Although we can rely on cpuacct to present the CPU usage of task
groups, it is hard to tell how intense the competition is between
these groups on CPU resources.
Monitoring the wait time or sched_debug of each process could be
very expensive, and there is no good way to accurately represent the
conflict with these info, we need the wait time on group dimension.
Thus we introduce group's wait_sum to represent the resource conflict
between task groups, which is simply the sum of the wait time of
the group's cfs_rq.
The 'cpu.stat' is modified to show the statistic, like:
nr_periods 0
nr_throttled 0
throttled_time 0
wait_sum 2035098795584
Now we can monitor the changes of wait_sum to tell how much a
a task group is suffering in the fight of CPU resources.
For example:
(wait_sum - last_wait_sum) * 100 / (nr_cpu * period_ns) == X%
means the task group paid X percentage of period on waiting
for the CPU.
Signed-off-by: Michael Wang <yun.wang@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/ff7dae3b-e5f9-7157-1caa-ff02c6b23dc1@linux.alibaba.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
NO_RT_RUNTIME_SHARE feature is used to prevent a CPU borrow enough
runtime with a spin-rt-task.
However, if RT_RUNTIME_SHARE feature is enabled and rt_rq has borrowd
enough rt_runtime at the beginning, rt_runtime can't be restored to
its initial bandwidth rt_runtime after we disable RT_RUNTIME_SHARE.
E.g. on my PC with 4 cores, procedure to reproduce:
1) Make sure RT_RUNTIME_SHARE is enabled
cat /sys/kernel/debug/sched_features
GENTLE_FAIR_SLEEPERS START_DEBIT NO_NEXT_BUDDY LAST_BUDDY
CACHE_HOT_BUDDY WAKEUP_PREEMPTION NO_HRTICK NO_DOUBLE_TICK
LB_BIAS NONTASK_CAPACITY TTWU_QUEUE NO_SIS_AVG_CPU SIS_PROP
NO_WARN_DOUBLE_CLOCK RT_PUSH_IPI RT_RUNTIME_SHARE NO_LB_MIN
ATTACH_AGE_LOAD WA_IDLE WA_WEIGHT WA_BIAS
2) Start a spin-rt-task
./loop_rr &
3) set affinity to the last cpu
taskset -p 8 $pid_of_loop_rr
4) Observe that last cpu have borrowed enough runtime.
cat /proc/sched_debug | grep rt_runtime
.rt_runtime : 950.000000
.rt_runtime : 900.000000
.rt_runtime : 950.000000
.rt_runtime : 1000.000000
5) Disable RT_RUNTIME_SHARE
echo NO_RT_RUNTIME_SHARE > /sys/kernel/debug/sched_features
6) Observe that rt_runtime can not been restored
cat /proc/sched_debug | grep rt_runtime
.rt_runtime : 950.000000
.rt_runtime : 900.000000
.rt_runtime : 950.000000
.rt_runtime : 1000.000000
This patch help to restore rt_runtime after we disable
RT_RUNTIME_SHARE.
Signed-off-by: Hailong Liu <liu.hailong6@zte.com.cn>
Signed-off-by: Jiang Biao <jiang.biao2@zte.com.cn>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: zhong.weidong@zte.com.cn
Link: http://lkml.kernel.org/r/1531874815-39357-1-git-send-email-liu.hailong6@zte.com.cn
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This commit:
9fb8d5dc4b ("stop_machine, Disable preemption when waking two stopper threads")
does not fully address the race condition that can occur
as follows:
On one CPU, call it CPU 3, thread 1 invokes
cpu_stop_queue_two_works(2, 3,...), and the execution is such
that thread 1 queues the works for migration/2 and migration/3,
and is preempted after releasing the locks for migration/2 and
migration/3, but before waking the threads.
Then, On CPU 2, a kworker, call it thread 2, is running,
and it invokes cpu_stop_queue_two_works(1, 2,...), such that
thread 2 queues the works for migration/1 and migration/2.
Meanwhile, on CPU 3, thread 1 resumes execution, and wakes
migration/2 and migration/3. This means that when CPU 2
releases the locks for migration/1 and migration/2, but before
it wakes those threads, it can be preempted by migration/2.
