commit c40f7d74c7 upstream.
Zhipeng Xie, Xie XiuQi and Sargun Dhillon reported lockups in the
scheduler under high loads, starting at around the v4.18 time frame,
and Zhipeng Xie tracked it down to bugs in the rq->leaf_cfs_rq_list
manipulation.
Do a (manual) revert of:
a9e7f6544b ("sched/fair: Fix O(nr_cgroups) in load balance path")
It turns out that the list_del_leaf_cfs_rq() introduced by this commit
is a surprising property that was not considered in followup commits
such as:
9c2791f936 ("sched/fair: Fix hierarchical order in rq->leaf_cfs_rq_list")
As Vincent Guittot explains:
"I think that there is a bigger problem with commit a9e7f6544b and
cfs_rq throttling:
Let take the example of the following topology TG2 --> TG1 --> root:
1) The 1st time a task is enqueued, we will add TG2 cfs_rq then TG1
cfs_rq to leaf_cfs_rq_list and we are sure to do the whole branch in
one path because it has never been used and can't be throttled so
tmp_alone_branch will point to leaf_cfs_rq_list at the end.
2) Then TG1 is throttled
3) and we add TG3 as a new child of TG1.
4) The 1st enqueue of a task on TG3 will add TG3 cfs_rq just before TG1
cfs_rq and tmp_alone_branch will stay on rq->leaf_cfs_rq_list.
With commit a9e7f6544b, we can del a cfs_rq from rq->leaf_cfs_rq_list.
So if the load of TG1 cfs_rq becomes NULL before step 2) above, TG1
cfs_rq is removed from the list.
Then at step 4), TG3 cfs_rq is added at the beginning of rq->leaf_cfs_rq_list
but tmp_alone_branch still points to TG3 cfs_rq because its throttled
parent can't be enqueued when the lock is released.
tmp_alone_branch doesn't point to rq->leaf_cfs_rq_list whereas it should.
So if TG3 cfs_rq is removed or destroyed before tmp_alone_branch
points on another TG cfs_rq, the next TG cfs_rq that will be added,
will be linked outside rq->leaf_cfs_rq_list - which is bad.
In addition, we can break the ordering of the cfs_rq in
rq->leaf_cfs_rq_list but this ordering is used to update and
propagate the update from leaf down to root."
Instead of trying to work through all these cases and trying to reproduce
the very high loads that produced the lockup to begin with, simplify
the code temporarily by reverting a9e7f6544b - which change was clearly
not thought through completely.
This (hopefully) gives us a kernel that doesn't lock up so people
can continue to enjoy their holidays without worrying about regressions. ;-)
[ mingo: Wrote changelog, fixed weird spelling in code comment while at it. ]
Analyzed-by: Xie XiuQi <xiexiuqi@huawei.com>
Analyzed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reported-by: Zhipeng Xie <xiezhipeng1@huawei.com>
Reported-by: Sargun Dhillon <sargun@sargun.me>
Reported-by: Xie XiuQi <xiexiuqi@huawei.com>
Tested-by: Zhipeng Xie <xiezhipeng1@huawei.com>
Tested-by: Sargun Dhillon <sargun@sargun.me>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Cc: <stable@vger.kernel.org> # v4.13+
Cc: Bin Li <huawei.libin@huawei.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: a9e7f6544b ("sched/fair: Fix O(nr_cgroups) in load balance path")
Link: http://lkml.kernel.org/r/1545879866-27809-1-git-send-email-xiexiuqi@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit eb4c238227 upstream.
The srcu_gp_start() function is called with the srcu_struct structure's
->lock held, but not with the srcu_data structure's ->lock. This is
problematic because this function accesses and updates the srcu_data
structure's ->srcu_cblist, which is protected by that lock. Failing to
hold this lock can result in corruption of the SRCU callback lists,
which in turn can result in arbitrarily bad results.
This commit therefore makes srcu_gp_start() acquire the srcu_data
structure's ->lock across the calls to rcu_segcblist_advance() and
rcu_segcblist_accelerate(), thus preventing this corruption.
Reported-by: Bart Van Assche <bvanassche@acm.org>
Reported-by: Christoph Hellwig <hch@infradead.org>
Reported-by: Sebastian Kuzminsky <seb.kuzminsky@gmail.com>
Signed-off-by: Dennis Krein <Dennis.Krein@netapp.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Tested-by: Dennis Krein <Dennis.Krein@netapp.com>
Cc: <stable@vger.kernel.org> # 4.16.x
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a95c90f1e2 upstream.
The last step before devm_memremap_pages() returns success is to allocate
a release action, devm_memremap_pages_release(), to tear the entire setup
down. However, the result from devm_add_action() is not checked.
Checking the error from devm_add_action() is not enough. The api
currently relies on the fact that the percpu_ref it is using is killed by
the time the devm_memremap_pages_release() is run. Rather than continue
this awkward situation, offload the responsibility of killing the
percpu_ref to devm_memremap_pages_release() directly. This allows
devm_memremap_pages() to do the right thing relative to init failures and
shutdown.
Without this change we could fail to register the teardown of
devm_memremap_pages(). The likelihood of hitting this failure is tiny as
small memory allocations almost always succeed. However, the impact of
the failure is large given any future reconfiguration, or disable/enable,
of an nvdimm namespace will fail forever as subsequent calls to
devm_memremap_pages() will fail to setup the pgmap_radix since there will
be stale entries for the physical address range.
An argument could be made to require that the ->kill() operation be set in
the @pgmap arg rather than passed in separately. However, it helps code
readability, tracking the lifetime of a given instance, to be able to grep
the kill routine directly at the devm_memremap_pages() call site.
Link: http://lkml.kernel.org/r/154275558526.76910.7535251937849268605.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Fixes: e8d5134833 ("memremap: change devm_memremap_pages interface...")
Reviewed-by: "Jérôme Glisse" <jglisse@redhat.com>
Reported-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 808153e118 upstream.
devm_memremap_pages() is a facility that can create struct page entries
for any arbitrary range and give drivers the ability to subvert core
aspects of page management.
Specifically the facility is tightly integrated with the kernel's memory
hotplug functionality. It injects an altmap argument deep into the
architecture specific vmemmap implementation to allow allocating from
specific reserved pages, and it has Linux specific assumptions about page
structure reference counting relative to get_user_pages() and
get_user_pages_fast(). It was an oversight and a mistake that this was
not marked EXPORT_SYMBOL_GPL from the outset.
