| Age | Commit message (Collapse) | Author |
|---|
|
Nikolay Aleksandrov says:
====================
inet: frag: cleanup and update
The end goal of this patchset is to remove the LRU list and to move the
frag eviction to a work queue. It also does a couple of necessary cleanups
and fixes. Brief patch descriptions:
Patches 1 - 3 inclusive: necessary clean ups
Patch 4 moves the eviction from the softirqs to a workqueue.
Patch 5 removes the nqueues counter which was protected by the LRU lock
Patch 6 removes the, by now unused, lru list.
Patch 7 moves the rebuild timer to the workqueue and schedules the rebuilds
only if we've hit the maximum queue length on some of the chains.
Patch 8 migrate the rwlock to a seqlock since the rehash is usually a rare
operation.
Patch 9 introduces an artificial global memory limit based on the value of
init_net's high_thresh which is used to cap the high_thresh of the
other namespaces. Also introduces some sane limits on the other
tunables, and makes it impossible to have low_thresh > high_thresh.
Here are some numbers from running netperf before and after the patchset:
Each test consists of the following setting: -I 95,5 -i 15,10
1. Bound test (-T 4,4)
1.1 Virtio before the patchset -
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf. : cpu bind
Socket Message Elapsed Messages CPU Service
Size Size Time Okay Errors Throughput Util Demand
bytes bytes secs # # 10^6bits/sec % SS us/KB
212992 64000 30.00 722177 0 12325.1 34.55 2.025
212992 30.00 368020 6280.9 34.05 0.752
1.2 Virtio after the patchset -
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf. : cpu bind
Socket Message Elapsed Messages CPU Service
Size Size Time Okay Errors Throughput Util Demand
bytes bytes secs # # 10^6bits/sec % SS us/KB
212992 64000 30.00 727030 0 12407.9 35.45 1.876
212992 30.00 505405 8625.5 34.92 0.693
2. Virtio unbound test
2.1 Before the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # 10^6bits/sec
212992 64000 30.00 730008 0 12458.77
212992 30.00 416721 7112.02
2.2 After the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.122.177 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # 10^6bits/sec
212992 64000 30.00 731129 0 12477.89
212992 30.00 487707 8323.50
3. 10 gig unbound tests
3.1 Before the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.133.1 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # 10^6bits/sec
212992 64000 30.00 417209 0 7120.33
212992 30.00 416740 7112.33
3.2 After the patchset
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.133.1 () port 0 AF_INET : +/-2.500% @ 95% conf.
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # 10^6bits/sec
212992 64000 30.00 438009 0 7475.33
212992 30.00 437630 7468.87
Given the options each netperf ran between 10 and 15 times for 30 seconds
to get the necessary confidence, also the tests themselves ran 3 times and
were consistent.
Another set of tests that I ran were parallel stress tests which consisted
of flooding the machine with fragmented packets from different sources with
frag timeout set to 0 (so there're lots of timeouts) and low_thresh set to
1 byte (so evictions are happening all the time) and on top of that running
a namespace create/destroy endless loop with network interfaces and
addresses that got flooded (for the brief periods they were up) in parallel.
This test ran for an hour without any issues.
====================
|
|
This patch makes init_net's high_thresh limit to be the maximum for all
namespaces, thus introducing a global memory limit threshold equal to the
sum of the individual high_thresh limits which are capped.
It also introduces some sane minimums for low_thresh as it shouldn't be
able to drop below 0 (or > high_thresh in the unsigned case), and
overall low_thresh should not ever be above high_thresh, so we make the
following relations for a namespace:
init_net:
high_thresh - max(not capped), min(init_net low_thresh)
low_thresh - max(init_net high_thresh), min (0)
all other namespaces:
high_thresh = max(init_net high_thresh), min(namespace's low_thresh)
low_thresh = max(namespace's high_thresh), min(0)
The major issue with having low_thresh > high_thresh is that we'll
schedule eviction but never evict anything and thus rely only on the
timers.
Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
rehash is rare operation, don't force readers to take
the read-side rwlock.
Instead, we only have to detect the (rare) case where
the secret was altered while we are trying to insert
a new inetfrag queue into the table.
If it was changed, drop the bucket lock and recompute
the hash to get the 'new' chain bucket that we have to
insert into.