If thread 2 is preempted by migration/2, then migration/2 will
execute the first work item successfully, since migration/3
was woken up by CPU 3, but when it goes to execute the second
work item, it disables preemption, calls multi_cpu_stop(),
and thus, CPU 2 will wait forever for migration/1, which should
have been woken up by thread 2. However migration/1 cannot be
woken up by thread 2, since it is a kworker, so it is affine to
CPU 2, but CPU 2 is running migration/2 with preemption
disabled, so thread 2 will never run.
Disable preemption after queueing works for stopper threads
to ensure that the operation of queueing the works and waking
the stopper threads is atomic.
Co-Developed-by: Prasad Sodagudi <psodagud@codeaurora.org>
Co-Developed-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Signed-off-by: Prasad Sodagudi <psodagud@codeaurora.org>
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bigeasy@linutronix.de
Cc: gregkh@linuxfoundation.org
Cc: matt@codeblueprint.co.uk
Fixes: 9fb8d5dc4b ("stop_machine, Disable preemption when waking two stopper threads")
Link: http://lkml.kernel.org/r/1531856129-9871-1-git-send-email-isaacm@codeaurora.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Ingo Molnar:
"Two fixes: a stop-machine preemption fix and a SCHED_DEADLINE fix"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/deadline: Fix switched_from_dl() warning
stop_machine: Disable preemption when waking two stopper threads
Like vm_area_dup(), it initializes the anon_vma_chain head, and the
basic mm pointer.
The rest of the fields end up being different for different users,
although the plan is to also initialize the 'vm_ops' field to a dummy
entry.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
.. and re-initialize th eanon_vma_chain head.
This removes some boiler-plate from the users, and also makes it clear
why it didn't need use the 'zalloc()' version.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The vm_area_struct is one of the most fundamental memory management
objects, but the management of it is entirely open-coded evertwhere,
ranging from allocation and freeing (using kmem_cache_[z]alloc and
kmem_cache_free) to initializing all the fields.
We want to unify this in order to end up having some unified
initialization of the vmas, and the first step to this is to at least
have basic allocation functions.
Right now those functions are literally just wrappers around the
kmem_cache_*() calls. This is a purely mechanical conversion:
# new vma:
kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL) -> vm_area_alloc()
# copy old vma
kmem_cache_alloc(vm_area_cachep, GFP_KERNEL) -> vm_area_dup(old)
# free vma
kmem_cache_free(vm_area_cachep, vma) -> vm_area_free(vma)
to the point where the old vma passed in to the vm_area_dup() function
isn't even used yet (because I've left all the old manual initialization
alone).
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull networking fixes from David Miller:
"Lots of fixes, here goes:
1) NULL deref in qtnfmac, from Gustavo A. R. Silva.
2) Kernel oops when fw download fails in rtlwifi, from Ping-Ke Shih.
3) Lost completion messages in AF_XDP, from Magnus Karlsson.
4) Correct bogus self-assignment in rhashtable, from Rishabh
Bhatnagar.
5) Fix regression in ipv6 route append handling, from David Ahern.
6) Fix masking in __set_phy_supported(), from Heiner Kallweit.
7) Missing module owner set in x_tables icmp, from Florian Westphal.
8) liquidio's timeouts are HZ dependent, fix from Nicholas Mc Guire.
9) Link setting fixes for sh_eth and ravb, from Vladimir Zapolskiy.
10) Fix NULL deref when using chains in act_csum, from Davide Caratti.
11) XDP_REDIRECT needs to check if the interface is up and whether the
MTU is sufficient. From Toshiaki Makita.
12) Net diag can do a double free when killing TCP_NEW_SYN_RECV
connections, from Lorenzo Colitti.
13) nf_defrag in ipv6 can unnecessarily hold onto dst entries for a
full minute, delaying device unregister. From Eric Dumazet.
14) Update MAC entries in the correct order in ixgbe, from Alexander
Duyck.
15) Don't leave partial mangles bpf program in jit_subprogs, from
Daniel Borkmann.
16) Fix pfmemalloc SKB state propagation, from Stefano Brivio.
