For device opened/closed, we check the D0i3 capability for the device
and invoke skl_tplg_d0i3_get/put, which counts the use case based on the
mode supported.
These counters are then used to decide if the device can enter D0i3 mode
of streaming or non-streaming or no D0i3.
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Not all use cases can support Doi3. Only certain use cases like hot word
detection, deep buffering can support D0i3 based on resource requirement.
So, pass the D0i3 capability for the FE/BE copier using topology. This will
be used to take a decision for D0i3 mode entry/exit.
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
For D0i3, we need to tell DSP to run the pipelines in LP mode. This
information is kept in topology and passed to driver as an attribute
for pipe.
So add a new tuple for lpmode and program the pipe based on value set.
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The driver needs two DSP callback, one to set D0i0 (active) and D0i3
(low-power) states.
Add these callbacks in dsp ops and implement them for broxton platforms.
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
To set the controller in D0i3 mode, the driver needs to set D0i3C
register after DSP is quiesced. Since the D0iX entry/exit is done by IPC,
add this as callback so that it can be invoked from IPC module.
Signed-off-by: Pardha Saradhi K <pardha.saradhi.kesapragada@intel.com>
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The audio DSP supports intermediate power states between D0 and D3
states. These states are D0i0 and D0i3 states.
Collectively we refer these two states as D0iX states.
To set or wake up from these states, driver also needs to send an IPC "Set D0iX
IPC" before doing anything else.
Add support for this new IPC messages.
Signed-off-by: Pardha Saradhi K <pardha.saradhi.kesapragada@intel.com>
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
If the DSP is in low power mode, it needs to be woken up by a "wake" IPC
to set it into the D0 state before we can send any other IPC command.
The call flow is that the driver calls sst_ipc_tx_message_wait() to send any
IPC and this call checks if the device is in low power mode and in that
case we need to send the wake IPC.
So add a new IPC nopm variant which can be called from driver and
doesn't check for power state (as we already know that) and avoids
circular dependency of again checking power state.
Signed-off-by: Jayachandran B <jayachandran.b@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
We used to use generic implementation of dma_map_ops.mmap which is
dma_common_mmap() but that only worked for simpler cached mappings when
vaddr = paddr.
If a driver requests uncached DMA buffer kernel maps it to virtual
address so that MMU gets involved and page uncached status takes into
account. In that case usage of dma_common_mmap() lead to mapping of
vaddr to vaddr for user-space which is obviously wrong. For more detals
please refer to verbose explanation here [1].
So here we implement our own version of mmap() which always deals
with dma_addr and maps underlying memory to user-space properly
(note that DMA buffer mapped to user-space is always uncached
because there's no way to properly manage cache from user-space).
[1] https://lkml.org/lkml/2016/10/26/973
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: <stable@vger.kernel.org> #4.5+
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
The i2s driver was only implementing playback for now. Implement capture to
make sure that's not a limitation anymore.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
This is the remaining update to PCM ABI object of version 5.
The flags will be applied to FE (Front End) links and can also be used
by physical links. The private data is reserved for future extension, so
offset update will add the private data size.
Now user space is using ABI v4, and the previous patch "ASoC: topology:
make PCM backward compatible from ABI v4" can assure the backward
compatibility.
Signed-off-by: Mengdong Lin <mengdong.lin@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Users start to use topology ABI from v4. ABI v5 updated existing manifest
and PCM elements. Two previous patches can support these ABI updates in a
backward compatible way. So if the topology file from user space is
generated by ABI v4, kernel will no longer quit but continue parsing.
Signed-off-by: Mengdong Lin <mengdong.lin@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
User space may not always set a valid FE DAI driver's name, FE DAI link's
name, stream name or cpu DAI name. In such cases, there are all ZERO in
these name string buffers of a topology PCM object. This patch will only
duplicate valid name strings for kernel FE DAI driver and DAI link when
creating them from topology, and free the name strings when destroying
them.
