffer->buffers[cpu] in ring_buffer_wake_waiters() can be NULL,
and thus the related checks are added.
However, in the same call stack, these variables are also used in
ring_buffer_free_read_page():
tracing_buffers_release()
ring_buffer_wake_waiters(iter->array_buffer->buffer)
cpu_buffer = buffer->buffers[cpu] -> Add checks by previous commit
ring_buffer_free_read_page(iter->array_buffer->buffer)
cpu_buffer = buffer->buffers[cpu] -> No check
Thus, to avod possible null-pointer derefernces, the related checks
should be added.
These results are reported by a static tool designed by myself.
Link: https://lkml.kernel.org/r/20230113125501.760324-1-baijiaju1990@gmail.com
Reported-by: TOTE Robot <oslab@tsinghua.edu.cn>
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit f1c97a1b4ef709e3f066f82e3ba3108c3b133ae6 upstream.
When arch_prepare_optimized_kprobe calculating jump destination address,
it copies original instructions from jmp-optimized kprobe (see
__recover_optprobed_insn), and calculated based on length of original
instruction.
arch_check_optimized_kprobe does not check KPROBE_FLAG_OPTIMATED when
checking whether jmp-optimized kprobe exists.
As a result, setup_detour_execution may jump to a range that has been
overwritten by jump destination address, resulting in an inval opcode error.
For example, assume that register two kprobes whose addresses are
<func+9> and <func+11> in "func" function.
The original code of "func" function is as follows:
0xffffffff816cb5e9 <+9>: push %r12
0xffffffff816cb5eb <+11>: xor %r12d,%r12d
0xffffffff816cb5ee <+14>: test %rdi,%rdi
0xffffffff816cb5f1 <+17>: setne %r12b
0xffffffff816cb5f5 <+21>: push %rbp
1.Register the kprobe for <func+11>, assume that is kp1, corresponding optimized_kprobe is op1.
After the optimization, "func" code changes to:
0xffffffff816cc079 <+9>: push %r12
0xffffffff816cc07b <+11>: jmp 0xffffffffa0210000
0xffffffff816cc080 <+16>: incl 0xf(%rcx)
0xffffffff816cc083 <+19>: xchg %eax,%ebp
0xffffffff816cc084 <+20>: (bad)
0xffffffff816cc085 <+21>: push %rbp
Now op1->flags == KPROBE_FLAG_OPTIMATED;
2. Register the kprobe for <func+9>, assume that is kp2, corresponding optimized_kprobe is op2.
register_kprobe(kp2)
register_aggr_kprobe
alloc_aggr_kprobe
__prepare_optimized_kprobe
arch_prepare_optimized_kprobe
__recover_optprobed_insn // copy original bytes from kp1->optinsn.copied_insn,
// jump address = <func+14>
3. disable kp1:
disable_kprobe(kp1)
__disable_kprobe
...
if (p == orig_p || aggr_kprobe_disabled(orig_p)) {
ret = disarm_kprobe(orig_p, true) // add op1 in unoptimizing_list, not unoptimized
orig_p->flags |= KPROBE_FLAG_DISABLED; // op1->flags == KPROBE_FLAG_OPTIMATED | KPROBE_FLAG_DISABLED
...
4. unregister kp2
__unregister_kprobe_top
...
if (!kprobe_disabled(ap) && !kprobes_all_disarmed) {
optimize_kprobe(op)
...
if (arch_check_optimized_kprobe(op) < 0) // because op1 has KPROBE_FLAG_DISABLED, here not return
return;
p->kp.flags |= KPROBE_FLAG_OPTIMIZED; // now op2 has KPROBE_FLAG_OPTIMIZED
}
"func" code now is:
0xffffffff816cc079 <+9>: int3
0xffffffff816cc07a <+10>: push %rsp
0xffffffff816cc07b <+11>: jmp 0xffffffffa0210000
0xffffffff816cc080 <+16>: incl 0xf(%rcx)
0xffffffff816cc083 <+19>: xchg %eax,%ebp
0xffffffff816cc084 <+20>: (bad)
0xffffffff816cc085 <+21>: push %rbp
5. if call "func", int3 handler call setup_detour_execution:
if (p->flags & KPROBE_FLAG_OPTIMIZED) {
...
regs->ip = (unsigned long)op->optinsn.insn + TMPL_END_IDX;
...
