February 2020 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] drm/mediatek: fix race condition for HDMI jack status reporting
by Tzung-Bi Shih 17 Feb '20

17 Feb '20

hdmi_conn_detect and mtk_hdmi_audio_hook_plugged_cb would be called by different threads. Imaging the following calling sequence: Thread A Thread B -------------------------------------------------------------------- mtk_hdmi_audio_hook_plugged_cb() mtk_cec_hpd_high() -> disconnected hdmi_conn_detect() mtk_cec_hpd_high() -> connected plugged_cb(connected) plugged_cb(disconnected) The latest disconnected is false reported. Makes mtk_cec_hpd_high and plugged_cb atomic to fix. plugged_cb and codec_dev are also in danger of race condition. Instead of using mutex to protect them: - Checks NULLs first. - Uses WRITE_ONCE() to prevent store tearing (i.e. write to plugged_cb after codec_dev). - Uses codec_dev as a signal to report HDMI jack status. Fixes: 5d3c64477392 ("drm/mediatek: support HDMI jack status reporting") Signed-off-by: Tzung-Bi Shih <tzungbi(a)google.com> --- Previous discussion: https://patchwork.kernel.org/patch/11367625/ Previous attempt: https://patchwork.kernel.org/patch/11378413/ drivers/gpu/drm/mediatek/mtk_hdmi.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.c b/drivers/gpu/drm/mediatek/mtk_hdmi.c index 03aeb73005ef..b1e5d0c538fa 100644 --- a/drivers/gpu/drm/mediatek/mtk_hdmi.c +++ b/drivers/gpu/drm/mediatek/mtk_hdmi.c @@ -12,6 +12,7 @@ #include <linux/io.h> #include <linux/kernel.h> #include <linux/mfd/syscon.h> +#include <linux/mutex.h> #include <linux/of_platform.h> #include <linux/of.h> #include <linux/of_gpio.h> @@ -171,6 +172,7 @@ struct mtk_hdmi { bool enabled; hdmi_codec_plugged_cb plugged_cb; struct device *codec_dev; + struct mutex update_plugged_status_lock; }; static inline struct mtk_hdmi *hdmi_ctx_from_bridge(struct drm_bridge *b) @@ -1199,10 +1201,13 @@ static void mtk_hdmi_clk_disable_audio(struct mtk_hdmi *hdmi) static enum drm_connector_status mtk_hdmi_update_plugged_status(struct mtk_hdmi *hdmi) { - bool connected = mtk_cec_hpd_high(hdmi->cec_dev); + bool connected; - if (hdmi->plugged_cb && hdmi->codec_dev) + mutex_lock(&hdmi->update_plugged_status_lock); + connected = mtk_cec_hpd_high(hdmi->cec_dev); + if (hdmi->codec_dev) hdmi->plugged_cb(hdmi->codec_dev, connected); + mutex_unlock(&hdmi->update_plugged_status_lock); return connected ? connector_status_connected : connector_status_disconnected; @@ -1669,8 +1674,12 @@ static int mtk_hdmi_audio_hook_plugged_cb(struct device *dev, void *data, { struct mtk_hdmi *hdmi = data; - hdmi->plugged_cb = fn; - hdmi->codec_dev = codec_dev; + if (!fn || !codec_dev) + return -EINVAL; + + /* Use WRITE_ONCE() to prevent store tearing. */ + WRITE_ONCE(hdmi->plugged_cb, fn); + WRITE_ONCE(hdmi->codec_dev, codec_dev); mtk_hdmi_update_plugged_status(hdmi); return 0; @@ -1729,6 +1738,7 @@ static int mtk_drm_hdmi_probe(struct platform_device *pdev) return ret; } + mutex_init(&hdmi->update_plugged_status_lock); platform_set_drvdata(pdev, hdmi); ret = mtk_hdmi_output_init(hdmi); -- 2.25.0.225.g125e21ebc7-goog

2 5

[alsa-devel] [PATCH v2 00/10] ASoC: soc-pcm cleanup step3
by Kuninori Morimoto 17 Feb '20

17 Feb '20

Hi Mark These are v2 of soc-pcm cleanup step3. I removed unneeded Multi-Codec patch. And re-ordered / merged my local tree and add [09/10] patch to make [10/10] more effective. Kuninori Morimoto (10): 1) ASoC: soc-pcm: move dai_get_widget() 2) ASoC: soc-pcm: use dai_get_widget() at dpcm_get_be() 3) ASoC: soc-pcm: use dai_get_widget() at dpcm_end_walk_at_be() 4) ASoC: soc-pcm: use dpcm_get_be() at dpcm_end_walk_at_be() 5) ASoC: soc-pcm: remove soc_dpcm_be_digital_mute() 6) ASoC: soc-pcm: remove snd_soc_dpcm_be_get/set_state() 7) ASoC: soc-pcm: add snd_soc_dpcm_can_be() and remove duplicate code 8) ASoC: soc-pcm: use goto and remove multi return 9) ASoC: soc-pcm: merge playback/cature_active into stream_active 10) ASoC: soc.h: add for_each_pcm_streams() include/sound/pcm.h | 5 + include/sound/soc-dai.h | 3 +- include/sound/soc-dpcm.h | 9 - sound/soc/codecs/cs4271.c | 4 +- sound/soc/dwc/dwc-i2s.c | 8 +- sound/soc/fsl/fsl_asrc_dma.c | 4 +- sound/soc/qcom/lpass-platform.c | 2 +- sound/soc/soc-core.c | 48 ++-- sound/soc/soc-generic-dmaengine-pcm.c | 8 +- sound/soc/soc-pcm.c | 327 +++++++++----------------- sound/soc/sof/sof-audio.c | 19 +- sound/usb/usx2y/usbusx2yaudio.c | 9 +- 12 files changed, 162 insertions(+), 284 deletions(-) -- 2.17.1

