During firmware and library download, sometimes it is observed that
firmware and library download is timed-out resulting into probe failure.

This patch disables dynamic clock gating while firmware and library
download.
Signed-off-by: NPardha Saradhi K <pardha.saradhi.kesapragada@intel.com>
Signed-off-by: NSanyog Kale <sanyog.r.kale@intel.com>
Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

DSP clock source is configured by sending the A-State table to the FW.
Add the large config set IPC to configure the desired clock source
Signed-off-by: NPradeep Tewani <pradeep.d.tewani@intel.com>
Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Set dma control ipc can be used to set the M/N divider, enable the clks. It
takes different payload for different configuration. So modify the
skl_dsp_set_dma_control API to take the size and node_id as argument.
Signed-off-by: NJaikrishna Nemallapudi <jaikrishnax.nemallapudi@intel.com>
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Copier can support upto 4 output pins. However, only pin 0 is configured
as a part of copier initialization. Configuring rest of pins require the
separate IPC to be sent to fw.
Signed-off-by: NPradeep Tewani <pradeep.d.tewani@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

A dsp path and the modules in the path can support various pcm
configurations. The list of supported pcm configurations from topology
manifest would be stored and later selected runtime based on the hw
pcm params. For legacy, module data is filled in the 0th index of
resource and interface table.

To accommodate both models, change the relevant structures and populate
them by parsing newly defined tokens. This change is backward compatible
with the existing model where driver computes the resources required by
each dsp module.
Signed-off-by: NRamesh Babu <ramesh.babu@intel.com>
Signed-off-by: NShreyas NC <shreyas.nc@intel.com>
Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

On failure to get dsp_ops, dsp_init returns error without assigning ret. ret
is assigned in code path which will never be executed. Fix it.

Fixes: f77d443c ("ASoC: Intel: Skylake: Fix to free resources for dsp_init failure"
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

unmap mmio and free memory resources if dsp_init fails.
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Add cannonlake dsp support by adding its dsp_ops.
Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

For different platforms, number of dsp cores can vary. So instead of
creating array of size SKL_DSP_CORES_MAX, use num_core in dsp_ops() of
the respective platform to allocate core usage counts and states.
Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Number of dsp cores may differ for different platforms hence
adding it in dsp ops.
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Trivial fix to typo in debug message, dst_instacne should be
dst_instance
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

If the dma_buffer_size is not defined in topology, fix it to 2ms default
value to make backward compatible.

Fixes: f6e6ab1d ("ASoC: Intel: Skylake: Fix dma buffer size calculation")
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

DMA buffer size for gateway copier will be calculated based on:

For host DMA copier:
Input buffer size (ibs) for output direction (playback)
Output buffer size (obs) for input direction (capture)

For link DMA copier:
IBS for input direction (capture)
OBS for output direction (playback)

Update the driver to use the above.
Signed-off-by: NRamesh Babu <ramesh.babu@intel.com>
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Acked-By: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

mic-select module is a DSP module, which is used to select one or more
input channels.

This patch adds mic-select module type.
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

For audio kabylake is same as skylake except the module load approach.
This patch registers different dsp_fw_ops for kabylake and next patch
adds the module load support for kabylake.
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Acked-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

When DSP module is unbound, the module state needs to be in INIT_DONE
state instead of UNINT. Also the state needs to be set to UNINIT after
module is deleted from DSP pipeline.

So, set the module state to INIT_DONE after unbind and then UNINIT after
module is deleted.
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Acked-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

dsp ops is already set in init, so use this in cleanup routine
instead of again retrieving it. Also constify struct skl_dsp_ops.
Signed-off-by: NG Kranthi <gudishax.kranthikumar@intel.com>
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Acked-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

DSP firmware sends notification every 1ms, which is disabled in runtime
suspend. But if a system has no runtime pm, we keep getting
notification, so disable after FW init as well.
Signed-off-by: NG Kranthi <gudishax.kranthikumar@intel.com>
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Acked-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

DSP pipe needs to stopped before deleting the pipe. Currently check is
for pipe state > STARTED, which is incorrect. So changed to include
pipe state STARTED to stop the pipe if it started.
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Acked-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Geminilake is next gen SoC, so add the IDs for Geminilake.
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

