31ef83dc05 "nfsd: add trace events" had a typo that dropped a trace
event and replaced it by an incorrect recursive call to
nfsd4_cb_layout_fail. 133d558216 "Subject: nfsd: don't recursively
call nfsd4_cb_layout_fail" fixed the crash, this restores the
tracepoint.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Eric Dumazet says:
====================
inet: tcp listener refactoring part 14
OK, we have serious patches here.
We get rid of the central timer handling SYNACK rtx,
which is killing us under even medium SYN flood.
We still use the listener specific hash table.
This will be done in next round ;)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
sk_ack_backlog & sk_max_ack_backlog were 16bit fields, meaning
listen() backlog was limited to 65535.
It is time to increase the width to allow much bigger backlog,
if admins change /proc/sys/net/core/somaxconn &
/proc/sys/net/ipv4/tcp_max_syn_backlog default values.
Tested:
echo 5000000 >/proc/sys/net/core/somaxconn
echo 5000000 >/proc/sys/net/ipv4/tcp_max_syn_backlog
Ran a SYNFLOOD test against a listener using listen(fd, 5000000)
myhost~# grep request_sock_TCP /proc/slabinfo
request_sock_TCP 4185642 4411940 304 13 1 : tunables 54 27 8 : slabdata 339380 339380 0
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
One of the major issue for TCP is the SYNACK rtx handling,
done by inet_csk_reqsk_queue_prune(), fired by the keepalive
timer of a TCP_LISTEN socket.
This function runs for awful long times, with socket lock held,
meaning that other cpus needing this lock have to spin for hundred of ms.
SYNACK are sent in huge bursts, likely to cause severe drops anyway.
This model was OK 15 years ago when memory was very tight.
We now can afford to have a timer per request sock.
Timer invocations no longer need to lock the listener,
and can be run from all cpus in parallel.
With following patch increasing somaxconn width to 32 bits,
I tested a listener with more than 4 million active request sockets,
and a steady SYNFLOOD of ~200,000 SYN per second.
Host was sending ~830,000 SYNACK per second.
This is ~100 times more what we could achieve before this patch.
Later, we will get rid of the listener hash and use ehash instead.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When request sock are put in ehash table, the whole notion
of having a previous request to update dl_next is pointless.
Also, following patch will get rid of big purge timer,
so we want to delete a request sock without holding listener lock.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As noted in earlier commit logs, CPU hotplug operations running
concurrently with grace-period initialization can result in a given
leaf rcu_node structure having all CPUs offline and no blocked readers,
but with this rcu_node structure nevertheless blocking the current
grace period. Therefore, the quiescent-state forcing code now checks
for this situation and repairs it.
Unfortunately, this checking can result in false positives, for example,
when the last task has just removed itself from this leaf rcu_node
structure, but has not yet started clearing the ->qsmask bits further
up the structure. This means that the grace-period kthread (which
forces quiescent states) and some other task might be attempting to
concurrently clear these ->qsmask bits. This is usually not a problem:
One of these tasks will be the first to acquire the upper-level rcu_node
structure's lock and with therefore clear the bit, and the other task,
seeing the bit already cleared, will stop trying to clear bits.
Sadly, this means that the following unusual sequence of events -can-
result in a problem:
1. The grace-period kthread wins, and clears the ->qsmask bits.
2. This is the last thing blocking the current grace period, so
that the grace-period kthread clears ->qsmask bits all the way
to the root and finds that the root ->qsmask field is now zero.
3. Another grace period is required, so that the grace period kthread
initializes it, including setting all the needed qsmask bits.
4. The leaf rcu_node structure (the one that started this whole
mess) is blocking this new grace period, either because it
has at least one online CPU or because there is at least one
task that had blocked within an RCU read-side critical section
while running on one of this leaf rcu_node structure's CPUs.
(And yes, that CPU might well have gone offline before the
grace period in step (3) above started, which can mean that
there is a task on the leaf rcu_node structure's ->blkd_tasks
list, but ->qsmask equal to zero.)
