November 2018 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH v7 3/3] ALSA: hda: add support for Huawei WMI micmute LED
by Ayman Bagabas 27 Nov '18

27 Nov '18

Some of Huawei laptops come with a LED in the micmute key. This patch enables the use of micmute LED for these devices: 1. Matebook X (19e5:3200), (19e5:3201) 2. Matebook X Pro (19e5:3204) Signed-off-by: Ayman Bagabas <ayman.bagabas(a)gmail.com> --- sound/pci/hda/patch_realtek.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 1326f32f4574..9766fd249bdf 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5776,7 +5776,9 @@ static const struct hda_fixup alc269_fixups[] = { {0x1e, 0x411111f0}, {0x21, 0x04211020}, { } - } + }, + .chained = true, + .chain_id = ALC255_FIXUP_MIC_MUTE_LED }, [ALC269_FIXUP_ASUS_X101_FUNC] = { .type = HDA_FIXUP_FUNC, @@ -6608,6 +6610,8 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x511f, "Thinkpad", ALC298_FIXUP_TPT470_DOCK), SND_PCI_QUIRK(0x17aa, 0x3bf8, "Quanta FL1", ALC269_FIXUP_PCM_44K), SND_PCI_QUIRK(0x17aa, 0x9e54, "LENOVO NB", ALC269_FIXUP_LENOVO_EAPD), + SND_PCI_QUIRK(0x19e5, 0x3200, "Huawei MBX", ALC255_FIXUP_MIC_MUTE_LED), + SND_PCI_QUIRK(0x19e5, 0x3201, "Huawei MBX", ALC255_FIXUP_MIC_MUTE_LED), SND_PCI_QUIRK(0x19e5, 0x3204, "Huawei MBXP", ALC256_FIXUP_HUAWEI_MBXP_PINS), SND_PCI_QUIRK(0x1b7d, 0xa831, "Ordissimo EVE2 ", ALC269VB_FIXUP_ORDISSIMO_EVE2), /* Also known as Malata PC-B1303 */ -- 2.19.1

1 0

[alsa-devel] [PATCH v7 1/3] ALSA: hda: fix front speakers on Huawei MBXP.
by Ayman Bagabas 27 Nov '18

27 Nov '18

This patch solves bug 200501 'Only 2 of 4 speakers playing sound.' https://bugzilla.kernel.org/show_bug.cgi?id=200501 It enables the front speakers on Huawei Matebook X Pro laptops. These laptops come with Dolby Atmos sound system and these pins configuration enables the front speakers. Signed-off-by: Ayman Bagabas <ayman.bagabas(a)gmail.com> Reviewed-by: Takashi Iwai <tiwai(a)suse.de> --- sound/pci/hda/patch_realtek.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 993d34c141c2..1326f32f4574 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5490,6 +5490,7 @@ enum { ALC298_FIXUP_TPT470_DOCK, ALC255_FIXUP_DUMMY_LINEOUT_VERB, ALC255_FIXUP_DELL_HEADSET_MIC, + ALC256_FIXUP_HUAWEI_MBXP_PINS, ALC295_FIXUP_HP_X360, ALC221_FIXUP_HP_HEADSET_MIC, }; @@ -5761,6 +5762,22 @@ static const struct hda_fixup alc269_fixups[] = { .chained = true, .chain_id = ALC269_FIXUP_HEADSET_MIC }, + [ALC256_FIXUP_HUAWEI_MBXP_PINS] = { + .type = HDA_FIXUP_PINS, + .v.pins = (const struct hda_pintbl[]) { + {0x12, 0x90a60130}, + {0x13, 0x40000000}, + {0x14, 0x90170110}, + {0x18, 0x411111f0}, + {0x19, 0x04a11040}, + {0x1a, 0x411111f0}, + {0x1b, 0x90170112}, + {0x1d, 0x40759a05}, + {0x1e, 0x411111f0}, + {0x21, 0x04211020}, + { } + } + }, [ALC269_FIXUP_ASUS_X101_FUNC] = { .type = HDA_FIXUP_FUNC, .v.func = alc269_fixup_x101_headset_mic, @@ -6591,6 +6608,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x511f, "Thinkpad", ALC298_FIXUP_TPT470_DOCK), SND_PCI_QUIRK(0x17aa, 0x3bf8, "Quanta FL1", ALC269_FIXUP_PCM_44K), SND_PCI_QUIRK(0x17aa, 0x9e54, "LENOVO NB", ALC269_FIXUP_LENOVO_EAPD), + SND_PCI_QUIRK(0x19e5, 0x3204, "Huawei MBXP", ALC256_FIXUP_HUAWEI_MBXP_PINS), SND_PCI_QUIRK(0x1b7d, 0xa831, "Ordissimo EVE2 ", ALC269VB_FIXUP_ORDISSIMO_EVE2), /* Also known as Malata PC-B1303 */ #if 0 -- 2.19.1