Joint work with Nikolay Aleksandrov.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
merge functionality into the eviction workqueue.
Instead of rebuilding every n seconds, take advantage of the upper
hash chain length limit.
If we hit it, mark table for rebuild and schedule workqueue.
To prevent frequent rebuilds when we're completely overloaded,
don't rebuild more than once every 5 seconds.
ipfrag_secret_interval sysctl is now obsolete and has been marked as
deprecated, it still can be changed so scripts won't be broken but it
won't have any effect. A comment is left above each unused secret_timer
variable to avoid confusion.
Joint work with Nikolay Aleksandrov.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
no longer used.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The 'nqueues' counter is protected by the lru list lock,
once thats removed this needs to be converted to atomic
counter. Given this isn't used for anything except for
reporting it to userspace via /proc, just remove it.
We still report the memory currently used by fragment
reassembly queues.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When the high_thresh limit is reached we try to toss the 'oldest'
incomplete fragment queues until memory limits are below the low_thresh
value. This happens in softirq/packet processing context.
This has two drawbacks:
1) processors might evict a queue that was about to be completed
by another cpu, because they will compete wrt. resource usage and
resource reclaim.
2) LRU list maintenance is expensive.
But when constantly overloaded, even the 'least recently used' element is
recent, so removing 'lru' queue first is not 'fairer' than removing any
other fragment queue.
This moves eviction out of the fast path:
When the low threshold is reached, a work queue is scheduled
which then iterates over the table and removes the queues that exceed
the memory limits of the namespace. It sets a new flag called
INET_FRAG_EVICTED on the evicted queues so the proper counters will get
incremented when the queue is forcefully expired.
When the high threshold is reached, no more fragment queues are
created until we're below the limit again.
The LRU list is now unused and will be removed in a followup patch.
Joint work with Nikolay Aleksandrov.
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
First step to move eviction handling into a work queue.
We lose two spots that accounted evicted fragments in MIB counters.
Accounting will be restored since the upcoming work-queue evictor
invokes the frag queue timer callbacks instead.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
hide actual hash size from individual users: The _find
function will now fold the given hash value into the required range.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If we have to copy data we must drop i_data_sem because of
get_blocks() will be called inside mext_page_mkuptodate(), but later we must
reacquire it again because we are about to change extent's tree
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
|
|
Inode's depth can be changed from here:
ext4_ext_try_to_merge() ->ext4_ext_try_to_merge_up()
We must use correct value.
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
|
|
Each caller of ext4_ext_dirty must hold i_data_sem,
The only exception is migration code, let's make it convenient.
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
|
|
As the member fe_len defined in struct ext4_free_extent is expressed as
number of clusters, the variable "size" computation is wrong, we need to
first translate fe_len to block number, then to bytes.
Signed-off-by: Xiaoguang Wang <wangxg.fnst@cn.fujitsu.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Lukas Czerner <lczerner@redhat.com>
|
|
The function rtas_flash_firmware passes the address of a data structure,
flash_block_list, when making the update-flash-64-and-reboot rtas call.
While the endianness of the address is handled correctly, the endianness
of the data is not. This patch ensures that the data in flash_block_list
is big endian when passed to rtas on little endian hosts.
Signed-off-by: Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
|
|
We can continue to read the error log (up to MAX size) even if
we get the elog size more than MAX size. Hence change BUG_ON to
WARN_ON.
Also updated error message.
Reported-by: Gopesh Kumar Chaudhary <gopchaud@in.ibm.com>
Signed-off-by: Vasant Hegde <hegdevasant@linux.vnet.ibm.com>
Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
Acked-by: Stewart Smith <stewart@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
|
|
STDDEV_THRESH was once defined and used in menu governor. But now its no longer
used anywhere. So removing the define.