5. The other kthread didn't get around to trying to clear the upper
level ->qsmask bits until all the above had happened. This means
that it now sees bits set in the upper-level ->qsmask field, so it
proceeds to clear them. Too bad that it is doing so on behalf of
a quiescent state that does not apply to the current grace period!
This sequence of events can result in the new grace period being too
short. It can also result in the new grace period ending before the
leaf rcu_node structure's ->qsmask bits have been cleared, which will
result in splats during initialization of the next grace period. In
addition, it can result in tasks blocking the new grace period still
being queued at the start of the next grace period, which will result
in other splats. Sasha's testing turned up another of these splats,
as did rcutorture testing. (And yes, rcutorture is being adjusted to
make these splats show up more quickly. Which probably is having the
undesirable side effect of making other problems show up less quickly.
Can't have everything!)
Reported-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: <stable@vger.kernel.org> # 4.0.x
Tested-by: Sasha Levin <sasha.levin@oracle.com>
Instead of looking up the destination station twice in the TX path
(first to build the header, and then for control processing), save
it when building the header and use it later in the TX path.
To avoid having to look up the station in the many callers, allow
those to pass %NULL which keeps the existing lookup.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
In ieee80211_build_hdr(), the station is looked up to build the
header correctly (QoS field) and to check for authorization. For
mesh, authorization isn't checked here, and QoS capability is
mandatory, so the station lookup can be avoided.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
If there's no station on the 4-addr VLAN interface, then frames
cannot be transmitted. Drop such frames earlier, before setting
up all the information for them.
We should keep the old check though since that code might be used
for other internally-generated frames.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
There's no need to look up the destination station twice while
building the 802.11 header for a given frame if the frame will
actually be transmitted to the station we initially looked up.
This happens for 4-addr VLAN interfaces and TDLS connections, which
both directly send the frame to the station they looked up, though
in the case of TDLS some station conditions need to be checked.
To avoid that, add a variable indicating that we've looked up the
station that the frame is going to be transmitted to, and avoid the
lookup/flag checking if it already has been done.
In the TDLS case, also move the authorized/wme_sta flag assignment
to the correct place, i.e. only when that station is really used.
Before this change, the new lookup should always have succeeded so
that the potentially erroneous data would be overwritten.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
As noted earlier, the following sequence of events can occur when
running PREEMPT_RCU and HOTPLUG_CPU on a system with a multi-level
rcu_node combining tree:
1. A group of tasks block on CPUs corresponding to a given leaf
rcu_node structure while within RCU read-side critical sections.
2. All CPUs corrsponding to that rcu_node structure go offline.
3. The next grace period starts, but because there are still tasks
blocked, the upper-level bits corresponding to this leaf rcu_node
structure remain set.
4. All the tasks exit their RCU read-side critical sections and
remove themselves from the leaf rcu_node structure's list,
leaving it empty.
5. But because there now is code to check for this condition at
force-quiescent-state time, the upper bits are cleared and the
grace period completes.
However, there is another complication that can occur following step 4 above:
4a. The grace period starts, and the leaf rcu_node structure's
gp_tasks pointer is set to NULL because there are no tasks
blocked on this structure.
4b. One of the CPUs corresponding to the leaf rcu_node structure
comes back online.
4b. An endless stream of tasks are preempted within RCU read-side
critical sections on this CPU, such that the ->blkd_tasks
list is always non-empty.
The grace period will never end.
This commit therefore makes the force-quiescent-state processing check only
for absence of tasks blocking the current grace period rather than absence
of tasks altogether. This will cause a quiescent state to be reported if
the current leaf rcu_node structure is not blocking the current grace period
and its parent thinks that it is, regardless of how RCU managed to get
itself into this state.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: <stable@vger.kernel.org> # 4.0.x
Tested-by: Sasha Levin <sasha.levin@oracle.com>
The current OPAL firmware exposes the different sensors of an IBM Power
system using node names such as :
sensors/amb-temp#1-data
sensors/amb-temp#1-thrs
cooling-fan#1-data
cooling-fan#1-faulted
cooling-fan#1-thrs
cooling-fan#2-data
...