}
The code for the destination address is
0xffffffffa021072c: push %r12
0xffffffffa021072e: xor %r12d,%r12d
0xffffffffa0210731: jmp 0xffffffff816cb5ee <func+14>
However, <func+14> is not a valid start instruction address. As a result, an error occurs.
Link: https://lore.kernel.org/all/20230216034247.32348-3-yangjihong1@huawei.com/
Fixes: f66c0447cca1 ("kprobes: Set unoptimized flag after unoptimizing code")
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Cc: stable@vger.kernel.org
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 868a6fc0ca2407622d2833adefe1c4d284766c4c upstream.
Since the following commit:
commit f66c0447cca1 ("kprobes: Set unoptimized flag after unoptimizing code")
modified the update timing of the KPROBE_FLAG_OPTIMIZED, a optimized_kprobe
may be in the optimizing or unoptimizing state when op.kp->flags
has KPROBE_FLAG_OPTIMIZED and op->list is not empty.
The __recover_optprobed_insn check logic is incorrect, a kprobe in the
unoptimizing state may be incorrectly determined as unoptimizing.
As a result, incorrect instructions are copied.
The optprobe_queued_unopt function needs to be exported for invoking in
arch directory.
Link: https://lore.kernel.org/all/20230216034247.32348-2-yangjihong1@huawei.com/
Fixes: f66c0447cca1 ("kprobes: Set unoptimized flag after unoptimizing code")
Cc: stable@vger.kernel.org
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 9f76d59173d9d146e96c66886b671c1915a5c5e5 ]
The nanosleep syscalls use the restart_block mechanism, with a quirk:
The `type` and `rmtp`/`compat_rmtp` fields are set up unconditionally on
syscall entry, while the rest of the restart_block is only set up in the
unlikely case that the syscall is actually interrupted by a signal (or
pseudo-signal) that doesn't have a signal handler.
If the restart_block was set up by a previous syscall (futex(...,
FUTEX_WAIT, ...) or poll()) and hasn't been invalidated somehow since then,
this will clobber some of the union fields used by futex_wait_restart() and
do_restart_poll().
If userspace afterwards wrongly calls the restart_syscall syscall,
futex_wait_restart()/do_restart_poll() will read struct fields that have
been clobbered.
This doesn't actually lead to anything particularly interesting because
none of the union fields contain trusted kernel data, and
futex(..., FUTEX_WAIT, ...) and poll() aren't syscalls where it makes much
sense to apply seccomp filters to their arguments.
So the current consequences are just of the "if userspace does bad stuff,
it can damage itself, and that's not a problem" flavor.
But still, it seems like a hazard for future developers, so invalidate the
restart_block when partly setting it up in the nanosleep syscalls.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20230105134403.754986-1-jannh@google.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 2d7f00b2f01301d6e41fd4a28030dab0442265be ]
The normal grace period's RCU CPU stall warnings are invoked from the
scheduling-clock interrupt handler, and can thus invoke smp_processor_id()
with impunity, which allows them to directly invoke dump_cpu_task().
In contrast, the expedited grace period's RCU CPU stall warnings are
invoked from process context, which causes the dump_cpu_task() function's
calls to smp_processor_id() to complain bitterly in debug kernels.
This commit therefore causes synchronize_rcu_expedited_wait() to disable
preemption around its call to dump_cpu_task().
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 9b00f1b78809309163dda2d044d9e94a3c0248a3 upstream.