Signed-off-by: Mohammad Merajul Islam Molla <meraj.enigma@gmail.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
|
* pm-opp:
PM / OPP: Remove ARCH_HAS_OPP
* pm-general:
MAINTAINERS: power_supply: update maintainership
|
|
* pm-apm:
x86, apm: Remove unused variable
* pm-sleep:
PM / sleep: Move platform suspend operations to separate functions
PM / sleep: Simplify sleep states sysfs interface code
|
|
* acpi-video:
ACPI: move models with win8 brightness problems from win8 blacklist to use_native_backlight
ACPI / video: Fix backlight taking 2 steps on a brightness up/down keypress
* acpi-hotplug:
ACPI / hotplug / PCI: Fix sparse non static symbol warning
ACPI / hotplug: Simplify acpi_set_hp_context()
ACPI / hotplug / PCI: Eliminate acpiphp_dev_to_bridge()
|
|
* acpi-pnp:
ACPI / PNP: Use ACPI_COMPANION() instead of ACPI_HANDLE()
ACPI / PNP: do ACPI binding directly
* acpi-pci:
ACPI / PCI: Use ACPI_COMPANION() instead of ACPI_HANDLE()
|
|
* acpi-pm:
ACPI / PM: Use ACPI_COMPANION() instead of ACPI_HANDLE()
ACPI / PM: Always enable wakeup GPEs when enabling device wakeup
ACPI / PM: Revork the handling of ACPI device wakeup notifications
PM: Create PM workqueue if runtime PM is not configured too
* acpi-sleep:
ACPI / sleep: Do not save NVS for new machines to accelerate S3
* acpi-button:
ACPI / button: Do not propagate wakeup-from-suspend events
|
|
* acpi-gpe:
ACPI / scan: No implicit wake notification for buttons
|
|
* acpi-config:
ACPI / processor: Introduce ARCH_MIGHT_HAVE_ACPI_PDC
ACPI: Don't use acpi_lapic in ACPI core code
ACPI: add config for BIOS table scan
|
|
* acpi-headers:
ACPI: Add support to force header inclusion rules for <acpi/acpi.h>.
ACPI / SFI: Fix wrong <acpi/acpi.h> inclusion in SFI/ACPI wrapper - table definitions.
ACPICA: Linux: Allow ACPICA inclusion for CONFIG_ACPI=n builds.
ACPICA: Linux: Add support to exclude <asm/acenv.h> inclusion.
ACPICA: Linux: Add stub implementation of ACPICA 64-bit mathematics.
ACPICA: Linux: Add stub support for Linux specific variables and functions.
|
|
* acpica: (30 commits)
ACPICA: Add new GPE public interface - acpi_mark_gpe_for_wake.
ACPICA: GPEs: Do not allow enable for GPEs that have no handler(s).
ACPICA: Fix a regression for deletion of Alias() objects.
ACPICA: Update version to 20140627
ACPICA: Tables: Merge DMAR table structure updates
ACPICA: Hardware: back port of a recursive locking fix
ACPICA: utprint/oslibcfs: cleanup - no functional change
ACPICA: Executer: Fix trivial issues in acpi_get_serial_access_bytes()
ACPICA: OSL: Update acpidump to reduce source code differences
ACPICA: acpidump: Reduce freopen() invocations to improve portability
ACPICA: acpidump: Replace file IOs with new APIs to improve portability
ACPICA: acpidump: Remove exit() from generic layer to improve portability
ACPICA: acpidump: Add memory/string OSL usage to improve portability
ACPICA: Common: Enhance acpi_getopt() to improve portability
ACPICA: Common: Enhance cm_get_file_size() to improve portability
ACPICA: Application: Enhance ACPI_USAGE_xxx/ACPI_OPTION with acpi_os_printf() to improve portability
ACPICA: Utilities: Introduce acpi_log_error() to improve portability
ACPICA: Utilities: Add formatted printing APIs
ACPICA: OSL: Add portable file IO to improve portability
ACPICA: OSL: Clean up acpi_os_printf()/acpi_os_vprintf() stubs
...
|
|
drivers/media/pci/solo6x10/solo6x10-disp.c:221 solo_set_motion_block() error: potential null dereference
'buf'. (kzalloc returns null)
Also propagate this error up the chain.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Commit 958c7c7e65 ("[media] v4l2-ctrls: fix corner case in round-to-range code") broke
controls that use a negative range.
The cause was a s32/u32 mixup: ctrl->step is unsigned while all others are signed. So
the result type of the expression '(ctrl)->maximum - ((ctrl)->step / 2)' became unsigned,
making 'val >= (ctrl)->maximum - ((ctrl)->step / 2)' true, since '((u32)-128) > 128'
(if val = -128, maximum = 128 and step = 1).
So carefully cast (step / 2) to s32.
