- 28 Oct, 2019 20 commits
-
-
Keyon Jie authored
Memory windows could be powered off before receiving PM_GATE IPC reply from FW, we can't read the mailbox to get reply. Signed-off-by:
Keyon Jie <yang.jie@linux.intel.com> Signed-off-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://lore.kernel.org/r/20191025224122.7718-15-pierre-louis.bossart@linux.intel.comSigned-off-by:
Mark Brown <broonie@kernel.org>
-
Keyon Jie authored
Add new ipc messages which will be sent from driver to FW, to ask FW to enter specific power saving state. Signed-off-by:
-
Keyon Jie authored
The helper sof_send_pm_ipc() is only suitable for context save/restore IPCs' sending, so rename it to sof_send_pm_ctx_ipc here. Signed-off-by:
-
Keyon Jie authored
Define macro and use it for the register polling retry count. Signed-off-by:
-
Keyon Jie authored
Align the logs for CIP timeout at D0I3C.I3 updating. Signed-off-by:
-
Keyon Jie authored
Parses the token from tplg file and store it to snd_sof_pcm_stream d0i3_compatible flag, which can be used later for d0ix transition management. Signed-off-by:
-
Keyon Jie authored
Add stream token SOF_TKN_STREAM_PLAYBACK_COMPATIBLE_D0I3 and SOF_TKN_STREAM_CAPTURE_COMPATIBLE_D0I3 to denote if the stream can be opened at low power d0i3 status or not. Signed-off-by:
-
Keyon Jie authored
Add flag d0i3_compatible to struct snd_sof_pcm_stream to denote if the stream can tolerate a transition to the D0i3 substate while opened (thus seen as 'active' by pm_runtime). Signed-off-by:
-
Keyon Jie authored
Using hda_dsp_set_power_state() as set_power_state() ops for apl to do d0ix platform configuration updates. Signed-off-by:
-
Keyon Jie authored
Using hda_dsp_set_power_state() as set_power_state() ops for cnl to do d0ix platform configuration updates. Signed-off-by:
-
Keyon Jie authored
Adding helper to implement setting dsp to d0i3 or d0i0 status, this will be needed for driver D0ix support. Signed-off-by:
-
Keyon Jie authored
D0i3 is a platform-defined substate of D0, so we need a platform-specific callback in dsp_ops to handle the relevant configurations. Signed-off-by:
-
Keyon Jie authored
We initialize/reset d0_substate to default d0i0 value when doing transition D3-->D0, e.g. at success of probing and resuming. Signed-off-by:
-
Keyon Jie authored
Add field d0_substate to struct snd_sof_dev to store the current DSP D0 sub-state(only meaningful when DSP in D0), which could be D0I0 or D0I3. Signed-off-by:
-
Kuninori Morimoto authored
chan > 8 or chan < 0 shouldn't happen. This patch uses WARN_ON() for such case. Reported-by:
Pavel Machek <pavel@denx.de> Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Link: https://lore.kernel.org/r/87y2x530a4.wl-kuninori.morimoto.gx@renesas.comSigned-off-by:
-
Shengjiu Wang authored
xrun may happen at the end of stream, the trigger->fsl_esai_trigger_stop maybe called in the middle of fsl_esai_hw_reset, this may cause esai in wrong state after stop, and there may be endless xrun interrupt. This issue may also happen with trigger->fsl_esai_trigger_start. So Add spin lock to lock those functions. Fixes: 7ccafa2b ("ASoC: fsl_esai: recover the channel swap after xrun") Signed-off-by:
Shengjiu Wang <shengjiu.wang@nxp.com> Acked-by:
Nicolin Chen <nicoleotsuka@gmail.com> Link: https://lore.kernel.org/r/52e92c4221a83e39a84a6cd92fc3d5479b44894c.1572252321.git.shengjiu.wang@nxp.comSigned-off-by:
-
Shengjiu Wang authored
The output divider should align with the output sample rate, if use ideal sample rate, there will be a lot of overload, which would cause underrun. The maximum divider of asrc clock is 1024, but there is no judgement for this limitation in driver, which may cause the divider setting not correct. For non-ideal ratio mode, the clock rate should divide the sample rate with no remainder, and the quotient should be less than 1024. Signed-off-by:
-
Brent Lu authored
The first DMIC capture always fail (zero sequence data from PCM port) after using DSP hotwording function (i.e. Google assistant). This rt5514 codec requires to control mclk directly in the set_bias_level function. Implement this function in machine driver to control the ssp1_mclk clock explicitly could fix this issue. Signed-off-by:
