- Mar 21, 2019
-
-
Charles Keepax authored
The wm_adsp_ops structures should be static and correct two printf specifiers. Fixes: 170b1e12 ("ASoC: wm_adsp: Add support for new Halo core DSPs") Fixes: 4e08d50d ("ASoC: wm_adsp: Factor out DSP specific operations") Reported-by:
kbuild test robot <lkp@intel.com> Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Sylwester Nawrocki authored
We can simplify the code by caching the CPU DAI master/slave information rather than reading previously set register bit. Signed-off-by:
Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Annaliese McDermond authored
Break the clock setting logic out from the main hw_params. It's rather large and unweildy and makes for a large function. This also better enables some of the following changes to the clock tree access in the driver. Signed-off-by:
Annaliese McDermond <nh6z@nh6z.net> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Paul Cercueil authored
Depending on MACH_JZ4740 prevent us from creating a generic kernel that works on more than one MIPS board. Instead, we just depend on MIPS being set. Signed-off-by:
Paul Cercueil <paul@crapouillou.net> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
- Mar 19, 2019
-
-
Stuart Henderson authored
Signed-off-by:
Stuart Henderson <stuarth@opensource.cirrus.com> Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Richard Fitzgerald authored
A Halo Core DSP has a memory protection unit that can trap and signal memory access faults. This patch adds a function that dumps the fault information. The interrupt reaches the host via the parent codec interrupt controller so this fault function is exported to be called by the codec driver. Signed-off-by:
Richard Fitzgerald <rf@opensource.cirrus.com> Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Wen Shi authored
The Halo core is a new generation of audio DSP architecture from Cirrus Logic. A new iteration of the WMFW file format (v3) is also added, for this new architecture. Currently this format is not supported on the old ADSP2 architecture however support may be added for it in the future. Signed-off-by:
Wen Shi <wenshi@opensource.cirrus.com> Signed-off-by:
Piotr Stankiewicz <piotrs@opensource.cirrus.com> Signed-off-by:
Richard Fitzgerald <rf@opensource.cirrus.com> Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Pi-Hsun Shih authored
Change the signature of mtk_regmap_update_bits to also take a shift, and warn when reg >= 0 but shift < 0. This reduce the code repetition on the calling side, and prevent future UBSAN warning when some of the xxx_shift and xxx_reg are both set to -1. Signed-off-by:
Pi-Hsun Shih <pihsun@chromium.org> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Pi-Hsun Shih authored
In sound/soc/mediatek/common/mtk-afe-fe-dai.c, when xxx_reg is -1, it's a no-op to call mtk_regmap_update_bits, but since both xxx_reg and xxx_shift are set to -1, the (1 << xxx_shift) in the argument would trigger a UBSAN warning. Fix the warning by setting those xxx_shift to 0 instead. Note that since the code explicitly checks .mono_shift >= 0 and .fs_shift >= 0 before using them in '<<' operator, those two members are not set to 0. Signed-off-by:
Pi-Hsun Shih <pihsun@chromium.org> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
In preparation for the addition of more types of DSP core refactor the handling of DSP specific operations such as starting the memory or enabling the core into a set of callbacks. This should make it easier to add new core types and allow for more code reuse between them. Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
There is no need to duplicate this code for both ADSP1 and 2 as the handling is exactly the same. Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
In preparation for further additions refactor the reading of the firmware status. Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Richard Fitzgerald authored
The original wm_adsp2_early_event took an additional frequency argument for clocking control so could not be used directly as a DAPM callback. But this setup could equally be done by the codec driver function wrapping wm_adsp2_early event. In preparation for adding support for new core types wm_adsp2_set_dspclk has been exported, and the freq argument removed so that it can be used directly as a DAPM callback. Signed-off-by:
