| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The "nodir" mode (statically assign master nodes instead
of using the resource directory) has always been highly
experimental, and never seriously used. This commit
fixes a number of problems, making nodir much more usable.
- Major change to recovery: recover all locks and restart
all in-progress operations after recovery. In some
cases it's not possible to know which in-progess locks
to recover, so recover all. (Most require recovery
in nodir mode anyway since rehashing changes most
master nodes.)
- Change the way nodir mode is enabled, from a command
line mount arg passed through gfs2, into a sysfs
file managed by dlm_controld, consistent with the
other config settings.
- Allow recovering MSTCPY locks on an rsb that has not
yet been turned into a master copy.
- Ignore RCOM_LOCK and RCOM_LOCK_REPLY recovery messages
from a previous, aborted recovery cycle. Base this
on the local recovery status not being in the state
where any nodes should be sending LOCK messages for the
current recovery cycle.
- Hold rsb lock around dlm_purge_mstcpy_locks() because it
may run concurrently with dlm_recover_master_copy().
- Maintain highbast on process-copy lkb's (in addition to
the master as is usual), because the lkb can switch
back and forth between being a master and being a
process copy as the master node changes in recovery.
- When recovering MSTCPY locks, flag rsb's that have
non-empty convert or waiting queues for granting
at the end of recovery. (Rename flag from LOCKS_PURGED
to RECOVER_GRANT and similar for the recovery function,
because it's not only resources with purged locks
that need grant a grant attempt.)
- Replace a couple of unnecessary assertion panics with
error messages.
Signed-off-by: David Teigland <[email protected]>
|
|
This resolves the conflict in:
drivers/staging/vt6656/ioctl.c
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
The common clock framework allocates clocks dynamically. Provide a
set of helpers to streamline the clkdev registration of the clock
lookups to avoid repetitive code sequences.
Reviewed-by: Viresh Kumar <[email protected]>
Tested-by: Shawn Guo <[email protected]>
Signed-off-by: Russell King <[email protected]>
|
|
git://people.freedesktop.org/~danvet/drm-intel into drm-core-next
Daniel Vetter writes:
A new drm-intel-next pull. Highlights:
- More gmbus patches from Daniel Kurtz, I think gmbus is now ready, all
known issues fixed.
- Fencing cleanup and pipelined fencing removal from Chris.
- rc6 residency interface from Ben, useful for powertop.
- Cleanups and code reorg around the ringbuffer code (Ben&me).
- Use hw semaphores in the pageflip code from Ben.
- More vlv stuff from Jesse, unfortunately his vlv cpu is doa, so less
merged than I've hoped for - we still have the unused function warning :(
- More hsw patches from Eugeni, again, not yet enabled fully.
- intel_pm.c refactoring from Eugeni.
- Ironlake sprite support from Chris.
- And various smaller improvements/fixes all over the place.
Note that this pull request also contains a backmerge of -rc3 to sort out
a few things in -next. I've also had to frob the shortlog a bit to exclude
anything that -rc3 brings in with this pull.
Regression wise we have a few strange bugs going on, but for all of them
closer inspection revealed that they've been pre-existing, just now
slightly more likely to be hit. And for most of them we have a patch
already. Otherwise QA has not reported any regressions, and I'm also not
aware of anything bad happening in 3.4.
* tag 'drm-intel-next-2012-04-23' of git://people.freedesktop.org/~danvet/drm-intel: (420 commits)
drm/i915: rc6 residency (fix the fix)
drm/i915/tv: fix open-coded ARRAY_SIZE.