3 13

[alsa-devel] [PATCH v2 0/3] simple-audio-card codec2codec support
by Samuel Holland 17 Feb '20

17 Feb '20

We are currently using simple-audio-card on the Allwinner A64 SoC. The digital audio codec there (sun8i-codec) has 3 AIFs, one each for the CPU, the modem, and Bluetooth. Adding support for the secondary AIFs requires adding codec2codec DAI links. Since the modem and bt-sco codec DAI drivers only have one set of possible PCM parameters (namely, 8kHz mono S16LE), there's no real need for a machine driver to specify the DAI link configuration. The parameters for these "simple" DAI links can be chosen automatically. This series adds codec2codec DAI link support to simple-audio-card. Codec to codec links are automatically detected when DAIs in the link belong to codec components. I tried to reuse as much code as possible, so the first two patches refactor a couple of helper functions to be more generic. The last patch adds the new feature and its documentation. Samuel Holland (3): ALSA: pcm: Add a non-runtime version of snd_pcm_limit_hw_rates ASoC: pcm: Export parameter intersection logic ASoC: simple-card: Add support for codec to codec DAI links Documentation/sound/soc/codec-to-codec.rst | 9 +++- include/sound/pcm.h | 9 +++- include/sound/soc.h | 3 ++ sound/core/pcm_misc.c | 18 +++---- sound/soc/generic/simple-card-utils.c | 50 ++++++++++++++++++++ sound/soc/soc-pcm.c | 55 +++++++++++++++------- 6 files changed, 116 insertions(+), 28 deletions(-) -- 2.24.1

1 3

Re: [alsa-devel] Duet Firewire
by Takashi Sakamoto 16 Feb '20

16 Feb '20

Hi, (C.C.ed to alsa-devel) On Sun, Feb 16, 2020 at 03:52:27AM +0000, roberto lopez torres wrote: > Hi, i just got a Duet Firewire, and i'm trying to make it work with 5.3.0-29-gen > eric #31-Ubuntu > > I have inputs and outputs from the interface but can't config anything > at the moment, and i got a noise every 16 seconds on headphones > monitoring. > > I'm pretty new using linux. The ALSA support for Apogee Duet FireWire starts at Linux kernel v4.21. As you can see, the support is just for packet streaming, thus it's just possible to transmit/receive PCM frames and MIDI messages. However, the device is designed to run just depending on its internal clock (e.g. 47994 frames/sec), thus at present it generates periodical noise since the ALSA driver is designed to transmit just the same number of PCM frames as sampling rate (e.g. 48000 frames/sec). The support of any control is expected to be implemented in userspace application. This is the reason that you can see no elements in ALSA control application (e.g. alsamixer). However, the protocol to control the device is not clear yet. As long as I know, it's complicated and requires the batch of codes for the application. This is the reason that ALSA in-kernel driver doesn't have the code. Regards Takashi Sakamoto

1 0

[alsa-devel] Best way to interface with equalizer
by Manusha Wijekoon 16 Feb '20

16 Feb '20

Hello I want to control ALSA equalizer from a mobile phone GUI. Essentially I would like to build a small GUI that runs on a phone to allow a user to change equalizer visually. I want this to be very smooth and stable. Few options that I can think of are: 1. Have an agent program running in the linux machine which controls alsa by sending various commands through linux shell. 2. Agent program running in the linux machine directly interfaces to ALSA through its C interface. 3. Is there a control interface provided by ALSA, for example REST like service? 3 seem to be the best option as it involves less work. 1 looks fragile to me. What are your recommendations? Thank you S

1 0

[alsa-devel] [PATCH v3] ALSA: usb-audio: Add support for Presonus Studio 1810c
by mickflemm＠gmail.com 15 Feb '20

