There are a few leftover CONFIG_PROC_FS forgotten to replace with
CONFIG_SND_PROC_FS.
Fixes: cd6a65036f0e ('ALSA: replace CONFIG_PROC_FS with CONFIG_SND_PROC_FS') Reported-by: Jim Davis <jim.epost@gmail.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jie Yang [Wed, 27 May 2015 11:45:45 +0000 (19:45 +0800)]
ALSA: replace CONFIG_PROC_FS with CONFIG_SND_PROC_FS
We may disable proc fs only for sound part, to reduce ALSA
memory footprint. So add CONFIG_SND_PROC_FS and replace the
old CONFIG_PROC_FSs in alsa code.
With sound proc fs disabled, we can save about 9KB memory
size on X86_64 platform.
Signed-off-by: Jie Yang <yang.jie@intel.com> Reviewed-by: Mark Brown <broonie@kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jie Yang [Wed, 27 May 2015 11:45:44 +0000 (19:45 +0800)]
ALSA: Kconfig: add config item SND_PROC_FS for expert
For some embedded devices, we need reduce code size and data
footprint as much as possible, e.g. disabling procfs, hw/sw
params refinement, mmap, dpcm, dapm, compressed API...
Here add SND_PROC_FS item for expert, we can unselect it to
disable sound proc FS and reduce memory footprint.
Signed-off-by: Jie Yang <yang.jie@intel.com> Reviewed-by: Mark Brown <broonie@kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Sakamoto [Wed, 27 May 2015 15:02:59 +0000 (00:02 +0900)]
ALSA: firewire-lib: fix buffer-over-run when detecting packet discontinuity
When detecting packet discontinuity, handle_in_packet() returns minus value
and this value is assigned to unsigned int variable, then the variable has
huge value. As a result, the variable causes buffer-over-run in
handle_out_packet(). This brings invalid page request and system hangup.
This commit fixes the bug to add a new argument into handle_in_packet()
and the number of handled data blocks is assignd to it. The function
return value is just used to check error.
I also considered to change the type of local variable to 'int' in
in_stream_callback(). This idea is based on type-conversion in C standard,
while it may cause future problems when adding more works. Thus, I dropped
this idea.
Fixes: 6fc6b9ce41c6('ALSA: firewire-lib: pass the number of data blocks in incoming packets to outgoing packets') Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Iwai [Tue, 26 May 2015 10:56:21 +0000 (12:56 +0200)]
ALSA: lx6464es: Use NULL for pointers
Spotted by sparse:
sound/pci/lx6464es/lx6464es.c:415:47: warning: Using plain integer as NULL pointer
sound/pci/lx6464es/lx6464es.c:417:48: warning: Using plain integer as NULL pointer
Takashi Iwai [Tue, 26 May 2015 10:52:54 +0000 (12:52 +0200)]
ALSA: lx6464es: Fix duplicated SSID entries
PCI_DEVICE() macro itself defines SSID entries while we override them
after that. Replace with a more appropriate macro, PCI_DEVICE_SUB()
to define all ID values properly.
Spotted by sparse:
sound/pci/lx6464es/lx6464es.c:60:11: warning: Initializer entry defined twice
sound/pci/lx6464es/lx6464es.c:61:12: also defined here
sound/pci/lx6464es/lx6464es.c:64:11: warning: Initializer entry defined twice
sound/pci/lx6464es/lx6464es.c:65:12: also defined here
Takashi Iwai [Tue, 26 May 2015 10:51:35 +0000 (12:51 +0200)]
ALSA: dummy: make local data static
Add missing prefix to make some local data static.
Spotted by sparse:
sound/drivers/dummy.c:159:20: warning: symbol 'model_emu10k1' was not declared. Should it be static?
sound/drivers/dummy.c:165:20: warning: symbol 'model_rme9652' was not declared. Should it be static?
....
With this patch, ALSA BeBoB driver can start packet streaming to/from
this model, while as a default, internal multiplexer of this model is
not initialized and generates no sound even if the driver transfers
any packets with PCM samples. To hear any sounds from this model,
userspace applications should be developed to set parameters to the
internal multiplexer. You can see raw information in FFADO website:
http://subversion.ffado.org/wiki/AvcModels/DigiDesignMboxPro2
Takashi Sakamoto [Fri, 22 May 2015 14:21:14 +0000 (23:21 +0900)]
ALSA: firewire-lib: use protocol error when detecting wrong value in CIP header
When detecting zero in 'dbs' field of CIP header, this packet streaming
should be aborted because of avoiding division-by-zero. This is an error
in an aspect of IEC 61883-1, thus protocol error.