17) Fix ACK handling in DCTCP congestion control, from Yuchung Cheng.
18) Use after free in tun XDP_TX, from Toshiaki Makita.
19) Stale ipv6 header pointer in ipv6 gre code, from Prashant Bhole.
20) Don't reuse remainder of RX page when XDP is set in mlx4, from
Saeed Mahameed.
21) Fix window probe handling of TCP rapair sockets, from Stefan
Baranoff.
22) Missing socket locking in smc_ioctl(), from Ursula Braun.
23) IPV6_ILA needs DST_CACHE, from Arnd Bergmann.
24) Spectre v1 fix in cxgb3, from Gustavo A. R. Silva.
25) Two spots in ipv6 do a rol32() on a hash value but ignore the
result. Fixes from Colin Ian King"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (176 commits)
tcp: identify cryptic messages as TCP seq # bugs
ptp: fix missing break in switch
hv_netvsc: Fix napi reschedule while receive completion is busy
MAINTAINERS: Drop inactive Vitaly Bordug's email
net: cavium: Add fine-granular dependencies on PCI
net: qca_spi: Fix log level if probe fails
net: qca_spi: Make sure the QCA7000 reset is triggered
net: qca_spi: Avoid packet drop during initial sync
ipv6: fix useless rol32 call on hash
ipv6: sr: fix useless rol32 call on hash
net: sched: Using NULL instead of plain integer
net: usb: asix: replace mii_nway_restart in resume path
net: cxgb3_main: fix potential Spectre v1
lib/rhashtable: consider param->min_size when setting initial table size
net/smc: reset recv timeout after clc handshake
net/smc: add error handling for get_user()
net/smc: optimize consumer cursor updates
net/nfc: Avoid stalls when nfc_alloc_send_skb() returned NULL.
ipv6: ila: select CONFIG_DST_CACHE
net: usb: rtl8150: demote allmulti message to dev_dbg()
...
Way back in 4.9, we committed 4cd13c21b2 ("softirq: Let ksoftirqd do
its job"), and ever since we've had small nagging issues with it. For
example, we've had:
1ff688209e ("watchdog: core: make sure the watchdog_worker is not deferred")
8d5755b3f7 ("watchdog: softdog: fire watchdog even if softirqs do not get to run")
217f697436 ("net: busy-poll: allow preemption in sk_busy_loop()")
all of which worked around some of the effects of that commit.
The DVB people have also complained that the commit causes excessive USB
URB latencies, which seems to be due to the USB code using tasklets to
schedule USB traffic. This seems to be an issue mainly when already
living on the edge, but waiting for ksoftirqd to handle it really does
seem to cause excessive latencies.
Now Hanna Hawa reports that this issue isn't just limited to USB URB and
DVB, but also causes timeout problems for the Marvell SoC team:
"I'm facing kernel panic issue while running raid 5 on sata disks
connected to Macchiatobin (Marvell community board with Armada-8040
SoC with 4 ARMv8 cores of CA72) Raid 5 built with Marvell DMA engine
and async_tx mechanism (ASYNC_TX_DMA [=y]); the DMA driver (mv_xor_v2)
uses a tasklet to clean the done descriptors from the queue"
The latency problem causes a panic:
mv_xor_v2 f0400000.xor: dma_sync_wait: timeout!
Kernel panic - not syncing: async_tx_quiesce: DMA error waiting for transaction
We've discussed simply just reverting the original commit entirely, and
also much more involved solutions (with per-softirq threads etc). This
patch is intentionally stupid and fairly limited, because the issue
still remains, and the other solutions either got sidetracked or had
other issues.
We should probably also consider the timer softirqs to be synchronous
and not be delayed to ksoftirqd (since they were the issue with the
earlier watchdog problems), but that should be done as a separate patch.
This does only the tasklet cases.
Reported-and-tested-by: Hanna Hawa <hannah@marvell.com>
Reported-and-tested-by: Josef Griebichler <griebichler.josef@gmx.at>
Reported-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
get_cpu() disables preemption for the entire sched_fork() function.
This get_cpu() was introduced in commit:
dd41f596cd ("sched: cfs core code")
... which also invoked sched_balance_self() and this function
required preemption do be off.