15 Feb '20

From: Nick Kossifidis <mickflemm(a)gmail.com> This patch adds support for Presonus Studio 1810c, a usb interface that's UAC2 compliant with a few quirks and a few extra hw-specific controls. I've tested all 3 altsettings and the added switch controls and they work as expected. More infos on the card: https://www.presonus.com/products/Studio-1810c Note that this work is based on packet inspection with usbmon. I just wanted to get this card to work for using it on our open-source radio station: https://github.com/UoC-Radio v2 address issues reported by Takashi: * Properly get/set enum type controls * Prevent race condition on switch_get/set * Various control naming changes * Various coding style fixes v3 improve readability of sample rate filtering and some other minor changes. Signed-off-by: Nick Kossifidis <mickflemm(a)gmail.com> --- sound/usb/Makefile | 1 + sound/usb/format.c | 37 +++ sound/usb/mixer_quirks.c | 5 + sound/usb/mixer_s1810c.c | 595 +++++++++++++++++++++++++++++++++++++++ sound/usb/mixer_s1810c.h | 7 + sound/usb/quirks.c | 36 +++ 6 files changed, 681 insertions(+) create mode 100644 sound/usb/mixer_s1810c.c create mode 100644 sound/usb/mixer_s1810c.h diff --git a/sound/usb/Makefile b/sound/usb/Makefile index 78edd7d2f..56031026b 100644 --- a/sound/usb/Makefile +++ b/sound/usb/Makefile @@ -13,6 +13,7 @@ snd-usb-audio-objs := card.o \ mixer_scarlett.o \ mixer_scarlett_gen2.o \ mixer_us16x08.o \ + mixer_s1810c.o \ pcm.o \ power.o \ proc.o \ diff --git a/sound/usb/format.c b/sound/usb/format.c index d79db7130..462187953 100644 --- a/sound/usb/format.c +++ b/sound/usb/format.c @@ -226,6 +226,36 @@ static int parse_audio_format_rates_v1(struct snd_usb_audio *chip, struct audiof return 0; } + +/* + * Presonus Studio 1810c supports a limited set of sampling + * rates per altsetting but reports the full set each time. + * If we don't filter out the unsupported rates and attempt + * to configure the card, it will hang refusing to do any + * further audio I/O until a hard reset is performed. + * + * The list of supported rates per altsetting (set of available + * I/O channels) is described in the owner's manual, section 2.2. + */ +static bool s1810c_valid_sample_rate(struct audioformat *fp, + unsigned int rate) +{ + switch (fp->altsetting) { + case 1: + /* All ADAT ports available */ + return rate <= 48000; + case 2: + /* Half of ADAT ports available */ + return (rate == 88200 || rate == 96000); + case 3: + /* Analog I/O only (no S/PDIF nor ADAT) */ + return rate >= 176400; + default: + return false; + } + return false; +} + /* * Helper function to walk the array of sample rate triplets reported by * the device. The problem is that we need to parse whole array first to @@ -262,6 +292,12 @@ static int parse_uac2_sample_rate_range(struct snd_usb_audio *chip, } for (rate = min; rate <= max; rate += res) { + + /* Filter out invalid rates on Presonus Studio 1810c */ + if (chip->usb_id == USB_ID(0x0194f, 0x010c) && + !s1810c_valid_sample_rate(fp, rate)) + goto skip_rate; + if (fp->rate_table) fp->rate_table[nr_rates] = rate; if (!fp->rate_min || rate < fp->rate_min) @@ -276,6 +312,7 @@ static int parse_uac2_sample_rate_range(struct snd_usb_audio *chip, break; } +skip_rate: /* avoid endless loop */ if (res == 0) break; diff --git a/sound/usb/mixer_quirks.c b/sound/usb/mixer_quirks.c index 39e27ae6c..71e705fab 100644 --- a/sound/usb/mixer_quirks.c +++ b/sound/usb/mixer_quirks.c @@ -34,6 +34,7 @@ #include "mixer_scarlett.h" #include "mixer_scarlett_gen2.h" #include "mixer_us16x08.h" +#include "mixer_s1810c.h" #include "helper.h" struct std_mono_table { @@ -2277,6 +2278,10 @@ int snd_usb_mixer_apply_create_quirk(struct usb_mixer_interface *mixer) case USB_ID(0x2a39, 0x3fd4): /* RME */ err = snd_rme_controls_create(mixer); break; + + case USB_ID(0x0194f, 0x010c): /* Presonus Studio 1810c */ + err = snd_sc1810_init_mixer(mixer); + break; } return err; diff --git a/sound/usb/mixer_s1810c.c b/sound/usb/mixer_s1810c.c new file mode 100644 index 000000000..8e2a05f80 --- /dev/null +++ b/sound/usb/mixer_s1810c.c @@ -0,0 +1,595 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Presonus Studio 1810c driver for ALSA + * Copyright (C) 2019 Nick Kossifidis <mickflemm(a)gmail.com> + * + * Based on reverse engineering of the communication protocol + * between the windows driver / Univeral Control (UC) program + * and the device, through usbmon. + * + * For now this bypasses the mixer, with all channels split, + * so that the software can mix with greater flexibility. + * It also adds controls for the 4 buttons on the front of + * the device. + */ + +#include <linux/usb.h> +#include <linux/usb/audio-v2.h> +#include <linux/slab.h> +#include <sound/core.h> +#include <sound/control.h> + +#include "usbaudio.h" +#include "mixer.h" +#include "mixer_quirks.h" +#include "helper.h" +#include "mixer_s1810c.h" + +#define SC1810C_CMD_REQ 160 +#define SC1810C_CMD_REQTYPE \ + (USB_TYPE_VENDOR | USB_RECIP_DEVICE | USB_DIR_OUT) +#define SC1810C_CMD_F1 0x50617269 +#define SC1810C_CMD_F2 0x14 + +/* + * DISCLAIMER: These are just guesses based on the + * dumps I got. + * + * It seems like a selects between + * device (0), mixer (0x64) and output (0x65) + * + * For mixer (0x64): + * * b selects an input channel (see below). + * * c selects an output channel pair (see below). + * * d selects left (0) or right (1) of that pair. + * * e 0-> disconnect, 0x01000000-> connect, + * 0x0109-> used for stereo-linking