Recently noticed that when mod32 with a known src reg of 0 is performed,
then the dst register is 32-bit truncated in verifier:
0: R1=ctx(id=0,off=0,imm=0) R10=fp0
0: (b7) r0 = 0
1: R0_w=inv0 R1=ctx(id=0,off=0,imm=0) R10=fp0
1: (b7) r1 = -1
2: R0_w=inv0 R1_w=inv-1 R10=fp0
2: (b4) w2 = -1
3: R0_w=inv0 R1_w=inv-1 R2_w=inv4294967295 R10=fp0
3: (9c) w1 %= w0
4: R0_w=inv0 R1_w=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2_w=inv4294967295 R10=fp0
4: (b7) r0 = 1
5: R0_w=inv1 R1_w=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2_w=inv4294967295 R10=fp0
5: (1d) if r1 == r2 goto pc+1
R0_w=inv1 R1_w=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2_w=inv4294967295 R10=fp0
6: R0_w=inv1 R1_w=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2_w=inv4294967295 R10=fp0
6: (b7) r0 = 2
7: R0_w=inv2 R1_w=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2_w=inv4294967295 R10=fp0
7: (95) exit
7: R0=inv1 R1=inv(id=0,umin_value=4294967295,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R2=inv4294967295 R10=fp0
7: (95) exit
However, as a runtime result, we get 2 instead of 1, meaning the dst
register does not contain (u32)-1 in this case. The reason is fairly
straight forward given the 0 test leaves the dst register as-is:
# ./bpftool p d x i 23
0: (b7) r0 = 0
1: (b7) r1 = -1
2: (b4) w2 = -1
3: (16) if w0 == 0x0 goto pc+1
4: (9c) w1 %= w0
5: (b7) r0 = 1
6: (1d) if r1 == r2 goto pc+1
7: (b7) r0 = 2
8: (95) exit
This was originally not an issue given the dst register was marked as
completely unknown (aka 64 bit unknown). However, after 468f6eafa6c4
("bpf: fix 32-bit ALU op verification") the verifier casts the register
output to 32 bit, and hence it becomes 32 bit unknown. Note that for
the case where the src register is unknown, the dst register is marked
64 bit unknown. After the fix, the register is truncated by the runtime
and the test passes:
# ./bpftool p d x i 23
0: (b7) r0 = 0
1: (b7) r1 = -1
2: (b4) w2 = -1
3: (16) if w0 == 0x0 goto pc+2
4: (9c) w1 %= w0
5: (05) goto pc+1
6: (bc) w1 = w1
7: (b7) r0 = 1
8: (1d) if r1 == r2 goto pc+1
9: (b7) r0 = 2
10: (95) exit
Semantics also match with {R,W}x mod{64,32} 0 -> {R,W}x. Invalid div
has always been {R,W}x div{64,32} 0 -> 0. Rewrites are as follows:
mod32: mod64:
(16) if w0 == 0x0 goto pc+2 (15) if r0 == 0x0 goto pc+1
(9c) w1 %= w0 (9f) r1 %= r0
(05) goto pc+1
(bc) w1 = w1
[Salvatore Bonaccorso: This is an earlier version based on work by
Daniel and John which does not rely on availability of the BPF_JMP32
instruction class. This means it is not even strictly a backport of the
upstream commit mentioned but based on Daniel's and John's work to
address the issue and was finalized by Thadeu Lima de Souza Cascardo.]
Fixes: 468f6eafa6c4 ("bpf: fix 32-bit ALU op verification")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: John Fastabend <john.fastabend@gmail.com>
Tested-by: Salvatore Bonaccorso <carnil@debian.org>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Edward Liaw <edliaw@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit e88b2c6e5a4d9ce30d75391e4d950da74bb2bd90 upstream.
While reviewing a different fix, John and I noticed an oddity in one of the
BPF program dumps that stood out, for example:
# bpftool p d x i 13
0: (b7) r0 = 808464450
1: (b4) w4 = 808464432
2: (bc) w0 = w0
3: (15) if r0 == 0x0 goto pc+1
4: (9c) w4 %= w0
[...]