Today, sched_balance_self() seems to be moved to ->task_fork callback
which is invoked while the ->pi_lock is held.
set_load_weight() could invoke reweight_task() which then via $callchain
might end up in smp_processor_id() but since `update_load' is false
this won't happen.
I didn't find any this_cpu*() or similar usage during the initialisation
of the task_struct.
The `cpu' value (from get_cpu()) is only used later in __set_task_cpu()
while the ->pi_lock lock is held.
Based on this it is possible to remove get_cpu() and use
smp_processor_id() for the `cpu' variable without breaking anything.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20180706130615.g2ex2kmfu5kcvlq6@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Mark noticed that syzkaller is able to reliably trigger the following warning:
dl_rq->running_bw > dl_rq->this_bw
WARNING: CPU: 1 PID: 153 at kernel/sched/deadline.c:124 switched_from_dl+0x454/0x608
Kernel panic - not syncing: panic_on_warn set ...
CPU: 1 PID: 153 Comm: syz-executor253 Not tainted 4.18.0-rc3+ #29
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x0/0x458
show_stack+0x20/0x30
dump_stack+0x180/0x250
panic+0x2dc/0x4ec
__warn_printk+0x0/0x150
report_bug+0x228/0x2d8
bug_handler+0xa0/0x1a0
brk_handler+0x2f0/0x568
do_debug_exception+0x1bc/0x5d0
el1_dbg+0x18/0x78
switched_from_dl+0x454/0x608
__sched_setscheduler+0x8cc/0x2018
sys_sched_setattr+0x340/0x758
el0_svc_naked+0x30/0x34
syzkaller reproducer runs a bunch of threads that constantly switch
between DEADLINE and NORMAL classes while interacting through futexes.
The splat above is caused by the fact that if a DEADLINE task is setattr
back to NORMAL while in non_contending state (blocked on a futex -
inactive timer armed), its contribution to running_bw is not removed
before sub_rq_bw() gets called (!task_on_rq_queued() branch) and the
latter sees running_bw > this_bw.
Fix it by removing a task contribution from running_bw if the task is
not queued and in non_contending state while switched to a different
class.
Reported-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Juri Lelli <juri.lelli@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Bristot de Oliveira <bristot@redhat.com>
Reviewed-by: Luca Abeni <luca.abeni@santannapisa.it>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: claudio@evidence.eu.com
Cc: rostedt@goodmis.org
Link: http://lkml.kernel.org/r/20180711072948.27061-1-juri.lelli@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When cpu_stop_queue_two_works() begins to wake the stopper threads, it does
so without preemption disabled, which leads to the following race
condition:
The source CPU calls cpu_stop_queue_two_works(), with cpu1 as the source
CPU, and cpu2 as the destination CPU. When adding the stopper threads to
the wake queue used in this function, the source CPU stopper thread is
added first, and the destination CPU stopper thread is added last.
When wake_up_q() is invoked to wake the stopper threads, the threads are
woken up in the order that they are queued in, so the source CPU's stopper
thread is woken up first, and it preempts the thread running on the source
CPU.
The stopper thread will then execute on the source CPU, disable preemption,
and begin executing multi_cpu_stop(), and wait for an ack from the
destination CPU's stopper thread, with preemption still disabled. Since the
worker thread that woke up the stopper thread on the source CPU is affine
to the source CPU, and preemption is disabled on the source CPU, that
thread will never run to dequeue the destination CPU's stopper thread from
the wake queue, and thus, the destination CPU's stopper thread will never
run, causing the source CPU's stopper thread to wait forever, and stall.
Disable preemption when waking the stopper threads in
cpu_stop_queue_two_works().
