Commit graph

533,100 commits

Author SHA1 Message Date
Julia Lawall
801421eb73 drivers/pcmcia/electra_cf.c: add missing iounmap and kfree
cf and cf->mem_base have been allocated at the point of this failure, so
they should be freed before leaving the function.

[linux@dominikbrodowski.net: limit call to device_init_wakeup() to the
same error paths as before]
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
2015-05-30 15:14:39 +02:00
Jim Cromie
e632cd9472 pcmcia: replace open-coded ARRAY_SIZE with macro
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
2015-05-30 15:10:47 +02:00
Mauro Carvalho Chehab
704db8050d [media] DocBook: some fixes at FE_GET_INFO
There are a few issues at FE_GET_INFO documentation:
- name is a string, not a pointer to a string;
- the return text should be after the paragraph.

While here, better to bold that two fields of the structure used
by FE_GET_INFO are actually deprecated.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:48:21 -03:00
Mauro Carvalho Chehab
997eb9039d [media] DocBook: Better document enum fe_modulation
Instead of using programlisting, use a table, as this provides
a better view of the structure.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:46:40 -03:00
Mauro Carvalho Chehab
2c2bc8ff97 [media] DocBook: add a proper description for dvb_frontend_info.fe_type
The fe_type is deprecated at the DVB API. However, it may still
be used by legacy DVBv3 applications. While this works with old
devices, modern devices may support more than one delivery
system.

Add an explanation about that and a point to what should be
used, instead, in order for legacy apps to support newer hardware.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:44:47 -03:00
Mauro Carvalho Chehab
be1ad6e76e [media] DocBook: Provide a high-level description for DVB frontend
Instead of just showing ioctls, let's add an introdutory text
briefly explaining the DVB frontend API.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:44:14 -03:00
Mauro Carvalho Chehab
d27518ca0c [media] DocBook: fix FE_READ_STATUS argument description
What's written there about the arguments for this ioctl
is bogus: it doesn't return an enum (or a typedef)
for enum fe_status. Instead, it returns a bitmask with the
values defined by enum fe_status.

Also, the size of the integer returned is not 16 bits, but,
instead, sizeof(fe_status_t), e. g. sizeof(enum), with is
arch-dependent.

This should of course be fixed, but this should be done on
a separate patch.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:43:09 -03:00
Mauro Carvalho Chehab
b60da5b203 [media] DocBook: better organize the function descriptions for frontend
Move the function ioctl definitions to the end of the chapter,
at their importance. That makes the document better organized,
as the DVB frontend system call index will look like:

	open()
	close()
	ioctl FE_GET_INFO — Query DVB frontend capabilities and returns information about the front-end. This call only requires read-only access to the device
	ioctl FE_READ_STATUS — Returns status information about the front-end. This call only requires read-only access to the device
	ioctl FE_SET_PROPERTY, FE_GET_PROPERTY — FE_SET_PROPERTY sets one or more frontend properties. FE_GET_PROPERTY returns one or more frontend properties.
	ioctl FE_DISEQC_RESET_OVERLOAD — Restores the power to the antenna subsystem, if it was powered off due to power overload.
	ioctl FE_DISEQC_SEND_MASTER_CMD — Sends a DiSEqC command
	ioctl FE_DISEQC_RECV_SLAVE_REPLY — Receives reply from a DiSEqC 2.0 command
	ioctl FE_DISEQC_SEND_BURST — Sends a 22KHz tone burst for 2x1 mini DiSEqC satellite selection.
	ioctl FE_SET_TONE — Sets/resets the generation of the continuous 22kHz tone.
	ioctl FE_SET_VOLTAGE — Allow setting the DC level sent to the antenna subsystem.
	ioctl FE_ENABLE_HIGH_LNB_VOLTAGE — Select output DC level between normal LNBf voltages or higher LNBf voltages.
	ioctl FE_SET_FRONTEND_TUNE_MODE — Allow setting tuner mode flags to the frontend.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:42:55 -03:00
Mauro Carvalho Chehab
26b2574fbb [media] DocBook: better document FE_DISEQC_RESET_OVERLOAD
Use the new format for the ioctl documentation and put the
struct dvb_diseqc_slave_reply together with the ioctl.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:42:31 -03:00
Mauro Carvalho Chehab
db2a4ee870 [media] DocBook: better document FE_DISEQC_SEND_MASTER_CMD
Use the new format for the ioctl documentation and put the
struct dvb_diseqc_slave_reply together with the ioctl.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:41:28 -03:00
Mauro Carvalho Chehab
32b6443658 [media] DocBook: better document FE_DISEQC_RECV_SLAVE_REPLY
Use the new format for the ioctl documentation and put the
struct dvb_diseqc_slave_reply together with the ioctl.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:39:27 -03:00
Mauro Carvalho Chehab
81959d996a [media] DocBook: better document FE_DISEQC_SEND_BURST ioctl
Use the proper format for FE_DISEQC_SEND_BURST documentation
and  improve the documentation.