Brent Lu <brent.lu@intel.com> Reviewed-by:
-
YueHaibing authored
remove unneeded semicolon. Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Link: https://lore.kernel.org/r/20191025120801.16236-1-yuehaibing@huawei.comSigned-off-by:
-
Michael Ellerman authored
Commit 4ac85de9 ("ASoC: fsl: fsl_dma: remove snd_pcm_ops") removed fsl_dma_ops but left a usage, leading to a build error for some configs, eg. mpc85xx_defconfig: sound/soc/fsl/fsl_dma.c: In function ‘fsl_soc_dma_probe’: sound/soc/fsl/fsl_dma.c:905:18: error: ‘fsl_dma_ops’ undeclared (first use in this function) dma->dai.ops = &fsl_dma_ops; ^~~~~~~~~~~ Remove the usage to fix the build. Fixes: 4ac85de9 ("ASoC: fsl: fsl_dma: remove snd_pcm_ops") Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20191025051353.2878-1-mpe@ellerman.id.auSigned-off-by:
-
- 25 Oct, 2019 10 commits
-
-
Pierre-Louis Bossart authored
The current position update is not precise enough for PulseAudio to work reliably with the timer-based scheduling on Baytrail, Cherrytrail, Broadwell. Disable the NO_PERIOD_WAKEUP capability and use BATCH to signal that the position is only reliable and updated during period_elapsed events. This will be reverted when the firmware provides a more accurate position for those platforms. Reviewed-by:
Jaska Uimonen <jaska.uimonen@intel.com> Reviewed-by:
Ranjani Sridharan <ranjani.sridharan@linux.intel.com> Signed-off-by:
-
Pierre-Louis Bossart authored
Currently the INFO_ flags such as PAUSE/NO_PERIOD_WAKEUP are defined in the SOF PCM core, which doesn't scale. To account for platform variations, these flags need to be set in DSP ops. This patch only moves the definitions and does not change any functionality. Reviewed-by:
-
Ben Dooks authored
We see odd FIFO overruns with this, we assume the best thing to do is to disable the RX I2S frontend first, and then disable the FIFO that is using it. This also fixes an issue where using multi-word frames (TDM) have partial samples stuck in the FIFO which then get read out when the next capture is started. Signed-off-by:
Ben Dooks <ben.dooks@codethink.co.uk> Acked-by:
Jon Hunter <jonathanh@nvidia.com> Link: https://lore.kernel.org/r/20191018154833.7560-5-ben.dooks@codethink.co.ukSigned-off-by:
-
Edward Cragg authored
Add a callback to configure TDM settings for the Tegra30 I2S ASoC 'platform' driver. Signed-off-by:
Edward Cragg <edward.cragg@codethink.co.uk> [ben.dooks@codethink.co.uk: merge fix for power management] [ben.dooks@codethink.co.uk: add review change for fsync of 1 clock] Signed-off-by:
-
Geert Uytterhoeven authored
Fix misspelling of "endpoints" and "connection". Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Link: https://lore.kernel.org/r/20191024153130.31082-1-geert+renesas@glider.beSigned-off-by:
-
Geert Uytterhoeven authored
Fix misspellings of "configuration" and "configure". Signed-off-by:
-
Kuninori Morimoto authored
for_each_dpcm_xx() macro is using "dpcm" as parameter (1), but, it is also struct member (2). #define for_each_dpcm_fe(be, stream, dpcm) \ list_for_each_entry(dpcm, &(be)->dpcm[stream]...) ^^^^(1) ^^^^(2) Thus, it will be compile error if user not used "dpcm" as parameter for_each_dpcm_fe(be, stream, dp) ^^ This patch fixup it. Reported-by:
-
Edward Cragg authored
The tegra3 audio can support 24 and 32 bit sample sizes so add the option to the tegra30_i2s_hw_params to configure the S24_LE or S32_LE formats when requested. Signed-off-by:
-
Curtis Malainey authored
Fix bugs reported by kbuild test robot Fixes: a0e0d135 ("ASoC: rt5677: Add a PCM device for streaming hotword via SPI") Reported-by:
kbuild test robot <lkp@intel.com> Signed-off-by:
Curtis Malainey <cujomalainey@chromium.org> Link: https://lore.kernel.org/r/20191024184026.183913-1-cujomalainey@chromium.orgSigned-off-by:
-
Colin Ian King authored