drm/i915: invalidate render cache on gen2
drm/i915: Silence the change of LVDS sync polarity
drm/i915: add generic power management initialization
drm/i915: move clock gating functionality into intel_pm module
drm/i915: move emon functionality into intel_pm module
drm/i915: move drps, rps and rc6-related functions to intel_pm
drm/i915: fix line breaks in intel_pm
drm/i915: move watermarks settings into intel_pm module
drm/i915: move fbc-related functionality into intel_pm module
drm/i915: Refactor get_fence() to use the common fence writing routine
drm/i915: Refactor fence clearing to use the common fence writing routine
drm/i915: Refactor put_fence() to use the common fence writing routine
drm/i915: Prepare to consolidate fence writing
drm/i915: Remove the unsightly "optimisation" from flush_fence()
drm/i915: Simplify fence finding
drm/i915: Discard the unused obj->last_fenced_ring
drm/i915: Remove unused ring->setup_seqno
drm/i915: Remove fence pipelining
...
|
|
EMIF is an SDRAM controller used in various Texas Instruments
SoCs. EMIF supports, based on its revision, one or more of
LPDDR2/DDR2/DDR3 protocols.
Add the basic infrastructure for EMIF driver that includes
driver registration, probe, parsing of platform data etc.
Signed-off-by: Aneesh V <[email protected]>
Reviewed-by: Santosh Shilimkar <[email protected]>
Reviewed-by: Benoit Cousson <[email protected]>
[[email protected]: Moved to drivers/memory from drivers/misc]
Signed-off-by: Santosh Shilimkar <[email protected]>
Tested-by: Lokesh Vutla <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Create a struct clk_init_data to hold all data that needs to be passed from
the platfrom specific driver to the common clock framework during clock
registration. Add a pointer to this struct inside clk_hw.
This has several advantages:
* Completely hides struct clk from many clock platform drivers and static
clock initialization code that don't care for static initialization of
the struct clks.
* For platforms that want to do complete static initialization, it removed
the need to directly mess with the struct clk's fields while still
allowing to statically allocate struct clk. This keeps the code more
future proof even if they include clk-private.h.
* Simplifies the generic clk_register() function and allows adding optional
fields in the future without modifying the function signature.
* Simplifies the static initialization of clocks on all platforms by
removing the need for forward delcarations or convoluted macros.
Signed-off-by: Saravana Kannan <[email protected]>
[[email protected]: kept DEFINE_CLK_* macros and __clk_init]
Signed-off-by: Mike Turquette <[email protected]>
Cc: Andrew Lunn <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Russell King <[email protected]>
Cc: Jeremy Kerr <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Arnd Bergman <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: Sascha Hauer <[email protected]>
Cc: Jamie Iles <[email protected]>
Cc: Richard Zhao <[email protected]>
Cc: Saravana Kannan <[email protected]>
Cc: Magnus Damm <[email protected]>
Cc: Mark Brown <[email protected]>
Cc: Linus Walleij <[email protected]>
Cc: Stephen Boyd <[email protected]>
Cc: Amit Kucheria <[email protected]>
Cc: Deepak Saxena <[email protected]>
Cc: Grant Likely <[email protected]>
|
|
parent name array is now expected to be const char *, make
the relevent changes in the clk macros which define
default clock types.
Signed-off-by: Rajendra Nayak <[email protected]>
Signed-off-by: Mike Turquette <[email protected]>
|
|
While the use of READDIRPLUS is significantly more efficient than
READDIR followed by many LOOKUP calls, it is still less efficient
than just READDIR if the attributes are not required.
This patch tracks when lookups are attempted on the directory,
and uses that information to selectively disable READDIRPLUS
on that directory.
The first 'readdir' call is always served using READDIRPLUS.
Subsequent calls only use READDIRPLUS if there was a successful
lookup or revalidation on a child in the mean time.
Credit for the original idea should go to Neil Brown. See:
http://www.spinics.net/lists/linux-nfs/msg19996.html
However, the implementation in this patch differs from Neil's
in that it focuses on tracking lookups rather than calls to
stat().
Signed-off-by: Trond Myklebust <[email protected]>
Cc: Neil Brown <[email protected]>
|
|
Currently most code to get child count in kernel are almost same,
add a helper to implement this function for dt to use.