This commit use EPROTO instead of EIO. Actually, the returned value is
not used for userspace and this commit has no effect.
Takashi Sakamoto [Fri, 22 May 2015 14:21:13 +0000 (23:21 +0900)]
ALSA: firewire-lib: use dev_err() when detecting incoming streaming error
When detecting invalid value in 'dbs' field of CIP header or packet
discontinuity, current implementation reports the status by err_info().
In most cases this state is caused by model-specific issue due to
vendor's customization and should be reported to developers.
This commit use dev_err() instead of dev_info() for this purpose.
In the cases, packet streaming is aborted, thus no message floading
occurs.
Takashi Sakamoto [Fri, 22 May 2015 14:00:52 +0000 (23:00 +0900)]
ALSA: firewire-lib: pass the number of data blocks in incoming packets to outgoing packets
Current implementation reuses the value of syt field in incoming packet to
outgoing packet for full duplex timestamp synchronization, while the number
of data blocks in outgoing packets refers to hard-coded table and the
synchronization cannot be applied to non-blocking stream.
This commit passes the number of data blocks from incoming packet
processing to outgoing packet processing for the synchronization. For
normal mode, isochronous callback handler is changed to generate the values
of syt and data blocks.
Takashi Sakamoto [Fri, 22 May 2015 14:00:50 +0000 (23:00 +0900)]
ALSA: firewire-lib: add buffer-over-run protection at receiving more data blocks than expected
In IEC 61883-6, the number of data blocks in a packet is limited up to
the value of SYT_INTERVAL. Current implementation is compliant to the
limitation, while it can cause buffer-over-run when the value of dbs
field in received packet is illegally large.
This commit adds a validator to detect such illegal packets to prevent
the buffer-over-run. Actually, the buffer is aligned to the size of memory
page, thus this issue hardly causes system errors due to the room to page
alignment, as long as a few packets includes such jumbo payload; i.e.
a packet to several received packets.
Here, Behringer F-Control Audio 202 (based on OXFW 960) has a quirk to
postpone transferring isochronous packet till finish handling any
asynchronous packets. In this case, this model is lazy, transfers no
packets according to several cycle-start packets. After finishing, this
model pushes required data in next isochronous packet. As a result, the
packet include more data blocks than IEC 61883-6 defines.
To continue to support this model, this commit adds a new flag to extend
the length of calculated payload. This flag allows the size of payload
5 times as large as IEC 61883-6 defines. As a result, packets from this
model passed the validator successfully.
Russell King [Sat, 9 May 2015 10:26:42 +0000 (11:26 +0100)]
ALSA: pcm: add DRM ELD helper
Add a helper for the EDID like data structure, which is typically passed
from a HDMI adapter to its associated audio driver. This informs the
audio driver of the capabilities of the attached HDMI sink.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Reviewed-by: Mark Brown <broonie@kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Mengdong Lin [Tue, 19 May 2015 14:29:30 +0000 (22:29 +0800)]
ALSA: hda - Move hda_i915.c from sound/pci/hda to sound/hda
The file is moved to hda core and renamed to hdac_i915.c, so can be used
by both legacy HDA driver and new Skylake audio driver.
- Add snd_hdac_ prefix to the public APIs.
- The i915 audio component is moved to core bus and dynamically allocated.
- A static pointer hdac_acomp is used to help bind/unbind callbacks to get
this component, because the sound card's private_data is used by the azx
chip pointer, which is a legacy structure. It could be removed if private
_data changes to some core structure which can be extended to find the
bus.
- snd_hdac_get_display_clk() is added to get the display core clock for
HSW/BDW.
- haswell_set_bclk() is moved to hda_intel.c because it needs to write the
controller registers EM4/EM5, and only legacy HD-A needs it for HSW/BDW.
- Move definition of HSW/BDW-specific registers EM4/EM5 to hda_register.h
and rename them to HSW_EM4/HSW_EM5, because other HD-A controllers have
different layout for the extended mode registers.
