The NUMA PTE scan rate is controlled with a combination of the
numa_balancing_scan_period_min, numa_balancing_scan_period_max and
numa_balancing_scan_size. This scan rate is independent of the size
of the task and as an aside it is further complicated by the fact that
numa_balancing_scan_size controls how many pages are marked pte_numa and
not how much virtual memory is scanned.
In combination, it is almost impossible to meaningfully tune the min and
max scan periods and reasoning about performance is complex when the time
to complete a full scan is is partially a function of the tasks memory
size. This patch alters the semantic of the min and max tunables to be
about tuning the length time it takes to complete a scan of a tasks occupied
virtual address space. Conceptually this is a lot easier to understand. There
is a "sanity" check to ensure the scan rate is never extremely fast based on
the amount of virtual memory that should be scanned in a second. The default
of 2.5G seems arbitrary but it is to have the maximum scan rate after the
patch roughly match the maximum scan rate before the patch was applied.
On a similar note, numa_scan_period is in milliseconds and not
jiffies. Properly placed pages slow the scanning rate but adding 10 jiffies
to numa_scan_period means that the rate scanning slows depends on HZ which
is confusing. Get rid of the jiffies_to_msec conversion and treat it as ms.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-18-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
PTE scanning and NUMA hinting fault handling is expensive so commit
5bca2303 ("mm: sched: numa: Delay PTE scanning until a task is scheduled
on a new node") deferred the PTE scan until a task had been scheduled on
another node. The problem is that in the purely shared memory case that
this may never happen and no NUMA hinting fault information will be
captured. We are not ruling out the possibility that something better
can be done here but for now, this patch needs to be reverted and depend
entirely on the scan_delay to avoid punishing short-lived processes.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-16-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge Linux v3.12-rc4 to fix a conflict and also to refresh the tree
before applying more scheduler patches.
Conflicts:
arch/avr32/include/asm/Kbuild
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The unpacked_lun field in the SCSI target tracepoints should be
initialized with cmd->orig_fe_lun rather than cmd->se_lun->unpacked_lun
for two reasons:
- most importantly, if we are in the cmd_complete tracepoint
returning a check condition due to no LUN found, cmd->se_lun will
be NULL and we'll crash trying to dereference it.
- also, in any case, cmd->se_lun->unpacked_lun is an internal index
into the target's internal set of LUNs; cmd->orig_fe_lun is much
more useful and interesting, since it's the value the initiator
actually sent.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Cc: <stable@vger.kernel.org> # 3.11+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
It is incorrect to refer to this as 11d as 802.11d was just a
proposed amendment, 802.11d was merged to the standard so
use proper terminology.
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
drm_vblank_init() is too ugly. Make it a bit easier on the eye by
collecting all the per-crtc vblank counters, timestamps etc. to
a structure and just allocate an array of those.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Analog to drm_dev_register(), we now provide drm_dev_unregister() which
does the reverse. drm_dev_put() is still in place and combines the calls
to drm_dev_unregister() and drm_dev_free() so buses don't have to change.
*_get() and *_put() are used for reference-counting in the kernel.
However, drm_dev_put() definitely does not do any kind of ref-counting.
Hence, use the more appropriate *_register(), *_unregister(), *_alloc()
and *_free() names.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The error paths in DRM bus drivers currently leak memory as they don't
correctly revert drm_dev_alloc(). Introduce drm_dev_free() to free DRM
devices which haven't been registered, yet.
We must be careful not to introduce any side-effects with cleanups done in
drm_dev_free(). drm_ht_remove(), drm_ctxbitmap_cleanup() and
drm_gem_destroy() are all fine in that regard.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
All bus drivers do device setup themselves. This requires us to adjust all
of them if we introduce new core features. Thus, merge all these into a
uniform drm_dev_register() helper.
Note that this removes the drm_lastclose() error path for AGP as it is
horribly broken. Moreover, no bus driver called this in any other error
path either. Instead, we use the recently introduced AGP cleanup helpers.
We also keep a DRIVER_MODESET condition around pci_set_drvdata() to keep
semantics.
[airlied: keep passing flags through so drivers don't oops on load]
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Instead of managing device allocation+initialization in each bus-driver,
we should do that in a central place. drm_fill_in_dev() already does most
of it, but also requires the global drm lock for partial AGP device
registration.