Cc: Grant Likely <[email protected]>
Acked-by: Rob Herring <[email protected]>
Acked-by: Stephen Warren <[email protected]>
Signed-off-by: Dong Aisheng <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
|
|
This platform has been obsoleted and was only available inside of
ST-Ericsson, no users of this code are left in the world. This
deletes the core U5500 support entirely in the same manner as the
obsoleted U8500 silicon was previously deleted.
Cc: Srinidhi Kasagar <[email protected]>
Cc: Rabin Vincent <[email protected]>
Cc: Jonas Aberg <[email protected]>
Cc: Per Forlin <[email protected]>
Cc: Ulf Hansson <[email protected]>
Acked-by: Samuel Ortiz <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
|
|
This platform has been obsoleted and was only available inside of
ST-Ericsson, no users of this code are left in the world. This
deletes the core U5500 support entirely in the same manner as the
obsoleted U8500 silicon was previously deleted.
Cc: Srinidhi Kasagar <[email protected]>
Cc: Rabin Vincent <[email protected]>
Cc: Jonas Aberg <[email protected]>
Cc: Per Forlin <[email protected]>
Cc: Ulf Hansson <[email protected]>
Acked-by: Samuel Ortiz <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
|
|
Get rid of the post-op GETATTR on the directory in order to reduce
the amount of processing done on the server.
Signed-off-by: Trond Myklebust <[email protected]>
|
|
Get rid of the post-op GETATTR on the directory in order to reduce
the amount of processing done on the server.
Signed-off-by: Trond Myklebust <[email protected]>
|
|
Get rid of the post-op GETATTR on the directory in order to reduce
the amount of processing done on the server.
The cost is that if we later need to stat() the directory, then we
know that the ctime and mtime are likely to be invalid.
Signed-off-by: Trond Myklebust <[email protected]>
|
|
Use the ctime to simulate a change attribute for NFSv2 and NFSv3.
Signed-off-by: Trond Myklebust <[email protected]>
|
|
After the previous changes in default_stop_ok() and
default_power_down_ok() for PM domains, there are two fields in
struct dev_pm_info that aren't necessary any more, suspend_time
and max_time_suspended_ns.
Remove those fields along with all of the code that accesses them,
which simplifies the runtime PM framework quite a bit.
Signed-off-by: Rafael J. Wysocki <[email protected]>
|
|
The existing default domain power down governor function for PM
domains, default_power_down_ok(), is supposed to check whether or not
the PM QoS latency constraints of the devices in the domain will be
violated if the domain is turned off by pm_genpd_poweroff().
However, the computations carried out by it don't reflect the
definition of the PM QoS latency constrait in
Documentation/ABI/testing/sysfs-devices-power.
Make default_power_down_ok() follow the definition of the PM QoS
latency constrait. In particular, make it only take latencies into
account, because it doesn't matter how much time has elapsed since
the domain's devices were suspended for the computation.
Remove the break_even_ns and power_off_time fields from
struct generic_pm_domain, because they are not necessary any more.
Signed-off-by: Rafael J. Wysocki <[email protected]>
|
|
The existing default device stop governor function for PM domains,
default_stop_ok(), is supposed to check whether or not the device's
PM QoS latency constraint will be violated if the device is stopped
by pm_genpd_runtime_suspend(). However, the computations carried out
by it don't reflect the definition of the PM QoS latency constrait in
Documentation/ABI/testing/sysfs-devices-power.
Make default_stop_ok() follow the definition of the PM QoS latency
constrait. In particular, make it take the device's start and stop
latencies correctly.
Add a new field, effective_constraint_ns, to struct gpd_timing_data
and use it to store the difference between the device's PM QoS
constraint and its resume latency for use by the device's parent
(the effective_constraint_ns values for the children are used for
computing the parent's one along with its PM QoS constraint).
Remove the break_even_ns field from struct gpd_timing_data, because
it's not used any more.
Signed-off-by: Rafael J. Wysocki <[email protected]>
|
|
Android uses one wakelock statistics that is only necessary for
opportunistic sleep. Namely, the prevent_suspend_time field
accumulates the total time the given wakelock has been locked
while "automatic suspend" was enabled. Add an analogous field,
prevent_sleep_time, to wakeup sources and make it behave in a similar
way.
Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Greg Kroah-Hartman <[email protected]>
|
|
Introduce a mechanism by which the kernel can trigger global
transitions to a sleep state chosen by user space if there are no
active wakeup sources.
It consists of a new sysfs attribute, /sys/power/autosleep, that
can be written one of the strings returned by reads from
/sys/power/state, an ordered workqueue and a work item carrying out
the "suspend" operations. If a string representing the system's
sleep state is written to /sys/power/autosleep, the work item
triggering transitions to that state is queued up and it requeues
itself after every execution until user space writes "off" to
/sys/power/autosleep.
That work item enables the detection of wakeup events using the
functions already defined in drivers/base/power/wakeup.c (with one
small modification) and calls either pm_suspend(), or hibernate() to
put the system into a sleep state. If a wakeup event is reported
while the transition is in progress, it will abort the transition and
the "system suspend" work item will be queued up again.
Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Greg Kroah-Hartman <[email protected]>
Reviewed-by: NeilBrown <[email protected]>
|
|
Wakeup statistics used by Android are slightly different from what we
have in wakeup sources at the moment and there aren't any known
users of those statistics other than Android, so modify them to make
it easier for Android to switch to wakeup sources.
This removes the struct wakeup_source's hit_cout field, which is very
rough and therefore not very useful, and adds two new fields,
wakeup_count and expire_count. The first one tracks how many times
the wakeup source is activated with events_check_enabled set (which
roughly corresponds to the situations when a system power transition
to a sleep state is in progress and would be aborted by this wakeup
source if it were the only active one at that time) and the second
one is the number of times the wakeup source has been activated with
a timeout that expired.
Additionally, the last_time field is now updated when the wakeup
source is deactivated too (previously it was only updated during
the wakeup source's activation), which seems to be what Android does
with the analogous counter for wakelocks.
Signed-off-by: Rafael J. Wysocki <[email protected]>
Acked-by: Greg Kroah-Hartman <[email protected]>
|
|
This new driver registers the NXP ISP1301 chip via the I2C subsystem. The chip
is the USB transceiver shared by ohci-nxp, lpc32xx_udc (gadget) and
isp1301_omap.
ISP1301 is a very low-level driver that primarily separates out the I2C client
registration of the ISP1301 chip (including instantiation via DT), used by
other drivers, and declares the chip's registers. It's only a helper driver for
some OHCI and USB device drivers. The driver can be considered as a register
set extension of ohci-nxp, lpc32xx-udc and isp1301_omap, which in turn know
best what to do with the low level functionality (individual ISP1301 registers
and timing, see the different initialization strategies in those drivers).
Those drivers previously internally duplicated ISP1301 register definitions
which is solved by this new isp1301 driver. The ISP1301 registers exposed via
isp1301.h can be accessed by other drivers using it with standard i2c_smbus_*()
accesses.
Following patches let the respective USB host and gadget drivers use this
driver, instead of duplicating ISP1301 handling.
Signed-off-by: Roland Stigge <[email protected]>
Acked-by: Arnd Bergmann <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
On some HCDs usb_unlink_urb() can directly call the
completion handler. That limits the spinlocks that can
be taken in the handler to locks not held while calling
usb_unlink_urb()
To prevent a race with resubmission, this patch exposes
usbcore's infrastructure for blocking submission, uses it
and so drops the lock without causing a race in usbhid.
Signed-off-by: Oliver Neukum <[email protected]>
Acked-by: Jiri Kosina <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
fix kernel doc typos in function names
Signed-off-by: Eric Dumazet <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Add ECN (Explicit Congestion Notification) marking capability to netem
tc qdisc add dev eth0 root netem drop 0.5 ecn
Instead of dropping packets, try to ECN mark them.