Again, devm_memremap_pagex() exposes and relies upon core kernel internal
assumptions and will continue to evolve along with 'struct page', memory
hotplug, and support for new memory types / topologies. Only an in-kernel
GPL-only driver is expected to keep up with this ongoing evolution. This
interface, and functionality derived from this interface, is not suitable
for kernel-external drivers.
Link: http://lkml.kernel.org/r/154275557457.76910.16923571232582744134.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7b55851367 upstream.
This changes the fork(2) syscall to record the process start_time after
initializing the basic task structure but still before making the new
process visible to user-space.
Technically, we could record the start_time anytime during fork(2). But
this might lead to scenarios where a start_time is recorded long before
a process becomes visible to user-space. For instance, with
userfaultfd(2) and TLS, user-space can delay the execution of fork(2)
for an indefinite amount of time (and will, if this causes network
access, or similar).
By recording the start_time late, it much closer reflects the point in
time where the process becomes live and can be observed by other
processes.
Lastly, this makes it much harder for user-space to predict and control
the start_time they get assigned. Previously, user-space could fork a
process and stall it in copy_thread_tls() before its pid is allocated,
but after its start_time is recorded. This can be misused to later-on
cycle through PIDs and resume the stalled fork(2) yielding a process
that has the same pid and start_time as a process that existed before.
This can be used to circumvent security systems that identify processes
by their pid+start_time combination.
Even though user-space was always aware that start_time recording is
flaky (but several projects are known to still rely on start_time-based
identification), changing the start_time to be recorded late will help
mitigate existing attacks and make it much harder for user-space to
control the start_time a process gets assigned.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Tom Gundersen <teg@jklm.no>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e9d81a1bc2 upstream.
CSS_TASK_ITER_PROCS implements process-only iteration by making
css_task_iter_advance() skip tasks which aren't threadgroup leaders;
however, when an iteration is started css_task_iter_start() calls the
inner helper function css_task_iter_advance_css_set() instead of
css_task_iter_advance(). As the helper doesn't have the skip logic,
when the first task to visit is a non-leader thread, it doesn't get
skipped correctly as shown in the following example.
# ps -L 2030
PID LWP TTY STAT TIME COMMAND
2030 2030 pts/0 Sl+ 0:00 ./test-thread
2030 2031 pts/0 Sl+ 0:00 ./test-thread
# mkdir -p /sys/fs/cgroup/x/a/b
# echo threaded > /sys/fs/cgroup/x/a/cgroup.type
# echo threaded > /sys/fs/cgroup/x/a/b/cgroup.type
# echo 2030 > /sys/fs/cgroup/x/a/cgroup.procs
# cat /sys/fs/cgroup/x/a/cgroup.threads
2030
2031
# cat /sys/fs/cgroup/x/cgroup.procs
2030
# echo 2030 > /sys/fs/cgroup/x/a/b/cgroup.threads
# cat /sys/fs/cgroup/x/cgroup.procs
2031
2030
The last read of cgroup.procs is incorrectly showing non-leader 2031
in cgroup.procs output.
This can be fixed by updating css_task_iter_advance() to handle the
first advance and css_task_iters_tart() to call
css_task_iter_advance() instead of the inner helper. After the fix,
the same commands result in the following (correct) result:
# ps -L 2062
PID LWP TTY STAT TIME COMMAND
2062 2062 pts/0 Sl+ 0:00 ./test-thread
2062 2063 pts/0 Sl+ 0:00 ./test-thread
# mkdir -p /sys/fs/cgroup/x/a/b
# echo threaded > /sys/fs/cgroup/x/a/cgroup.type
# echo threaded > /sys/fs/cgroup/x/a/b/cgroup.type
# echo 2062 > /sys/fs/cgroup/x/a/cgroup.procs
# cat /sys/fs/cgroup/x/a/cgroup.threads
2062
2063
# cat /sys/fs/cgroup/x/cgroup.procs
2062
# echo 2062 > /sys/fs/cgroup/x/a/b/cgroup.threads
# cat /sys/fs/cgroup/x/cgroup.procs
2062
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com>
Fixes: 8cfd8147df ("cgroup: implement cgroup v2 thread support")
Cc: stable@vger.kernel.org # v4.14+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c7c3f05e34 upstream.
From printk()/serial console point of view panic() is special, because
it may force CPU to re-enter printk() or/and serial console driver.
Therefore, some of serial consoles drivers are re-entrant. E.g. 8250:
serial8250_console_write()
{
if (port->sysrq)
locked = 0;
else if (oops_in_progress)
locked = spin_trylock_irqsave(&port->lock, flags);
else
spin_lock_irqsave(&port->lock, flags);
...
}
panic() does set oops_in_progress via bust_spinlocks(1), so in theory
we should be able to re-enter serial console driver from panic():
CPU0
<NMI>
uart_console_write()
serial8250_console_write() // if (oops_in_progress)
// spin_trylock_irqsave()
call_console_drivers()
console_unlock()
console_flush_on_panic()
bust_spinlocks(1) // oops_in_progress++
panic()
<NMI/>
spin_lock_irqsave(&port->lock, flags) // spin_lock_irqsave()
serial8250_console_write()
call_console_drivers()
console_unlock()
printk()
...
However, this does not happen and we deadlock in serial console on
port->lock spinlock. And the problem is that console_flush_on_panic()
called after bust_spinlocks(0):
void panic(const char *fmt, ...)
{
bust_spinlocks(1);
...
bust_spinlocks(0);
console_flush_on_panic();
...
}
bust_spinlocks(0) decrements oops_in_progress, so oops_in_progress
can go back to zero. Thus even re-entrant console drivers will simply
spin on port->lock spinlock. Given that port->lock may already be
locked either by a stopped CPU, or by the very same CPU we execute
panic() on (for instance, NMI panic() on printing CPU) the system
deadlocks and does not reboot.
Fix this by removing bust_spinlocks(0), so oops_in_progress is always
set in panic() now and, thus, re-entrant console drivers will trylock
the port->lock instead of spinning on it forever, when we call them
from console_flush_on_panic().
Link: http://lkml.kernel.org/r/20181025101036.6823-1-sergey.senozhatsky@gmail.com
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Daniel Wang <wonderfly@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Alan Cox <gnomes@lxorguk.ukuu.org.uk>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: Peter Feiner <pfeiner@google.com>
Cc: linux-serial@vger.kernel.org
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 9b6f7e163c ("mm: rework memcg kernel stack accounting") will
result in fork failing if allocating a kernel stack for a task in
dup_task_struct exceeds the kernel memory allowance for that cgroup.