Fixes: 0b26351b91 ("stop_machine, sched: Fix migrate_swap() vs. active_balance() deadlock")
Co-Developed-by: Prasad Sodagudi <psodagud@codeaurora.org>
Signed-off-by: Prasad Sodagudi <psodagud@codeaurora.org>
Co-Developed-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: peterz@infradead.org
Cc: matt@codeblueprint.co.uk
Cc: bigeasy@linutronix.de
Cc: gregkh@linuxfoundation.org
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1530655334-4601-1-git-send-email-isaacm@codeaurora.org
Pull timer fixes from Ingo Molnar:
"A clocksource driver fix and a revert"
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
clocksource: arm_arch_timer: Set arch_mem_timer cpumask to cpu_possible_mask
Revert "tick: Prefer a lower rating device only if it's CPU local device"
Pull rseq fixes from Ingo Molnar:
"Various rseq ABI fixes and cleanups: use get_user()/put_user(),
validate parameters and use proper uapi types, etc"
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
rseq/selftests: cleanup: Update comment above rseq_prepare_unload
rseq: Remove unused types_32_64.h uapi header
rseq: uapi: Declare rseq_cs field as union, update includes
rseq: uapi: Update uapi comments
rseq: Use get_user/put_user rather than __get_user/__put_user
rseq: Use __u64 for rseq_cs fields, validate user inputs
Pull tracing fixlet from Steven Rostedt:
"Joel Fernandes asked to add a feature in tracing that Android had its
own patch internally for. I took it back in 4.13. Now he realizes that
he had a mistake, and swapped the values from what Android had. This
means that the old Android tools will break when using a new kernel
that has the new feature on it.
The options are:
1. To swap it back to what Android wants.
2. Add a command line option or something to do the swap
3. Just let Android carry a patch that swaps it back
Since it requires setting a tracing option to enable this anyway, I
doubt there are other users of this than Android. Thus, I've decided
to take option 1. If someone else is actually depending on the order
that is in the kernel, then we will have to revert this change and go
to option 2 or 3"
* tag 'trace-v4.18-rc3-3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Reorder display of TGID to be after PID
Currently ftrace displays data in trace output like so:
_-----=> irqs-off
/ _----=> need-resched
| / _---=> hardirq/softirq
|| / _--=> preempt-depth
||| / delay
TASK-PID CPU TGID |||| TIMESTAMP FUNCTION
| | | | |||| | |
bash-1091 [000] ( 1091) d..2 28.313544: sched_switch:
However Android's trace visualization tools expect a slightly different
format due to an out-of-tree patch patch that was been carried for a
decade, notice that the TGID and CPU fields are reversed:
_-----=> irqs-off
/ _----=> need-resched
| / _---=> hardirq/softirq
|| / _--=> preempt-depth
||| / delay
TASK-PID TGID CPU |||| TIMESTAMP FUNCTION
| | | | |||| | |
bash-1091 ( 1091) [002] d..2 64.965177: sched_switch:
From kernel v4.13 onwards, during which TGID was introduced, tracing
with systrace on all Android kernels will break (most Android kernels
have been on 4.9 with Android patches, so this issues hasn't been seen
yet). From v4.13 onwards things will break.
The chrome browser's tracing tools also embed the systrace viewer which
uses the legacy TGID format and updates to that are known to be
difficult to make.
Considering this, I suggest we make this change to the upstream kernel
and backport it to all Android kernels. I believe this feature is merged
recently enough into the upstream kernel that it shouldn't be a problem.
Also logically, IMO it makes more sense to group the TGID with the
TASK-PID and the CPU after these.
Link: http://lkml.kernel.org/r/20180626000822.113931-1-joel@joelfernandes.org
Cc: jreck@google.com
Cc: tkjos@google.com
Cc: stable@vger.kernel.org
Fixes: 441dae8f2f ("tracing: Add support for display of tgid in trace output")
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
syzkaller managed to trigger the following bug through fault injection:
[...]
