- 04 2月, 2016 6 次提交
-
-
由 Jeeja KP 提交于
Some modules may be directly connected to a pipeline without a mixer module. For these modules, we require PRE_PMU and POST_PMU handler which will do bind between the pipelines, so add these missing handlers. 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>
-
由 Jeeja KP 提交于
skl_tplg_bind_sinks() takes only the first sink widget. This breaks in case we have multiple sinks for a module. So pass source widget to skl_tplg_bind_sinks() and bind for all sinks by calling this recursively 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>
-
由 Jeeja KP 提交于
We should not stop the sink pipe in it's pmd handler for a mixin module as this module may still be connected to other pipes. This will be stopped and freed by current implementation on last connected pipe unbind. 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>
-
由 Dharageswari.R 提交于
We check and allocate pipeline resources in one shot. That causes leaks if module creation fails later as that is not freed. So split the resource allocation into two, first check if resources are available and then add the resources upon successful creation. So two new functions are added for checking and current functions are re-purposed to only add the resources for memory and MCPS. 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>
-
由 Guneshwor Singh 提交于
While cleaning resources on module pmd event, we check for return of skl_unbind_modules(). On failure this causes leak as all modules attached do not have resources freed. So ignore return value of module unbind and continue freeing resources. This makes dapm state and resources correct. Signed-off-by: NGuneshwor Singh <guneshwor.o.singh@intel.com> Signed-off-by: NVinod Koul <vinod.koul@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
由 Vinod Koul 提交于
TLV buffer can be smaller than the module data, so update the size of data to be copied before doing the copy. Also TLV header consists of two unsigned ints, this is also taken into account here and size modified to reflect this Suggested-by: NTakashi Iwai <tiwai@suse.de> Signed-off-by: NVinod Koul <vinod.koul@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
- 23 1月, 2016 1 次提交
-
-
由 Sudip Mukherjee 提交于
If snd_soc_tplg_component_load() fails we just printed an error message and returned the error code but we missed releasing the firmware. Signed-off-by: NSudip Mukherjee <sudip@vectorindia.org> Signed-off-by: NMark Brown <broonie@kernel.org>
-
- 05 1月, 2016 2 次提交
-
-
由 Vinod Koul 提交于
This provide the fix for firmware memory by freeing the pointer in driver remove where it is safe to do so Signed-off-by: NVinod Koul <vinod.koul@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
由 Vinod Koul 提交于
This reverts commit 87b5ed8e ("ASoC: Intel: Skylake: fix memory leak") as it causes regression on Skylake devices The SKL drivers can be deferred probe. The topology file based widgets can have references to topology file so this can't be freed until card is fully created, so revert this patch for now [ 66.682767] BUG: unable to handle kernel paging request at ffffc900001363fc [ 66.690735] IP: [<ffffffff806c94dd>] strnlen+0xd/0x40 [ 66.696509] PGD 16e035067 PUD 16e036067 PMD 16e038067 PTE 0 [ 66.702925] Oops: 0000 [#1] PREEMPT SMP [ 66.768390] CPU: 3 PID: 57 Comm: kworker/u16:3 Tainted: G O 4.4.0-rc7-skl #62 [ 66.778869] Hardware name: Intel Corporation Skylake Client platform [ 66.793201] Workqueue: deferwq deferred_probe_work_func [ 66.799173] task: ffff88008b700f40 ti: ffff88008b704000 task.ti: ffff88008b704000 [ 66.807692] RIP: 0010:[<ffffffff806c94dd>] [<ffffffff806c94dd>] strnlen+0xd/0x40 [ 66.816243] RSP: 0018:ffff88008b707878 EFLAGS: 00010286 [ 66.822293] RAX: ffffffff80e60a82 RBX: 000000000000000e RCX: fffffffffffffffe [ 66.830406] RDX: ffffc900001363fc RSI: ffffffffffffffff RDI: ffffc900001363fc [ 66.838520] RBP: ffff88008b707878 R08: 000000000000ffff R09: 000000000000ffff [ 66.846649] R10: 0000000000000001 R11: ffffffffa01c6368 R12: ffffc900001363fc [ 66.854765] R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000 [ 66.862910] FS: 0000000000000000(0000) GS:ffff88016ecc0000(0000) knlGS:0000000000000000 [ 66.872150] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 66.878696] CR2: ffffc900001363fc CR3: 