In line 2 we noticed that the mov32 would 32 bit truncate the original src
register for the div/mod operation. While for the two operations the dst
register is typically marked unknown e.g. from adjust_scalar_min_max_vals()
the src register is not, and thus verifier keeps tracking original bounds,
simplified:
0: R1=ctx(id=0,off=0,imm=0) R10=fp0
0: (b7) r0 = -1
1: R0_w=invP-1 R1=ctx(id=0,off=0,imm=0) R10=fp0
1: (b7) r1 = -1
2: R0_w=invP-1 R1_w=invP-1 R10=fp0
2: (3c) w0 /= w1
3: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R1_w=invP-1 R10=fp0
3: (77) r1 >>= 32
4: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R1_w=invP4294967295 R10=fp0
4: (bf) r0 = r1
5: R0_w=invP4294967295 R1_w=invP4294967295 R10=fp0
5: (95) exit
processed 6 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0
Runtime result of r0 at exit is 0 instead of expected -1. Remove the
verifier mov32 src rewrite in div/mod and replace it with a jmp32 test
instead. After the fix, we result in the following code generation when
having dividend r1 and divisor r6:
div, 64 bit: div, 32 bit:
0: (b7) r6 = 8 0: (b7) r6 = 8
1: (b7) r1 = 8 1: (b7) r1 = 8
2: (55) if r6 != 0x0 goto pc+2 2: (56) if w6 != 0x0 goto pc+2
3: (ac) w1 ^= w1 3: (ac) w1 ^= w1
4: (05) goto pc+1 4: (05) goto pc+1
5: (3f) r1 /= r6 5: (3c) w1 /= w6
6: (b7) r0 = 0 6: (b7) r0 = 0
7: (95) exit 7: (95) exit
mod, 64 bit: mod, 32 bit:
0: (b7) r6 = 8 0: (b7) r6 = 8
1: (b7) r1 = 8 1: (b7) r1 = 8
2: (15) if r6 == 0x0 goto pc+1 2: (16) if w6 == 0x0 goto pc+1
3: (9f) r1 %= r6 3: (9c) w1 %= w6
4: (b7) r0 = 0 4: (b7) r0 = 0
5: (95) exit 5: (95) exit
x86 in particular can throw a 'divide error' exception for div
instruction not only for divisor being zero, but also for the case
when the quotient is too large for the designated register. For the
edx:eax and rdx:rax dividend pair it is not an issue in x86 BPF JIT
since we always zero edx (rdx). Hence really the only protection
needed is against divisor being zero.
[Salvatore Bonaccorso: This is an earlier version of the patch provided
by Daniel Borkmann which does not rely on availability of the BPF_JMP32
instruction class. This means it is not even strictly a backport of the
upstream commit mentioned but based on Daniel's and John's work to
address the issue.]
Fixes: 68fda450a7df ("bpf: fix 32-bit divide by zero")
Co-developed-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Salvatore Bonaccorso <carnil@debian.org>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Edward Liaw <edliaw@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit f6b1b3bf0d5f681631a293cfe1ca934b81716f1e upstream.
One of the ugly leftovers from the early eBPF days is that div/mod
operations based on registers have a hard-coded src_reg == 0 test
in the interpreter as well as in JIT code generators that would
return from the BPF program with exit code 0. This was basically
adopted from cBPF interpreter for historical reasons.
There are multiple reasons why this is very suboptimal and prone
to bugs. To name one: the return code mapping for such abnormal
program exit of 0 does not always match with a suitable program
type's exit code mapping. For example, '0' in tc means action 'ok'
where the packet gets passed further up the stack, which is just
undesirable for such cases (e.g. when implementing policy) and
also does not match with other program types.
While trying to work out an exception handling scheme, I also
noticed that programs crafted like the following will currently
pass the verifier:
0: (bf) r6 = r1
1: (85) call pc+8
caller:
R6=ctx(id=0,off=0,imm=0) R10=fp0,call_-1
callee:
frame1: R1=ctx(id=0,off=0,imm=0) R10=fp0,call_1
10: (b4) (u32) r2 = (u32) 0
11: (b4) (u32) r3 = (u32) 1
12: (3c) (u32) r3 /= (u32) r2
13: (61) r0 = *(u32 *)(r1 +76)
14: (95) exit
returning from callee:
frame1: R0_w=pkt(id=0,off=0,r=0,imm=0)
R1=ctx(id=0,off=0,imm=0) R2_w=inv0
R3_w=inv(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff))
R10=fp0,call_1
to caller at 2:
R0_w=pkt(id=0,off=0,r=0,imm=0) R6=ctx(id=0,off=0,imm=0)
R10=fp0,call_-1
from 14 to 2: R0=pkt(id=0,off=0,r=0,imm=0)
R6=ctx(id=0,off=0,imm=0) R10=fp0,call_-1
2: (bf) r1 = r6
3: (61) r1 = *(u32 *)(r1 +80)
4: (bf) r2 = r0
5: (07) r2 += 8
6: (2d) if r2 > r1 goto pc+1
R0=pkt(id=0,off=0,r=8,imm=0) R1=pkt_end(id=0,off=0,imm=0)
R2=pkt(id=0,off=8,r=8,imm=0) R6=ctx(id=0,off=0,imm=0)
R10=fp0,call_-1
7: (71) r0 = *(u8 *)(r0 +0)
8: (b7) r0 = 1
9: (95) exit
from 6 to 8: safe
processed 16 insns (limit 131072), stack depth 0+0
Basically what happens is that in the subprog we make use of a
div/mod by 0 exception and in the 'normal' subprog's exit path
we just return skb->data back to the main prog. This has the
implication that the verifier thinks we always get a pkt pointer
in R0 while we still have the implicit 'return 0' from the div
as an alternative unconditional return path earlier. Thus, R0
then contains 0, meaning back in the parent prog we get the
address range of [0x0, skb->data_end] as read and writeable.