Signed-off-by: Eric Dumazet <[email protected]>
Cc: Neal Cardwell <[email protected]>
Cc: Tom Herbert <[email protected]>
Cc: Hagen Paul Pfeifer <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Acked-by: Hagen Paul Pfeifer <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
remove useless casts and rename variables for less confusion.
Signed-off-by: Eric Dumazet <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
L2TPv3 defines an IP encapsulation packet format where data is carried
directly over IP (no UDP). The kernel already has support for L2TP IP
encapsulation over IPv4 (l2tp_ip). This patch introduces support for
L2TP IP encapsulation over IPv6.
The implementation is derived from ipv6/raw and ipv4/l2tp_ip.
Signed-off-by: Chris Elston <[email protected]>
Signed-off-by: James Chapman <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This patch adds support for unmanaged L2TPv3 tunnels over IPv6 using
the netlink API. We already support unmanaged L2TPv3 tunnels over
IPv4. A patch to iproute2 to make use of this feature will be
submitted separately.
Signed-off-by: Chris Elston <[email protected]>
Signed-off-by: James Chapman <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Checkpatch warns about the use of __attribute__((packed)). So use the
recommended __packed syntax instead.
Signed-off-by: James Chapman <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
drivers/infiniband/ulp/srp/ib_srp.c #defines pr_fmt() PFX fmt, but PFX
is not #defined until after <linux/*> headers are included.
This results in a bad expansion of the pr_warn() in the stub function.
2084c2084
< printk("<4>" PFX "dyndbg supported only in " "CONFIG_DYNAMIC_DEBUG builds\n")
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
The core branch is behind driver commits that we want to build
on for 3.5, hence I'm pulling in a later -rc.
Linux 3.4-rc5
Conflicts:
Documentation/feature-removal-schedule.txt
Signed-off-by: Jens Axboe <[email protected]>
|
|
If we are in the middle of an I2C transfer we need to deny suspend
of the AB8500 core. Implement an atomic reference counter for the
I2C operations to make sure we don't do this.
Signed-off-by: Jonas Aaberg <[email protected]>
Reviewed-by: Mattias Wallin <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
The AB8505 and AB9540 has extended support for micro USB
resistance detection, used for detecting chargers. Let's
register resources for this resource. Let's also split off the
separate codec device for AB9540.
Signed-off-by: Virupax Sadashivpetimath <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
Switch the driver over to device group handling. By adding the
HID_GROUP_MULTITOUCH group to hid-core, hid-generic will no longer
match multitouch devices. By adding the HID_GROUP_MULTITOUCH entry to
the device list, hid-multitouch will match all unknown multitouch
devices, and udev will automatically load the module.
Since HID_QUIRK_MULTITOUCH never gets set, the special quirks handling
can be removed. Since all HID MT devices have HID_DG_CONTACTID, they
can be removed from the hid_have_special_driver list.
With this patch, the unknown device ids are no longer NULL, so the code
is modified to check for the generic entry instead.
Signed-off-by: Henrik Rydberg <[email protected]>
Acked-by: Benjamin Tissoires <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
|
|
Devices that do not have a special driver are handled by the generic
driver. This patch does the same thing using device groups; Instead of
forcing a particular driver, the appropriate driver is picked up by
udev. As a consequence, one can now move a device from generic to
specific handling by a simple rebind. By adding a new device id to the
generic driver, the same thing can be done in reverse.
Signed-off-by: Henrik Rydberg <[email protected]>
Acked-by: Benjamin Tissoires <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
|
|
Most HID drivers do not need to know what bus driver is in use.
A generic group driver can drive any hid device, and the device
list should not need to be duplicated for each new bus.
This patch adds wildcard matching to the HID bus, simplifying device
list handling for group drivers.
Signed-off-by: Henrik Rydberg <[email protected]>
Acked-by: Benjamin Tissoires <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
|
|
In order to allow the report descriptor to influence the hid device
properties, one needs to parse the descriptor early, without reference
to any driver. Scan the descriptor for group information during device
add, before the device has been broadcast to userland. The device
modalias will contain group information which can be used to
differentiate between modules. For starters, just handle the generic
group.