[ 141.043668] verifier bug. No program starts at insn 3
[ 141.044648] WARNING: CPU: 3 PID: 4072 at kernel/bpf/verifier.c:1613
get_callee_stack_depth kernel/bpf/verifier.c:1612 [inline]
[ 141.044648] WARNING: CPU: 3 PID: 4072 at kernel/bpf/verifier.c:1613
fixup_call_args kernel/bpf/verifier.c:5587 [inline]
[ 141.044648] WARNING: CPU: 3 PID: 4072 at kernel/bpf/verifier.c:1613
bpf_check+0x525e/0x5e60 kernel/bpf/verifier.c:5952
[ 141.047355] CPU: 3 PID: 4072 Comm: a.out Not tainted 4.18.0-rc4+ #51
[ 141.048446] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),BIOS 1.10.2-1 04/01/2014
[ 141.049877] Call Trace:
[ 141.050324] __dump_stack lib/dump_stack.c:77 [inline]
[ 141.050324] dump_stack+0x1c9/0x2b4 lib/dump_stack.c:113
[ 141.050950] ? dump_stack_print_info.cold.2+0x52/0x52 lib/dump_stack.c:60
[ 141.051837] panic+0x238/0x4e7 kernel/panic.c:184
[ 141.052386] ? add_taint.cold.5+0x16/0x16 kernel/panic.c:385
[ 141.053101] ? __warn.cold.8+0x148/0x1ba kernel/panic.c:537
[ 141.053814] ? __warn.cold.8+0x117/0x1ba kernel/panic.c:530
[ 141.054506] ? get_callee_stack_depth kernel/bpf/verifier.c:1612 [inline]
[ 141.054506] ? fixup_call_args kernel/bpf/verifier.c:5587 [inline]
[ 141.054506] ? bpf_check+0x525e/0x5e60 kernel/bpf/verifier.c:5952
[ 141.055163] __warn.cold.8+0x163/0x1ba kernel/panic.c:538
[ 141.055820] ? get_callee_stack_depth kernel/bpf/verifier.c:1612 [inline]
[ 141.055820] ? fixup_call_args kernel/bpf/verifier.c:5587 [inline]
[ 141.055820] ? bpf_check+0x525e/0x5e60 kernel/bpf/verifier.c:5952
[...]
What happens in jit_subprogs() is that kcalloc() for the subprog func
buffer is failing with NULL where we then bail out. Latter is a plain
return -ENOMEM, and this is definitely not okay since earlier in the
loop we are walking all subprogs and temporarily rewrite insn->off to
remember the subprog id as well as insn->imm to temporarily point the
call to __bpf_call_base + 1 for the initial JIT pass. Thus, bailing
out in such state and handing this over to the interpreter is troublesome
since later/subsequent e.g. find_subprog() lookups are based on wrong
insn->imm.
Therefore, once we hit this point, we need to jump to out_free path
where we undo all changes from earlier loop, so that interpreter can
work on unmodified insn->{off,imm}.
Another point is that should find_subprog() fail in jit_subprogs() due
to a verifier bug, then we also should not simply defer the program to
the interpreter since also here we did partial modifications. Instead
we should just bail out entirely and return an error to the user who is
trying to load the program.
Fixes: 1c2a088a66 ("bpf: x64: add JIT support for multi-function programs")
Reported-by: syzbot+7d427828b2ea6e592804@syzkaller.appspotmail.com
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
When extracting bitfield from a number, btf_int_bits_seq_show() builds
a mask and accesses least significant byte of the number in a way
specific to little-endian. This patch fixes that by checking endianness
of the machine and then shifting left and right the unneeded bits.
Thanks to Martin Lau for the help in navigating potential pitfalls when
dealing with endianess and for the final solution.
Fixes: b00b8daec8 ("bpf: btf: Add pretty print capability for data with BTF type info")
Signed-off-by: Okash Khawaja <osk@fb.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Pull kprobe fix from Steven Rostedt:
"This fixes a memory leak in the kprobe code"
* tag 'trace-v4.18-rc3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing/kprobe: Release kprobe print_fmt properly
Pull scheduler fixes from Thomas Gleixner:
- The hopefully final fix for the reported race problems in
kthread_parkme(). The previous attempt still left a hole and was
partially wrong.
- Plug a race in the remote tick mechanism which triggers a warning
about updates not being done correctly. That's a false positive if
the race condition is hit as the remote CPU is idle. Plug it by
checking the condition again when holding run queue lock.
- Fix a bug in the utilization estimation of a run queue which causes
the estimation to be 0 when a run queue is throttled.
- Advance the global expiration of the period timer when the timer is
restarted after a idle period. Otherwise the expiry time is stale and
the timer fires prematurely.