Unfortunately, it also results in a crash.
This is due to the code jumping to free_stack and calling
free_thread_stack when the memcg kernel stack charge fails, but without
tsk->stack pointing at the freshly allocated stack.
This in turn results in the vfree_atomic in free_thread_stack oopsing
with a backtrace like this:
#5 [ffffc900244efc88] die at ffffffff8101f0ab
#6 [ffffc900244efcb8] do_general_protection at ffffffff8101cb86
#7 [ffffc900244efce0] general_protection at ffffffff818ff082
[exception RIP: llist_add_batch+7]
RIP: ffffffff8150d487 RSP: ffffc900244efd98 RFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff88085ef55980 RCX: 0000000000000000
RDX: ffff88085ef55980 RSI: 343834343531203a RDI: 343834343531203a
RBP: ffffc900244efd98 R8: 0000000000000001 R9: ffff8808578c3600
R10: 0000000000000000 R11: 0000000000000001 R12: ffff88029f6c21c0
R13: 0000000000000286 R14: ffff880147759b00 R15: 0000000000000000
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018
#8 [ffffc900244efda0] vfree_atomic at ffffffff811df2c7
#9 [ffffc900244efdb8] copy_process at ffffffff81086e37
#10 [ffffc900244efe98] _do_fork at ffffffff810884e0
#11 [ffffc900244eff10] sys_vfork at ffffffff810887ff
#12 [ffffc900244eff20] do_syscall_64 at ffffffff81002a43
RIP: 000000000049b948 RSP: 00007ffcdb307830 RFLAGS: 00000246
RAX: ffffffffffffffda RBX: 0000000000896030 RCX: 000000000049b948
RDX: 0000000000000000 RSI: 00007ffcdb307790 RDI: 00000000005d7421
RBP: 000000000067370f R8: 00007ffcdb3077b0 R9: 000000000001ed00
R10: 0000000000000008 R11: 0000000000000246 R12: 0000000000000040
R13: 000000000000000f R14: 0000000000000000 R15: 000000000088d018
ORIG_RAX: 000000000000003a CS: 0033 SS: 002b
The simplest fix is to assign tsk->stack right where it is allocated.
Link: http://lkml.kernel.org/r/20181214231726.7ee4843c@imladris.surriel.com
Fixes: 9b6f7e163c ("mm: rework memcg kernel stack accounting")
Signed-off-by: Rik van Riel <riel@surriel.com>
Acked-by: Roman Gushchin <guro@fb.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull timer fix from Ingo Molnar:
"Fix a division by zero crash in the posix-timers code"
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
posix-timers: Fix division by zero bug
Pull futex fix from Ingo Molnar:
"A single fix for a robust futexes race between sys_exit() and
sys_futex_lock_pi()"
* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Cure exit race
Pull networking fixes from David Miller:
1) Off by one in netlink parsing of mac802154_hwsim, from Alexander
Aring.
2) nf_tables RCU usage fix from Taehee Yoo.
3) Flow dissector needs nhoff and thoff clamping, from Stanislav
Fomichev.
4) Missing sin6_flowinfo initialization in SCTP, from Xin Long.
5) Spectrev1 in ipmr and ip6mr, from Gustavo A. R. Silva.
6) Fix r8169 crash when DEBUG_SHIRQ is enabled, from Heiner Kallweit.
7) Fix SKB leak in rtlwifi, from Larry Finger.
8) Fix state pruning in bpf verifier, from Jakub Kicinski.
9) Don't handle completely duplicate fragments as overlapping, from
Michal Kubecek.
10) Fix memory corruption with macb and 64-bit DMA, from Anssi Hannula.
11) Fix TCP fallback socket release in smc, from Myungho Jung.
12) gro_cells_destroy needs to napi_disable, from Lorenzo Bianconi.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (130 commits)
rds: Fix warning.
neighbor: NTF_PROXY is a valid ndm_flag for a dump request
net: mvpp2: fix the phylink mode validation
net/sched: cls_flower: Remove old entries from rhashtable
net/tls: allocate tls context using GFP_ATOMIC
iptunnel: make TUNNEL_FLAGS available in uapi
gro_cell: add napi_disable in gro_cells_destroy
lan743x: Remove MAC Reset from initialization
net/mlx5e: Remove the false indication of software timestamping support
net/mlx5: Typo fix in del_sw_hw_rule
net/mlx5e: RX, Fix wrong early return in receive queue poll
ipv6: explicitly initialize udp6_addr in udp_sock_create6()
bnxt_en: Fix ethtool self-test loopback.
net/rds: remove user triggered WARN_ON in rds_sendmsg
net/rds: fix warn in rds_message_alloc_sgs
ath10k: skip sending quiet mode cmd for WCN3990
mac80211: free skb fraglist before freeing the skb
nl80211: fix memory leak if validate_pae_over_nl80211() fails
net/smc: fix TCP fallback socket release
vxge: ensure data0 is initialized in when fetching firmware version information
...
Stefan reported, that the glibc tst-robustpi4 test case fails
occasionally. That case creates the following race between
sys_exit() and sys_futex_lock_pi():
CPU0 CPU1
sys_exit() sys_futex()
do_exit() futex_lock_pi()
exit_signals(tsk) No waiters:
tsk->flags |= PF_EXITING; *uaddr == 0x00000PID
mm_release(tsk) Set waiter bit
exit_robust_list(tsk) { *uaddr = 0x80000PID;
Set owner died attach_to_pi_owner() {
*uaddr = 0xC0000000; tsk = get_task(PID);
} if (!tsk->flags & PF_EXITING) {
... attach();
tsk->flags |= PF_EXITPIDONE; } else {
if (!(tsk->flags & PF_EXITPIDONE))
return -EAGAIN;
return -ESRCH; <--- FAIL
}
ESRCH is returned all the way to user space, which triggers the glibc test
case assert. Returning ESRCH unconditionally is wrong here because the user
space value has been changed by the exiting task to 0xC0000000, i.e. the
FUTEX_OWNER_DIED bit is set and the futex PID value has been cleared. This
is a valid state and the kernel has to handle it, i.e. taking the futex.
Cure it by rereading the user space value when PF_EXITING and PF_EXITPIDONE
is set in the task which 'owns' the futex. If the value has changed, let
the kernel retry the operation, which includes all regular sanity checks
and correctly handles the FUTEX_OWNER_DIED case.