Richard Fitzgerald <rf@opensource.cirrus.com> Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
This function is not presently called from outside the adsp code and nor should it be, as such stop exporting it. Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Mark Brown authored
-
Charles Keepax authored
If a watchdog timeout is received from the DSP it is safe to assume the DSP is not functioning anymore and as such any active compressed streams should be put into an error state. Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
Best to lock across handling the bus error to ensure the DSP doesn't change power state as we are reading the status registers. Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
During recent logging improvements it seems two error messages lost their updates during patch application/rebasing. Add these back in. Fixes: 0d3fba3e ("ASoC: wm_adsp: Improve logging messages") Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Charles Keepax authored
Previously support was added to allow streams to be stopped and started again without the DSP being power cycled and this was done by clearing the buffer state in trigger start. Another supported use-case is using the DSP for a trigger event then opening the compressed stream later to receive the audio, unfortunately clearing the buffer state in trigger start destroys the data received from such a trigger. Correct this issue by moving the call to wm_adsp_buffer_clear to be in trigger stop instead. Fixes: 61fc060c ("ASoC: wm_adsp: Support streams which can start/stop with DSP active") Signed-off-by:
Charles Keepax <ckeepax@opensource.cirrus.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Bard liao authored
Some strings are allocated by kstrdup, but not freed when error happened. Signed-off-by:
Bard liao <yung-chuan.liao@linux.intel.com> Acked-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Bard liao authored
The stream_name is allocated by kstrdup. We have to free it when the dai is removed or return from error. Signed-off-by:
Bard liao <yung-chuan.liao@linux.intel.com> Acked-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
- Mar 18, 2019
-
-
Mark Brown authored
Merge branch 'for-5.1' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into asoc-5.2
-
Kuninori Morimoto authored
It is parsing mclk_fs at many places, but it should be same operation. This patch adds graph_parse_mclk_fs() and parse it. This patch also renames similar function graph_get_conversion() to graph_parse_convert(). Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Kuninori Morimoto authored
use same naming rule, and this patch add missing of_node_put() on it Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Kuninori Morimoto authored
It is parsing mclk_fs at many places, but it should be same operation. This patch adds simple_parse_mclk_fs() and parse it. Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Kuninori Morimoto authored
use same naming rule Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Shuming Fan authored
If ASRC turns on, HW will use clk_dac as the reference clock whether recording or playback. Both of clk_dac and clk_adc should set proper clock while using ASRC. Signed-off-by:
Shuming Fan <shumingf@realtek.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Shuming Fan authored
The jack type detection needs the main bias power of analog. The modification makes sure the main bias power on/off while jack plug/unplug. Signed-off-by:
Shuming Fan <shumingf@realtek.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Shuming Fan authored
The IRQ function may not work when system suspend. We remove snd_soc_dapm_force_enable_pin function call to make sure the bias off when idle and run into suspend/resume function. Signed-off-by:
Shuming Fan <shumingf@realtek.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Tzung-Bi Shih authored
Skip for i2s5 in mck_disable which is also bypassed in mck_enable. Signed-off-by:
Tzung-Bi Shih <tzungbi@google.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Hans de Goede authored