Signed-off-by: Henrik Rydberg <[email protected]>
Acked-by: Benjamin Tissoires <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
|
|
HID devices are only partially presented to userland. Hotplugged
devices emit events containing a modalias based on the basic bus,
vendor and product entities. However, in practise a hid device can
depend on details such as a single usb interface or a particular item
in a report descriptor.
This patch adds a device group to the hid device id, and broadcasts it
using uevent and the device modalias. The module alias generation is
modified to match. As a consequence, a device with a non-zero group
will be processed by the corresponding group driver instead of by the
generic hid driver.
Signed-off-by: Henrik Rydberg <[email protected]>
Acked-by: Benjamin Tissoires <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
|
|
The low-level driver can read the report descriptor, but it cannot
determine driver-specific changes to it. The hid core can fixup
and parse the report descriptor during driver attach, but does
not have direct access to the descriptor when doing so.
To be able to handle attach/detach of hid drivers properly,
a semantic change to hid_parse_report() is needed. This function has
been used in two ways, both as descriptor reader in the ll drivers and
as a parsor in the probe of the drivers. This patch splits the usage
by introducing hid_open_report(), and modifies the hid_parse() macro
to call hid_open_report() instead. The only usage of hid_parse_report()
is then to read and store the device descriptor. As a consequence, we
can handle the report fixups automatically inside the hid core.
Signed-off-by: Henrik Rydberg <[email protected]>
Tested-by: Nikolai Kondrashov <[email protected]>
Tested-by: Benjamin Tissoires <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
|
|
Adding support for device sleep through the external input control
signal "SLEEP".
Changing the SLEEP signal state can switch the device into SLEEP and
ACTIVE state.
Also adding sleep configuration for different resources so that they
should be keep on during sleep state of device.
Signed-off-by: Laxman Dewangan <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
The mfd/asic3 driver does not set the ds1wm_driver_data clock_rate field
before passing the structure to the DS1WM w1 busmaster driver.
This was not noticed before commit 26a6afb, because ds1wm_find_divisor()
unintentionally returned the correct divisor when a zero clock_rate was
passed in. However after that commit DS1WM fails a zero clock_rate:
ds1wm ds1wm: no suitable divisor for 0Hz clock
This patch sets the ds1wm_driver_data clock_rate field.
Signed-off-by: Paul Parsons <[email protected]>
Acked-by: Philipp Zabel <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
This driver currently creates resources for use by a forthcoming ICH
chipset GPIO driver. It could be expanded to create the resources for
converting the esb2rom (mtd) and iTCO_wdt (wdt), and potentially more,
drivers to use the mfd model.
Signed-off-by: Aaron Sierra <[email protected]>
Signed-off-by: Guenter Roeck <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
|
|
GRO can check if skb to be merged has its skb->head mapped to a page
fragment, instead of a kmalloc() area.
We 'upgrade' skb->head as a fragment in itself
This avoids the frag_list fallback, and permits to build true GRO skb
(one sk_buff and up to 16 fragments), using less memory.
This reduces number of cache misses when user makes its copy, since a
single sk_buff is fetched.
This is a followup of patch "net: allow skb->head to be a page fragment"
Signed-off-by: Eric Dumazet <[email protected]>
Cc: Ilpo Järvinen <[email protected]>
Cc: Herbert Xu <[email protected]>
Cc: Maciej Żenczykowski <[email protected]>
Cc: Neal Cardwell <[email protected]>
Cc: Tom Herbert <[email protected]>
Cc: Jeff Kirsher <[email protected]>
Cc: Ben Hutchings <[email protected]>
Cc: Matt Carlson <[email protected]>
Cc: Michael Chan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
skb->head is currently allocated from kmalloc(). This is convenient but
has the drawback the data cannot be converted to a page fragment if
needed.