If it hasn't changed, then return ESRCH as there is no way to distinguish
this case from malfunctioning user space. This happens when the exiting
task did not have a robust list, the robust list was corrupted or the user
space value in the futex was simply bogus.
Reported-by: Stefan Liebler <stli@linux.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Darren Hart <dvhart@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Sasha Levin <sashal@kernel.org>
Cc: stable@vger.kernel.org
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200467
Link: https://lkml.kernel.org/r/20181210152311.986181245@linutronix.de
The dma_direct_supported() function intends to check the DMA mask against
specific values. However, the phys_to_dma() function includes the SME
encryption mask, which defeats the intended purpose of the check. This
results in drivers that support less than 48-bit DMA (SME encryption mask
is bit 47) from being able to set the DMA mask successfully when SME is
active, which results in the driver failing to initialize.
Change the function used to check the mask from phys_to_dma() to
__phys_to_dma() so that the SME encryption mask is not part of the check.
Fixes: c1d0af1a1d ("kernel/dma/direct: take DMA offset into account in dma_direct_supported")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Alexei Starovoitov says:
====================
pull-request: bpf 2018-12-15
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) fix liveness propagation of callee saved registers, from Jakub.
2) fix overflow in bpf_jit_limit knob, from Daniel.
3) bpf_flow_dissector api fix, from Stanislav.
4) bpf_perf_event api fix on powerpc, from Sandipan.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently for liveness and state pruning the register parentage
chains don't include states of the callee. This makes some sense
as the callee can't access those registers. However, this means
that READs done after the callee returns will not propagate into
the states of the callee. Callee will then perform pruning
disregarding differences in caller state.
Example:
0: (85) call bpf_user_rnd_u32
1: (b7) r8 = 0
2: (55) if r0 != 0x0 goto pc+1
3: (b7) r8 = 1
4: (bf) r1 = r8
5: (85) call pc+4
6: (15) if r8 == 0x1 goto pc+1
7: (05) *(u64 *)(r9 - 8) = r3
8: (b7) r0 = 0
9: (95) exit
10: (15) if r1 == 0x0 goto pc+0
11: (95) exit
Here we acquire unknown state with call to get_random() [1]. Then
we store this random state in r8 (either 0 or 1) [1 - 3], and make
a call on line 5. Callee does nothing but a trivial conditional
jump (to create a pruning point). Upon return caller checks the
state of r8 and either performs an unsafe read or not.
Verifier will first explore the path with r8 == 1, creating a pruning
point at [11]. The parentage chain for r8 will include only callers
states so once verifier reaches [6] it will mark liveness only on states
in the caller, and not [11]. Now when verifier walks the paths with
r8 == 0 it will reach [11] and since REG_LIVE_READ on r8 was not
propagated there it will prune the walk entirely (stop walking
the entire program, not just the callee). Since [6] was never walked
with r8 == 0, [7] will be considered dead and replaced with "goto -1"
causing hang at runtime.
This patch weaves the callee's explored states onto the callers
parentage chain. Rough parentage for r8 would have looked like this
before:
[0] [1] [2] [3] [4] [5] [10] [11] [6] [7]
| | ,---|----. | | |
sl0: sl0: / sl0: \ sl0: sl0: sl0:
fr0: r8 <-- fr0: r8<+--fr0: r8 `fr0: r8 ,fr0: r8<-fr0: r8
\ fr1: r8 <- fr1: r8 /
\__________________/
after:
[0] [1] [2] [3] [4] [5] [10] [11] [6] [7]
| | | | | |
sl0: sl0: sl0: sl0: sl0: sl0:
fr0: r8 <-- fr0: r8 <- fr0: r8 <- fr0: r8 <-fr0: r8<-fr0: r8
fr1: r8 <- fr1: r8
Now the mark from instruction 6 will travel through callees states.
Note that we don't have to connect r0 because its overwritten by
callees state on return and r1 - r5 because those are not alive
any more once a call is made.
v2:
- don't connect the callees registers twice (Alexei: suggestion & code)
- add more details to the comment (Ed & Alexei)
v1: don't unnecessarily link caller saved regs (Jiong)
Fixes: f4d7e40a5b ("bpf: introduce function calls (verification)")
Reported-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pull tracing fixes from Steven Rostedt:
"While running various ftrace tests on new development code, the
kmemleak detector found some allocations that were not freed
correctly.
This fixes a couple of leaks in the event trigger code as well as in
adding function trace filters in trace instances"
* tag 'trace-v4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Fix memory leak of instance function hash filters
tracing: Fix memory leak in set_trigger_filter()
tracing: Fix memory leak in create_filter()
Michael and Sandipan report:
Commit ede95a63b5 introduced a bpf_jit_limit tuneable to limit BPF
JIT allocations. At compile time it defaults to PAGE_SIZE * 40000,
and is adjusted again at init time if MODULES_VADDR is defined.
For ppc64 kernels, MODULES_VADDR isn't defined, so we're stuck with
the compile-time default at boot-time, which is 0x9c400000 when
using 64K page size. This overflows the signed 32-bit bpf_jit_limit
value:
root@ubuntu:/tmp# cat /proc/sys/net/core/bpf_jit_limit
-1673527296
and can cause various unexpected failures throughout the network
stack. In one case `strace dhclient eth0` reported:
setsockopt(5, SOL_SOCKET, SO_ATTACH_FILTER, {len=11, filter=0x105dd27f8},
16) = -1 ENOTSUPP (Unknown error 524)
and similar failures can be seen with tools like tcpdump. This doesn't
always reproduce however, and I'm not sure why. The more consistent
failure I've seen is an Ubuntu 18.04 KVM guest booted on a POWER9
host would time out on systemd/netplan configuring a virtio-net NIC
with no noticeable errors in the logs.
Given this and also given that in near future some architectures like
arm64 will have a custom area for BPF JIT image allocations we should
get rid of the BPF_JIT_LIMIT_DEFAULT fallback / default entirely. For
4.21, we have an overridable bpf_jit_alloc_exec(), bpf_jit_free_exec()
so therefore add another overridable bpf_jit_alloc_exec_limit() helper
function which returns the possible size of the memory area for deriving
the default heuristic in bpf_jit_charge_init().
Like bpf_jit_alloc_exec() and bpf_jit_free_exec(), the new
bpf_jit_alloc_exec_limit() assumes that module_alloc() is the default
JIT memory provider, and therefore in case archs implement their custom
module_alloc() we use MODULES_{END,_VADDR} for limits and otherwise for
vmalloc_exec() cases like on ppc64 we use VMALLOC_{END,_START}.