Similar can be crafted with other pointer register types.
Since i) BPF_ABS/IND is not allowed in programs that contain
BPF to BPF calls (and generally it's also disadvised to use in
native eBPF context), ii) unknown opcodes don't return zero
anymore, iii) we don't return an exception code in dead branches,
the only last missing case affected and to fix is the div/mod
handling.
What we would really need is some infrastructure to propagate
exceptions all the way to the original prog unwinding the
current stack and returning that code to the caller of the
BPF program. In user space such exception handling for similar
runtimes is typically implemented with setjmp(3) and longjmp(3)
as one possibility which is not available in the kernel,
though (kgdb used to implement it in kernel long time ago). I
implemented a PoC exception handling mechanism into the BPF
interpreter with porting setjmp()/longjmp() into x86_64 and
adding a new internal BPF_ABRT opcode that can use a program
specific exception code for all exception cases we have (e.g.
div/mod by 0, unknown opcodes, etc). While this seems to work
in the constrained BPF environment (meaning, here, we don't
need to deal with state e.g. from memory allocations that we
would need to undo before going into exception state), it still
has various drawbacks: i) we would need to implement the
setjmp()/longjmp() for every arch supported in the kernel and
for x86_64, arm64, sparc64 JITs currently supporting calls,
ii) it has unconditional additional cost on main program
entry to store CPU register state in initial setjmp() call,
and we would need some way to pass the jmp_buf down into
___bpf_prog_run() for main prog and all subprogs, but also
storing on stack is not really nice (other option would be
per-cpu storage for this, but it also has the drawback that
we need to disable preemption for every BPF program types).
All in all this approach would add a lot of complexity.
Another poor-man's solution would be to have some sort of
additional shared register or scratch buffer to hold state
for exceptions, and test that after every call return to
chain returns and pass R0 all the way down to BPF prog caller.
This is also problematic in various ways: i) an additional
register doesn't map well into JITs, and some other scratch
space could only be on per-cpu storage, which, again has the
side-effect that this only works when we disable preemption,
or somewhere in the input context which is not available
everywhere either, and ii) this adds significant runtime
overhead by putting conditionals after each and every call,
as well as implementation complexity.
Yet another option is to teach verifier that div/mod can
return an integer, which however is also complex to implement
as verifier would need to walk such fake 'mov r0,<code>; exit;'
sequeuence and there would still be no guarantee for having
propagation of this further down to the BPF caller as proper
exception code. For parent prog, it is also is not distinguishable
from a normal return of a constant scalar value.
The approach taken here is a completely different one with
little complexity and no additional overhead involved in
that we make use of the fact that a div/mod by 0 is undefined
behavior. Instead of bailing out, we adapt the same behavior
as on some major archs like ARMv8 [0] into eBPF as well:
X div 0 results in 0, and X mod 0 results in X. aarch64 and
aarch32 ISA do not generate any traps or otherwise aborts
of program execution for unsigned divides. I verified this
also with a test program compiled by gcc and clang, and the
behavior matches with the spec. Going forward we adapt the
eBPF verifier to emit such rewrites once div/mod by register
was seen. cBPF is not touched and will keep existing 'return 0'
semantics. Given the options, it seems the most suitable from
all of them, also since major archs have similar schemes in
place. Given this is all in the realm of undefined behavior,
we still have the option to adapt if deemed necessary and
this way we would also have the option of more flexibility
from LLVM code generation side (which is then fully visible
to verifier). Thus, this patch i) fixes the panic seen in
above program and ii) doesn't bypass the verifier observations.
[0] ARM Architecture Reference Manual, ARMv8 [ARM DDI 0487B.b]
http://infocenter.arm.com/help/topic/com.arm.doc.ddi0487b.b/DDI0487B_b_armv8_arm.pdf
1) aarch64 instruction set: section C3.4.7 and C6.2.279 (UDIV)
"A division by zero results in a zero being written to
the destination register, without any indication that
the division by zero occurred."