0000000002c09000 CR4: 00000000003406e0 [ 66.886820] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 66.894938] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 66.903052] Stack: [ 66.905346] ffff88008b7078b0 ffffffff806cb1db 000000000000000e 0000000000000000 [ 66.913854] ffff88008b707928 ffffffffa00d1050 ffffffffa00d104e ffff88008b707918 [ 66.922353] ffffffff806ccbd6 ffff88008b707948 0000000000000046 ffff88008b707940 [ 66.930855] Call Trace: [ 66.933646] [<ffffffff806cb1db>] string.isra.4+0x3b/0xd0 [ 66.939793] [<ffffffff806ccbd6>] vsnprintf+0x116/0x540 [ 66.945742] [<ffffffff806d02f0>] kvasprintf+0x40/0x80 [ 66.951591] [<ffffffff806d0370>] kasprintf+0x40/0x50 [ 66.957359] [<ffffffffa00c085f>] dapm_create_or_share_kcontrol+0x1cf/0x300 [snd_soc_core] [ 66.966771] [<ffffffff8057dd1e>] ? __kmalloc+0x16e/0x2a0 [ 66.972931] [<ffffffffa00c0dab>] snd_soc_dapm_new_widgets+0x41b/0x4b0 [snd_soc_core] [ 66.981857] [<ffffffffa00be8c0>] ? snd_soc_dapm_add_routes+0xb0/0xd0 [snd_soc_core] [ 67.007828] [<ffffffffa00b92ed>] soc_probe_component+0x23d/0x360 [snd_soc_core] [ 67.016244] [<ffffffff80b14e69>] ? mutex_unlock+0x9/0x10 [ 67.022405] [<ffffffffa00ba02f>] snd_soc_instantiate_card+0x47f/0xd10 [snd_soc_core] [ 67.031329] [<ffffffff8049eeb2>] ? debug_mutex_init+0x32/0x40 [ 67.037973] [<ffffffffa00baa92>] snd_soc_register_card+0x1d2/0x2b0 [snd_soc_core] [ 67.046619] [<ffffffffa00c8b54>] devm_snd_soc_register_card+0x44/0x80 [snd_soc_core] [ 67.055539] [<ffffffffa01c303b>] skylake_audio_probe+0x1b/0x20 [snd_soc_skl_rt286] [ 67.064292] [<ffffffff808aa887>] platform_drv_probe+0x37/0x90 Signed-off-by: NVinod Koul <vinod.koul@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
- 17 12月, 2015 1 次提交
-
-
由 Dan Carpenter 提交于
"data" is a u32 pointer so this copies the information to wrong place entirely. Fixes: 140adfba ('ASoC: Intel: Skylake: Add tlv byte kcontrols') Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com> Acked-by: NVinod Koul <vinod.koul@intel.com> Tested-by: NDharageswari R <dharageswari.r@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
- 09 12月, 2015 4 次提交
-
-
由 Omair M Abdullah 提交于
If a module is ON then we should read the module parameters from DSP rather than driver cached values Signed-off-by: NOmair M Abdullah <omair.m.abdullah@intel.com> Signed-off-by: NVinod Koul <vinod.koul@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
由 Jeeja KP 提交于
Module initialization parameter data can be set by - INIT_INSTANCE IPC by using the default value - SET_PARAMS immediately after INIT_INSTANCE - SET_PARAMS data from kcontrol values set this patch add param type to identify the parameters has to be sent to DSP. 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>
-
由 Jeeja KP 提交于
A module can require extra memory for processing, like audio algorithms. The memory for these modules needs to be represented in base module configuration and passed to DSP on init, so add the memory pages as a field in widget data 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>
-
由 Dharageswari R 提交于
A module is loaded when the path consisting the module is opened. The module binary(ies) is loaded from file system and cached in kernel memory for future use. This is downloaded to DSP using DMA and invoking Load module IPCs This patch adds support for load/unload module IPCs, DMAing modules and manging the modules 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>
-
- 02 12月, 2015 3 次提交
-
-
由 Jeeja KP 提交于
This adds tlv bytes topology control creation and control load to initialize kcontrol data. And this also adds the callbacks for the these tlv byte kcontrols Signed-off-by: NMythri P K <mythri.p.k@intel.com> Signed-off-by: NDivya Prakash <divya1.prakash@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>
-
由 Jeeja KP 提交于
This adds support to configure module parameter during module initialization or after module init using set module param required by the DSP firmware sequence. 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>
-
由 Jeeja KP 提交于
Some DSP modules have user configurable parameters. These parameters are required by modules in the following scenario - during initialization - after initialization using set parameter This patch adds helper routine to set module parameters using large config set IPC message and removes params to be passed as init module routine. 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>