Additionally, for archs supporting large page sizes, we should change
the sysctl to be handled as long to not run into sysctl restrictions
in future.
Fixes: ede95a63b5 ("bpf: add bpf_jit_limit knob to restrict unpriv allocations")
Reported-by: Sandipan Das <sandipan@linux.ibm.com>
Reported-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The following commands will cause a memory leak:
# cd /sys/kernel/tracing
# mkdir instances/foo
# echo schedule > instance/foo/set_ftrace_filter
# rmdir instances/foo
The reason is that the hashes that hold the filters to set_ftrace_filter and
set_ftrace_notrace are not freed if they contain any data on the instance
and the instance is removed.
Found by kmemleak detector.
Cc: stable@vger.kernel.org
Fixes: 591dffdade ("ftrace: Allow for function tracing instance to filter functions")
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
When create_event_filter() fails in set_trigger_filter(), the filter may
still be allocated and needs to be freed. The caller expects the
data->filter to be updated with the new filter, even if the new filter
failed (we could add an error message by setting set_str parameter of
create_event_filter(), but that's another update).
But because the error would just exit, filter was left hanging and
nothing could free it.
Found by kmemleak detector.
Cc: stable@vger.kernel.org
Fixes: bac5fb97a1 ("tracing: Add and use generic set_trigger_filter() implementation")
Reviewed-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
The create_filter() calls create_filter_start() which allocates a
"parse_error" descriptor, but fails to call create_filter_finish() that
frees it.
The op_stack and inverts in predicate_parse() were also not freed.
Found by kmemleak detector.
Cc: stable@vger.kernel.org
Fixes: 80765597bc ("tracing: Rewrite filter logic to be simpler and faster")
Reviewed-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Pull networking fixes from David Miller:
"A decent batch of fixes here. I'd say about half are for problems that
have existed for a while, and half are for new regressions added in
the 4.20 merge window.
1) Fix 10G SFP phy module detection in mvpp2, from Baruch Siach.
2) Revert bogus emac driver change, from Benjamin Herrenschmidt.
3) Handle BPF exported data structure with pointers when building
32-bit userland, from Daniel Borkmann.
4) Memory leak fix in act_police, from Davide Caratti.
5) Check RX checksum offload in RX descriptors properly in aquantia
driver, from Dmitry Bogdanov.
6) SKB unlink fix in various spots, from Edward Cree.
7) ndo_dflt_fdb_dump() only works with ethernet, enforce this, from
Eric Dumazet.
8) Fix FID leak in mlxsw driver, from Ido Schimmel.
9) IOTLB locking fix in vhost, from Jean-Philippe Brucker.
10) Fix SKB truesize accounting in ipv4/ipv6/netfilter frag memory
limits otherwise namespace exit can hang. From Jiri Wiesner.
11) Address block parsing length fixes in x25 from Martin Schiller.
12) IRQ and ring accounting fixes in bnxt_en, from Michael Chan.
13) For tun interfaces, only iface delete works with rtnl ops, enforce
this by disallowing add. From Nicolas Dichtel.
14) Use after free in liquidio, from Pan Bian.
15) Fix SKB use after passing to netif_receive_skb(), from Prashant
Bhole.
16) Static key accounting and other fixes in XPS from Sabrina Dubroca.
17) Partially initialized flow key passed to ip6_route_output(), from
Shmulik Ladkani.
18) Fix RTNL deadlock during reset in ibmvnic driver, from Thomas
Falcon.
19) Several small TCP fixes (off-by-one on window probe abort, NULL
deref in tail loss probe, SNMP mis-estimations) from Yuchung
Cheng"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (93 commits)
net/sched: cls_flower: Reject duplicated rules also under skip_sw
bnxt_en: Fix _bnxt_get_max_rings() for 57500 chips.
bnxt_en: Fix NQ/CP rings accounting on the new 57500 chips.
bnxt_en: Keep track of reserved IRQs.
bnxt_en: Fix CNP CoS queue regression.
net/mlx4_core: Correctly set PFC param if global pause is turned off.
Revert "net/ibm/emac: wrong bit is used for STA control"
neighbour: Avoid writing before skb->head in neigh_hh_output()
ipv6: Check available headroom in ip6_xmit() even without options
tcp: lack of available data can also cause TSO defer
ipv6: sr: properly initialize flowi6 prior passing to ip6_route_output
mlxsw: spectrum_switchdev: Fix VLAN device deletion via ioctl
mlxsw: spectrum_router: Relax GRE decap matching check
mlxsw: spectrum_switchdev: Avoid leaking FID's reference count
mlxsw: spectrum_nve: Remove easily triggerable warnings
ipv4: ipv6: netfilter: Adjust the frag mem limit when truesize changes
sctp: frag_point sanity check
tcp: fix NULL ref in tail loss probe
tcp: Do not underestimate rwnd_limited
net: use skb_list_del_init() to remove from RX sublists
...
Pull gcc stackleak plugin fixes from Kees Cook:
- Remove tracing for inserted stack depth marking function (Anders
Roxell)
- Move gcc-plugin pass location to avoid objtool warnings (Alexander
Popov)
* tag 'gcc-plugins-v4.20-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
stackleak: Register the 'stackleak_cleanup' pass before the '*free_cfg' pass
stackleak: Mark stackleak_track_stack() as notrace
Pull tracing fix from Steven Rostedt:
"This is a single commit that fixes a bug in uprobes SDT code due to a
missing mutex protection"
* tag 'trace-v4.20-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
Uprobes: Fix kernel oops with delayed_uprobe_remove()
Function graph tracing recurses into itself when stackleak is enabled,
causing the ftrace graph selftest to run for up to 90 seconds and
trigger the softlockup watchdog.