Keep the enum fe_sec_mini_cmd description together with
the ioctl, as both are used together.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:36:37 -03:00
Mauro Carvalho Chehab
6dc59e7a19 [media] DocBook: better document FE_SET_TONE ioctl
Use the proper format for FE_SET_TONE documentation and
improve the documentation.

Keep the enum fe_sec_tone_mode description together with
the ioctl, as both are used together.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:35:51 -03:00
Mauro Carvalho Chehab
d6b6d346e5 [media] DocBook: better document FE_SET_VOLTAGE ioctl
Use the proper format for FE_SET_VOLTAGE documentation and fix
the documentation. The description for the enum is not 100%,
and it is missing the voltage off value.

Also, it is better to keep the enum description together with
the ioctl, as both are used together.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:33:45 -03:00
Mauro Carvalho Chehab
50f88f3da0 [media] DocBook: reformat FE_ENABLE_HIGH_LNB_VOLTAGE ioctl
Use the proper format for FE_ENABLE_HIGH_LNB_VOLTAGE documentation.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:32:13 -03:00
Mauro Carvalho Chehab
3d7f9c9b17 [media] DocBook: reformat FE_SET_FRONTEND_TUNE_MODE ioctl
Use the proper format for FE_SET_FRONTEND_TUNE_MODE documentation.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:31:02 -03:00
Mauro Carvalho Chehab
0f16290e6f [media] DocBook: move FE_GET_PROPERTY to its own xml file
That helps the xref logic at the Makefile to point to the
right place. Also, it becomes more organized and easier to
maintain if each ioctl have its own xml file.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:28:57 -03:00
Mauro Carvalho Chehab
09c3b4a0af [media] DocBook: Rename ioctl xml files
for the xml files describing ioctls, use the same nomenclature
as on V4L2: the ioctl name, in lower case, using - instead of _.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:27:55 -03:00
Mauro Carvalho Chehab
55d4423a2d [media] DocBook: Fix false positive undefined ioctl references
The new code that detects undocumented ioctls hits some false
positives:

This one is not documented, nor it should, as this is
there just to reserve namespace:

	Warning: can't find reference for VIDIOC_RESERVED ioctl

But those are already documented together with other ioctls:

	Warning: can't find reference for VIDIOC_UNSUBSCRIBE_EVENT ioctl
	Warning: can't find reference for FE_GET_PROPERTY ioctl
	Warning: can't find reference for VIDIOC_SUBDEV_G_EDID ioctl
	Warning: can't find reference for VIDIOC_SUBDEV_S_EDID ioctl
	Warning: can't find reference for VIDIOC_SUBDEV_S_DV_TIMINGS ioctl
	Warning: can't find reference for VIDIOC_SUBDEV_G_DV_TIMINGS ioctl
	Warning: can't find reference for VIDIOC_SUBDEV_QUERY_DV_TIMINGS ioctl

So, we need to just be sure to point to the right documentation.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 09:25:20 -03:00
Eric Wong
2f80b2958a ALSA: usb-audio: don't try to get Outlaw RR2150 sample rate
This quirk allows us to avoid the noisy:

	current rate 0 is different from the runtime rate

message every time playback starts.  While USB DAC in the RR2150
supports reading the sample rate, it never returns a sample rate
other than zero in my observation with common sample rates.