Signed-off-by: Mengdong Lin <mengdong.lin@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch adds support for old version 4 of PCMs (FE DAI & DAI links).
Topology ABI v5 added new fields to stream caps and thus changed PCMs.
Since user space may still uses v4, kernel will check the ABI version by
comparing the object size. If user space uses v4 of PCMs, kernel will
create the latest version of PCMs from the old version, and use the new
version internally to create FE DAI & DAI links. Because these new created
PCM elements will be freed later, kernel need duplicate the name strings
of DAI driver and DAI links when creating them.
Signed-off-by: Mengdong Lin <mengdong.lin@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
bsg_validate_sgv4_hdr() doesn't care about the request_queue, so drop it
from it's arguments.
Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
RFC 2734 defines the datagram_size field in fragment encapsulation
headers thus:
datagram_size: The encoded size of the entire IP datagram. The
value of datagram_size [...] SHALL be one less than the value of
Total Length in the datagram's IP header (see STD 5, RFC 791).
Accordingly, the eth1394 driver of Linux 2.6.36 and older set and got
this field with a -/+1 offset:
ether1394_tx() /* transmit */
ether1394_encapsulate_prep()
hdr->ff.dg_size = dg_size - 1;
ether1394_data_handler() /* receive */
if (hdr->common.lf == ETH1394_HDR_LF_FF)
dg_size = hdr->ff.dg_size + 1;
else
dg_size = hdr->sf.dg_size + 1;
Likewise, I observe OS X 10.4 and Windows XP Pro SP3 to transmit 1500
byte sized datagrams in fragments with datagram_size=1499 if link
fragmentation is required.
Only firewire-net sets and gets datagram_size without this offset. The
result is lacking interoperability of firewire-net with OS X, Windows
XP, and presumably Linux' eth1394. (I did not test with the latter.)
For example, FTP data transfers to a Linux firewire-net box with max_rec
smaller than the 1500 bytes MTU
- from OS X fail entirely,
- from Win XP start out with a bunch of fragmented datagrams which
time out, then continue with unfragmented datagrams because Win XP
temporarily reduces the MTU to 576 bytes.
So let's fix firewire-net's datagram_size accessors.
Note that firewire-net thereby loses interoperability with unpatched
firewire-net, but only if link fragmentation is employed. (This happens
with large broadcast datagrams, and with large datagrams on several
FireWire CardBus cards with smaller max_rec than equivalent PCI cards,
and it can be worked around by setting a small enough MTU.)
Cc: stable@vger.kernel.org
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The IP-over-1394 driver firewire-net lacked input validation when
handling incoming fragmented datagrams. A maliciously formed fragment
with a respectively large datagram_offset would cause a memcpy past the
datagram buffer.
So, drop any packets carrying a fragment with offset + length larger
than datagram_size.
In addition, ensure that
- GASP header, unfragmented encapsulation header, or fragment
encapsulation header actually exists before we access it,
- the encapsulated datagram or fragment is of nonzero size.
Reported-by: Eyal Itkin <eyal.itkin@gmail.com>
Reviewed-by: Eyal Itkin <eyal.itkin@gmail.com>
Fixes: CVE 2016-8633
Cc: stable@vger.kernel.org
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The interrupts can now be delivered as platform MSI interrupts on newer
platforms. The code looks for a new OF and ACPI strings in order to enable
the functionality.
Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
When MSI interrupts are supported, error and the transfer interrupt can
come from multiple processor contexts.
Each error interrupt is an MSI interrupt. If the channel is disabled by
the first error interrupt, the remaining error interrupts will gracefully
return in the interrupt handler.
If an error is observed while servicing the completions in success case,
the posting of the completions will be aborted as soon as channel disabled
state is observed. The error interrupt handler will take it from there and
finish the remaining completions. We don't want to create multiple success
and error messages to be delivered to the client in mixed order.
Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Bring out the interrupt cause to the top level so that MSI interrupts
can be hooked at a later stage.
Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Getting ready for the MSI interrupts. The pending_tre_count is used
in the interrupt handler to make sure all outstanding requests are
serviced.
The driver will allocate 11 MSI interrupts. Each MSI interrupt can be
assigned to a different CPU. Then, we have a race condition for common
variables as they share the same interrupt handler with a different
cause bit and they can potentially be executed in parallel. Making this
variable atomic so that it can be updated from multiple processor
contexts.
Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
During randconfig builds you can get the following warning
"warning: (ST_FDMA) selects ST_SLIM_REMOTEPROC which has unmet direct
dependencies (REMOTEPROC)"
randconfig builds should always build without any warnings so
update fdma to depend on REMOTEPROC so this can not happen.
Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Tested-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
On some benchmarks (e.g. netperf with ioeventfd disabled), APICv
posted interrupts turn out to be slower than interrupt injection via
KVM_REQ_EVENT.
This patch optimizes a bit the IRR update, avoiding expensive atomic
operations in the common case where PI.ON=0 at vmentry or the PIR vector
is mostly zero. This saves at least 20 cycles (1%) per vmexit, as
measured by kvm-unit-tests' inl_from_qemu test (20 runs):
| enable_apicv=1 | enable_apicv=0
| mean stdev | mean stdev
----------|-----------------|------------------
before | 5826 32.65 | 5765 47.09
after | 5809 43.42 | 5777 77.02
Of course, any change in the right column is just placebo effect. :)
The savings are bigger if interrupts are frequent.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
NF_REPEAT is only needed from nf_conntrack_in() under a very specific
case required by the TCP protocol tracker, we can handle this case
without returning to the core hook path. Handling of NF_REPEAT from the
nf_reinject() is left untouched.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
nf_iterate() has become rather simple, we can integrate this code into
nf_hook_slow() to reduce the amount of LOC in the core path.
However, we still need nf_iterate() around for nf_queue packet handling,
so move this function there where we only need it. I think it should be
possible to refactor nf_queue code to get rid of it definitely, but
given this is slow path anyway, let's have a look this later.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This field is only useful for nf_queue, so store it in the
nf_queue_entry structure instead, away from the core path. Pass
hook_head to nf_hook_slow().
Since we always have a valid entry on the first iteration in
nf_iterate(), we can use 'do { ... } while (entry)' loop instead.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use switch() for verdict handling and add explicit handling for
NF_STOLEN and other non-conventional verdicts.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Don't copy relevant fields from hook state structure, instead use the
one that is already available in struct xt_action_param.
This patch also adds a set of new wrapper functions to fetch relevant
hook state structure fields.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Move has_64bit_reloc into dev_priv->info. This will make it visible
in the feature listing debug output.
v2:
- Keep the struct member to keep GCC fragile but happy (Chris)
v3:
- More detailed commit message (Chris)
- Include forgotten CHV and BXT (Chris)
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1478162386-5018-1-git-send-email-joonas.lahtinen@linux.intel.com
Now that we don't have automatic syntax highlighting, use the code-block
directive with the explicitly selected language, where appropriate.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Now that we don't have automatic syntax highlighting, use the code-block
directive with the explicitly selected language, where appropriate.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Now that we don't have automatic syntax highlighting, use the code-block
directive with the explicitly selected language, where appropriate.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Set the default highlight language to "none", i.e. do not try to guess
the language and do automatic syntax highlighting on literal blocks.
Eyeballing around the generated documentation, we don't seem to actually
have a lot of literal blocks that would benefit from syntax
highlighting. The C code blocks we do have are typically very short, and
most of the literal blocks are things that shouldn't be highlighted (or,
do not have a pygments lexer). This seems to be true for literal blocks
both in the rst source files and in source code comments.
Not highlighting code is never wrong, but guessing the language wrong
almost invariably leads to silly or confusing highlighting.