Breakpoint 2, ftrace_graph_caller () at ../arch/arm64/kernel/entry-ftrace.S:200
200 mcount_get_lr_addr x0 // pointer to function's saved lr
(gdb) bt
\#0 ftrace_graph_caller () at ../arch/arm64/kernel/entry-ftrace.S:200
\#1 0xffffff80081d5280 in ftrace_caller () at ../arch/arm64/kernel/entry-ftrace.S:153
\#2 0xffffff8008555484 in stackleak_track_stack () at ../kernel/stackleak.c:106
\#3 0xffffff8008421ff8 in ftrace_ops_test (ops=0xffffff8009eaa840 <graph_ops>, ip=18446743524091297036, regs=<optimized out>) at ../kernel/trace/ftrace.c:1507
\#4 0xffffff8008428770 in __ftrace_ops_list_func (regs=<optimized out>, ignored=<optimized out>, parent_ip=<optimized out>, ip=<optimized out>) at ../kernel/trace/ftrace.c:6286
\#5 ftrace_ops_no_ops (ip=18446743524091297036, parent_ip=18446743524091242824) at ../kernel/trace/ftrace.c:6321
\#6 0xffffff80081d5280 in ftrace_caller () at ../arch/arm64/kernel/entry-ftrace.S:153
\#7 0xffffff800832fd10 in irq_find_mapping (domain=0xffffffc03fc4bc80, hwirq=27) at ../kernel/irq/irqdomain.c:876
\#8 0xffffff800832294c in __handle_domain_irq (domain=0xffffffc03fc4bc80, hwirq=27, lookup=true, regs=0xffffff800814b840) at ../kernel/irq/irqdesc.c:650
\#9 0xffffff80081d52b4 in ftrace_graph_caller () at ../arch/arm64/kernel/entry-ftrace.S:205
Rework so we mark stackleak_track_stack as notrace
Co-developed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Alexei Starovoitov says:
====================
pull-request: bpf 2018-12-05
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) fix bpf uapi pointers for 32-bit architectures, from Daniel.
2) improve verifer ability to handle progs with a lot of branches, from Alexei.
3) strict btf checks, from Yonghong.
4) bpf_sk_lookup api cleanup, from Joe.
5) other misc fixes
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
malicious bpf program may try to force the verifier to remember
a lot of distinct verifier states.
Put a limit to number of per-insn 'struct bpf_verifier_state'.
Note that hitting the limit doesn't reject the program.
It potentially makes the verifier do more steps to analyze the program.
It means that malicious programs will hit BPF_COMPLEXITY_LIMIT_INSNS sooner
instead of spending cpu time walking long link list.
The limit of BPF_COMPLEXITY_LIMIT_STATES==64 affects cilium progs
with slight increase in number of "steps" it takes to successfully verify
the programs:
before after
bpf_lb-DLB_L3.o 1940 1940
bpf_lb-DLB_L4.o 3089 3089
bpf_lb-DUNKNOWN.o 1065 1065
bpf_lxc-DDROP_ALL.o 28052 | 28162
bpf_lxc-DUNKNOWN.o 35487 | 35541
bpf_netdev.o 10864 10864
bpf_overlay.o 6643 6643
bpf_lcx_jit.o 38437 38437
But it also makes malicious program to be rejected in 0.4 seconds vs 6.5
Hence apply this limit to unprivileged programs only.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
pathological bpf programs may try to force verifier to explode in
the number of branch states:
20: (d5) if r1 s<= 0x24000028 goto pc+0
21: (b5) if r0 <= 0xe1fa20 goto pc+2
22: (d5) if r1 s<= 0x7e goto pc+0
23: (b5) if r0 <= 0xe880e000 goto pc+0
24: (c5) if r0 s< 0x2100ecf4 goto pc+0
25: (d5) if r1 s<= 0xe880e000 goto pc+1
26: (c5) if r0 s< 0xf4041810 goto pc+0
27: (d5) if r1 s<= 0x1e007e goto pc+0
28: (b5) if r0 <= 0xe86be000 goto pc+0
29: (07) r0 += 16614
30: (c5) if r0 s< 0x6d0020da goto pc+0
31: (35) if r0 >= 0x2100ecf4 goto pc+0
Teach verifier to recognize always taken and always not taken branches.
This analysis is already done for == and != comparison.
Expand it to all other branches.
It also helps real bpf programs to be verified faster:
before after
bpf_lb-DLB_L3.o 2003 1940
bpf_lb-DLB_L4.o 3173 3089
bpf_lb-DUNKNOWN.o 1080 1065
bpf_lxc-DDROP_ALL.o 29584 28052
bpf_lxc-DUNKNOWN.o 36916 35487
bpf_netdev.o 11188 10864
bpf_overlay.o 6679 6643
bpf_lcx_jit.o 39555 38437
Reported-by: Anatoly Trosinenko <anatoly.trosinenko@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Malicious user space may try to force the verifier to use as much cpu
time and memory as possible. Hence check for pending signals
while verifying the program.
Note that suspend of sys_bpf(PROG_LOAD) syscall will lead to EAGAIN,
since the kernel has to release the resources used for program verification.
Reported-by: Anatoly Trosinenko <anatoly.trosinenko@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Pull STIBP fallout fixes from Thomas Gleixner:
"The performance destruction department finally got it's act together
and came up with a cure for the STIPB regression:
- Provide a command line option to control the spectre v2 user space
mitigations. Default is either seccomp or prctl (if seccomp is
disabled in Kconfig). prctl allows mitigation opt-in, seccomp
enables the migitation for sandboxed processes.
- Rework the code to handle the conditional STIBP/IBPB control and
remove the now unused ptrace_may_access_sched() optimization
attempt
- Disable STIBP automatically when SMT is disabled
- Optimize the switch_to() logic to avoid MSR writes and invocations
of __switch_to_xtra().
- Make the asynchronous speculation TIF updates synchronous to
prevent stale mitigation state.
As a general cleanup this also makes retpoline directly depend on
compiler support and removes the 'minimal retpoline' option which just
pretended to provide some form of security while providing none"
* 'x86-pti-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (31 commits)
x86/speculation: Provide IBPB always command line options
x86/speculation: Add seccomp Spectre v2 user space protection mode
x86/speculation: Enable prctl mode for spectre_v2_user
x86/speculation: Add prctl() control for indirect branch speculation
x86/speculation: Prepare arch_smt_update() for PRCTL mode
x86/speculation: Prevent stale SPEC_CTRL msr content
x86/speculation: Split out TIF update
ptrace: Remove unused ptrace_may_access_sched() and MODE_IBRS
x86/speculation: Prepare for conditional IBPB in switch_mm()
x86/speculation: Avoid __switch_to_xtra() calls
x86/process: Consolidate and simplify switch_to_xtra() code
x86/speculation: Prepare for per task indirect branch speculation control
x86/speculation: Add command line control for indirect branch speculation
x86/speculation: Unify conditional spectre v2 print functions
x86/speculataion: Mark command line parser data __initdata
x86/speculation: Mark string arrays const correctly
x86/speculation: Reorder the spec_v2 code
x86/l1tf: Show actual SMT state
x86/speculation: Rework SMT state change
sched/smt: Expose sched_smt_present static key
...
