Currently the lifetime of bpf programs attached to a cgroup is bound
to the lifetime of the cgroup itself. It means that if a user
forgets (or intentionally avoids) to detach a bpf program before
removing the cgroup, it will stay attached up to the release of the
cgroup. Since the cgroup can stay in the dying state (the state
between being rmdir()'ed and being released) for a very long time, it
leads to a waste of memory. Also, it blocks a possibility to implement
the memcg-based memory accounting for bpf objects, because a circular
reference dependency will occur. Charged memory pages are pinning the
corresponding memory cgroup, and if the memory cgroup is pinning
the attached bpf program, nothing will be ever released.
A dying cgroup can not contain any processes, so the only chance for
an attached bpf program to be executed is a live socket associated
with the cgroup. So in order to release all bpf data early, let's
count associated sockets using a new percpu refcounter. On cgroup
removal the counter is transitioned to the atomic mode, and as soon
as it reaches 0, all bpf programs are detached.
Because cgroup_bpf_release() can block, it can't be called from
the percpu ref counter callback directly, so instead an asynchronous
work is scheduled.
The reference counter is not socket specific, and can be used for any
other types of programs, which can be executed from a cgroup-bpf hook
outside of the process context, had such a need arise in the future.
Signed-off-by: Roman Gushchin <guro@fb.com>
Cc: jolsa@redhat.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Kasan reported a use after free in cpia2_usb_disconnect()
It first freed everything and then woke up those waiting.
The reverse order is correct.
Fixes: 6c493f8b28 ("[media] cpia2: major overhaul to get it in a working state again")
Signed-off-by: Oliver Neukum <oneukum@suse.com>
Reported-by: syzbot+0c90fc937c84f97d0aa6@syzkaller.appspotmail.com
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Add a paragraph to make it more clear that video equipment will transmit
fields in the same order the fields were captured, and replace some of
the "is transmitted first" language with "is the older field", since the
latter is the important info for motion compensation applications.
Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Unbinding the driver while streaming caused the driver to hang.
The cause of this was failing to use the v4l2_device release
function and the use of devm_kmalloc for the state structure.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Make sure the default value at least passes the std_validate() tests.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
This commit clears the interrupt reason (INT_REASON) register
on the interrupt handler. Without this clearing, the CODA hardware
has been observed to get completely stalled on CODA980 variants,
requiring a pretty deep hardware reset.
The datasheet specifies that the INT_REASON register is set
by the CODA hardware, and should be cleared by the host.
While the CODA versions that are currently supported by this driver
don't seem to need this change, it's a really small change,
so it seems a wise thing to do to avoid hitting some
rare race-condition in the hardware.
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
The current interrupt handler is doing very little, and not doing
any non-atomic operations. Pretty much all it does is accessing
a couple registers, taking a couple spinlocks and then signalling
a completion.
There is no reason this should be a threaded interrupt handler,
so move the handler to regular hard interrupt context.
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
The mutex unlock in the threaded interrupt handler is not paired
with any mutex lock. Remove it.
This bug has been here for a really long time, so it applies
to any stable repo.
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Cc: stable@vger.kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
This is just a cosmetic change to print a more descriptive
message, to distinguish decoder from encoder:
So, instead of printing
coda 2040000.vpu: codec registered as /dev/video[4-5]
With this change, the driver now prints
coda 2040000.vpu: encoder registered as /dev/video4
coda 2040000.vpu: decoder registered as /dev/video5
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
When the pvrusb2 driver detects that there's something wrong with the
device, it prints a warning message. Right now those message are
printed in two different formats:
1. ***WARNING*** message here
2. WARNING: message here
There's an issue with the second format. Syzkaller recognizes it as a
message produced by a WARN_ON(), which is used to indicate a bug in the
kernel. However pvrusb2 prints those warnings to indicate an issue with
the device, not the bug in the kernel.
This patch changes the pvrusb2 driver to consistently use the first
warning message format. This will unblock syzkaller testing of this
driver.
Reported-by: syzbot+af8f8d2ac0d39b0ed3a0@syzkaller.appspotmail.com
Reported-by: syzbot+170a86bf206dd2c6217e@syzkaller.appspotmail.com
Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
It is really weird that the media controller sources are all top-level
in drivers/media. It is a bit of a left-over from long ago when most
media sources were all at the top-level. At some point we reorganized
the directory structure, but the media-*.c sources where never moved
to their own directory.