channels, + * e is also used for setting volume levels + * in which case b is also set so I guess + * this way it is possible to set the volume + * level from the specified input to the + * specified output. + * + * IN Channels: + * 0 - 7 Mic/Inst/Line (Analog inputs) + * 8 - 9 S/PDIF + * 10 - 17 ADAT + * 18 - 35 DAW (Inputs from the host) + * + * OUT Channels (pairs): + * 0 -> Main out + * 1 -> Line1/2 + * 2 -> Line3/4 + * 3 -> S/PDIF + * 4 -> ADAT? + * + * For device (0): + * * b and c are not used, at least not on the + * dumps I got. + * * d sets the control id to be modified + * (see below). + * * e sets the setting for that control. + * (so for the switches I was interested + * in it's 0/1) + * + * For output (0x65): + * * b is the output channel (see above). + * * c is zero. + * * e I guess the same as with mixer except 0x0109 + * which I didn't see in my dumps. + * + * The two fixed fields have the same values for + * mixer and output but a different set for device. + */ +struct s1810c_ctl_packet { + u32 a; + u32 b; + u32 fixed1; + u32 fixed2; + u32 c; + u32 d; + u32 e; +}; + +#define SC1810C_CTL_LINE_SW 0 +#define SC1810C_CTL_MUTE_SW 1 +#define SC1810C_CTL_AB_SW 3 +#define SC1810C_CTL_48V_SW 4 + +#define SC1810C_SET_STATE_REQ 161 +#define SC1810C_SET_STATE_REQTYPE SC1810C_CMD_REQTYPE +#define SC1810C_SET_STATE_F1 0x64656D73 +#define SC1810C_SET_STATE_F2 0xF4 + +#define SC1810C_GET_STATE_REQ 162 +#define SC1810C_GET_STATE_REQTYPE \ + (USB_TYPE_VENDOR | USB_RECIP_DEVICE | USB_DIR_IN) +#define SC1810C_GET_STATE_F1 SC1810C_SET_STATE_F1 +#define SC1810C_GET_STATE_F2 SC1810C_SET_STATE_F2 + +#define SC1810C_STATE_F1_IDX 2 +#define SC1810C_STATE_F2_IDX 3 + +/* + * This packet includes mixer volumes and + * various other fields, it's an extended + * version of ctl_packet, with a and b + * being zero and different f1/f2. + */ +struct s1810c_state_packet { + u32 fields[63]; +}; + +#define SC1810C_STATE_48V_SW 58 +#define SC1810C_STATE_LINE_SW 59 +#define SC1810C_STATE_MUTE_SW 60 +#define SC1810C_STATE_AB_SW 62 + +struct s1810_mixer_state { + uint16_t seqnum; + struct mutex usb_mutex; + struct mutex data_mutex; +}; + +static int +snd_s1810c_send_ctl_packet(struct usb_device *dev, u32 a, + u32 b, u32 c, u32 d, u32 e) +{ + struct s1810c_ctl_packet pkt = { 0 }; + int ret = 0; + + pkt.fixed1 = SC1810C_CMD_F1; + pkt.fixed2 = SC1810C_CMD_F2; + + pkt.a = a; + pkt.b = b; + pkt.c = c; + pkt.d = d; + /* + * Value for settings 0/1 for this + * output channel is always 0 (probably because + * there is no ADAT output on 1810c) + */ + pkt.e = (c == 4) ? 0 : e; + + ret = snd_usb_ctl_msg(dev, usb_sndctrlpipe(dev, 0), + SC1810C_CMD_REQ, + SC1810C_CMD_REQTYPE, 0, 0, &pkt, sizeof(pkt)); + if (ret < 0) { + dev_warn(&dev->dev, "could not send ctl packet\n"); + return ret; + } + return 0; +} + +/* + * When opening Universal Control the program periodicaly + * sends and receives state packets for syncinc state between + * the device and the host. + * + * Note that if we send only the request to get data back we'll + * get an error, we need to first send an empty state packet and + * then ask to receive a filled. Their seqnumbers must also match. + */ +static int +snd_sc1810c_get_status_field(struct usb_device *dev, + u32 *field, int field_idx, uint16_t *seqnum) +{ + struct s1810c_state_packet pkt_out = { 0 }; + struct s1810c_state_packet pkt_in = { 0 }; + int ret = 0; + + pkt_out.fields[SC1810C_STATE_F1_IDX] = SC1810C_SET_STATE_F1; + pkt_out.fields[SC1810C_STATE_F2_IDX] = SC1810C_SET_STATE_F2; + ret = snd_usb_ctl_msg(dev, usb_rcvctrlpipe(dev, 0), + SC1810C_SET_STATE_REQ, + SC1810C_SET_STATE_REQTYPE, + (*seqnum), 0, &pkt_out, sizeof(pkt_out)); + if (ret < 0) { + dev_warn(&dev->dev, "could not send state packet (%d)\n", ret); + return ret; + } + + ret = snd_usb_ctl_msg(dev, usb_rcvctrlpipe(dev, 0), + SC1810C_GET_STATE_REQ, + SC1810C_GET_STATE_REQTYPE, + (*seqnum), 0, &pkt_in, sizeof(pkt_in)); + if (ret < 0) { + dev_warn(&dev->dev, "could not get state field %u (%d)\n", + field_idx, ret); + return ret; + } + + (*field) = pkt_in.fields[field_idx]; + (*seqnum)++; + return 0; +} + +/* + * This is what I got when bypassing the mixer with + * all channels split. I'm not 100% sure of what's going + * on, I could probably clean this up based on my observations + * but I prefer to keep the same behavior as the windows driver. + */ +static int snd_s1810c_init_mixer_maps(struct snd_usb_audio *chip) +{ + u32 a, b, c, e, n, off; + struct usb_device *dev = chip->dev; + + /* Set initial volume levels ? */ + a = 0x64; + e = 0xbc; + for (n = 0; n < 2; n++) { + off = n * 18; + for (b = off, c = 0; b < 18 + off; b++) { + /* This channel to all outputs ? */ + for (c = 0; c <= 8; c++) { + snd_s1810c_send_ctl_packet(dev, a, b, c, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, c, 1, e); + } + /* This channel to main output (again) */ + snd_s1810c_send_ctl_packet(dev, a, b, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, 0, 1, e); + } + /* + * I noticed on UC that DAW channels have different + * initial volumes, so this makes sense. + */ + e = 0xb53bf0; + } + + /* Connect analog outputs ? */ + a = 0x65; + e = 0x01000000; + for (b = 1; b < 3; b++) { + snd_s1810c_send_ctl_packet(dev, a, b, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, 0, 1, e); + } + snd_s1810c_send_ctl_packet(dev, a, 0, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, 0, 0, 1, e); + + /* Set initial volume levels for S/PDIF mappings ? */ + a = 0x64; + e = 0xbc; + c = 3; + for (n = 0; n < 2; n++) { + off = n * 18; + for (b = off; b < 18 + off; b++) { + snd_s1810c_send_ctl_packet(dev, a, b, c, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, c, 1, e); + } + e = 0xb53bf0; + } + + /* Connect S/PDIF output ? */ + a = 0x65; + e = 0x01000000; + snd_s1810c_send_ctl_packet(dev, a, 3, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, 3, 0, 1, e); + + /* Connect all outputs (again) ? */ + a = 0x65; + e = 0x01000000; + for (b = 0; b < 4; b++) { + snd_s1810c_send_ctl_packet(dev, a, b, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, 0, 1, e); + } + + /* Basic routing to get sound out of the device */ + a = 0x64; + e = 0x01000000; + for (c = 0; c < 4; c++) { + for (b = 0; b < 36; b++) { + if ((c == 0 && b == 18) || /* DAW1/2 -> Main */ + (c == 1 && b == 20) || /* DAW3/4 -> Line3/4 */ + (c == 2 && b == 22) || /* DAW4/5 -> Line5/6 */ + (c == 3 && b == 24)) { /* DAW5/6 -> S/PDIF */ + /* Left */ + snd_s1810c_send_ctl_packet(dev, a, b, c, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, c, 1, 0); + b++; + /* Right */ + snd_s1810c_send_ctl_packet(dev, a, b, c, 0, 0); + snd_s1810c_send_ctl_packet(dev, a, b, c, 1, e); + } else { + /* Leave the rest disconnected */ + snd_s1810c_send_ctl_packet(dev, a, b, c, 0, 0); + snd_s1810c_send_ctl_packet(dev, a, b, c, 1, 0); + } + } + } + + /* Set initial volume levels for S/PDIF (again) ? */ + a = 0x64; + e = 0xbc; + c = 3; + for (n = 0; n < 2; n++) { + off = n * 18; + for (b = off; b < 18 + off; b++) { + snd_s1810c_send_ctl_packet(dev, a, b, c, 0, e); + snd_s1810c_send_ctl_packet(dev, a, b, c, 1, e); + } + e = 0xb53bf0; + } + + /* Connect S/PDIF outputs (again) ? */ + a = 0x65; + e = 0x01000000; + snd_s1810c_send_ctl_packet(dev, a, 3, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, 3, 0, 1, e); + + /* Again ? */ + snd_s1810c_send_ctl_packet(dev, a, 3, 0, 0, e); + snd_s1810c_send_ctl_packet(dev, a, 3, 0, 1, e); + + return 0; +} + +/* + * Sync state with the device and retrieve the requested field, + * whose index is specified in (kctl->private_value & 0xFF), + * from the received fields array. + */ +static int +snd_s1810c_get_switch_state(struct usb_mixer_interface *mixer, + struct snd_kcontrol *kctl, u32 *state) +{ + struct snd_usb_audio *chip = mixer->chip; + struct s1810_mixer_state *private = mixer->private_data; + u32 field = 0; + u32 ctl_idx = (u32) (kctl->private_value & 0xFF); + int ret = 0; + + mutex_lock(&private->usb_mutex); + ret = snd_sc1810c_get_status_field(chip->dev, &field, + ctl_idx, &private->seqnum); + if (ret < 0) + goto unlock; + + *state = field; + unlock: + mutex_unlock(&private->usb_mutex); + return ret ? ret : 0; +} + +/* + * Send a control packet to the device for the control id + * specified in (kctl->private_value >> 8) with value + * specified in (kctl->private_value >> 16). + */ +static int +snd_s1810c_set_switch_state(struct usb_mixer_interface *mixer, + struct snd_kcontrol *kctl) +{ + struct snd_usb_audio *chip = mixer->chip; + struct s1810_mixer_state *private = mixer->private_data; + u32 pval = (u32) kctl->private_value; + u32 ctl_id = (pval >> 8) & 0xFF; + u32 ctl_val = (pval >> 16) & 0x1; + int ret = 0; + + mutex_lock(&private->usb_mutex); + ret = snd_s1810c_send_ctl_packet(chip->dev, 0, 0, 0, ctl_id, ctl_val); + mutex_unlock(&private->usb_mutex); + return ret; +} + +/* Generic get/set/init functions for switch controls */ + +static int +snd_s1810c_switch_get(struct snd_kcontrol *kctl, + struct snd_ctl_elem_value *ctl_elem) +{ + struct usb_mixer_elem_list *list = snd_kcontrol_chip(kctl); + struct usb_mixer_interface *mixer = list->mixer; + struct s1810_mixer_state *private = mixer->private_data; + u32 pval = (u32) kctl->private_value; + u32 ctl_idx = pval & 0xFF; + u32 state = 0; + int ret = 0; + + mutex_lock(&private->data_mutex); + ret = snd_s1810c_get_switch_state(mixer, kctl, &state); + if (ret < 0) + goto unlock; + + switch (ctl_idx) { + case SC1810C_STATE_LINE_SW: + case SC1810C_STATE_AB_SW: + ctl_elem->value.enumerated.item[0] = (int)state; + break; + default: + ctl_elem->value.integer.value[0] = (long)state; + } + + unlock: + mutex_unlock(&private->data_mutex); + return (ret < 0) ? ret : 0; +} + +static int +snd_s1810c_switch_set(struct snd_kcontrol *kctl, + struct snd_ctl_elem_value *ctl_elem) +{ + struct usb_mixer_elem_list *list = snd_kcontrol_chip(kctl); + struct usb_mixer_interface *mixer = list->mixer; + struct s1810_mixer_state *private = mixer->private_data; + u32 pval = (u32) kctl->private_value; + u32 ctl_idx = pval & 0xFF; + u32 curval = 0; + u32 newval = 0; + int ret = 0; + + mutex_lock(&private->data_mutex); + ret = snd_s1810c_get_switch_state(mixer, kctl, &curval); + if (ret < 0) + goto unlock; + + switch (ctl_idx) { + case SC1810C_STATE_LINE_SW: + case SC1810C_STATE_AB_SW: + newval = (u32) ctl_elem->value.enumerated.item[0]; + break; + default: + newval = (u32) ctl_elem->value.integer.value[0]; + } + + if (curval == newval) + goto unlock; + + kctl->private_value &= ~(0x1 << 16); + kctl->private_value |= (unsigned int)(newval & 0x1) << 16; + ret = snd_s1810c_set_switch_state(mixer, kctl); + + unlock: + mutex_unlock(&private->data_mutex); + return (ret < 0) ? 