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Iwai [Tue, 19 May 2015 09:35:13 +0000 (11:35 +0200)]
ALSA: hda - Fix NULL dereference from CA0132 DSP loader
The CA0132 DSP loader leads to NULL deference since the recent
transition to HDA core code, as it unconditionally accesses
hdac_stream->substream->runtime. For DSP loading, the substream
shouldn't be assigned.
This patch addresses the NULL dereference above in addition to assure
the substream is cleared while DSP loading.
Takashi Iwai [Tue, 19 May 2015 08:20:13 +0000 (10:20 +0200)]
ALSA: hda - Add headset-mode* model options for ALC269 & co
Modern machines tend to have only one headset jack nowadays, and they
often need these quirks. Let's allow them applicable via model
option for ease of debugging.
Gabriele Martino [Mon, 18 May 2015 19:15:13 +0000 (21:15 +0200)]
ALSA: hda/ca0132: support for Alienware 15 Creative Sound Core3D-EX
This patch adds quirks detection to the Creative CA0132 codec, and the
quirk for Alienware 15 (2015).
Some quirks may need different pin configuration, so the relevant
compile-time configuration has been removed.
The pin configuration and related initialization verbs are generated at
runtime instead, in ca0132_config() and ca0132_prepare_verbs().
Thierry Reding [Tue, 5 May 2015 12:56:20 +0000 (14:56 +0200)]
ALSA: hda/hdmi - Implement Tegra-specific patch
The HDMI codec on NVIDIA Tegra SoCs has a feature that doesn't exist on
the MCP or GPU variants. The highest bit in the vendor-defined scratch
registers can be used to trigger an interrupt in the HDMI codec, which
is signalled to the HDMI driver. This can be used to pass information,
such as the HDA format, to the HDMI driver so that it can reconfigure
itself accordingly.
While at it, change the name of the codec to Tegra124 since there are no
other SoCs in the Tegra12x family. There isn't really a Tegra12x family.
Thierry Reding [Tue, 5 May 2015 12:56:19 +0000 (14:56 +0200)]
ALSA: hda/tegra - Set CORBRP self-clear flag
This is set for the MCP variants of the NVIDIA HDA controller, which the
Tegra variant was derived from. This fixes the following warning at boot
time:
Takashi Iwai [Mon, 18 May 2015 07:20:24 +0000 (09:20 +0200)]
ALSA: info: Register proc entries recursively, too
The commit [c560a6797e3b: ALSA: core: Remove child proc file elements
recursively] converted snd_card_proc_new() with the normal
snd_info_*() call and removed snd_device chain for such info
entries. However, it misses one point: the creation of the proc entry
was managed by snd_device chain in the former code, and now it's also
gone, which results in no proc files creation at all. Mea culpa.
This patch makes snd_info_card_register() creating the all pending
child proc entries in a shot. Also, since snd_card_register() might
be called multiple times, this function is also changed to be callable
multiple times.
Along with the changes above, now the linked list of snd_info_entry is
added at creation time instead of snd_info_register() for keeping eyes
of pending info entries.
Takashi Iwai [Mon, 18 May 2015 06:59:12 +0000 (08:59 +0200)]
ALSA: info: Fix leaks of child entries at snd_info_free_entry()
snd_info_free_entry() releases the all children nodes as well, but due
to the wrong timing of releasing the link, the children nodes may be
disconnected but left unreleased. This patch fixes it by moving the
link free at the right position. Also it eases list_for_each_entry()
without _safe option in snd_info_disconnect() because it no longer
frees the children nodes there.
Thierry Reding [Tue, 5 May 2015 12:45:57 +0000 (14:45 +0200)]
ALSA: hda/tegra - Fix oops
Commit a41d122449be ("ALSA: hda - Embed bus into controller object")
introduced a regression in the Tegra HDA driver that causes the
following oops during boot:
This is caused by azx_bus_init() trying to dereference chip->card, which
for the Tegra driver doesn't get initialized until sometime later during
the call to hda_tegra_create().
Fix this by mimicking the behaviour of the Intel driver and defer HDA
bus initialization until right before the call to snd_device_new().