Split both apart so we have a clean device initialization/allocation
phase, and a registration phase.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
TCP listener refactoring, part 4 :
To speed up inet lookups, we moved IPv4 addresses from inet to struct
sock_common
Now is time to do the same for IPv6, because it permits us to have fast
lookups for all kind of sockets, including upcoming SYN_RECV.
Getting IPv6 addresses in TCP lookups currently requires two extra cache
lines, plus a dereference (and memory stall).
inet6_sk(sk) does the dereference of inet_sk(__sk)->pinet6
This patch is way bigger than its IPv4 counter part, because for IPv4,
we could add aliases (inet_daddr, inet_rcv_saddr), while on IPv6,
it's not doable easily.
inet6_sk(sk)->daddr becomes sk->sk_v6_daddr
inet6_sk(sk)->rcv_saddr becomes sk->sk_v6_rcv_saddr
And timewait socket also have tw->tw_v6_daddr & tw->tw_v6_rcv_saddr
at the same offset.
We get rid of INET6_TW_MATCH() as INET6_MATCH() is now the generic
macro.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP listener refactoring, part 3 :
Our goal is to hash SYN_RECV sockets into main ehash for fast lookup,
and parallel SYN processing.
Current inet_ehash_bucket contains two chains, one for ESTABLISH (and
friend states) sockets, another for TIME_WAIT sockets only.
As the hash table is sized to get at most one socket per bucket, it
makes little sense to have separate twchain, as it makes the lookup
slightly more complicated, and doubles hash table memory usage.
If we make sure all socket types have the lookup keys at the same
offsets, we can use a generic and faster lookup. It turns out TIME_WAIT
and ESTABLISHED sockets already have common lookup fields for IPv4.
[ INET_TW_MATCH() is no longer needed ]
I'll provide a follow-up to factorize IPv6 lookup as well, to remove
INET6_TW_MATCH()
This way, SYN_RECV pseudo sockets will be supported the same.
A new sock_gen_put() helper is added, doing either a sock_put() or
inet_twsk_put() [ and will support SYN_RECV later ].
Note this helper should only be called in real slow path, when rcu
lookup found a socket that was moved to another identity (freed/reused
immediately), but could eventually be used in other contexts, like
sock_edemux()
Before patch :
dmesg | grep "TCP established"
TCP established hash table entries: 524288 (order: 11, 8388608 bytes)
After patch :
TCP established hash table entries: 524288 (order: 10, 4194304 bytes)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
include/linux/netdevice.h
net/core/sock.c
Trivial merge issues.
Removal of "extern" for functions declaration in netdevice.h
at the same time "const" was added to an argument.
Two parallel line additions in net/core/sock.c
Signed-off-by: David S. Miller <davem@davemloft.net>
The since the removal of the routing cache computing
fib_compute_spec_dst() does a fib_table lookup for each UDP multicast
packet received. This has introduced a performance regression for some
UDP workloads.
This change skips populating the packet info for sockets that do not have
IP_PKTINFO set.
Benchmark results from a netperf UDP_RR test:
Before 89789.68 transactions/s
After 90587.62 transactions/s
Benchmark results from a fio 1 byte UDP multicast pingpong test
(Multicast one way unicast response):
Before 12.63us RTT
After 12.48us RTT
Signed-off-by: Shawn Bohrer <sbohrer@rgmadvisors.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The removal of the routing cache introduced a performance regression for
some UDP workloads since a dst lookup must be done for each packet.
This change caches the dst per socket in a similar manner to what we do
for TCP by implementing early_demux.
For UDP multicast we can only cache the dst if there is only one
receiving socket on the host. Since caching only works when there is
one receiving socket we do the multicast socket lookup using RCU.
For UDP unicast we only demux sockets with an exact match in order to
not break forwarding setups. Additionally since the hash chains may be
long we only check the first socket to see if it is a match and not
waste extra time searching the whole chain when we might not find an
exact match.
Benchmark results from a netperf UDP_RR test:
Before 87961.22 transactions/s
After 89789.68 transactions/s
Benchmark results from a fio 1 byte UDP multicast pingpong test
(Multicast one way unicast response):
Before 12.97us RTT
After 12.63us RTT
Signed-off-by: Shawn Bohrer <sbohrer@rgmadvisors.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The wake-up interrupt bit is available on omap3/4/5 processors
unlike what we claim. Without fixing it we cannot use it on
omap3 and the system configured for wake-up events will just
hang on wake-up.