- Cure the drift between the bandwidth timer and the runqueue
accounting, which leads to bogus throttling of runqueues
- Place the call to cpufreq_update_util() correctly so the function
will observe the correct number of running RT tasks and not a stale
one.
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
kthread, sched/core: Fix kthread_parkme() (again...)
sched/util_est: Fix util_est_dequeue() for throttled cfs_rq
sched/fair: Advance global expiration when period timer is restarted
sched/fair: Fix bandwidth timer clock drift condition
sched/rt: Fix call to cpufreq_update_util()
sched/nohz: Skip remote tick on idle task entirely
Otherwise we end up with attempting to send packets from down devices
or to send oversized packets, which may cause unexpected driver/device
behaviour. Generic XDP has already done this check, so reuse the logic
in native XDP.
Fixes: 814abfabef ("xdp: add bpf_redirect helper function")
Signed-off-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
In commit
'bpf: bpf_compute_data uses incorrect cb structure' (8108a77515)
we added the routine bpf_compute_data_end_sk_skb() to compute the
correct data_end values, but this has since been lost. In kernel
v4.14 this was correct and the above patch was applied in it
entirety. Then when v4.14 was merged into v4.15-rc1 net-next tree
we lost the piece that renamed bpf_compute_data_pointers to the
new function bpf_compute_data_end_sk_skb. This was done here,
e1ea2f9856 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net")
When it conflicted with the following rename patch,
6aaae2b6c4 ("bpf: rename bpf_compute_data_end into bpf_compute_data_pointers")
Finally, after a refactor I thought even the function
bpf_compute_data_end_sk_skb() was no longer needed and it was
erroneously removed.
However, we never reverted the sk_skb_convert_ctx_access() usage of
tcp_skb_cb which had been committed and survived the merge conflict.
Here we fix this by adding back the helper and *_data_end_sk_skb()
usage. Using the bpf_skc_data_end mapping is not correct because it
expects a qdisc_skb_cb object but at the sock layer this is not the
case. Even though it happens to work here because we don't overwrite
any data in-use at the socket layer and the cb structure is cleared
later this has potential to create some subtle issues. But, even
more concretely the filter.c access check uses tcp_skb_cb.
And by some act of chance though,
struct bpf_skb_data_end {
struct qdisc_skb_cb qdisc_cb; /* 0 28 */
/* XXX 4 bytes hole, try to pack */
void * data_meta; /* 32 8 */
void * data_end; /* 40 8 */
/* size: 48, cachelines: 1, members: 3 */
/* sum members: 44, holes: 1, sum holes: 4 */
/* last cacheline: 48 bytes */
};
and then tcp_skb_cb,
struct tcp_skb_cb {
[...]
struct {
__u32 flags; /* 24 4 */
struct sock * sk_redir; /* 32 8 */
void * data_end; /* 40 8 */
} bpf; /* 24 */
};
So when we use offset_of() to track down the byte offset we get 40 in
either case and everything continues to work. Fix this mess and use
correct structures its unclear how long this might actually work for
until someone moves the structs around.
Reported-by: Martin KaFai Lau <kafai@fb.com>
Fixes: e1ea2f9856 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net")
Fixes: 6aaae2b6c4 ("bpf: rename bpf_compute_data_end into bpf_compute_data_pointers")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently, when a sock is closed and the bpf_tcp_close() callback is
used we remove memory but do not free the skb. Call consume_skb() if
the skb is attached to the buffer.
Reported-by: syzbot+d464d2c20c717ef5a6a8@syzkaller.appspotmail.com
Fixes: 1aa12bdf1b ("bpf: sockmap, add sock close() hook to remove socks")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
After latest lock updates there is no longer anything preventing a
close and recvmsg call running in parallel. Additionally, we can
race update with close if we close a socket and simultaneously update
if via the BPF userspace API (note the cgroup ops are already run
with sock_lock held).
To resolve this take sock_lock in close and update paths.
Reported-by: syzbot+b680e42077a0d7c9a0c4@syzkaller.appspotmail.com
Fixes: e9db4ef6bf ("bpf: sockhash fix omitted bucket lock in sock_close")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