There was one cast of step to s32 where it should have been u32 because both offset and
step are unsigned, so casting to signed makes no sense there. You do need a cast to u32
there, because otherwise architectures that have no 64-bit division start complaining
(step is a u64).
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Reported-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Some vendors have their on keymap table that are used on
all (or almost all) models for that vendor.
So, instead of specifying the keymap table per USB ID,
let's use the Vendor ID's table by default.
At the end, this will mean less code to be added when newer
devices for those vendors are added.
Of course, if rc_map is specified per board, it takes
precedence.
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Looks like no one's working on the driver anymore, so remove it for now.
If someone wants to work on moving it out of staging, this commit can be
reverted.
Signed-off-by: Kristina Martšenko <kristina.martsenko@gmail.com>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Mark A. Allyn <mark.a.allyn@intel.com>
Cc: Jayant Mangalampalli <jayant.mangalampalli@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Add support for the si2165-based cx231xx devices:
[2013:025e] PCTV QuatroStick 522e
[2013:0259] PCTV QuatroStick 521e
[2040:b131] Hauppauge WinTV 930C-HD (model 1114xx)
They're similar to the already supported:
[2040:b130] Hauppauge WinTV 930C-HD (model 1113xx)
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Fix the following warnings:
drivers/media/usb/cx231xx/cx231xx-cards.c: In function 'read_eeprom':
drivers/media/usb/cx231xx/cx231xx-cards.c:979:6: warning: variable 'ret' set but not used [-Wunused-but-set-variable]
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Add support for:
[2013:025e] PCTV QuatroStick 522e
The hardware is identical to Hauppauge WinTV 930C-HD (model 1114xx)
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Add support for:
[2013:0259] PCTV QuatroStick 521e
The hardware is identical to Hauppauge WinTV 930C-HD (model 1113xx)
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Add support for:
[2040:b131] Hauppauge WinTV 930C-HD (model 1114xx)
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Tested-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Add support for:
[2040:b130] Hauppauge WinTV 930C-HD (model 1113xx)
After loading the driver the first open to dvb device node fails.
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
This is needed to support PCTV QuatroStick 522e which uses a si2157.
The si2157 driver is written using i2c_client attachment.
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Tested-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
The same card entry is used for HVR-4400 and HVR-5500.
Only HVR-5500 has been tested.
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
DVB-T was tested with 8MHz BW channels in germany
This driver is the simplest possible, it uses automatic mode for all
parameters (TPS).
Added spectrum inversion support.
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Tested-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
We don't have license to redistribute this firmware, so extract
it from the driver, and add the needed headers.
Signed-off-by: Matthias Schwarzott <zzam@gentoo.org>
Tested-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
Instead of using some random value, return an error if the
PCB config is not available or doesn't match a know profile
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
That makes easier to understand the code. It would also help
to add support for having boards with just digital support
on some latter patch, as allowed by some PCB configs.
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
The initialization is already too complex. Use devm_ functions
to make the code simpler and easier to modify.
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
On some newer boards, like HVR-930C HD, the information at
the PCB tables are sometimes higher than the ones actually
available on the device. That causes the probing code to
go past the interfaces array.
Add checks to the interface number before going past the
array.
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
The max number of interfaces was read from the wrong descriptor.
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
|
|
|
|
The s->clk never gets setup in sc16is7xx_probe() and instead was using a
local clk variable, but then testing the uninitialized s->clk during
teardown
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jon Ringle <jringle@gridpoint.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
This patch started as a fix to some checkpatch complaints in ft1000
procfs code but Greg suggested to remove the procfs entries altogether:
http://driverdev.linuxdriverproject.org/pipermail/driverdev-devel/2014-July/055594.html
Signed-off-by: Nicolas Thery <nthery@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
This patch fixes the warning generated by sparse: "Using plain integer as NULL pointer" by replacing the offending 0s with NULL.
drivers/staging/lustre/lustre/lov/lov_obd.c:902:48: warning: Using plain integer as NULL pointer
drivers/staging/lustre/lustre/lov/lov_obd.c:946:54: warning: Using plain integer as NULL pointer
drivers/staging/lustre/lustre/lov/lov_obd.c:2819:46: warning: Using plain integer as NULL pointer
Signed-off-by: Marc Fite <marc@fite.cat>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Fixed a coding style issue.
Signed-off-by: Sylvain Calador <sylvain.calador@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