-
- 26 11月, 2015 2 次提交
-
-
由 Vinod Koul 提交于
The MCPS is freed first thing in pmd events but non memory. So if we face error during teardown we leak this mem, so move the code up 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>
-
由 Jeeja KP 提交于
Module hw param fixup will change the valid bit depth based on the fixup flag. If valid bit depth changes, need to set the bit depth according to valid bit depth. This patch fixes this issue of updating bit depth correctly. 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>
-
- 23 11月, 2015 1 次提交
-
-
由 Sudip Mukherjee 提交于
We have requested the firmware but missed releasing it. Signed-off-by: NSudip Mukherjee <sudip@vectorindia.org> Signed-off-by: NMark Brown <broonie@kernel.org>
-
- 19 11月, 2015 1 次提交
-
-
由 Jeeja KP 提交于
This patch fixes the below warning form smatch and makes the skl_tplg_bind_sinks take the next sink as argument which is true when the current sink is valid sound/soc/intel/skylake/skl-topology.c:453 skl_tplg_bind_sinks() error: we previously assumed 'sink' could be null (see line 452) sound/soc/intel/skylake/skl-topology.c 451 452 if (!sink) ^^^^ New check. Reversed? 453 return skl_tplg_bind_sinks(sink, skl, src_mconfig); ^^^^ This is dereferenced inside the function. 454 455 return 0; Reported-by: NDan Carpenter <dan.carpenter@oracle.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>
-
- 16 11月, 2015 11 次提交
-
-
由 Vinod Koul 提交于
MCPS free was being done from PGA context which will free up MCPS for only last modules in a pipe and not the rest causing MCPS leak and eventual audio loss due to no "free" MCPS. This needs to be freed for every module while cleaning up the modules, so move the check to skl_tplg_mixer_dapm_post_pmd_event() Signed-off-by: NMohan Krishna Velaga <mohan.krishnax.velaga@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>
-
由 Hardik T Shah 提交于
The DSP FW specifies loadable modules using GUIDs so add support to specify the GUIDs from topology Signed-off-by: NHardik T Shah <hardik.t.shah@intel.com> Signed-off-by: NOmair M Abdullah <omair.m.abdullah@intel.com> Signed-off-by: NJeeja KP <jeeja.kp@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>
-
由 Hardik T Shah 提交于
The module pin formats are considered homogeneous, but some modules can have different pcm formats on different pins, like reference signal for a module. This patch add support for configuration of each pin of module and allows us to specify if pins and homogeneous or heterogeneous Signed-off-by: NHardik T Shah <hardik.t.shah@intel.com> Signed-off-by: NOmair M Abdullah <omair.m.abdullah@intel.com> Signed-off-by: NJeeja KP <jeeja.kp@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>
-
由 Jeeja KP 提交于
ASoC core already checks if BE is active. If BE is active, hw_params callback is ignored. This patch removes the redundant check in driver for copier widget power check in update be hw_params. 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>
-
由 Jeeja KP 提交于
While rigourous testing of SKL drivers, we noticed underuns and overuns and on debug realized that we need to change driver handling of FE pipe startup and shutdown We need to start DMA and then run pipe together and not split these up. Similarly while stopping we should stop pipe and then DMA in a sequence. 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>
-
由 Jeeja KP 提交于
In SKL topology routes, some paths can be connected by a widget which are not a DSP FW widget and virtual with respect to firmware. In these case when module has to bind, then the virtual DSP modules needs to skipped till a actual DSP module is found which connects the pipelines. So we need to walk the graph and find a widget which is real in nature. This patch adds that support and splits skl_tplg_pga_dapm_pre_pmu_event() fn with parsing code to skl_tplg_bind_sinks() fn and call that recursively as well as while parsing The patch moves code a bit while splitting so diffstat doesn't tell real picture 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>
-
由 Jeeja KP 提交于