1 0

[alsa-devel] [PATCH v9 0/8] Add support for audiocodec in Allwinner A64
by Chen-Yu Tsai 26 Nov '18

26 Nov '18

Hi everyone, This is v9 of the A64 audio codec series, originally started by Vasily. These are the last changes to make the A64 audio codec work correctly, and be accurately represented in the device tree and in software. Notably, the new driver changes cover regulator supplies for the simple audio amplifier, which is used in the Pinebook for the internal speakers, and the audio codec's internal headphone amp. There's also an extra patch that enables SDM fractional clock support for the audio PLL on the A64, which gives us accurate audio playback. Patch 1 & 2 add the cpvdd regulator supply to the sun50i-codec-analog dt bindings & driver. Patch 3 & 4 add a regulator supply to the simple-audio-amplifier dt bindings & driver. Patch 5 adds SDM fractional clock support for the audio PLL on the A64. Patch 6 ~ 8 add the device tree bits and enables audio on several boards. I have one extra patch for the Bananapi M64 that's not included here, as I'm waiting for an accessory to test the microphone bias. I'll send that one out once it's tested. As always, we'll take the clock and dts patches after Mark has merged the ASoC ones. And a big thanks to Vasily for having worked on this for so long. Regards ChenYu v9: - Dropped simple amplifier headphone amp device tree changes - Dropped "make enable-gpio optional for simple amplifier" - Lowered oversampling rate from 512 to 128 - Add regulator supply for audio codec CPVDD (headphone amp charge pump) - Hook up CPVDD regulator in board device trees - Add patch to enable SDM fractional clock support for audio PLL on A64 v8: - drop merged patches - add regulator property for simple amplifier, make enable-gpio optional - add regulator to speaker amp, introduce headphone amp v7: - drop 'Mic' and 'Headset Mic' from dtsi - move AIF1 Slot 0 ADC to ADC route to sun8i-codec - drop BIAS routes for Pine64 and SoPine v6: - driver patches has been merged through ASoC tree - rebase onto 4.20-rc1 - Drop 'Speaker' from routes on sopine and pine64, they don't have speaker. v5: - collect all the tags v4: - dts: don't use 'Mic' and 'Headset Mic' widgets from sun8i-codec, define our board-level widgets instead. v3: - renamed sunxi-adda-pr-regmap to sun8i-adda-pr-regmap - use ilog2() to calculate reg value for LRCK div instead of using a table v2: - Use simple-amplifier for speaker amp on Pinebook - Rename sun50i-a64-i2s to sun50i-a64-codec-i2s to preserve compatible string for other 3 I2S modules in A64 in case if there's any incompatibility with H3 Chen-Yu Tsai (3): ASoC: dt-bindings: sun50i-codec-analog: Add headphone amp regulator supply ASoC: sunxi: sun50i-codec-analog: Add support for cpvdd regulator supply clk: sunxi-ng: sun50i: a64: Use sigma-delta modulation for audio PLL Vasily Khoruzhick (5): ASoC: dt-bindings: add regulator property to simple amplifier ASoC: simple-amplifier: add VCC regulator widget arm64: dts: allwinner: a64: add nodes necessary for analog sound support arm64: dts: allwinner: a64: enable sound on Pine64 and SoPine arm64: dts: allwinner: a64: enable sound on Pinebook .../bindings/sound/simple-amplifier.txt | 5 ++ .../bindings/sound/sun50i-codec-analog.txt | 2 + .../boot/dts/allwinner/sun50i-a64-pine64.dts | 27 ++++++++++ .../dts/allwinner/sun50i-a64-pinebook.dts | 49 +++++++++++++++++ .../allwinner/sun50i-a64-sopine-baseboard.dts | 26 +++++++++ .../boot/dts/allwinner/sun50i-a64-sopine.dtsi | 4 ++ arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 54 +++++++++++++++++++ drivers/clk/sunxi-ng/ccu-sun50i-a64.c | 37 ++++++++----- sound/soc/codecs/simple-amplifier.c | 4 ++ sound/soc/sunxi/sun50i-codec-analog.c | 2 + 10 files changed, 197 insertions(+), 13 deletions(-) -- 2.19.1