Fixes: a41d122449be ('ALSA: hda - Embed bus into controller object') Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Lu, Han [Tue, 5 May 2015 01:05:48 +0000 (09:05 +0800)]
ALSA: hda - reset display codec when power on
In SKL, HDMI/DP codec and PCH HD Audio Controller are in different power wells,
so it's necessary to reset display audio codecs when power well on, otherwise
display audio codecs will disappear when resume from low power state.
Reset steps when power on:
enable codec wakeup -> azx_init_chip() -> disable codec wakeup
The callback for codec wakeup enable/disable is in drivers/gpu/drm/i915/.
Signed-off-by: Lu, Han <han.lu@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Add support for enabling codec wakeup override signal to allow
re-enumeration of the controller on SKL after resume from low power state.
In SKL, HDMI/DP codec and PCH HD Audio Controller are in different power
wells, so it's necessary to reset display audio codecs when power well on,
otherwise display audio codecs will disappear when resume from low power
state.
Reset steps when power on:
enable codec wakeup -> azx_init_chip() -> disable codec wakeup
v3 by Jani: Simplify to only support toggling the appropriate chicken bit.
v4 by Han: add explanation and specify the hw swquence.
Signed-off-by: Lu, Han <han.lu@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jie Yang [Thu, 30 Apr 2015 12:22:46 +0000 (20:22 +0800)]
ALSA: jack: don't report input event for phantom jack
There is no input_dev for phantom jack, we should not report
input event for it, otherwise, NULL pointer dereference error
will occur.
Fixes: 2ba2dfa1fcc7 ('ALSA: hda - Update to use the new jack kctls method') Signed-off-by: Jie Yang <yang.jie@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
We're providing dummy functions for CONFIG_SND_HDA_i915=n, thus ifdef
can be reduced. (But hda_i915_init() has to be fixed to return zero.)
This automatically fixes a compile warning:
sound/pci/hda/hda_intel.c: In function 'azx_probe_continue':
sound/pci/hda/hda_intel.c:1966:1: warning: label 'i915_power_fail' defined but not used [-Wunused-label]
Reported-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Mengdong Lin [Wed, 29 Apr 2015 09:43:43 +0000 (17:43 +0800)]
ALSA: hda - remove controller dependency on i915 power well for Baytrail/Braswell
For Baytrail (Valleyview) and Braswell (Cherryview), only the HDMI codec is
in the display power well while the HD-A controller isn't. So the controller
flag 'need_i915_power' is not set to release the display power after probe,
and the codec flag 'link_power_control" is set to request/release the display
power via bus link_power ops.
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Mengdong Lin [Wed, 29 Apr 2015 09:43:36 +0000 (17:43 +0800)]
ALSA: hda - divide controller and codec dependency on i915 gfx power well
This patch can improve power saving for Intel platforms on which only the
display audio codec is in the shared i915 power well:
- Add a flag "need_i915_power" to indicate whether the controller needs the
i915 power well.
- The driver will always request the i915 power when probing the controller
and codecs if AZX_DCAPS_I915_POWERWELL is set (either the controller or a
codec needs this power).
- If the controller needs the i915 power, the power will be held after probe
until the controller is runtime suspended or S3. If the controller doesn't
need the power, the power will be released the after probe, and a codec
that needs the power can request/release the power via bus link_power ops.
Background:
- For Haswell/Broadwell, which has a separate HD-A controller for display audio,
both the controller and the display codec are in the i915 power well.
- For Baytrail/Braswell, the display and analog audio share the same HDA
controller and link, and only the display codec is in the i915 power well.
- For Skylake, the display and analog audio share the same HDA controller but
use separate links. Only the display codec is in the i915 power well. And in
legacy mode we take the two links as one. So it can follow Baytrail/Braswell.
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Mengdong Lin [Wed, 29 Apr 2015 09:43:29 +0000 (17:43 +0800)]
ALSA: hda - implement link_power ops for i915 display power control
This patch implements the bus link_power ops to request/release i915 display
power well. It can be used by the display codec which shares this power well
with GPU on Intel platforms.
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Mengdong Lin [Wed, 29 Apr 2015 09:43:20 +0000 (17:43 +0800)]
ALSA: hda - allow a codec to control the link power
A flag "link_power_control" is added to indicate whether a codec needs to
control the link power. And a new bus ops link_power() is defined for the
codec to request to enable/disable the link power.