Add BYT_RT5651_JD_NOT_INV quirk for devices with an inverted (active-high instead of the normal active-low) jack-detect switch. And add a quirk for the Complet Electro Serv MY8307 tablet which has an inverted jack-detect switch (and a mono-speaker). Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Hans de Goede authored
Some boards use a jack-receptacle with a switch which reports the jack-inserted status as active-high, rather then the standard active-low reporting most jacks use. This commit adds support for it. This is activated by a boolean "realtek,jack-detect-not-inverted" device-property. The not-inverted in the device-property name, rather then active-high, was chosen to keep the device-property naming consistent with the rt5640 codec driver. Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Olivier Moysan authored
Add suspend and resume sleep callbacks to STM32 SPDIFRX driver, to support system low power modes. Signed-off-by:
Olivier Moysan <olivier.moysan@st.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
KaiChieh Chuang authored
The dpcm get from fe_clients/be_clients may be free before use Add a spin lock at snd_soc_card level, to protect the dpcm instance. The lock may be used in atomic context, so use spin lock. possible race condition between void dpcm_be_disconnect( ... list_del(&dpcm->list_be); list_del(&dpcm->list_fe); kfree(dpcm); ... and for_each_dpcm_fe() for_each_dpcm_be*() race condition example Thread 1: snd_soc_dapm_mixer_update_power() -> soc_dpcm_runtime_update() -> dpcm_be_disconnect() -> kfree(dpcm); Thread 2: dpcm_fe_dai_trigger() -> dpcm_be_dai_trigger() -> snd_soc_dpcm_can_be_free_stop() -> if (dpcm->fe == fe) Excpetion Scenario: two FE link to same BE FE1 -> BE FE2 -> Thread 1: switch of mixer between FE2 -> BE Thread 2: pcm_stop FE1 Exception: Unable to handle kernel paging request at virtual address dead0000000000e0 pc=<> [<ffffff8960e2cd10>] dpcm_be_dai_trigger+0x29c/0x47c sound/soc/soc-pcm.c:3226 if (dpcm->fe == fe) lr=<> [<ffffff8960e2f694>] dpcm_fe_dai_do_trigger+0x94/0x26c Backtrace: [<ffffff89602dba80>] notify_die+0x68/0xb8 [<ffffff896028c7dc>] die+0x118/0x2a8 [<ffffff89602a2f84>] __do_kernel_fault+0x13c/0x14c [<ffffff89602a27f4>] do_translation_fault+0x64/0xa0 [<ffffff8960280cf8>] do_mem_abort+0x4c/0xd0 [<ffffff8960282ad0>] el1_da+0x24/0x40 [<ffffff8960e2cd10>] dpcm_be_dai_trigger+0x29c/0x47c [<ffffff8960e2f694>] dpcm_fe_dai_do_trigger+0x94/0x26c [<ffffff8960e2edec>] dpcm_fe_dai_trigger+0x3c/0x44 [<ffffff8960de5588>] snd_pcm_do_stop+0x50/0x5c [<ffffff8960dded24>] snd_pcm_action+0xb4/0x13c [<ffffff8960ddfdb4>] snd_pcm_drop+0xa0/0x128 [<ffffff8960de69bc>] snd_pcm_common_ioctl+0x9d8/0x30f0 [<ffffff8960de1cac>] snd_pcm_ioctl_compat+0x29c/0x2f14 [<ffffff89604c9d60>] compat_SyS_ioctl+0x128/0x244 [<ffffff8960283740>] el0_svc_naked+0x34/0x38 [<ffffffffffffffff>] 0xffffffffffffffff Signed-off-by:
KaiChieh Chuang <kaichieh.chuang@mediatek.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Mylène Josserand authored
Some amplifier may not have a GPIO to control the power, but instead simply rely on the regulator to power up and down the amplifier. In order to support those setups, let's make the GPIO optional. Signed-off-by:
Mylène Josserand <mylene.josserand@bootlin.com> Signed-off-by:
Maxime Ripard <maxime.ripard@bootlin.com> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Kuninori Morimoto authored
If SoC has extended Audio DMAC peri peri address, DT needs to select it. Otherwise, it can select normal Audio DMAC peri peri address. Extended Audio DMAC peri peri address can use BUSIF4-7, normal address can't. Reported-by:
Jiada Wang <jiada_wang@mentor.com> Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Reviewed-by:
Geert Uytterhoeven <geert+renesas@glider.be> Reviewed-by:
Simon Horman <horms+renesas@verge.net.au> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Kuninori Morimoto authored
reg is missing "Audio DMAC peri peri" part, and reg-name itself is missing at Document. This patch add these. Reported-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Reviewed-by:
Simon Horman <horms+renesas@verge.net.au> Signed-off-by:
Mark Brown <broonie@kernel.org>
-
Mark Brown authored
Linux 5.1-rc1
-
Mark Brown authored
Linux 5.1-rc1
-
Linus Torvalds authored
-