So create a new mc directory and move all sources there. Also rename
the prefix from media- to mc-.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Setting the media entity pads after the async register subdev can be
racy with probe complete callback. So, make sure that the media pads
are initialized before the probe complete is called.
For that move the media entity pads initialization to the registered
subdev internal operation.
Signed-off-by: Rui Miguel Silva <rui.silva@linaro.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
vdev is guaranteed to be equal to vpu->vfd_enc thanks a test done a few
lines above. Remove this useless test.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
In order to make the code more generic, introduce a pair of start/stop
codec operations, and use them to allocate and release the JPEG bounce
buffer.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Proc filesystem has special locking rules for various files. Thus
fanotify which opens files on event delivery can easily deadlock
against another process that waits for fanotify permission event to be
handled. Since permission events on /proc have doubtful value anyway,
just disallow them.
Link: https://lore.kernel.org/linux-fsdevel/20190320131642.GE9485@quack2.suse.cz/
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
We need to make the macrobock alignment generic, in order
to support multiple codecs.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Use the recently introduced v4l2_m2m_buf_copy_metadata helper
and get rid of some code.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Now that we've introduced the pixel format helpers, use them
in vpu driver, and get rid of the internal helpers.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
These two slice modes used by the V4L2_CID_MPEG_VIDEO_MULTI_SLICE_MODE
control had a silly typo:
V4L2_MPEG_VIDEO_MULTI_SICE_MODE_MAX_MB
V4L2_MPEG_VIDEO_MULTI_SICE_MODE_MAX_BYTES
SICE should be SLICE.
Rename these enum values, keeping the old ones (under #ifndef __KERNEL__)
for backwards compatibility reasons.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Various security techniques can obfuscate pointer printouts on the
console. Unfortunately, rcutorture relies on either "null" or all zeroes
to identify the last few statistics printouts at the end of the test.
These need to be identified because failing to do so will results in
false-positive complaints about grace-period hangs.
This commit therefore prints the "ver:" in capitals ("VER:") when
the RCU-protected pointer has been set to NULL, which causes rcutorture's
parse-console.sh script to correctly ignore these lines.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
When trace_printk() is used, a message including "BUG" is printed to
the console, which fools the rcutorture scripting into believing that
the corresponding test scenario failed. This commit therefore filters
out this particular instance of "BUG", thus avoiding the false-positive
test-failure report.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
The current rcutorture scripts unconditionally do "make clean", which is
a good way of getting the needed testing done despite any imperfections in
Makefile dependency tracking. However, this can be a bit irritating when
repeatedly running a single scenario after small changes, for example,
when debugging a problem that affects only a single scenario. This commit
therefore adds a --trust-make argument that suppresses the "make clean".
Even when using ccache, this speeds up kernel builds by up to almost an
order of magnitude on my laptop.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Currently, rcutorture will use relatively few CPUs to build the kernel
on a busy system, which is often as it should be. However, if the user
has used the --cpus argument to dedicate a specified number of CPUs to
this torture test, it would be good if the kernel build also made use
of them. This commit therefore changes the cpus2use.sh script to use
--cpus when specified and to do the idleness calculations otherwise.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
For historical reasons, rcutorture places its build products in a
tools/testing/selftests/rcutorture/b1 directory using the O= kbuild
command-line argument. However, doing this requires that the source
directory be pristine: Not just "make clean" pristine, but instead "make
mrproper" (or, equivalently, "make distclean") pristine. Therefore,
rcutorture executes a "make mrproper" before each build. Unfortunately,
"make mrproper" has the side effect of removing pretty much everything,
including tags files and cscope databases, which can be inconvenient
to people whose workflow centers around a single source tree.
This commit therefore makes rcutorture do the build directly in the
source directory, removing the need for "make mrproper". This works
because all needed build products are moved to their proper place in the
"res" directory immediately after the build completes, so that multiple
rcutorture kernels can still run concurrently.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Currently qemu output appears on standard output, but is inaccessible
later on. This commit therefore captures this output and causes
kvm-recheck.sh to output this output if QEMU gave a non-zero non-137
exit code. (And exit code of 137 indicates that QEMU was killed, in
which case we want to know about the hang rather than the fact that
QEMU was killed.)