0 : 1; +} + +static int +snd_s1810c_switch_init(struct usb_mixer_interface *mixer, + const struct snd_kcontrol_new *new_kctl) +{ + struct snd_kcontrol *kctl; + struct usb_mixer_elem_info *elem; + + elem = kzalloc(sizeof(struct usb_mixer_elem_info), GFP_KERNEL); + if (!elem) + return -ENOMEM; + + elem->head.mixer = mixer; + elem->control = 0; + elem->head.id = 0; + elem->channels = 1; + + kctl = snd_ctl_new1(new_kctl, elem); + if (!kctl) { + kfree(elem); + return -ENOMEM; + } + kctl->private_free = snd_usb_mixer_elem_free; + + return snd_usb_mixer_add_control(&elem->head, kctl); +} + +static int +snd_s1810c_line_sw_info(struct snd_kcontrol *kctl, + struct snd_ctl_elem_info *uinfo) +{ + static const char *const texts[2] = { + "Preamp On (Mic/Inst)", + "Preamp Off (Line in)" + }; + + return snd_ctl_enum_info(uinfo, 1, ARRAY_SIZE(texts), texts); +} + +static const struct snd_kcontrol_new snd_s1810c_line_sw = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Line 1/2 Source Type", + .info = snd_s1810c_line_sw_info, + .get = snd_s1810c_switch_get, + .put = snd_s1810c_switch_set, + .private_value = (SC1810C_STATE_LINE_SW | SC1810C_CTL_LINE_SW << 8) +}; + +static const struct snd_kcontrol_new snd_s1810c_mute_sw = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Mute Main Out Switch", + .info = snd_ctl_boolean_mono_info, + .get = snd_s1810c_switch_get, + .put = snd_s1810c_switch_set, + .private_value = (SC1810C_STATE_MUTE_SW | SC1810C_CTL_MUTE_SW << 8) +}; + +static const struct snd_kcontrol_new snd_s1810c_48v_sw = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "48V Phantom Power On Mic Inputs Switch", + .info = snd_ctl_boolean_mono_info, + .get = snd_s1810c_switch_get, + .put = snd_s1810c_switch_set, + .private_value = (SC1810C_STATE_48V_SW | SC1810C_CTL_48V_SW << 8) +}; + +static int +snd_s1810c_ab_sw_info(struct snd_kcontrol *kctl, + struct snd_ctl_elem_info *uinfo) +{ + static const char *const texts[2] = { + "1/2", + "3/4" + }; + + return snd_ctl_enum_info(uinfo, 1, ARRAY_SIZE(texts), texts); +} + +static const struct snd_kcontrol_new snd_s1810c_ab_sw = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Headphone 1 Source Route", + .info = snd_s1810c_ab_sw_info, + .get = snd_s1810c_switch_get, + .put = snd_s1810c_switch_set, + .private_value = (SC1810C_STATE_AB_SW | SC1810C_CTL_AB_SW << 8) +}; + +static void snd_sc1810_mixer_state_free(struct usb_mixer_interface *mixer) +{ + struct s1810_mixer_state *private = mixer->private_data; + kfree(private); + mixer->private_data = NULL; +} + +/* Entry point, called from mixer_quirks.c */ +int snd_sc1810_init_mixer(struct usb_mixer_interface *mixer) +{ + struct s1810_mixer_state *private = NULL; + struct snd_usb_audio *chip = mixer->chip; + struct usb_device *dev = chip->dev; + int ret = 0; + + /* Run this only once */ + if (!list_empty(&chip->mixer_list)) + return 0; + + dev_info(&dev->dev, + "Presonus Studio 1810c, device_setup: %u\n", chip->setup); + if (chip->setup == 1) + dev_info(&dev->dev, "(8out/18in @ 48KHz)\n"); + else if (chip->setup == 2) + dev_info(&dev->dev, "(6out/8in @ 192KHz)\n"); + else + dev_info(&dev->dev, "(8out/14in @ 96KHz)\n"); + + ret = snd_s1810c_init_mixer_maps(chip); + if (ret < 0) + return ret; + + private = kzalloc(sizeof(struct s1810_mixer_state), GFP_KERNEL); + if (!private) + return -ENOMEM; + + mutex_init(&private->usb_mutex); + mutex_init(&private->data_mutex); + + mixer->private_data = private; + mixer->private_free = snd_sc1810_mixer_state_free; + + private->seqnum = 1; + + ret = snd_s1810c_switch_init(mixer, &snd_s1810c_line_sw); + if (ret < 0) + return ret; + + ret = snd_s1810c_switch_init(mixer, &snd_s1810c_mute_sw); + if (ret < 0) + return ret; + + ret = snd_s1810c_switch_init(mixer, &snd_s1810c_48v_sw); + if (ret < 0) + return ret; + + ret = snd_s1810c_switch_init(mixer, &snd_s1810c_ab_sw); + if (ret < 0) + return ret; + return ret; +} diff --git a/sound/usb/mixer_s1810c.h b/sound/usb/mixer_s1810c.h new file mode 100644 index 000000000..a79a3743c --- /dev/null +++ b/sound/usb/mixer_s1810c.h @@ -0,0 +1,7 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Presonus Studio 1810c driver for ALSA + * Copyright (C) 2019 Nick Kossifidis <mickflemm(a)gmail.com> + */ + +int snd_sc1810_init_mixer(struct usb_mixer_interface *mixer); diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index a81c20664..9bbdea882 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -1227,6 +1227,38 @@ static int fasttrackpro_skip_setting_quirk(struct snd_usb_audio *chip, return 0; /* keep this altsetting */ } +static int s1810c_skip_setting_quirk(struct snd_usb_audio *chip, + int iface, int altno) +{ + /* + * Altno settings: + * + * Playback (Interface 1): + * 1: 6 Analog + 2 S/PDIF + * 2: 6 Analog + 2 S/PDIF + * 3: 6 Analog + * + * Capture (Interface 2): + * 1: 8 Analog + 2 S/PDIF + 8 ADAT + * 2: 8 Analog + 2 S/PDIF + 4 ADAT + * 3: 8 Analog + */ + + /* + * I'll leave 2 as the default one and + * use device_setup to switch to the + * other two. + */ + if ((chip->setup == 0 || chip->setup > 2) && altno != 2) + return 1; + else if (chip->setup == 1 && altno != 1) + return 1; + else if (chip->setup == 2 && altno != 3) + return 1; + + return 0; +} + int snd_usb_apply_interface_quirk(struct snd_usb_audio *chip, int iface, int altno) @@ -1240,6 +1272,10 @@ int snd_usb_apply_interface_quirk(struct snd_usb_audio *chip, /* fasttrackpro usb: skip altsets incompatible with device_setup */ if (chip->usb_id == USB_ID(0x0763, 0x2012)) return fasttrackpro_skip_setting_quirk(chip, iface, altno); + /* presonus studio 1810c: skip altsets incompatible with device_setup */ + if (chip->usb_id == USB_ID(0x0194f, 0x010c)) + return s1810c_skip_setting_quirk(chip, iface, altno); + return 0; } -- 2.23.0