Signed-off-by: Eric Wong <normalperson@yhbt.net>
Cc: Joe Turner <joe@oampo.co.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2015-05-30 14:14:40 +02:00
Mauro Carvalho Chehab
3246c385f3 [media] DocBook: remove duplicated ioctl from v4l2-subdev
Those ioctls are already parsed. No need to explicitly add
them.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:43:30 -03:00
Mauro Carvalho Chehab
87e4d63748 [media] DocBook: Improve xref check for undocumented ioctls
There are several badly documented undocumented ioctls.
Currently, it just generates an empty link. Instead of doing that,
only add references to the ones that exists, and add a warning
for all references that weren't found.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:42:59 -03:00
Mauro Carvalho Chehab
c569eb707a [media] DocBook: Add xref links for DTV propeties
Create xref links for all DTV properties and link the frontend.h
to each. Also use them at the DVB frontent API example.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:41:27 -03:00
Mauro Carvalho Chehab
e4aa18d33c [media] DocBook: Improve the description of the properties API
Make the text clearer about what the properties API does.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:40:09 -03:00
Mauro Carvalho Chehab
277dfcb4e0 [media] DocBook: Merge FE_SET_PROPERTY/FE_GET_PROPERTY ioctl description
Instead of having two refentries, merge them into just one,
like what's done with other similar ioctls at V4L2 side.

That makes the entry cleaner and will allow to add the associated
structures together with the refentry.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:37:26 -03:00
Mauro Carvalho Chehab
c130d88b01 [media] DocBook: fix xref to the FE open() function
Instead of going to the V4L2 open(), use the xref to the
proper place at the frontend ioctls that were already
reformatted.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:37:05 -03:00
Mauro Carvalho Chehab
ffec8bceac [media] DocBook: rewrite FE_GET_PROPERTY/FE_SET_PROPERTY to use the std way
Use the standard way of documenting ioctls for FE_GET_PROPERTY
and FE_SET_PROPERTY.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:36:03 -03:00
Mauro Carvalho Chehab
56599d4196 [media] DocBook: move DVB properties to happen earlier at the document
The DVBv5 API uses DVB properties as the main way to set the frontend
and collect statistics. Move the definition to happen earlier, in
order to reflect its importance.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:34:47 -03:00
Mauro Carvalho Chehab
5b233e1a91 [media] DocBook: improve documentation for FE_READ_STATUS
Move the ioctl and enum fe_status to a separate xml file and
put it into a better format.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:33:50 -03:00
Mauro Carvalho Chehab
002563e991 [media] DocBook: move FE_GET_INFO to a separate xml file
Keeping everything altogether makes harder to reorganize the
DocBook. So, move the FE_GET_INFO ioctl and the associated structures
into a separate file.

No functional changes.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:32:49 -03:00
Mauro Carvalho Chehab
bb38c5474e [media] DocBook: reformat FE_GET_INFO ioctl documentation
The DVB part of the docbook has a completely different format
than the V4L2 part, as it was written as a separate document.

As the V4L2 documentation is on better shape, and its format
allows adding more information, let's use it for FE_GET_INFO
and gradually update the non-legacy DVB ioctls using the new
format.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:30:53 -03:00
Mauro Carvalho Chehab
97f411d917 [media] dvb: split enum from typedefs at frontend.h
Using typedefs is already bad enough, but doing it together
with enum declaration is even worse.