I2S blob config structure can change between DSP copier module versions.
We are not modifying these structures anyway and NHLT blobs specify them
properly, so use the blob directly to send the dma_control ipc instead
of using the predefined I2S blob config structure.
Signed-off-by: NGuruprasadX Pawse <guruprasadx.pawse@intel.com>
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Acked-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

While going to suspend, if we have any pending D0i3 work scheduled,
flush that and force the DSP to goto D0i3 mode before going to suspend.
Signed-off-by: NJayachandran B <jayachandran.b@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark 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: NJayachandran B <jayachandran.b@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Post bind parameters of KPB module contains the instance id's of
neighbouring modules in the sink path

Now that module instance ids are generated dynamically we need to update
these parameters as well, so use the table created and update the ids
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Use private id's of module instances that are generated during
init_module for the IPC messages to DSP. These id's are freed
up during delete pipeline.
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Now that we have the bus parse capabilities moved to core, we need to
convert users.

The SKL driver and HDA extended lib needs to converted in single patch,
otherwise we regress on the functionality.
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

A module can be scheduled in deferent processing domains in DSP. Topology
specifies the module domain.
Signed-off-by: NSenthilnathan Veppur <senthilnathanx.veppur@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

The DSP instance creation also loads the firmware on DSPs. For library load
the firmware names come from topology so can't be loaded at object creation.

So split the firmware load and object creation. FW load is now called after
topology init in platform probe.
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

To query the ops used for a platform, we use skl_get_dsp_ops() which return
index and then we load the ops.

Rather than this return the ops, this way it cna be used later to query the
ops in rest of the driver.
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Kabylake is next generation Intel platform which has similar
audio controller to Skylake, so add the ID and driver data in
SKL driver.
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

There is log spam while doing playback, record or reloading the
audio firmware.

print_hex_dump uses printk(KERN_DEBUG,... which is different from
dev_dbg used elsewhere in the driver: it's always enabled at
compile-time. Change it to print_hex_dump_debug for logging consistency.

For consistency with other log statements, change dev_info to dev_dbg
for a kernel print which is frequently printed by the driver.
Signed-off-by: NVedang Patel <vedang.patel@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

If pipeline is not STARTED, we do not need to pause pipeline while
deleting.
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

When pipe is stopped/Paused, set the pipe state to paused instead
of created.
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

In case of XRUN recovery PCM prepare will be called. In this case
Host/Link DMAs are reset and reconfigured, hence the corresponding
FE/BE pipe needs to be reset in order to get to a clean state.
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

param_data variable is allocated during set module format of init
instance is not getting freed and hence can cause a memory leak.
So free it up.
Signed-off-by: NMousumi Jana <mousumix.jana@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Broxton DSP is mostly similar to Skylake one but with subtle
differences like no Code Load DMA and uses HDA DMA for code
loading, DSP D0 and D3 sequences are different.

These changes are comprehended by adding different DSP power up
and down handlers, and new loader ops and also adding prepare and
trigger which HDA DSP DMA requires
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Signed-off-by: NJayachandran B <jayachandran.b@intel.com>
Signed-off-by: NGuruprasadX Pawse <guruprasadx.pawse@intel.com>
Signed-off-by: NKranthi G <gudishax.kranthikumar@intel.com>
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NRamesh Babu <ramesh.babu@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

The code loading for Skylake and other platforms is different, so
add a dsp_ops and a loader_ops which can be defined for each
platform.

Move the dsp init, cleanup and loader ops (alloc and free dma) to
these ops
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

The SoC has MCLK output which is typically required by codecs.
The MCLK is controlled by DSP FW, so driver can configure that by
sending DMA_CONTROL IPC. The configuration for MCLK is present
in the endpoint blob.

So if block has this configuration, send IPC to DSP for MCLK
configuration. This is done by new function skl_dsp_set_dma_control()
which is invoked by BE prepare.
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

skl_setup_cpr_gateway_cfg() retrieves gateway settings which are
required for copier module configuration. For upcoming DMA
control IPC we need similar retrieval, so separate this piece into
skl_get_node_id() helper which will be common for these functions
Signed-off-by: NDharageswari R <dharageswari.r@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

For binding modules we should check if source or destination
module is in UNINT state. We canot bind even if one of them is
in this state.

So update the check from logical AND to logical OR and do not
bind modules for this case
Signed-off-by: NJeeja KP <jeeja.kp@intel.com>
Signed-off-by: NVinod Koul <vinod.koul@intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>