2) aarch32 instruction set: section F1.4.8 and F5.1.263 (UDIV)
"For the SDIV and UDIV instructions, division by zero
always returns a zero result."
Fixes: f4d7e40a5b71 ("bpf: introduce function calls (verification)")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Edward Liaw <edliaw@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit c348d806ed1d3075af52345344243824d72c4945 upstream.
Partially undo old commit 144cd91c4c2b ("bpf: move tmp variable into ax
register in interpreter"). The reason we need this here is because ax
register will be used for holding temporary state for div/mod instruction
which otherwise interpreter would corrupt. This will cause a small +8 byte
stack increase for interpreter, but with the gain that we can use it from
verifier rewrites as scratch register.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: John Fastabend <john.fastabend@gmail.com>
[cascardo: This partial revert is needed in order to support using AX for
the following two commits, as there is no JMP32 on 4.19.y]
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
[edliaw: Removed redeclaration of tmp introduced by patch differences
between 4.14 and 4.19]
Signed-off-by: Edward Liaw <edliaw@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d125d1349abeb46945dc5e98f7824bf688266f13 upstream.
syzbot reported a RCU stall which is caused by setting up an alarmtimer
with a very small interval and ignoring the signal. The reproducer arms the
alarm timer with a relative expiry of 8ns and an interval of 9ns. Not a
problem per se, but that's an issue when the signal is ignored because then
the timer is immediately rearmed because there is no way to delay that
rearming to the signal delivery path. See posix_timer_fn() and commit
58229a1899 ("posix-timers: Prevent softirq starvation by small intervals
and SIG_IGN") for details.
The reproducer does not set SIG_IGN explicitely, but it sets up the timers
signal with SIGCONT. That has the same effect as explicitely setting
SIG_IGN for a signal as SIGCONT is ignored if there is no handler set and
the task is not ptraced.
The log clearly shows that:
[pid 5102] --- SIGCONT {si_signo=SIGCONT, si_code=SI_TIMER, si_timerid=0, si_overrun=316014, si_int=0, si_ptr=NULL} ---
It works because the tasks are traced and therefore the signal is queued so
the tracer can see it, which delays the restart of the timer to the signal
delivery path. But then the tracer is killed:
[pid 5087] kill(-5102, SIGKILL <unfinished ...>
...
./strace-static-x86_64: Process 5107 detached
and after it's gone the stall can be observed:
syzkaller login: [ 79.439102][ C0] hrtimer: interrupt took 68471 ns
[ 184.460538][ C1] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
...
[ 184.658237][ C1] rcu: Stack dump where RCU GP kthread last ran:
[ 184.664574][ C1] Sending NMI from CPU 1 to CPUs 0:
[ 184.669821][ C0] NMI backtrace for cpu 0
[ 184.669831][ C0] CPU: 0 PID: 5108 Comm: syz-executor192 Not tainted 6.2.0-rc6-next-20230203-syzkaller #0
...
[ 184.670036][ C0] Call Trace:
[ 184.670041][ C0] <IRQ>
[ 184.670045][ C0] alarmtimer_fired+0x327/0x670
posix_timer_fn() prevents that by checking whether the interval for
timers which have the signal ignored is smaller than a jiffie and
artifically delay it by shifting the next expiry out by a jiffie. That's
accurate vs. the overrun accounting, but slightly inaccurate
vs. timer_gettimer(2).
The comment in that function says what needs to be done and there was a fix
available for the regular userspace induced SIG_IGN mechanism, but that did
not work due to the implicit ignore for SIGCONT and similar signals. This
needs to be worked on, but for now the only available workaround is to do
exactly what posix_timer_fn() does:
Increase the interval of self-rearming timers, which have their signal
ignored, to at least a jiffie.
Interestingly this has been fixed before via commit ff86bf0c65
("alarmtimer: Rate limit periodic intervals") already, but that fix got
lost in a later rework.
Reported-by: syzbot+b9564ba6e8e00694511b@syzkaller.appspotmail.com
Fixes: f2c45807d3 ("alarmtimer: Switch over to generic set/get/rearm routine")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/87k00q1no2.ffs@tglx
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