in_pin and out_pin list for a module has the information about the module that are bound together. So we can directly look at pin information of module for binding and unbind. As a result the preinitialized dapm_path_last we had is removed and code and memory optimzed. 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>
-
由 Jeeja KP 提交于
For supporting multiple dynamic pins, module state check is incorrect. In case of unbind, module state need to be changed to uninit if all pins in the module is is unbind state. To handle module state correctly add pin state and use pin state check to set module state correctly. 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>
-
由 Jeeja KP 提交于
If link type is HDA, NHLT blob is null, as NHLT defines non HDA links only. So we should ignore blob query for HDA links. 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>
-
由 Jeeja KP 提交于
In case of 32 bit, the FE update params returns error as it falls thru to default case. This patch adds 32 bit depth handling in update FE params. 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>
-
由 Jeeja KP 提交于
To get the FE copier module, the check to ignore non DSP widgets was wrong. This path corrects the check to ignore non DSP widget. 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>
-
- 24 10月, 2015 4 次提交
-
-
由 Jeeja KP 提交于
Some module pin connection are static and defined by the topology. This patch adds support for static pin definitions in topology widget private data 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>
-
由 Subhransu S. Prusty 提交于
Currently params only for first copier widget identified in the source/sink path is queried from NHLT. In the dapm route the playback/capture widget may be connected to more than one copier widget. This patch adds return check to return only for any error case. Signed-off-by: NSubhransu S. Prusty <subhransu.s.prusty@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>
-
由 Subhransu S. Prusty 提交于
In case of playback, for the BE dai source path should be iterated to find the pipe params. With sink path iterated, this resulted in a loop and kernel panic with page request failure. Similar are the cases for Capture and FE dais. Using correct macros to fix the panic 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>
-
由 Jeeja KP 提交于
The BE copier private data allowed endpoint configuration blobs, now these are queried from BIOS, we don't need to copy the blob, but only capability. Removing the blob from private data will not allocate memory for module specific config in which case memcpy will fail. Fix is to assign the ptr queried from the NHLT table for the endpoint configuration. 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>
-
- 08 10月, 2015 1 次提交
-
-
由 Vinod Koul 提交于
dev_info is too noisy for tplg wiget loading, so move it to debug level Signed-off-by: NVinod Koul <vinod.koul@intel.com> Signed-off-by: NMark Brown <broonie@kernel.org>
-
- 07 10月, 2015 3 次提交
-
-
由 Jeeja KP 提交于
Initialize and creates DSP controls if processing pipe capability is supported by HW. Updates the dma_id, hw_params to module param to be used when DSP module has to be configured. Signed-off-by: NJeeja KP <jeeja.kp@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>
-
由 Vinod Koul 提交于
The SKL driver does not code DSP topology in driver. It uses the newly added ASoC topology core to parse the topology information (controls, widgets and map) from topology binary. Each topology element passed private data which contains information that driver used to identify the module instance within firmware and send IPCs for that module to DSP firmware along with parameters. This patch adds init routine to invoke topology load and callback for topology creation. Signed-off-by: NJeeja KP <jeeja.kp@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>
-
由 Vinod Koul 提交于
For FE and BE, the PCM parameters come from FE and BE hw_params values passed. For a FE we convert the FE params to DSP expected module format and pass to DSP. For a BE we need to find the gateway settings (i2s/PDM) to be applied. These are queried from NHLT table and applied. Further for BE based on direction the settings are applied as either source or destination parameters. These helpers here allow the format to be calculated and queried as per firmware format. Signed-off-by: NJeeja KP <jeeja.kp@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>
-