Cc: Grygorii Strashko <grygorii.strashko@ti.com>
Cc: Benoît Cousson <bcousson@baylibre.com>
Cc: devicetree@vger.kernel.org
Signed-off-by: Tony Lindgren <tony@atomide.com>
Pull perf fixes from Ingo Molnar:
"Various fixlets:
On the kernel side:
- fix a race
- fix a bug in the handling of the perf ring-buffer data page
On the tooling side:
- fix the handling of certain corrupted perf.data files
- fix a bug in 'perf probe'
- fix a bug in 'perf record + perf sched'
- fix a bug in 'make install'
- fix a bug in libaudit feature-detection on certain distros"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf session: Fix infinite loop on invalid perf.data file
perf tools: Fix installation of libexec components
perf probe: Fix to find line information for probe list
perf tools: Fix libaudit test
perf stat: Set child_pid after perf_evlist__prepare_workload()
perf tools: Add default handler for mmap2 events
perf/x86: Clean up cap_user_time* setting
perf: Fix perf_pmu_migrate_context
Add minimal device tree source needed for DRA7 based SoCs.
Also add a board dts file for the dra7-evm (based on dra752)
which contains 1.5G of memory with 1G interleaved and 512MB
non-interleaved. Also added in the board file are pin configuration
details for i2c, mcspi and uart devices on board.
Signed-off-by: R Sricharan <r.sricharan@ti.com>
Signed-off-by: Rajendra Nayak <rnayak@ti.com>
Signed-off-by: Sourav Poddar <sourav.poddar@ti.com>
Signed-off-by: Benoit Cousson <bcousson@baylibre.com>
The names are prefixed incorrectly on the documentation.
Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com>
[also remove spurious blank line]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
This adds a new driver to support the s3c24xx dma using the dmaengine
and makes the old one in mach-s3c24xx obsolete in the long run.
Conceptually the s3c24xx-dma feels like a distant relative of the pl08x
with numerous virtual channels being mapped to a lot less physical ones.
The driver therefore borrows a lot from the amba-pl08x driver in this
regard. Functionality-wise the driver gains a memcpy ability in addition
to the slave_sg one.
The driver supports both the method for requesting the peripheral used
by SoCs before the S3C2443 and the different method for S3C2443 and later.
On earlier SoCs the hardware channels usable for specific peripherals is
constrainted while on later SoCs all channels can be used for any peripheral.
Tested on a s3c2416-based board, memcpy using the dmatest module and
slave_sg partially using the spi-s3c64xx driver.
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Conflicts:
drivers/net/wireless/brcm80211/brcmfmac/dhd_bus.h
drivers/net/wireless/rtlwifi/rtl8188ee/phy.h
drivers/net/wireless/rtlwifi/rtl8192ce/phy.h
drivers/net/wireless/rtlwifi/rtl8192de/phy.h
drivers/net/wireless/rtlwifi/rtl8723ae/phy.h
Just some minor conflicts between the wireless-next changes
and Joe Perches's "extern" removal from function prototypes
in header files.
John W. Linville says:
====================
Regarding the Bluetooth bits, Gustavo says:
"The big work here is from Marcel and Johan. They did a lot of work
in the L2CAP, HCI and MGMT layers. The most important ones are the
addition of a new MGMT command to enable/disable LE advertisement
and the introduction of the HCI user channel to allow applications
to get directly and exclusive access to Bluetooth devices."
As to the ath10k bits, Kalle says:
"Bartosz dropped support for qca98xx hw1.0 hardware from ath10k, it's
just too much to support it. Michal added support for the new firmware
interface. Marek fixed WEP in AP and IBSS mode. Rest of the changes are
minor fixes or cleanups."
And also:
"Major changes are:
* throughput improvements including aligning the RX frames correctly and
optimising HTT layer (Michal)
* remove qca98xx hw1.0 support (Bartosz)
* add support for firmware version 999.999.0.636 (Michal)
* firmware htt statistics support (Kalle)
* fix WEP in AP and IBSS mode (Marek)
* fix a mutex unlock balance in debugfs file (Shafi)
And of course there's a lot of smaller fixes and cleanup."
For the wl12xx bits, Luca says:
"Here are some patches intended for 3.13. Eliad is upstreaming a bunch
of patches that have been pending in the internal tree. Mostly bugfixes
and other small improvements."
Along with that...