Merge misc fixes from Andrew Morton:
"31 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (31 commits)
ocfs2: fix potential use after free
mm/khugepaged: fix the xas_create_range() error path
mm/khugepaged: collapse_shmem() do not crash on Compound
mm/khugepaged: collapse_shmem() without freezing new_page
mm/khugepaged: minor reorderings in collapse_shmem()
mm/khugepaged: collapse_shmem() remember to clear holes
mm/khugepaged: fix crashes due to misaccounted holes
mm/khugepaged: collapse_shmem() stop if punched or truncated
mm/huge_memory: fix lockdep complaint on 32-bit i_size_read()
mm/huge_memory: splitting set mapping+index before unfreeze
mm/huge_memory: rename freeze_page() to unmap_page()
initramfs: clean old path before creating a hardlink
kernel/kcov.c: mark funcs in __sanitizer_cov_trace_pc() as notrace
psi: make disabling/enabling easier for vendor kernels
proc: fixup map_files test on arm
debugobjects: avoid recursive calls with kmemleak
userfaultfd: shmem: UFFDIO_COPY: set the page dirty if VM_WRITE is not set
userfaultfd: shmem: add i_size checks
userfaultfd: shmem/hugetlbfs: only allow to register VM_MAYWRITE vmas
userfaultfd: shmem: allocate anonymous memory for MAP_PRIVATE shmem
...
Pull stackleak plugin fix from Kees Cook:
"Fix crash by not allowing kprobing of stackleak_erase() (Alexander
Popov)"
* tag 'gcc-plugins-v4.20-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
stackleak: Disable function tracing and kprobes for stackleak_erase()
Since __sanitizer_cov_trace_pc() is marked as notrace, function calls in
__sanitizer_cov_trace_pc() shouldn't be traced either.
ftrace_graph_caller() gets called for each function that isn't marked
'notrace', like canonicalize_ip(). This is the call trace from a run:
[ 139.644550] ftrace_graph_caller+0x1c/0x24
[ 139.648352] canonicalize_ip+0x18/0x28
[ 139.652313] __sanitizer_cov_trace_pc+0x14/0x58
[ 139.656184] sched_clock+0x34/0x1e8
[ 139.659759] trace_clock_local+0x40/0x88
[ 139.663722] ftrace_push_return_trace+0x8c/0x1f0
[ 139.667767] prepare_ftrace_return+0xa8/0x100
[ 139.671709] ftrace_graph_caller+0x1c/0x24
Rework so that check_kcov_mode() and canonicalize_ip() that are called
from __sanitizer_cov_trace_pc() are also marked as notrace.
Link: http://lkml.kernel.org/r/20181128081239.18317-1-anders.roxell@linaro.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signen-off-by: Anders Roxell <anders.roxell@linaro.org>
Co-developed-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mel Gorman reports a hackbench regression with psi that would prohibit
shipping the suse kernel with it default-enabled, but he'd still like
users to be able to opt in at little to no cost to others.
With the current combination of CONFIG_PSI and the psi_disabled bool set
from the commandline, this is a challenge. Do the following things to
make it easier:
1. Add a config option CONFIG_PSI_DEFAULT_DISABLED that allows distros
to enable CONFIG_PSI in their kernel but leave the feature disabled
unless a user requests it at boot-time.
To avoid double negatives, rename psi_disabled= to psi=.
2. Make psi_disabled a static branch to eliminate any branch costs
when the feature is disabled.
In terms of numbers before and after this patch, Mel says:
: The following is a comparision using CONFIG_PSI=n as a baseline against
: your patch and a vanilla kernel
:
: 4.20.0-rc4 4.20.0-rc4 4.20.0-rc4
: kconfigdisable-v1r1 vanilla psidisable-v1r1
: Amean 1 1.3100 ( 0.00%) 1.3923 ( -6.28%) 1.3427 ( -2.49%)
: Amean 3 3.8860 ( 0.00%) 4.1230 * -6.10%* 3.8860 ( -0.00%)
: Amean 5 6.8847 ( 0.00%) 8.0390 * -16.77%* 6.7727 ( 1.63%)
: Amean 7 9.9310 ( 0.00%) 10.8367 * -9.12%* 9.9910 ( -0.60%)
: Amean 12 16.6577 ( 0.00%) 18.2363 * -9.48%* 17.1083 ( -2.71%)
: Amean 18 26.5133 ( 0.00%) 27.8833 * -5.17%* 25.7663 ( 2.82%)
: Amean 24 34.3003 ( 0.00%) 34.6830 ( -1.12%) 32.0450 ( 6.58%)
: Amean 30 40.0063 ( 0.00%) 40.5800 ( -1.43%) 41.5087 ( -3.76%)
: Amean 32 40.1407 ( 0.00%) 41.2273 ( -2.71%) 39.9417 ( 0.50%)
:
: It's showing that the vanilla kernel takes a hit (as the bisection
: indicated it would) and that disabling PSI by default is reasonably
: close in terms of performance for this particular workload on this
: particular machine so;
Link: http://lkml.kernel.org/r/20181127165329.GA29728@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Tested-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull more tracing fixes from Steven Rostedt:
"Two more fixes:
- Change idx variable in DO_TRACE macro to __idx to avoid name
conflicts. A kvm event had "idx" as a parameter and it confused the
macro.
- Fix a race where interrupts would be traced when set_graph_function
was set. The previous patch set increased a race window that
tricked the function graph tracer to think it should trace
interrupts when it really should not have.
The bug has been there before, but was seldom hit. Only the last
patch series made it more common"
* tag 'trace-v4.20-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing/fgraph: Fix set_graph_function from showing interrupts
tracepoint: Use __idx instead of idx in DO_TRACE macro to make it unique
Pull tracing fixes from Steven Rostedt:
"While rewriting the function graph tracer, I discovered a design flaw
that was introduced by a patch that tried to fix one bug, but by doing
so created another bug.
As both bugs corrupt the output (but they do not crash the kernel), I
decided to fix the design such that it could have both bugs fixed. The
original fix, fixed time reporting of the function graph tracer when
doing a max_depth of one. This was code that can test how much the
kernel interferes with userspace. But in doing so, it could corrupt
the time keeping of the function profiler.
The issue is that the curr_ret_stack variable was being used for two
different meanings. One was to keep track of the stack pointer on the
ret_stack (shadow stack used by the function graph tracer), and the
other use case was the graph call depth. Although, the two may be
closely related, where they got updated was the issue that lead to the
two different bugs that required the two use cases to be updated
differently.