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Peter Zubaj [Tue, 28 Apr 2015 19:57:29 +0000 (21:57 +0200)]
ALSA: emu10k1: Emu10k2 32 bit DMA mode
Looks like audigy emu10k2 (probably emu10k1 - sb live too) support two
modes for DMA. Second mode is useful for 64 bit os with more then 2 GB
of ram (fixes problems with big soundfont loading)
1) 32MB from 2 GB address space using 8192 pages (used now as default)
2) 16MB from 4 GB address space using 4096 pages
Mode is set using HCFG_EXPANDED_MEM flag in HCFG register.
Also format of emu10k2 page table is then different.
This patch addresses the deadlock by reducing the rance taking
emux->register_mutex in snd_emux_open_seq_oss(). The lock is needed
for the refcount handling, so move it locally. The calls in
emux_seq.c are already with the mutex, thus they are replaced with the
version without mutex lock/unlock.
Jie Yang [Mon, 27 Apr 2015 13:20:58 +0000 (21:20 +0800)]
ALSA: jack: extend snd_jack_new to support phantom jack
Dont create input devices for phantom jacks.
Here, we extend snd_jack_new() to support phantom jack creating:
pass in a bool param for [non-]phantom flag, and a bool param
initial_jack to indicate whether we need to create a kctl at
this stage.
We can also add a kctl to the jack after its created meaning we
can now integrate the HDA and ASoC jacks.
Signed-off-by: Jie Yang <yang.jie@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jie Yang [Mon, 27 Apr 2015 13:20:57 +0000 (21:20 +0800)]
ALSA: Jack: handle jack embedded kcontrol creating within ctljack
This patch adds a static method get_available_index() to
allocate the index of new jack kcontrols and also adds
jack_kctl_name_gen() which is used to ensure compatibility
with jack naming by removing " Jack" from some incorrectly
passed names.
Signed-off-by: Jie Yang <yang.jie@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jie Yang [Mon, 27 Apr 2015 13:20:56 +0000 (21:20 +0800)]
ALSA: jack: implement kctl creating for jack devices
Currently the ALSA jack core registers only input devices for each jack
registered. These jack input devices are not readable by userspace devices
that run as non root. This patch series will implement kctls inside the
core jack part, including kctls creating, status changing report, for both
HD-Audio and ASoC jack. This allows non root userspace to read jack status
and act on it.
This patch adds a new API called snd_jack_add_new_kctl(), which will create
a kcontrol, add it to the card, and also attach it to the jack kctl list.
This patch also initialises the jack kctl list after jack is newed, and
reports kctl status when jack insertion/removal events occur.
snd_jack_new() is updated in the following patches to also support creating
phantom jacks and jack kcontrols. We then remove these duplicated features
from HDA jack and have jack kctls handled by core throughout HDA and ASoC.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com> Modified-by: Jie Yang <yang.jie@intel.com> Signed-off-by: Jie Yang <yang.jie@intel.com> Reveiwed-by: Mark Brown <broonie@kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Basically snd_emux_detach_seq() doesn't need a protection of
emu->register_mutex as it's already being unregistered. So, we can
get rid of this for avoiding the deadlock.
Some models provide too long string for the shortname that has 32bytes
including the terminator, and it results in a non-terminated string
exposed to the user-space. This isn't too critical, though, as the
string is stopped at the succeeding longname string.
This patch fixes such entries by dropping "SB" prefix (it's enough to
fit within 32 bytes, so far). Meanwhile, it also changes strcpy()
with strlcpy() to make sure that this kind of problem won't happen in
future, too.
ALSA: Close holes in struct snd_pcm_constraint_list
On a 64-bit system there is a 32-bit hole in struct snd_pcm_constraint_list
and then 32-bit padding at the end. Reordering things slightly gets rid of
the hole and padding, reducing the size of the struct by 50% from its
original size.
On a 64-bit system there are two 32-bit holes due to the alignment of 64-bit
fields. Reordering things slightly gets rid of those holes, reducing the
size of the struct by 17% percent of its original size.
The mute-LED mode control has the fixed on/off states that are
supposed to remain on/off regardless of the master switch. However,
this doesn't work actually because the vmaster hook is called in the
vmaster code itself.