2 1

[alsa-devel] [PATCH 0/4] simple-audio-card codec2codec support
by Samuel Holland 15 Feb '20

15 Feb '20

We are currently using simple-audio-card on the Allwinner A64 SoC. The digital audio codec there (sun8i-codec) has 3 AIFs, one each for the CPU, the modem, and Bluetooth. Adding support for the secondary AIFs requires adding codec2codec DAI links. Since the modem and bt-sco codec DAI drivers only have one set of possible PCM parameters (namely, 8kHz mono S16LE), there's no real need for a machine driver to specify the DAI link configuration. The parameters for these "simple" DAI links can be chosen automatically. This series adds a single "codec-to-codec" property to the simple-audio-card binding, which does exactly what it says. It works out rather nicely without making the device tree binding too complicated. The first patch fixes a bug I found while implementing this feature. I tried to reuse as much code as possible, so the middle two patches refactor a couple of helper functions to be more generic. Finally, the last patch adds the new feature and its documentation. Samuel Holland (4): ASoC: codec2codec: avoid invalid/double-free of pcm runtime ALSA: pcm: Make snd_pcm_limit_hw_rates take hw directly ASoC: pcm: Export parameter intersection logic ASoC: simple-card: Add support for codec-to-codec dai_links .../devicetree/bindings/sound/simple-card.txt | 1 + Documentation/sound/soc/codec-to-codec.rst | 9 ++- .../drm/bridge/synopsys/dw-hdmi-ahb-audio.c | 2 +- include/sound/pcm.h | 2 +- include/sound/simple_card_utils.h | 1 + include/sound/soc.h | 3 + sound/arm/aaci.c | 2 +- sound/arm/pxa2xx-ac97.c | 2 +- sound/core/pcm_misc.c | 14 ++--- sound/firewire/dice/dice-pcm.c | 2 +- sound/firewire/digi00x/digi00x-pcm.c | 2 +- sound/firewire/fireworks/fireworks_pcm.c | 2 +- sound/firewire/motu/motu-pcm.c | 2 +- sound/firewire/tascam/tascam-pcm.c | 2 +- sound/pci/atiixp.c | 2 +- sound/pci/cs5535audio/cs5535audio_pcm.c | 4 +- sound/pci/hda/hda_controller.c | 4 +- sound/pci/intel8x0.c | 2 +- sound/pci/sis7019.c | 2 +- sound/pci/via82xx.c | 4 +- sound/soc/generic/simple-card-utils.c | 39 +++++++++++++ sound/soc/generic/simple-card.c | 12 ++++ sound/soc/soc-dapm.c | 3 - sound/soc/soc-pcm.c | 57 +++++++++++++------ sound/usb/caiaq/audio.c | 4 +- 25 files changed, 130 insertions(+), 49 deletions(-) -- 2.24.1