Also, it breaks the scripts at DocBook that would be generating
reference pointers for the enums.

Well, we can't get rid of typedef right now, but let's at least
declare it on a separate line, and let the scripts to generate
the cross-reference, as this is needed for the next DocBook
patches.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:29:27 -03:00
Mauro Carvalho Chehab
a8d3e4b4d9 [media] DocBook: move DVBv3 frontend bits to a separate section
Although the recommended usage is the DVBv5 API calls, the
documentation doesn't make it clear about what's the recommended
calls and what's legacy.

So, move the legacy API bits to a separate xml, putting them into
a new section.

Please notice that more changes are needed, since some of the
bits there are cross-referenced elsewhere.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:28:28 -03:00
Mauro Carvalho Chehab
ee072ed7e3 [media] DocBook: Improve DVB frontend description
The DVB frontend API got bitrotten. Update it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:26:05 -03:00
Mauro Carvalho Chehab
0b0f1a0056 [media] DocBook: fix emphasis at the DVB documentation
Currently, it is using 'role="tt"', but this is not defined at
the DocBook 4.5 spec. The net result is that no emphasis happens.

So, replace them to bold emphasis.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:26:05 -03:00
Mauro Carvalho Chehab
a1ef3eeabe [media] DocBook: add drawing with a typical media device
An illustration of what's considered a typical media device
may help people to better understand the contents of the media
infrastructure API docbook. So, add it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:25:19 -03:00
Mauro Carvalho Chehab
9f67bc5203 [media] DocBook: add a note about the ALSA API
The Docbook mistakenly makes to believe that all needed APIs for
media devices are there. Add a note there pointing that some
sub-devices are actually be controlled via ALSA API.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:14:42 -03:00
Mauro Carvalho Chehab
39b9ce2c6c [media] DocBook: Update DVB supported standards at introduction
The list of standards at the media docbook is incomplete, and it
is mentioning that the DVB-S2 & friends is "currently being updated".
That's wrong, as such update occurred back in 2008.

So, provide a more complete list of supported standards and add
a reference to the actual list.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:13:20 -03:00
Jemma Denson
bf8de2d38a [media] cx24120: Assume ucb registers is a counter
The ucblocks register is probably a counter and not a rate; assume
it is so and change the calculations as required.

Signed-off-by: Jemma Denson <jdenson@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-05-30 08:12:39 -03:00
Andy Lutomirski
c2affbf9a5 x86/asm/entry/32, selftests: Add a selftest for kernel entries from VM86 mode
Test a couple of special cases in 32-bit kernels for entries
from vm86 mode.  This will OOPS both old kernels due to a bug
and and 4.1-rc5 due to a regression I introduced, and it should
make sure that the SYSENTER-from-vm86-mode hack in the kernel
keeps working.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Jan Beulich <JBeulich@suse.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Shuah Khan <shuahkh@osg.samsung.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/09a9916761e0a9e42d4922f147af45a0079cc1e8.1432936374.git.luto@kernel.org
Tests: 394838c960 x86/asm/entry/32: Fix user_mode() misuses
Tests: 7ba554b5ac x86/asm/entry/32: Really make user_mode() work correctly for VM86 mode
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-05-30 10:21:32 +02:00
Anda-Maria Nicolae
e86d69dd78 power_supply: Add support for Richtek RT9455 battery charger
Based on the datasheet found here:
http://www.richtek.com/download_ds.jsp?p=RT9455