5 20

[alsa-devel] [PATCH] ASoC: amd: Fix a NULL vs IS_ERR() check in probe
by Dan Carpenter 26 Nov '18

26 Nov '18

The platform_device_register_full() function doesn't return NULL, it returns error pointers. Fixes: 7894a7e7ea3d ("ASoC: amd: create ACP3x PCM platform device") Signed-off-by: Dan Carpenter <dan.carpenter(a)oracle.com> --- sound/soc/amd/raven/pci-acp3x.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/amd/raven/pci-acp3x.c b/sound/soc/amd/raven/pci-acp3x.c index c28457fd9b81..facec2472b34 100644 --- a/sound/soc/amd/raven/pci-acp3x.c +++ b/sound/soc/amd/raven/pci-acp3x.c @@ -97,10 +97,10 @@ static int snd_acp3x_probe(struct pci_dev *pci, pdevinfo.size_data = sizeof(irqflags); adata->pdev = platform_device_register_full(&pdevinfo); - if (!adata->pdev) { + if (IS_ERR(adata->pdev)) { dev_err(&pci->dev, "cannot register %s device\n", pdevinfo.name); - ret = -ENODEV; + ret = PTR_ERR(adata->pdev); goto unmap_mmio; } break; -- 2.11.0

2 1

[alsa-devel] [PATCH] ASoC: rsnd: tidyup for SSIU subnode
by Kuninori Morimoto 26 Nov '18

26 Nov '18

From: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> commit da48a6eb82ea2 ("ASoC: rsnd: add SSIU BUSIF support for Document") updated Documentation for SSIU, but 1) we want to keep old/deprecated DMA description, 2) it is missing SSIU subnode properties. This patch tidyup these Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> --- Documentation/devicetree/bindings/sound/renesas,rsnd.txt | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Documentation/devicetree/bindings/sound/renesas,rsnd.txt b/Documentation/devicetree/bindings/sound/renesas,rsnd.txt index 47dc819..4d19d2b 100644 --- a/Documentation/devicetree/bindings/sound/renesas,rsnd.txt +++ b/Documentation/devicetree/bindings/sound/renesas,rsnd.txt @@ -327,8 +327,13 @@ SSI subnode properties: - no-busif : BUSIF is not ussed when [mem -> SSI] via DMA case - dma : Should contain Audio DMAC entry - dma-names : SSI case "rx" (=playback), "tx" (=capture) + Deprecated: see SSIU subnode properties SSIU case "rxu" (=playback), "txu" (=capture) +SSIU subnode properties: +- dma : Should contain Audio DMAC entry +- dma-names : "rx" (=playback), "tx" (=capture) + SRC subnode properties: - dma : Should contain Audio DMAC entry - dma-names : "rx" (=playback), "tx" (=capture) -- 2.7.4