The ibmpowernv driver, when loaded, parses these names to extract the
sensor index and the sensor attribute name. Unfortunately, this scheme
makes it difficult to add sensors with a different layout (specially of
the same type, like temperature) as the sensor index calculated in OPAL
is directly used in the hwmon sysfs interface.
What this patch does is add a independent hwmon index for each sensor.
The increment of the hwmon index (temp, fan, power, etc.) is kept per
sensor type in the sensor_group table. The sensor_data table is used
to store the association of the hwmon and OPAL indexes, as we need to
have the same hwmon index for different attributes of a same sensor.
Signed-off-by: Cédric Le Goater <clg@fr.ibm.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
It simplifies the creation of the hwmon attributes and will help when
support for a new device tree layout is added. The patch also changes
the name of the routine to parse_opal_node_name().
Signed-off-by: Cédric Le Goater <clg@fr.ibm.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
It simplifies the create_hwmon_attr_name() routine and it clearly isolates
the conversion done between the OPAL node names and hwmon attributes names.
Signed-off-by: Cédric Le Goater <clg@fr.ibm.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
It will help in adding different compatible properties, coming from a
new device tree layout for example.
Signed-off-by: Cédric Le Goater <clg@fr.ibm.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Ambient is too restrictive as there can be other temperature channels :
core, memory, etc.
Signed-off-by: Cédric Le Goater <clg@fr.ibm.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Use the right array index to reference the last
element of rq->biotail->bi_io_vec[]
Signed-off-by: Wenbo Wang <wenbo.wang@memblaze.com>
Reviewed-by: Chong Yuan <chong.yuan@memblaze.com>
Fixes: 66cb45aa41 ("block: add support for limiting gaps in SG lists")
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <axboe@fb.com>
p9_mount_tag_show() uses '%s' format string to print
non-NULL terminated chan->tag string. This leads
to out of bounds memory read, because format '%s'
implies that string is NULL-terminated.
The length of string is know here, so its simpler and safer
to use memcpy instead of snprintf().
Signed-off-by: Andrey Ryabinin <a.ryabinin@samsung.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
As specification says, all integers in messages are unsigned. Let's fix
behaviour of p9pdu_vreadf()/p9pdu_vwritef() accordingly.
Fix for p9pdu_vreadf() is critical. If server replies with Rwalk, where
nwqid > SHRT_MAX, the value will be interpreted as negative. kmalloc, in
its order, will cast the value to (very big) size_t.
It should never happen in normal situation: we never submit Twalk with
nwname > 16, but malicious or broken server can still produce
problematic Rwalk.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Current 9p implementation will crash whole system if sees unknown lock
status code. It's trivial target for DOS: 9p server can produce such
code easily.
Let's fallback more gracefully: warning in dmesg + -ENOLCK.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
p9_client_lock_dotl() doesn't set status if p9_client_rpc() fails.
It can lead to 'default:' case in switch below and kernel crashes.
Let's bypass the switch if p9_client_lock_dotl() fails.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
p is initialized but unused.
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
options argument was removed from v9fs_session_info in commit 4b53e4b500
("9p: remove unnecessary v9fses->options which duplicates the mount string")
iov and nr_segs were removed from v9fs_direct_IO
in commit d8d3d94b80
("pass iov_iter to ->direct_IO()")
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Cc: Ron Minnich <rminnich@sandia.gov>
Cc: Latchesar Ionkov <lucho@ionkov.net>
Cc: v9fs-developer@lists.sourceforge.net
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Commit b4b55cda58 (Refine the way to release PCI IRQ resources)
introduced a regression in the PCI IRQ resource management by causing
the IRQ resource of a device, established when pci_enabled_device()
is called on a fully disabled device, to be released when the driver
is unbound from the device, regardless of the enable_cnt.
This leads to the situation that an ill-behaved driver can now make a
device unusable to subsequent drivers by an imbalance in their use of
pci_enable/disable_device(). That is a serious problem for secondary
drivers like vfio-pci, which are innocent of the transgressions of
the previous driver.