This patch fixes it by calling the hook indirectly after checking the
mute LED mode.
ALSA: hda - Fix click noise at start on Dell XPS13
Dell XPS13 produces a click noise at boot up, and Gabriele spotted out
that it's triggered by the initial pin control of the mic (NID 0x19).
This has to be set to Hi-Z Vref while the driver initializes to Vref
80% as a normal mic.
This patch fixes the generic parser code not to override the target
vref if it has been already set by the driver, and adds a proper
initialization of the target vref for this pin in the Realtek driver
side.
Andy Lutomirski [Sun, 26 Apr 2015 23:47:59 +0000 (16:47 -0700)]
x86_64, asm: Work around AMD SYSRET SS descriptor attribute issue
AMD CPUs don't reinitialize the SS descriptor on SYSRET, so SYSRET with
SS == 0 results in an invalid usermode state in which SS is apparently
equal to __USER_DS but causes #SS if used.
Work around the issue by setting SS to __KERNEL_DS __switch_to, thus
ensuring that SYSRET never happens with SS set to NULL.
This was exposed by a recent vDSO cleanup.
Fixes: e7d6eefaaa44 x86/vdso32/syscall.S: Do not load __USER32_DS to %ss Signed-off-by: Andy Lutomirski <luto@kernel.org> Cc: Peter Anvin <hpa@zytor.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Denys Vlasenko <vda.linux@googlemail.com> Cc: Brian Gerst <brgerst@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull intel drm fixes from Dave Airlie.
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/i915: vlv: fix save/restore of GFX_MAX_REQ_COUNT reg
drm/i915: Workaround to avoid lite restore with HEAD==TAIL
drm/i915: cope with large i2c transfers
Pull intel iommu updates from David Woodhouse:
"This lays a little of the groundwork for upcoming Shared Virtual
Memory support — fixing some bogus #defines for capability bits and
adding the new ones, and starting to use the new wider page tables
where we can, in anticipation of actually filling in the new fields
therein.
It also allows graphics devices to be assigned to VM guests again.
This got broken in 3.17 by disallowing assignment of RMRR-afflicted
devices. Like USB, we do understand why there's an RMRR for graphics
devices — and unlike USB, it's actually sane. So we can make an
exception for graphics devices, just as we do USB controllers.
Finally, tone down the warning about the X2APIC_OPT_OUT bit, due to
persistent requests. X2APIC_OPT_OUT was added to the spec as a nasty
hack to allow broken BIOSes to forbid us from using X2APIC when they
do stupid and invasive things and would break if we did.
Someone noticed that since Windows doesn't have full IOMMU support for
DMA protection, setting the X2APIC_OPT_OUT bit made Windows avoid
initialising the IOMMU on the graphics unit altogether.
This means that it would be available for use in "driver mode", where
the IOMMU registers are made available through a BAR of the graphics
device and the graphics driver can do SVM all for itself.
So they started setting the X2APIC_OPT_OUT bit on *all* platforms with
SVM capabilities. And even the platforms which *might*, if the
planets had been aligned correctly, possibly have had SVM capability
but which in practice actually don't"
* git://git.infradead.org/intel-iommu:
iommu/vt-d: support extended root and context entries
iommu/vt-d: Add new extended capabilities from v2.3 VT-d specification
iommu/vt-d: Allow RMRR on graphics devices too
iommu/vt-d: Print x2apic opt out info instead of printing a warning
iommu/vt-d: kill bogus ecap_niotlb_iunits()
Merge branch 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
"This has a mixture of merge window cleanups and bugfixes"
* 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: st: add include for pinctrl
i2c: mux: use proper dev when removing "channel-X" symlinks
i2c: digicolor: remove duplicate include
i2c: Mark adapter devices with pm_runtime_no_callbacks
i2c: pca-platform: fix broken email address
i2c: mxs: fix broken email address
i2c: rk3x: report number of messages transmitted
Merge branch 'for-linus-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
Pull btrfs fixes from Chris Mason:
"Filipe hit two problems in my block group cache patches. We finalized
the fixes last week and ran through more tests"
* 'for-linus-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: prevent list corruption during free space cache processing
Btrfs: fix inode cache writeout