Signed-off-by: Anda-Maria Nicolae <anda-maria.nicolae@intel.com>
Reviewed-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-05-30 03:34:55 +02:00
Anda-Maria Nicolae
21472a54a1 Documentation: devicetree: Add Richtek RT9455 bindings
Signed-off-by: Anda-Maria Nicolae <anda-maria.nicolae@intel.com>
Reviewed-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
[added default values to optional properties]
Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-05-30 03:34:11 +02:00
Anda-Maria Nicolae
eb82b9062b of: Add vendor prefix for Richtek Technology Corporation
Signed-off-by: Anda-Maria Nicolae <anda-maria.nicolae@intel.com>
Reviewed-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Signed-off-by: Sebastian Reichel <sre@kernel.org>
2015-05-30 03:33:16 +02:00
Guenter Roeck
c7bd6dc320 hwmon: (nct6683) Add missing sysfs attribute initialization
The following error message is seen when loading the nct6683 driver
with DEBUG_LOCK_ALLOC enabled.

BUG: key ffff88040b2f0030 not in .data!
------------[ cut here ]------------
WARNING: CPU: 0 PID: 186 at kernel/locking/lockdep.c:2988
				lockdep_init_map+0x469/0x630()
DEBUG_LOCKS_WARN_ON(1)

Caused by a missing call to sysfs_attr_init() when initializing
sysfs attributes.

Reported-by: Alexey Orishko <alexey.orishko@gmail.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Cc: stable@vger.kernel.org # v3.18+
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
2015-05-29 17:47:42 -07:00
Guenter Roeck
1b63bf6172 hwmon: (nct6775) Add missing sysfs attribute initialization
The following error message is seen when loading the nct6775 driver
with DEBUG_LOCK_ALLOC enabled.

BUG: key ffff88040b2f0030 not in .data!
------------[ cut here ]------------
WARNING: CPU: 0 PID: 186 at kernel/locking/lockdep.c:2988
				lockdep_init_map+0x469/0x630()
DEBUG_LOCKS_WARN_ON(1)

Caused by a missing call to sysfs_attr_init() when initializing
sysfs attributes.

Reported-by: Alexey Orishko <alexey.orishko@gmail.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Cc: stable@vger.kernel.org # v3.12+
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
2015-05-29 17:47:33 -07:00
Herton R. Krzesinski
47b98c74fa cpupower: mperf monitor: fix output in MAX_FREQ_SYSFS mode
There is clearly wrong output when mperf monitor runs in MAX_FREQ_SYSFS mode:
average frequency shows in kHz unit (despite the intended output to be in MHz),
and percentages for C state information are all wrong (including high/negative
values shown).

The problem is that the max_frequency read on initialization isn't used where it
should have been used on mperf_get_count_percent (to estimate the number of
ticks in the given time period), and the value we read from sysfs is in kHz, so
we must divide it to get the MHz value to use in current calculations.

While at it, also I fixed another small issues in the debug output of
max_frequency value in mperf_get_count_freq.

Signed-off-by: Herton R. Krzesinski <herton@redhat.com>
Acked-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-05-30 02:21:31 +02:00
Tony Lindgren
db874c7e10 PM / wakeirq: Fix typo in prototype for dev_pm_set_dedicated_wake_irq
Looks like I only built test the dev_pm_set_wake_irq and not the
dev_pm_set_dedicated_wake_irq case on x86.

Turns out there's a typo for the dev_pm_set_dedicated_wake_irq
prototype that causes a build error if CONFIG_COMPILE_TEST
and CONFIG_MMC_OMAP_HS are selected.

Reported-by: Jim Davis <jim.epost@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Reviewed-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-05-30 02:17:15 +02:00
Rafael J. Wysocki
7d51d97925 cpuidle: Do not use CPUIDLE_DRIVER_STATE_START in cpuidle.c
The CPUIDLE_DRIVER_STATE_START symbol is defined as 1 only if
CONFIG_ARCH_HAS_CPU_RELAX is set, otherwise it is defined as 0.
However, if CONFIG_ARCH_HAS_CPU_RELAX is set, the first (index 0)
entry in the cpuidle driver's table of states is overwritten with
the default "poll" entry by the core.  The "state" defined by the
"poll" entry doesn't provide ->enter_dead and ->enter_freeze
callbacks and its exit_latency is 0.