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
In one of my rcutorture tests the TSC clocksource got marked unstable
due to a large difference in the TSC value. I'm not sure if the guest
run for a long time with disabled interrupts or if the host was very
busy and didn't schedule the guest for some time.
I took a look on the qemu/KVM options and decided to update the options:
- Use kvm{32|64} as CPU. We could probably use `host' (like ARM does)
for maximum available features but since we don't run any userland I'm
not sure if it makes any difference.
- Drop the "noapic" option. There is no history why the APIC was disabled,
I see no reason for it. Once old qemu versions fade away, we can add
"x2apic=on,tsc-deadline=on,hypervisor=on,tsc_adjust=on".
- Additional config options. It ensures that the kernel knowns that it
runs as a kvm guest and can use virt devices like the kvm-clock as
clocksource. The kvm-clock was the main motivation here.
- I didn't add a random HW device. It would make the random device ready
earlier (not it doesn't complete the initialisation at all) but I
doubt that there is any need for this.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
[ paulmck: The world is not quite ready for CONFIG_PARAVIRT_SPINLOCKS=y
and x2apic, so they are omitted for the time being. ]
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
I have been showing off a trivial RCU implementation for non-preemptive
environments for some time now:
#define rcu_read_lock()
#define rcu_read_unlock()
#define rcu_dereference(p) READ_ONCE(p)
#define rcu_assign_pointer(p, v) smp_store_release(&(p), (v))
void synchronize_rcu(void)
{
int cpu;
for_each_online_cpu(cpu)
sched_setaffinity(current->pid, cpumask_of(cpu));
}
Trivial or not, as the old saying goes, "if it ain't tested, it don't
work!". This commit therefore adds a "trivial" flavor to rcutorture
and a corresponding TRIVIAL test scenario. This variant does not handle
CPU hotplug, which is unconditionally enabled on x86 for post-v5.1-rc3
kernels, which is why the TRIVIAL.boot says "rcutorture.onoff_interval=0".
This commit actually does handle CONFIG_PREEMPT=y kernels, but only
because it turns back the Linux-kernel clock in order to provide these
alternative definitions (or the moral equivalent thereof):
#define rcu_read_lock() preempt_disable()
#define rcu_read_unlock() preempt_enable()
In CONFIG_PREEMPT=n kernels without debugging, these are equivalent to
empty macros give or take a compiler barrier. However, the have been
successfully tested with actual empty macros as well.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
[ paulmck: Fix symbol issue reported by kbuild test robot <lkp@intel.com>. ]
[ paulmck: Work around sched_setaffinity() issue noted by Andrea Parri. ]
[ paulmck: Add rcutorture.shuffle_interval=0 to TRIVIAL.boot to fix
interaction with shuffler task noted by Peter Zijlstra. ]
Tested-by: Andrea Parri <andrea.parri@amarulasolutions.com>
Once removed, an rcu_torture element can be deferred-freed by a chain
of call_rcu() invocations, with each callback invoking another round of
call_rcu() until either a fixed number of call_rcu() invocations have
been chained or until the test ends. This means that if the test ends,
some of the rcu_torture elements will be "stranded" partway through the
deferred-free process, which results in false-positive warnings from
rcu_torture_writer() due to lack of forward progress should the test
end just at the end of a stutter interval.
This commit therefore suppresses rcu_torture_writer()'s forward-progress
checks when the test ends in order to avoid these false-positive reports..
Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
In !PREEMPT kernels, cond_resched() is a no-op. In NO_HZ_FULL kernels,
in-kernel execution (such as that of rcutorture's kthreads) might extend
indefinitely without the scheduler gaining the aid of a scheduling-clock
interrupt. This combination can make the interaction of an rcutorture
forward-progress test and a CPU-hotplug stop_machine operation make less
forward progress than one might like. Additionally, Sebastian Siewior
notes that NO_HZ_FULL kernels have a scheduler check upon return to
userspace execution, which suggests that in-kernel emulation of tight
userspace loops containing system calls doing call_rcu() might also need
explicit checks in the PREEMPT && NO_HZ_FULL case.