At the time of writing, admin-guide/oops-tracing.rst and
admin-guide/ramoops.rst contain good examples of 1) a small C code
snippet not highlighted, 2) a hex dump highligted as who knows what, 3)
device tree block highlighted as C or maybe Python, 4) a terminal
interaction highlighted as code in some language, and finally, 5) some C
code snippets correctly identified as C. I think we're better off
disabling language guessing, and going by explicitly identified
languages for longer code blocks.
It is still possible to enable highlighting on an rst source file basis
using the highlight directive:
.. higlight:: language
and on a literal block basis using the code-block directive:
.. code-block:: language
See http://www.sphinx-doc.org/en/latest/markup/code.html for details.
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Cc: Markus Heiser <markus.heiser@darmarit.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Add missing semicolon to fix pdf build with more than one SPHINXDIRS
directory specified. For example make SPHINXDIRS="gpu media" pdfdocs.
Fixes: cd21379b16 ("doc-rst: generic way to build PDF of sub-folders")
Signed-off-by: Markus Heiser <markus.heiser@darmarIT.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Radically reduce the noise on stdout. The full build logs will still be
available under Documentatio/output/latex/*.log.
Cc: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Reviewed-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Tested-by: Markus Heiser <markus.heiser@darmarIT.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
With the unnecessary ; removed, the terminal URL detection also works
better.
Reviewed-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Building latexdocs doesn't actually require $(PDFLATEX). Move the checks
for it to the pdfdocs target which does require it, and specifically
outside of the target in order to not depend on latexdocs when we can't
build pdfdocs anyway.
Reviewed-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Tested-by: Markus Heiser <markus.heiser@darmarIT.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Refer to xelatex and latex options via variables. This allows the user
to override the pdflatex and latex options to use on the make command
line for experimenting. As a side effect, this makes the makefile a bit
tidier.
Reviewed-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
The CLCD does not come up on Versatile Express as it does not
(currently) have a syscon node for controlling the block apart
from the CLCD itself. Make sure the .init() function can bail
out without an error making it probe again.
Reported-by: Amit Pundir <amit.pundir@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Amit Pundir <amit.pundir@linaro.org>
Tested-by: Nicolae Rosia <nicolae_rosia@mentor.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Place pointer to hook state in xt_action_param structure instead of
copying the fields that we need. After this change xt_action_param fits
into one cacheline.
This patch also adds a set of new wrapper functions to fetch relevant
hook state structure fields.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
NF_STOP is only used by br_netfilter these days, and it can be emulated
with a combination of NF_STOLEN plus explicit call to the ->okfn()
function as Florian suggests.
To retain binary compatibility with userspace nf_queue application, we
have to keep NF_STOP around, so libnetfilter_queue userspace userspace
applications still work if they use NF_STOP for some exotic reason.
Out of tree modules using NF_STOP would break, but we don't care about
those.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Patch c5136b15ea ("netfilter: bridge: add and use br_nf_hook_thresh")
introduced br_nf_hook_thresh().
Replace NF_HOOK_THRESH() by br_nf_hook_thresh from
br_nf_forward_finish(), so we have no more callers for this macro.
As a result, state->thresh and explicit thresh parameter in the hook
state structure is not required anymore. And we can get rid of
skip-hook-under-thresh loop in nf_iterate() in the core path that is
only used by br_netfilter to search for the filter hook.
Suggested-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
We cannot block/sleep on nf_iterate because netfilter runs under rcu
read lock these days, where blocking is well-known to be illegal. So
let's remove these old comments.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch remove compile time code to catch inconventional verdicts.
We have better ways to handle this case these days, eg. pr_debug() but
even though I don't think this is useful at all, so let's remove this.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Intel's BXT devices need to execute a _DSM method
during {runtime_,}{suspend,resume} in order to get a
chunk of dwc3 to power gate and save some extra
power.
Let's do that now.
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
We'll be tracking a little more information for PCI
drivers, it's about time we add a private structure
for that.
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
When we get a half-way processed request, we should
make sure to try to prepare further TRBs for it or
for any possibly queued up request held in our
pending_list. This will make sure our controller is
kept busy for as long as possible.
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>