For this reason, it is not necessary to use CPUIDLE_DRIVER_STATE_START
in cpuidle_play_dead() (->enter_dead is NULL, so the "poll state"
will be skipped by the loop).

It also is arguably unuseful to return states with exit_latency
equal to 0 from find_deepest_state(), so the function can be modified
to start the loop from index 0 and the "poll state" will be skipped by
it as a result of the check against latency_req.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Preeti U Murthy <preeti@linux.vnet.ibm.com>
2015-05-30 02:15:21 +02:00
Linus Torvalds
aaa20fc233 PCI / ACPI fix for v4.1-rc6
This fixes a bug uncovered by a recent driver core change that
 modified the implementation of the ACPI_COMPANION_SET() macro to
 strictly rely on its second argument to be either NULL or a valid
 pointer to struct acpi_device.
 
 As it turns out, pcibios_root_bridge_prepare() on x86 and ia64
 works with the assumption that the only code path calling
 pci_create_root_bus() is pci_acpi_scan_root() and therefore
 the sysdata argument passed to it will always match the
 expectations of pcibios_root_bridge_prepare().  That need not
 be the case, however, and in particular it is not the case for
 the Xen pcifront driver that passes a pointer to its own private
 data strcture as sysdata to pci_scan_bus_parented() which then
 passes it to pci_create_root_bus() and it ends up being used
 incorrectly by pcibios_root_bridge_prepare().
 
 /
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.22 (GNU/Linux)
 
 iQIcBAABCAAGBQJVaP1HAAoJEILEb/54YlRxpwEQAJn+6phfcLZiBkwRCvGAmWwv
 Hi5UhcLCZb8r0RX+CssrwDxmiTD+OO5Kume7yqzb9R1EJfEWcx7dOnvrtl6mfy9E
 sfWuU/ILF9TW57bqEHBde921VwtRh3BQ4fOCbp4qYEEXjFx+N/UOYHRefgWzvgYP
 0j9U7HvwBatIjOC6gPWDyoOI7AYZwBVh4lhjYTPfOTta7QqbjFPD1b+EaAqSyNsl
 TqrTQDgtxMTL02u50fP1hLJtOibdp7gOTRlEyfCJru4JjHBTRxBlaefymGi4e2Nw
 5bFyXlDv4SVAGpE2XFDIJftAkAs+lXPjfIt7aHPB+DSa01axBesyqLITQV9wDoL8
 AETnGomgoe+fsymJ1Pk4JOar6bKJmezAwDibi7bDcUoZkTD91VREwZiaLR+kEvzH
 msujMNV0SdVS+PIZUNIwYE6t/ffSDnHjdGjGnsAhNt2JtFFcDKyNkf21O4NYlV99
 iLLx/YY+kxmTBuV+L9pjEefS/u9g60cdjslPFBzffXIZ8N9g2uKOqB6ulfg2dSm/
 tB5u8gyTuR++jmm/f+rlpCCzQ4EMG9ciZnZJfce3m961Sac2kNacGARMMIAe6xMq
 ijuOgcVZZibGBviG5v64ntTQki9qgmImZiPCSoOn+3UgrkCExJfUNDSVl5tQA6MP
 oJEj4WRvH2LrFd/PjPlW
 =ulKW
 -----END PGP SIGNATURE-----

Merge tag 'acpi-pci-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull PCI / ACPI fix from Rafael Wysocki:
 "This fixes a bug uncovered by a recent driver core change that
  modified the implementation of the ACPI_COMPANION_SET() macro to
  strictly rely on its second argument to be either NULL or a valid
  pointer to struct acpi_device.

  As it turns out, pcibios_root_bridge_prepare() on x86 and ia64 works
  with the assumption that the only code path calling pci_create_root_bus()
  is pci_acpi_scan_root() and therefore the sysdata argument passed to
  it will always match the expectations of pcibios_root_bridge_prepare().