This commit therefore introduces a rcu_torture_fwd_prog_cond_resched()
function that explicitly invokes schedule() in such kernels whenever
need_resched() returns true, while retaining use of cond_resched()
for kernels that are either !PREEMPT or !NO_HZ_FULL.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Because TREE01 can end up running more vCPUs that physical CPUs,
hammering these shortchanged CPUs with tight loops containing call_rcu()
invocations seems a bit like overkill. This commit therefore exempts
TREE01 from rcutorture's forward-progress testing.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
After the end of each stutter pause interval, the rcu_torture_writer()
kthread checks to be sure that all prior callbacks have completed so
that all the test structures have been freed. This works fine except
for tasks RCU, in which grace periods can take one good long time.
This commit therefore exempts tasks RCU from this check.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
This commit provides a rudimentary Makefile that runs a 10-minute
rcutorture test on scenario TREE01. This must be run on a system capable
of spawning virtual machines and with everything installed to permit
building Linux kernels.
Reported-by: Shuah Khan <shuah@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
This commit causes both kvm-find-errors.sh and kvm-recheck.sh to provide
an exit status based on whether or not errors were located. In the
case of kvm-recheck.sh, this will be the error status of the last run.
This change allows these commands to be used in scripting and Makefiles
to automatically report failed rcutorture runs.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Currently, the inter-stutter interval is the same as the stutter duration,
that is, whatever number of jiffies is passed into torture_stutter_init().
This has worked well for quite some time, but the addition of
forward-progress testing to rcutorture can delay processes for several
seconds, which can triple the time that they are stuttered.
This commit therefore adds a second argument to torture_stutter_init()
that specifies the inter-stutter interval. While locktorture preserves
the current behavior, rcutorture uses the RCU CPU stall warning interval
to provide a wider inter-stutter interval.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
The stutter_wait() function is supposed to return true if it actually
waits and false otherwise, but it instead unconditionally returns false.
Which hides a bug in rcu_torture_writer() that fails to account for
the fact that one of the rcu_tortures[] array elements will normally be
referenced by rcu_torture_current, and thus not be on the freelist.
This commit therefore corrects the stutter_wait() return value and adds a
check for rcu_torture_current to rcu_torture_writer()'s check that things
get freed after everything goes quiescent. In addition, this commit
causes torture_stutter() to give a bit more than one second (instead of
only one jiffy) warning of the end of the stutter interval. Finally,
this commit disables long-delay readers and aggressive update-side
forward-progress checks while forward-progress testing is in flight.
Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
The rcu_torture_fwd_prog_cbfree() function frees callbacks used during
rcutorture's call_rcu() forward-progress test, but does so in a tight
loop. This could cause problems given a very long list of callbacks to be
freed, and actual testing produces lists with as many as 25M callbacks.
This commit therefore adds a cond_resched() to this loop. While in
the area, this commit also rearranges the lock releases to look a bit
more sane.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
jitter.sh currently does not add CPU0 to the list of CPUs for adding of
jitter. Let us add it to this list even when it is not hot-pluggable.
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
The rcutorture jitter.sh script selects a random CPU but does not check
if it is offline or online. This leads to taskset errors many times. On
my machine, hyper threading is disabled so half the cores are offline
causing taskset errors a lot of times. Let us fix this by checking from
only the online CPUs on the system.
Cc: rcu@vger.kernel.org
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
With this patch rcu_sync has a single state variable and the transition rules
become really simple:
GP_IDLE - owned by the first rcu_sync_enter() which moves it to
GP_ENTER - owned by rcu-callback which moves it to
GP_PASSED - owned by the last rcu_sync_exit() which moves it to
GP_EXIT - and this is the only "nontrivial" state.
rcu-callback moves it back to GP_IDLE unless another enter()
comes before a GP pass.
If rcu-callback is invoked before the next rcu_sync_exit() it
must see gp_count incremented by that enter() and set GP_PASSED.
Otherwise, if the next rcu_sync_exit() wins the race, it will
move it to
GP_REPLAY - owned by rcu-callback which moves it to GP_EXIT
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
[ paulmck: While here, apply READ_ONCE() and WRITE_ONCE() to ->gp_state. ]
[ paulmck: Tweaks to make htmldocs happy. (Reported by kbuild test robot.) ]
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Turn DEFINE_STATIC_PERCPU_RWSEM() into __DEFINE_PERCPU_RWSEM() with the
additional "is_static" argument to introduce DEFINE_PERCPU_RWSEM().