The big issue with this fix is that it requires changing each
architecture. The good news is, I was able to remove a lot of code
that was duplicated within the architectures and place it into a
single location. Then I could make the fix in one place.
I pushed this code into linux-next to let it settle over a week, and
before doing so, I cross compiled all the affected architectures to
make sure that they built fine.
In the mean time, I also pulled in a patch that fixes the sched_switch
previous tasks state output, that was not actually correct"
* tag 'trace-v4.20-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
sched, trace: Fix prev_state output in sched_switch tracepoint
function_graph: Have profiler use curr_ret_stack and not depth
function_graph: Reverse the order of pushing the ret_stack and the callback
function_graph: Move return callback before update of curr_ret_stack
function_graph: Use new curr_ret_depth to manage depth instead of curr_ret_stack
function_graph: Make ftrace_push_return_trace() static
sparc/function_graph: Simplify with function_graph_enter()
sh/function_graph: Simplify with function_graph_enter()
s390/function_graph: Simplify with function_graph_enter()
riscv/function_graph: Simplify with function_graph_enter()
powerpc/function_graph: Simplify with function_graph_enter()
parisc: function_graph: Simplify with function_graph_enter()
nds32: function_graph: Simplify with function_graph_enter()
MIPS: function_graph: Simplify with function_graph_enter()
microblaze: function_graph: Simplify with function_graph_enter()
arm64: function_graph: Simplify with function_graph_enter()
ARM: function_graph: Simplify with function_graph_enter()
x86/function_graph: Simplify with function_graph_enter()
function_graph: Create function_graph_enter() to consolidate architecture code
The stackleak_erase() function is called on the trampoline stack at the
end of syscall. This stack is not big enough for ftrace and kprobes
operations, e.g. it can be exhausted if we use kprobe_events for
stackleak_erase().
So let's disable function tracing and kprobes of stackleak_erase().
Reported-by: kernel test robot <lkp@intel.com>
Fixes: 10e9ae9fab ("gcc-plugins: Add STACKLEAK plugin for tracking the kernel stack")
Signed-off-by: Alexander Popov <alex.popov@linux.com>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
The tracefs file set_graph_function is used to only function graph functions
that are listed in that file (or all functions if the file is empty). The
way this is implemented is that the function graph tracer looks at every
function, and if the current depth is zero and the function matches
something in the file then it will trace that function. When other functions
are called, the depth will be greater than zero (because the original
function will be at depth zero), and all functions will be traced where the
depth is greater than zero.
The issue is that when a function is first entered, and the handler that
checks this logic is called, the depth is set to zero. If an interrupt comes
in and a function in the interrupt handler is traced, its depth will be
greater than zero and it will automatically be traced, even if the original
function was not. But because the logic only looks at depth it may trace
interrupts when it should not be.
The recent design change of the function graph tracer to fix other bugs
caused the depth to be zero while the function graph callback handler is
being called for a longer time, widening the race of this happening. This
bug was actually there for a longer time, but because the race window was so
small it seldom happened. The Fixes tag below is for the commit that widen
the race window, because that commit belongs to a series that will also help
fix the original bug.
Cc: stable@kernel.org
Fixes: 39eb456dac ("function_graph: Use new curr_ret_depth to manage depth instead of curr_ret_stack")
Reported-by: Joe Lawrence <joe.lawrence@redhat.com>
Tested-by: Joe Lawrence <joe.lawrence@redhat.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
This patch added name checking for the following types:
. BTF_KIND_PTR, BTF_KIND_ARRAY, BTF_KIND_VOLATILE,
BTF_KIND_CONST, BTF_KIND_RESTRICT:
the name must be null
. BTF_KIND_STRUCT, BTF_KIND_UNION: the struct/member name
is either null or a valid identifier
. BTF_KIND_ENUM: the enum type name is either null or a valid
identifier; the enumerator name must be a valid identifier.
. BTF_KIND_FWD: the name must be a valid identifier
. BTF_KIND_TYPEDEF: the name must be a valid identifier
For those places a valid name is required, the name must be
a valid C identifier. This can be relaxed later if we found
use cases for a different (non-C) frontend.
Fixes: 69b693f0ae ("bpf: btf: Introduce BPF Type Format (BTF)")
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Function btf_name_valid_identifier() have been implemented in
bpf-next commit 2667a2626f ("bpf: btf: Add BTF_KIND_FUNC and
BTF_KIND_FUNC_PROTO"). Backport this function so later patch
can use it.
Fixes: 69b693f0ae ("bpf: btf: Introduce BPF Type Format (BTF)")
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pull networking fixes from David Miller:
1) ARM64 JIT fixes for subprog handling from Daniel Borkmann.
2) Various sparc64 JIT bug fixes (fused branch convergance, frame
pointer usage detection logic, PSEODU call argument handling).
3) Fix to use BH locking in nf_conncount, from Taehee Yoo.
4) Fix race of TX skb freeing in ipheth driver, from Bernd Eckstein.
5) Handle return value of TX NAPI completion properly in lan743x
driver, from Bryan Whitehead.
6) MAC filter deletion in i40e driver clears wrong state bit, from
Lihong Yang.
7) Fix use after free in rionet driver, from Pan Bian.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (53 commits)
s390/qeth: fix length check in SNMP processing
net: hisilicon: remove unexpected free_netdev
rapidio/rionet: do not free skb before reading its length
i40e: fix kerneldoc for xsk methods
ixgbe: recognize 1000BaseLX SFP modules as 1Gbps
i40e: Fix deletion of MAC filters
igb: fix uninitialized variables
netfilter: nf_tables: deactivate expressions in rule replecement routine
lan743x: Enable driver to work with LAN7431
tipc: fix lockdep warning during node delete
lan743x: fix return value for lan743x_tx_napi_poll
net: via: via-velocity: fix spelling mistake "alignement" -> "alignment"
qed: fix spelling mistake "attnetion" -> "attention"
net: thunderx: fix NULL pointer dereference in nic_remove
sctp: increase sk_wmem_alloc when head->truesize is increased
firestream: fix spelling mistake: "Inititing" -> "Initializing"
net: phy: add workaround for issue where PHY driver doesn't bind to the device
usbnet: ipheth: fix potential recvmsg bug and recvmsg bug 2
sparc: Adjust bpf JIT prologue for PSEUDO calls.
bpf, doc: add entries of who looks over which jits
...