2 1

[alsa-devel] [PATCH] ASoC: soc.h: makes snd_soc_of_parse_audio_prefix() inline
by Kuninori Morimoto 26 Nov '18

26 Nov '18

From: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> commit 3b7103562c03c ("ASoC: soc-core: add snd_soc_of_parse_node_prefix()") maked snd_soc_of_parse_audio_prefix() as #define. But it'd be better to make this a static inline rather than a #define. It helps with error messages and type safety. This patch makes it inline. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> --- include/sound/soc.h | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/include/sound/soc.h b/include/sound/soc.h index 37155f9..8ec1de8 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -1481,9 +1481,15 @@ void snd_soc_of_parse_node_prefix(struct device_node *np, struct snd_soc_codec_conf *codec_conf, struct device_node *of_node, const char *propname); -#define snd_soc_of_parse_audio_prefix(card, conf, node, name) \ - snd_soc_of_parse_node_prefix((card)->dev->of_node, \ - (conf), (node), (name)) +static inline +void snd_soc_of_parse_audio_prefix(struct snd_soc_card *card, + struct snd_soc_codec_conf *codec_conf, + struct device_node *of_node, + const char *propname) +{ + snd_soc_of_parse_node_prefix(card->dev->of_node, + codec_conf, of_node, propname); +} int snd_soc_of_parse_audio_routing(struct snd_soc_card *card, const char *propname); -- 2.7.4

2 1

[alsa-devel] [RFC 0/9] ASoC/ARM: Merge the davinci and omap audio directories
by Peter Ujfalusi 26 Nov '18