Change cgroup.c to use DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem).
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Use DEFINE_STATIC_PERCPU_RWSEM() to initialize dup_mmap_sem.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Now that the RCU flavors have been consolidated, rcu_sync_type makes no
sense because none of internal update functions aside from .held() depend
on gp_type. This commit therefore removes this field and consolidates
the relevant code.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
[ paulmck: Added RCU and RCU-bh checks to rcu_sync_is_idle(). ]
[ paulmck: And applied subsequent feedback from Oleg Nesterov. ]
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Because __call_srcu() is not used outside kernel/rcu/srcutree.c,
this commit makes it static.
Signed-off-by: Jiang Biao <benbjiang@tencent.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Since commit title ("srcu: Allocate per-CPU data for DEFINE_SRCU() in
modules"), modules that call DEFINE_{STATIC,}SRCU will have a new array
of srcu_struct pointers, which is used by srcu code to initialize and
clean up these structures and save valuable per-cpu reserved space.
There is no reason for this array of pointers to be writable, and can
cause security or other hidden bugs. Mark these are read-only after the
module init has completed.
Tested with the following diff to ensure array not writable:
(diff is a bit reduced to avoid patch command getting confused)
a/kernel/module.c
b/kernel/module.c
-3506,6 +3506,14 static noinline int do_init_module [snip]
rcu_assign_pointer(mod->kallsyms, &mod->core_kallsyms);
#endif
module_enable_ro(mod, true);
+
+ if (mod->srcu_struct_ptrs) {
+ // Check if srcu_struct_ptrs access is possible
+ char x = *(char *)mod->srcu_struct_ptrs;
+ *(char *)mod->srcu_struct_ptrs = 0;
+ *(char *)mod->srcu_struct_ptrs = x;
+ }
+
mod_tree_remove_init(mod);
disable_ro_nx(&mod->init_layout);
module_arch_freeing_init(mod);
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: paulmck@linux.vnet.ibm.com
Cc: rostedt@goodmis.org
Cc: mathieu.desnoyers@efficios.com
Cc: rcu@vger.kernel.org
Cc: kernel-hardening@lists.openwall.com
Cc: kernel-team@android.com
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
The SRCU for modules optimization (commit title "srcu: Allocate per-CPU
data for DEFINE_SRCU() in modules") introduced vmlinux linker entries
which is unused since it applies only to the built-in vmlinux. So remove
it to prevent any space usage due to the 8 byte alignment it added.
vmlinux.lds.h has no effect on module loading and is not used for
building the module object, so the changes were not needed in the first
place since the optimization is specific to modules.
Tested with SRCU torture_type and rcutorture. Put prints in module
loader to confirm it is able to find and initialize the srcu structures.
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Cc: kernel-team@android.com
Cc: paulmck@linux.vnet.ibm.com
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Adding DEFINE_SRCU() or DEFINE_STATIC_SRCU() to a loadable module requires
that the size of the reserved region be increased, which is not something
we want to be doing all that often. One approach would be to require
that loadable modules define an srcu_struct and invoke init_srcu_struct()
from their module_init function and cleanup_srcu_struct() from their
module_exit function. However, this is more than a bit user unfriendly.
This commit therefore creates an ___srcu_struct_ptrs linker section,
and pointers to srcu_struct structures created by DEFINE_SRCU() and
DEFINE_STATIC_SRCU() within a module are placed into that module's
___srcu_struct_ptrs section. The required init_srcu_struct() and
cleanup_srcu_struct() functions are then automatically invoked as needed
when that module is loaded and unloaded, thus allowing modules to continue
to use DEFINE_SRCU() and DEFINE_STATIC_SRCU() while avoiding the need
to increase the size of the reserved region.
Many of the algorithms and some of the code was cheerfully cherry-picked
from other code making use of linker sections, perhaps most notably from
tracepoints. All bugs are nevertheless the sole property of the author.
Suggested-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
[ paulmck: Use __section() and use "default" in srcu_module_notify()'s
"switch" statement as suggested by Joel Fernandes. ]
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