Since the solution of this problem is not immediate and requires
further discussion, revert commit b4b55cda58 and the issue it was
supposed to address (a bug related to xen-pciback) will be taken
care of in a different way going forward.
Reported-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Kernel.org is only hosting patches and kernel compressed with xz,
so change the old gzip/bzip2 instances to xz.
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
When 4.0 is released, the README should reflect the new numbering.
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
This patch fix spelling typo in Documentation/PCI.
Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The documentation is trying to describe accessing a field through a
pointer, but it is using '-<' instead of '->'. Fix that.
Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Remove mentioning of block barriers since they were removed.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Leonid V. Fedorenchik <leonidsbox@gmail.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Fix one grammar mistake (Allows->Allow) and add extra information about
the external editor add-on of Thunderbird: the developer must make sure
that their editor doesn't fork (IOW it mustn't return before closing)
thus they should be careful how they configure the addon. Furthermore,
add a tip how to do it with gvim.
Signed-off-by: Giedrius Statkevičius <giedrius.statkevicius@gmail.com>
[jc: some minor wording/formatting tweaks]
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The patch 125e564("Move Kconfig.instrumentation to arch/Kconfig and
init/Kconfig") had removed the "Instrumentation Support" menu,
and the configurations under this had be moved to "General setup".
Update Documentation/kprobes.txt to reflect this change.
Signed-off-by: Li Bin <huawei.libin@huawei.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
We can remove the i2o documentation because a) the subsystem has been
moved to staging with commit 2cbf7fe2d5 (i2o: move to staging)
anyhow and b) the here removed files are present in the subsystem
directory again. There, README even has an additional paragraph and the
ioctl docs only differ in whitespaces. Well...
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The CONFIG_DEBUG_PAGEALLOC option now is located under "Kernel
hacking" / "Memory Debugging" / "Debug page memory allocations".
so we should update the description in kmemcheck.txt.
Signed-off-by: Wang Long <long.wanglong@huawei.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Fix a trivial typo in Documentation/memory-hotplug.txt.
Signed-off-by: Sheng Yong <shengyong1@huawei.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The documentation and the code disagrees; fix the former.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The IRQF_DISABLED is a NOOP and scheduled to be removed. According to Ingo
Molnar in commit e58aa3d2d0 (genirq: Run irq
handlers with interrupts disabled), running IRQ handlers with interrupts
enabled can cause stack overflows when the interrupt line of the issuing
device is still active.
This patch removes IRQF_DISABLED from this documentation. It was
mentioned to be a solution to avoid deadlocks when a device uses
multiple interrupts. As the flag is a NOOP this solution does not work
anymore.
Signed-off-by: Valentin Rothberg <Valentin.Rothberg@lip6.fr>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
max_ptes_none specifies how many extra small pages (that are
not already mapped) can be allocated when collapsing a group
of small pages into one large page.
/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none
A higher value leads to use additional memory for programs.
A lower value leads to gain less thp performance. Value of
max_ptes_none can waste cpu time very little, you can
ignore it.
Signed-off-by: Ebru Akagunduz <ebru.akagunduz@gmail.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
as the commit: "lib/vsprintf: implement bitmap printing through
'%*pb[l]'" add an easy way to print bitmaps. so printk-formats.txt
should reflect it.
Signed-off-by: Wang Long <long.wanglong@huawei.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
We have to check for IP6T_INV_PROTO in invflags, instead of flags.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Balazs Scheidler <bazsi@balabit.hu>
Fix CamelCase names:
isLocal => local
hasStandardPayloadHeader => standard_payload_header
tryAgain => retry
Away => cleanup
Fix spacing after typecasts, add missing braces to the if statement, and
eliminate the NULL comparison by just using the pointer value directly.
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert all references to PARSER_CONTEXT into struct parser_context.
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove unnecessary NULLs by just using the pointer value directly.
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Take out the variable used to compute a comparison against NULL, and
just use the parameter directly.
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix the name and references:
Poll_jiffies => poll_jiffies
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix all comparisons to NULL so that they don't use NULL anymore.
Signed-off-by: Benjamin Romer <benjamin.romer@unisys.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