26 Nov '18

Hi, This RFC series is on top of the omap-mcbsp cleanup series: http://mailman.alsa-project.org/pipermail/alsa-devel/2018-November/141604.h… For Texas Instruments audio support we have two directories at the moment: davinci omap In reality TI have the following SoC lines supported by Linux: daVinci OMAP Sitara Keystone We already needed to have cross directory 'links' for DRA7x/AM5 since they use McASP (davinci directory) and sDMA (omap directory). Keystone II line of SoCs are using drivers from davinci directory (McASP and EDMA). Which is kind of fine as they are in the same place. With the new AM654 (http://www.ti.com/lit/pdf/spruid7) we will have completely new DMA and there is no place to put that as it does not belong under davinci or omap, but it makes no sense to create yet another TI related directory just for the small udma-pcm. Another issue with the current support is how the Kconfigs are written. The DAI drivers are depending on the ASoC platform drivers and it is a bit of a mess at the moment. For simple-sound-card one need to know what DMA is used and then select the DAI driver (if the Kconfig is right). I have also cleaned up the Kconfig prompts and comments a bit to be more precise. As can be seen in the diffstat I tried to do the merger with minimal code change, resisting the urge to rename the davinci-i2c driver to davinci-asp. With the merger I have created a new Kconfig structure: - Submenu for TI related settings for clarity - Only the DAI drivers are selectable and the platform drivers are selected based on the available (enabled) DMA engines - The remaining ASoC machine drivers are selecting the DAIs they need - The davinci-evm is now a single entry, combining all the various daVinci EVMS it supports and selects the needed DAIs based on the enabled boards The series also includes patches to updated the arch/arm/ files (defconfigs, source files using audio Kconfig symbols). I believe with v1 all of this can be merged via ASoC tree as the chances of conflicts are low for the touched arch/arm files. But Sekhar and Tony can correct me if I'm wrong. I think I have catched all (I hope) issues regarding to Kconfig thanks to the various autobuild machines out there kindly testing my linux-next-wip branch. I would appreciate any testing comments before I send the v1! Regards, Peter --- Peter Ujfalusi (9): ASoC: ti: Merge davinci and omap directories MAINTAINERS: Add entry for sound/soc/ti and update the OMAP audio support ARM: OMAP1: Makefile: Update for new MCBSP Kconfig option ARM: OMAP2: Update for new MCBSP Kconfig option ARM: davinci: dm365-evm: Update for the new ASoC Kcofnig options ARM: omap2plus_defconfig: Update the audio options ARM: omap1_defconfig: Do not select ASoC by default ARM: davinci_all_defconfig: Update the audio options ASoC: ti: Kconfig: Remove the deprecated options MAINTAINERS | 11 +- arch/arm/configs/davinci_all_defconfig | 5 +- arch/arm/configs/omap1_defconfig | 2 - arch/arm/configs/omap2plus_defconfig | 14 +- arch/arm/mach-davinci/board-dm365-evm.c | 4 +- arch/arm/mach-omap1/Makefile | 2 +- arch/arm/mach-omap2/Makefile | 2 +- arch/arm/mach-omap2/pdata-quirks.c | 4 +- sound/soc/Kconfig | 3 +- sound/soc/Makefile | 3 +- sound/soc/davinci/Kconfig | 106 --------- sound/soc/davinci/Makefile | 16 -- sound/soc/omap/Kconfig | 120 ---------- sound/soc/omap/Makefile | 30 --- sound/soc/ti/Kconfig | 209 ++++++++++++++++++ sound/soc/ti/Makefile | 44 ++++ sound/soc/{omap => ti}/ams-delta.c | 0 sound/soc/{davinci => ti}/davinci-evm.c | 4 +- sound/soc/{davinci => ti}/davinci-i2s.c | 0 sound/soc/{davinci => ti}/davinci-i2s.h | 0 sound/soc/{davinci => ti}/davinci-mcasp.c | 14 +- sound/soc/{davinci => ti}/davinci-mcasp.h | 0 sound/soc/{davinci => ti}/davinci-vcif.c | 0 sound/soc/{davinci => ti}/edma-pcm.c | 0 sound/soc/{davinci => ti}/edma-pcm.h | 4 +- sound/soc/{omap => ti}/n810.c | 0 sound/soc/{omap => ti}/omap-abe-twl6040.c | 0 sound/soc/{omap => ti}/omap-dmic.c | 0 sound/soc/{omap => ti}/omap-dmic.h | 0 .../omap-hdmi-audio.c => ti/omap-hdmi.c} | 0 sound/soc/{omap => ti}/omap-mcbsp-priv.h | 0 sound/soc/{omap => ti}/omap-mcbsp-st.c | 0 sound/soc/{omap => ti}/omap-mcbsp.c | 0 sound/soc/{omap => ti}/omap-mcbsp.h | 0 sound/soc/{omap => ti}/omap-mcpdm.c | 0 sound/soc/{omap => ti}/omap-mcpdm.h | 0 sound/soc/{omap => ti}/omap-twl4030.c | 0 sound/soc/{omap => ti}/omap3pandora.c | 0 sound/soc/{omap => ti}/osk5912.c | 0 sound/soc/{omap => ti}/rx51.c | 0 sound/soc/{omap => ti}/sdma-pcm.c | 0 sound/soc/{omap => ti}/sdma-pcm.h | 4 +- 42 files changed, 294 insertions(+), 307 deletions(-) delete mode 100644 sound/soc/davinci/Kconfig delete mode 100644 sound/soc/davinci/Makefile delete mode 100644 sound/soc/omap/Kconfig delete mode 100644 sound/soc/omap/Makefile create mode 100644 sound/soc/ti/Kconfig create mode 100644 sound/soc/ti/Makefile rename sound/soc/{omap => ti}/ams-delta.c (100%) rename sound/soc/{davinci => ti}/davinci-evm.c (99%) rename sound/soc/{davinci => ti}/davinci-i2s.c (100%) rename sound/soc/{davinci => ti}/davinci-i2s.h (100%) rename sound/soc/{davinci => ti}/davinci-mcasp.c (99%) rename sound/soc/{davinci => ti}/davinci-mcasp.h (100%) rename sound/soc/{davinci => ti}/davinci-vcif.c (100%) rename sound/soc/{davinci => ti}/edma-pcm.c (100%) rename sound/soc/{davinci => ti}/edma-pcm.h (91%) rename sound/soc/{omap => ti}/n810.c (100%) rename sound/soc/{omap => ti}/omap-abe-twl6040.c (100%) rename sound/soc/{omap => ti}/omap-dmic.c (100%) rename sound/soc/{omap => ti}/omap-dmic.h (100%) rename sound/soc/{omap/omap-hdmi-audio.c => ti/omap-hdmi.c} (100%) rename sound/soc/{omap => ti}/omap-mcbsp-priv.h (100%) rename sound/soc/{omap => ti}/omap-mcbsp-st.c (100%) rename sound/soc/{omap => ti}/omap-mcbsp.c (100%) rename sound/soc/{omap => ti}/omap-mcbsp.h (100%) rename sound/soc/{omap => ti}/omap-mcpdm.c (100%) rename sound/soc/{omap => ti}/omap-mcpdm.h (100%) rename sound/soc/{omap => ti}/omap-twl4030.c (100%) rename sound/soc/{omap => ti}/omap3pandora.c (100%) rename sound/soc/{omap => ti}/osk5912.c (100%) rename sound/soc/{omap => ti}/rx51.c (100%) rename sound/soc/{omap => ti}/sdma-pcm.c (100%) rename sound/soc/{omap => ti}/sdma-pcm.h (85%) -- Peter Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki

5 14

[alsa-devel] Fwd: [BUG] The infamous MOTU Ultralite * bitcrunch issue
by Stanisław Drozd 26 Nov '18

26 Nov '18

[Originally posted in alsa-user, but that list doesn't appear to be very widely used] Hello everyone, I recently updated the firmware on my UltraLite mk4 (IF YOURS IS WORKING OK, DON'T UPGRADE or you'll repeat my mistake). Current firmware on my piece is 1.3.4+232, 07/27/18 17:46:22, BIOS is Aug-04-2017 b568a4c. After the update an issue appeared - a moment after initializing the device and playing normally, it starts sounding as if it was put through a bitcruncher effect. After a longer while the output loses gain only to disappear completely after some more seconds pass (the gain change is not smooth, it seems to go away in steps). What I have tried: * Switching internal clock to LTC on the interface - would hold the sync together for a little longer; I did NOT send the actual timecode because I had trouble finding something to generate it for me * Switching sample rates - No noticeable change in behavior * Testing with speaker-test - No change in behavior, looks like this has to happen on ALSA level or lower * Changing clock source in Cadence (a JACK helper) settings and reinitializing - No effect Possible clues I saw: * dmesg started showing this when I plug the device in: [ 533.419099] usb 1-2.3: new high-speed USB device number 8 using xhci_hcd [ 533.527638] usb 1-2.3: New USB device found, idVendor=07fd, idProduct=0005, bcdDevice= 2.00 [ 533.527646] usb 1-2.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 533.527650] usb 1-2.3: Product: UltraLite-mk4 [ 533.527654] usb 1-2.3: Manufacturer: MOTU [ 533.527658] usb 1-2.3: SerialNumber: 0001f2fffe008e19 [ 534.377873] usb 1-2.3: clock source 1 is not valid, cannot use <and about 50 more lines of the clock source message> Questions: * What other places can I go to to report this? * What other settings can I tweak? Thanks, Stanislaw PS: I have a basic understanding of driver work, I think I could help with creating a patch with some guidance

1 0

[alsa-devel] [PATCH] ALSA: hda/realtek - fix the pop noise on headphone for lenovo laptops
by Hui Wang 26 Nov '18