4 12

[alsa-devel] [RFC PATCH 0/2] soundwire: add master_device/driver support
by Pierre-Louis Bossart 15 Feb '20

15 Feb '20

The SoundWire master representation needs to evolve to take into account: a) a May 2019 recommendation from Greg KH to remove platform devices b) the need on Intel platforms to support hardware startup only once the power rail dependencies are settled. The SoundWire master is not an independent IP at all. c) the need to deal with external wakes handled by the PCI subsystem in specific low-power modes In case it wasn't clear, the SoundWire subsystem is currently unusable with v5.5 on devices hitting the shelves very soon (race conditions, power management, suspend/resume, etc). v5.6 will only provide interface changes and no functional improvement. We've circled on the same concepts since September 2019, and I hope this direction is now aligned with the suggestions from Vinod Koul and that we can target v5.7 as the 'SoundWire just works(tm)' version. This series is provided as an RFC since it depends on patches already for review. Pierre-Louis Bossart (2): soundwire: bus_type: add master_device/driver support soundwire: intel: transition to sdw_master_device/driver support drivers/soundwire/Makefile | 2 +- drivers/soundwire/bus_type.c | 141 +++++++++++++- drivers/soundwire/intel.c | 96 ++++++---- drivers/soundwire/intel.h | 8 +- drivers/soundwire/intel_init.c | 283 ++++++++++++++++++++++------- drivers/soundwire/master.c | 119 ++++++++++++ drivers/soundwire/slave.c | 7 +- include/linux/soundwire/sdw.h | 83 +++++++++ include/linux/soundwire/sdw_type.h | 36 +++- 9 files changed, 662 insertions(+), 113 deletions(-) create mode 100644 drivers/soundwire/master.c -- 2.20.1

2 5

[alsa-devel] [PATCH 0/3] ASoC: max98090: revert invalid fixes
by Tzung-Bi Shih 14 Feb '20

14 Feb '20

A critical side-effect was observed along with the commit 62d5ae4cafb7 ("ASoC: max98090: save and restore SHDN when changing sensitive registers"). When keep playing something, (e.g. sox -n -t s16 -r 48000 -c 2 - synth sine 440 vol 0.1 | \ aplay -D hw:0,0 -f S16_LE -r 48000 -c 2) the recorded sound has chance to break (clipping). (e.g. while [ 1 ]; do arecord -q -f S16_LE -d 2 -r 48000 -c 2 test.raw \ sox -t s16 -r 48000 -c 2 test.raw -n stat \ ... done) The series reverts all related commits. I feel really sorry for the whole mess. Tzung-Bi Shih (3): ASoC: max98090: revert invalid fix for handling SHDN ASoC: dapm: remove snd_soc_dapm_put_enum_double_locked ASoC: core: remove SND_SOC_BYTES_E include/sound/soc-dapm.h | 2 - include/sound/soc.h | 6 - sound/soc/codecs/max98090.c | 434 ++++++++++-------------------------- sound/soc/codecs/max98090.h | 3 +- sound/soc/soc-dapm.c | 54 ++--- 5 files changed, 137 insertions(+), 362 deletions(-) -- 2.25.0.265.gbab2e86ba0-goog

2 6

[alsa-devel] [PATCH 0/5] ASoC: meson: aiu: fixups
by Jerome Brunet 14 Feb '20

14 Feb '20

This patchset fixes some mistakes which slipped through my initial submission. Most of these issues have been reported by coccinelle and sparse. Jerome Brunet (5): ASoC: meson: aiu: remove unused encoder structure ASoC: meson: aiu: fix clk bulk size allocation ASoC: meson: aiu: fix irq registration ASoC: meson: aiu: fix acodec dai input name init ASoC: meson: codec-glue: fix pcm format cast warning sound/soc/meson/aiu-acodec-ctrl.c | 1 - sound/soc/meson/aiu-encoder-i2s.c | 7 ------- sound/soc/meson/aiu.c | 10 +++------- sound/soc/meson/aiu.h | 2 +- sound/soc/meson/meson-codec-glue.c | 2 +- 5 files changed, 5 insertions(+), 17 deletions(-) -- 2.24.1

2 10