Arend and friends bring us a batch of brcmfmac updates, Larry Finger
offers some rtlwifi refactoring, and Sujith sends the usual batch of
ath9k updates. As usual, there are a number of other small updates
from a variety of players as well.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Separate the unreg_list and the close_list in dev_close_many preventing
dev_close_many from permuting the unreg_list. The permutations of the
unreg_list have resulted in cases where the loopback device is accessed
it has been freed in code such as dst_ifdown. Resulting in subtle memory
corruption.
This is the second bug from sharing the storage between the close_list
and the unreg_list. The issues that crop up with sharing are
apparently too subtle to show up in normal testing or usage, so let's
forget about being clever and use two separate lists.
v2: Make all callers pass in a close_list to dev_close_many
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Provide tracepoints for the lifecycle of a message from submission to
completion and for the active time for masters to help with performance
analysis of SPI I/O.
Signed-off-by: Mark Brown <broonie@linaro.org>
From Sebastian Hasselbarth:
This is a patch set based on an RFC [1][2] sent earlier to provide
a common arch/arm init for DT clock providers. Currently, the call to
of_clk_init(NULL) to initialize DT clock providers is spread among several
mach-dirs. Since most machs require DT clocks initialized before timers,
no initcall can be used.
By adding of_clk_init(NULL) to arch/arm time_init(), we can remove all
mach-specific .init_time hooks that basically called of_clk_init and
clocksource_of_init.
In contrast to the RFC version, of_clk_init(NULL) is now only called if
no custom .init_time callback is set. This allows some machs to still
call clock init themselves, as not all can be converted now. Therefore,
this patch sets drops conversion of mach-mvebu and mach-zynq. New machs
that were introduced with v3.12-rc1 are also converted, except mach-u300
that requires clocks before irqs.
* 'clk-of-init-v2_for-3.13' of https://github.com/shesselba/linux-dove: (29 commits)
ARM: vt8500: remove custom .init_time hook
ARM: vexpress: remove custom .init_time hook
ARM: tegra: remove custom .init_time hook
ARM: sunxi: remove custom .init_time hook
ARM: sti: remove custom .init_time hook
ARM: socfpga: remove custom .init_time hook
ARM: rockchip: remove custom .init_time hook
ARM: prima2: remove custom .init_time hook
ARM: nspire: remove custom .init_time hook
ARM: nomadik: remove custom .init_time hook
ARM: mxs: remove custom .init_time hook
ARM: kirkwood: remove custom .init_time hook
ARM: imx: remove custom .init_time hook
ARM: highbank: remove custom .init_time hook
ARM: exynos: remove custom .init_time hook
ARM: dove: remove custom .init_time hook
ARM: bcm2835: remove custom .init_time hook
ARM: bcm: provide common arch init for DT clocks
ARM: call of_clk_init from default time_init handler
ARM: vt8500: prepare for arch-wide .init_time callback
...
Signed-off-by: Olof Johansson <olof@lixom.net>
Pull HID fixes from Jiri Kosina:
- fix for hidraw reference counting regression, by Manoj Chourasia
- fix for minor number allocation for uhid, by David Herrmann
- other small unsorted fixes / device ID additions
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
HID: wiimote: fix FF deadlock
HID: add Holtek USB ID 04d9:a081 SHARKOON DarkGlider
HID: hidraw: close underlying device at removal of last reader
HID: roccat: Fix "cannot create duplicate filename" problems
HID: uhid: allocate static minor
virtio wants to pass in cpumask_of(cpu), make parameter
const to avoid build warnings.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When initializing an AMP controller, read its current known location
data so that it can be analyzed later on.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
This patch adds support for virtual DAPM mixer controls. They are similar to
virtual DAPM enums. There is no hardware register backing the control, so
changing the control's value wont have any direct effect on the hardware. But it
still influences the DAPM graph by causing the path it sits on to be connected
or disconnected. This in turn can cause power changes for some of the widgets on
the DAPM graph, which will then modify the hardware state.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Tested-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@linaro.org>
The return value of mgmt_discovering() function is not used
and so just change it to return void.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
The return value of mgmt_remote_name() function is not used
and so just change it to return void.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
The return value of mgmt_device_found() function is not used
and so just change it to return void.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
The return value of mgmt_device_disconnected() function is not used
and so just change it to return void.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
The return value of mgmt_device_connected() function is not used
and so just change it to return void.
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