26 Nov '18

We have several Lenovo laptops with the codec alc285, when playing sound via headphone, we can hear click/pop noise in the headphone, if we let the headphone share the DAC of NID 0x2 with the speaker, the noise disappears. The Lenovo laptops here include P52, P72, X1 yoda2 and X1 carbon. I have tried to set preferred_dacs and override_conn, but neither of them worked. Thanks for Kailang, he told me to invalidate the NID 0x3 through override_wcaps. BugLink: https://bugs.launchpad.net/bugs/1805079 Cc: <stable(a)vger.kernel.org> Signed-off-by: Kailang Yang <kailang(a)realtek.com> Signed-off-by: Hui Wang <hui.wang(a)canonical.com> --- sound/pci/hda/patch_realtek.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 1118fd1bbf1a..e66da22272fd 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5358,6 +5358,16 @@ static void alc274_fixup_bind_dacs(struct hda_codec *codec, spec->gen.preferred_dacs = preferred_pairs; } +/* The DAC of NID 0x3 will introduce click/pop noise on headphones, so invalidate it */ +static void alc285_fixup_invalidate_dacs(struct hda_codec *codec, + const struct hda_fixup *fix, int action) +{ + if (action != HDA_FIXUP_ACT_PRE_PROBE) + return; + + snd_hda_override_wcaps(codec, 0x03, 0); +} + /* for hda_fixup_thinkpad_acpi() */ #include "thinkpad_helper.c" @@ -5495,6 +5505,7 @@ enum { ALC255_FIXUP_DELL_HEADSET_MIC, ALC295_FIXUP_HP_X360, ALC221_FIXUP_HP_HEADSET_MIC, + ALC285_FIXUP_LENOVO_HEADPHONE_NOISE, }; static const struct hda_fixup alc269_fixups[] = { @@ -6362,6 +6373,10 @@ static const struct hda_fixup alc269_fixups[] = { .chained = true, .chain_id = ALC269_FIXUP_HEADSET_MIC }, + [ALC285_FIXUP_LENOVO_HEADPHONE_NOISE] = { + .type = HDA_FIXUP_FUNC, + .v.func = alc285_fixup_invalidate_dacs, + }, }; static const struct snd_pci_quirk alc269_fixup_tbl[] = { @@ -7035,6 +7050,11 @@ static const struct snd_hda_pin_quirk alc269_pin_fixup_tbl[] = { {0x12, 0x90a60130}, {0x19, 0x03a11020}, {0x21, 0x0321101f}), + SND_HDA_PIN_QUIRK(0x10ec0285, 0x17aa, "Lenovo", ALC285_FIXUP_LENOVO_HEADPHONE_NOISE, + {0x12, 0x90a60130}, + {0x14, 0x90170110}, + {0x19, 0x04a11040}, + {0x21, 0x04211020}), SND_HDA_PIN_QUIRK(0x10ec0288, 0x1028, "Dell", ALC288_FIXUP_DELL1_MIC_NO_PRESENCE, {0x12, 0x90a60120}, {0x14, 0x90170110}, -- 2.17.1

2 1

Re: [alsa-devel] [LKP] [ALSA] 3a6d1980fe: kernel_BUG_at_mm/page_alloc.c
by Takashi Iwai 26 Nov '18

26 Nov '18

On Mon, 26 Nov 2018 01:28:05 +0100, kernel test robot wrote: > > FYI, we noticed the following commit (built with gcc-7): > > commit: 3a6d1980fe96dbbfe3ae58db0048867f5319cdbf ("ALSA: Replace snd_malloc_pages() and snd_free_pages() with standard helpers") > https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git for-next .... > [ 174.830284] kernel BUG at mm/page_alloc.c:2850! OK, this seem to be an issue about using __GFP_COMP and alloc_pages_exact(). I didn't hit this while testing locally maybe just because I used the buffer size in power-of-two casually. I dropped the patch until we sort this out. thanks, Takashi

1 0