  That need not be the case, however, and in particular it is not the
  case for the Xen pcifront driver that passes a pointer to its own
  private data strcture as sysdata to pci_scan_bus_parented() which then
  passes it to pci_create_root_bus() and it ends up being used incorrectly
  by pcibios_root_bridge_prepare()"

* tag 'acpi-pci-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  PCI / ACPI: Do not set ACPI companions for host bridges with parents
2015-05-29 17:09:39 -07:00
Linus Torvalds
1be44e234b xfs: update for 4.1-rc6
Changes in this update:
 o regression fix for new rename whiteout code
 o regression fixes for new superblock generic per-cpu counter code
 o fix for incorrect error return sign introduced in 3.17
 o metadata corruption fixes that need to go back to -stable kernels
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.12 (GNU/Linux)
 
 iQIcBAABAgAGBQJVaO0JAAoJEK3oKUf0dfodd4UQANRdfXnUrpyQGhVS7HFoFoVt
 FIQ52pPGbMu72+DqHc+Q41uvgAPe65LFB2VUL6CUGCMExstF72F5+QonzppMgkMo
 unPER3eB8ya03SY+Kp+803ZGgzI2Nl2M6w8Kof730/RUk56PTGYIx4eLXd6iZSli
 RsYjw8JDbeue5OQo5FPmLCSQ/Kr5ZJXbgWVPyWkKg9aCcXLN5YSJIV3xcMTK9Q2I
 LqqODkyatnGc6YxGAKddS7Xzt1ntlZgbe5mndQw04a2g0Lf6emPH5r8b0UJXIu96
 advOBX0pEbad4FeFS6Mo5D+nNCaaNP4WzN7wgdb+BYNVw3ss4Ebam7+yY6Gexg6y
 bzZOEkk9saL4YeBDgyYICNu7kG4BRVKRQiiX220G6SFXM3nqbl7qBPb3kVFyDpcI
 RRuFJ0ZV0kFJ+3IQ4xVnIh6nootceRk/mvZaK5HhLhQLzklpZ8fj4HF3oBDUAnvN
 wNd+7GoZy7zldjCkbF4BP3GjUeW+b9ngrCNc+bFXi5cUbdECXAa2krjxyY+MlQF2
 veNVVcsoRdfeM0VjJh2/piGJxMWIlRqXdKzPKsfMWnlIaJ6YyslfbSq+2K7LxgGR
 Ho3Sjt0oUuPMZ9F/Mjj+XDqwmzgooUHXNyDBxhGXBNBPjApcRLcb2vQ2SrWEmeGJ
 vZmC2R1ZoGdBJg8a55BT
 =w5SP
 -----END PGP SIGNATURE-----

Merge tag 'xfs-for-linus-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs

Pull xfs fixes from Dave Chinner:
 "This is a little larger than I'd like late in the release cycle, but
  all the fixes are for regressions introduced in the 4.1-rc1 merge, or
  are needed back in -stable kernels fairly quickly as they are
  filesystem corruption or userspace visible correctness issues.

  Changes in this update:

   - regression fix for new rename whiteout code

   - regression fixes for new superblock generic per-cpu counter code

   - fix for incorrect error return sign introduced in 3.17

   - metadata corruption fixes that need to go back to -stable kernels"

* tag 'xfs-for-linus-4.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs:
  xfs: fix broken i_nlink accounting for whiteout tmpfile inode
  xfs: xfs_iozero can return positive errno
  xfs: xfs_attr_inactive leaves inconsistent attr fork state behind
  xfs: extent size hints can round up extents past MAXEXTLEN
  xfs: inode and free block counters need to use __percpu_counter_compare
  percpu_counter: batch size aware __percpu_counter_compare()
  xfs: use percpu_counter_read_positive for mp->m_icount
2015-05-29 16:45:45 -07:00