The allocation of rt5677_dsp can potentially fail and return null, so add a null check and return -ENOMEM on a memory allocation failure. Addresses-Coverity: ("Dereference null return") Fixes: a0e0d135 ("ASoC: rt5677: Add a PCM device for streaming hotword via SPI") Signed-off-by:Colin Ian King <colin.king@canonical.com> Link: https://lore.kernel.org/r/20191024124610.18182-1-colin.king@canonical.comSigned-off-by:
-
- 24 Oct, 2019 1 commit
-
-
Akshu Agrawal authored
Add platform specific data for Kahlee project. Signed-off-by:
Akshu Agrawal <akshu.agrawal@amd.com> Signed-off-by:
Shirish S <shirish.s@amd.com> Signed-off-by:
-
- 23 Oct, 2019 9 commits
-
-
Russell King authored
Propagate the error code from request_irq(), rather than returning -EBUSY. Signed-off-by:
Russell King <rmk+kernel@armlinux.org.uk> Link: https://lore.kernel.org/r/E1iNIqh-0000tW-EZ@rmk-PC.armlinux.org.ukSigned-off-by:
-
Pan Xiuli authored
Add Kconfig, PCI ID and chip info for JSL platform. The DSP only has 2 cores for this platform. Signed-off-by:
Pan Xiuli <xiuli.pan@linux.intel.com> Signed-off-by:
-
Pan Xiuli authored
There are no upstream machine drivers just yet so just add dummy table for compilation in nocodec-mode. Signed-off-by:
-
Pierre-Louis Bossart authored
Such traces should be extremely rare but extremely useful for debug. Report errors for all calls to sdn_sof_dsp_read_poll_timeout(), but only on negative values for consistency. Add traces that enable each timeout to be uniquely identified. Signed-off-by:
-
Pierre-Louis Bossart authored
If a ROM timeout is detected, we still stop the DMA but will return the initial error should the DMA stop also fail. Likewise the cleanup is handled regardless of the status, but we return the initial error. Signed-off-by:
-
Mao Wenan authored
If SND_SOC_MT8183_MT6358_TS3A227E_MAX98357A=y, below errors can be seen: sound/soc/codecs/cros_ec_codec.o: In function `send_ec_host_command': cros_ec_codec.c:(.text+0x534): undefined reference to `cros_ec_cmd_xfer_status' cros_ec_codec.c:(.text+0x101c): undefined reference to `cros_ec_get_host_event' This is because it will select SND_SOC_CROS_EC_CODEC after commit 2cc3cd5f ("ASoC: mediatek: mt8183: support WoV"), but SND_SOC_CROS_EC_CODEC depends on CROS_EC. Fixes: 2cc3cd5f ("ASoC: mediatek: mt8183: support WoV") Signed-off-by:
Mao Wenan <maowenan@huawei.com> Link: https://lore.kernel.org/r/20191023063103.44941-1-maowenan@huawei.comSigned-off-by:
-
Kuninori Morimoto authored
This patch adds missing of_node_put() for rsnd_parse_tdm_split_mode() rsnd_parse_connect_graph() Reported-by:
-
Jiada Wang authored
According to R-Car3 HW manual 40.3.3 (Data Format on Audio Local Bus), in case of monaural data writing or reading through Audio-DMAC, it's always in Left Justified format, so both src and dst DMA Bus width should be equal to physical data width. Therefore set src and dst's DMA bus width to: - [monaural case] data width - [non-monaural case] 32bits (as prior applying the patch) Cc: Andrew Gabbasov <andrew_gabbasov@mentor.com> Cc: Timo Wischer <twischer@de.adit-jv.com> Signed-off-by:
Jiada Wang <jiada_wang@mentor.com> Signed-off-by:
Eugeniu Rosca <erosca@de.adit-jv.com> Link: https://lore.kernel.org/r/20191022185518.12838-1-erosca@de.adit-jv.comSigned-off-by:
-
Kuninori Morimoto authored
ALSA SoC has for_each_rtdcom() which is link list for rtd-component which is called as rtdcom. The relationship image is like below rtdcom rtdcom rtdcom component component component rtd->component_list -> list -> list -> list ... Here, the pointer get via normal link list is rtdcom, Thus, current for_each loop is like below, and need to get component via rtdcom->component for_each_rtdcom(rtd, rtdcom) { component = rtdcom->component; ... } but usually, user want to get pointer from for_each_xxx is component directly, like below. for_each_rtd_component(rtd, rtdcom, component) { ... } This patch expands list_for_each_entry manually, and enable to get component directly from for_each macro. Because of it, the macro becoming difficult to read, but macro itself becoming useful. Signed-off-by:
-