We have three spots were it hurts :
1) GRO aggregation
When a linear skb must be appended to another skb, GRO uses the
frag_list fallback, very inefficient since we keep all struct sk_buff
around. So drivers enabling GRO but delivering linear skbs to network
stack aren't enabling full GRO power.
2) splice(socket -> pipe).
We must copy the linear part to a page fragment.
This kind of defeats splice() purpose (zero copy claim)
3) TCP coalescing.
Recently introduced, this permits to group several contiguous segments
into a single skb. This shortens queue lengths and save kernel memory,
and greatly reduce probabilities of TCP collapses. This coalescing
doesnt work on linear skbs (or we would need to copy data, this would be
too slow)
Given all these issues, the following patch introduces the possibility
of having skb->head be a fragment in itself. We use a new skb flag,
skb->head_frag to carry this information.
build_skb() is changed to accept a frag_size argument. Drivers willing
to provide a page fragment instead of kmalloc() data will set a non zero
value, set to the fragment size.
Then, on situations we need to convert the skb head to a frag in itself,
we can check if skb->head_frag is set and avoid the copies or various
fallbacks we have.
This means drivers currently using frags could be updated to avoid the
current skb->head allocation and reduce their memory footprint (aka skb
truesize). (thats 512 or 1024 bytes saved per skb). This also makes
bpf/netfilter faster since the 'first frag' will be part of skb linear
part, no need to copy data.
Signed-off-by: Eric Dumazet <[email protected]>
Cc: Ilpo Järvinen <[email protected]>
Cc: Herbert Xu <[email protected]>
Cc: Maciej Żenczykowski <[email protected]>
Cc: Neal Cardwell <[email protected]>
Cc: Tom Herbert <[email protected]>
Cc: Jeff Kirsher <[email protected]>
Cc: Ben Hutchings <[email protected]>
Cc: Matt Carlson <[email protected]>
Cc: Michael Chan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This patch implements the directed yield hypercall found on other
System z hypervisors. It delegates execution time to the virtual cpu
specified in the instruction's parameter.
Useful to avoid long spinlock waits in the guest.
Christian Borntraeger: moved common code in virt/kvm/
Signed-off-by: Konstantin Weitz <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
Signed-off-by: Marcelo Tosatti <[email protected]>
|
|
struct nfs_direct_req can't compile when struct pnfs_ds_commit_info
is undefined.
Reported-by: Bryan Schumaker <[email protected]>
Signed-off-by: Trond Myklebust <[email protected]>
Cc: Fred Isaman <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"This is a set of SAS and SATA fixes; there are one or two longstanding
bug fixes, but most of this is regression fixes."
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
[SCSI] libfc: update mfs boundry checking
[SCSI] Revert "[SCSI] libsas: fix sas port naming"
[SCSI] libsas: fix false positive 'device attached' conditions
[SCSI] libsas, libata: fix start of life for a sas ata_port
[SCSI] libsas: fix ata_eh clobbering ex_phys via smp_ata_check_ready
[SCSI] libsas: unify domain_device sas_rphy lifetimes
[SCSI] libsas: fix sas_get_port_device regression
[SCSI] libsas: fix sas_find_bcast_phy() in the presence of 'vacant' phys
[SCSI] libsas: introduce sas_work to fix sas_drain_work vs sas_queue_work
[SCSI] libata: Pass correct DMA device to scsi host
[SCSI] scsi_lib: use correct DMA device in __scsi_alloc_queue
|
|
More recent versions of the UEFI spec have added new attributes for
variables. Add them.
Signed-off-by: Matthew Garrett <[email protected]>
Cc: [email protected]
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Some devices does not support bulk read and write operations, for them
we have series of single write and read operations.
Signed-off-by: Anthony Olech <[email protected]>
Signed-off-by: Ashish Jangam <[email protected]>
[Fixed coding style, don't check use_single_rw before assign --broonie ]
Signed-off-by: Mark Brown <[email protected]>
|
|
We need a hook to release host bridge resources allocated when creating
root bus.
Signed-off-by: Yinghai Lu <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
|