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

[alsa-devel] [PATCH v2 5/5] ALSA: emu10k1: add a IOMMU workaround
by Maciej S. Szmigiero 14 Feb '18

14 Feb '18

The Audigy 2 CA0102 chip (but most likely others from the emu10k1 family, too) has a problem that from time to time it likes to do few DMA reads a bit beyond its normal allocation and gets very confused if these reads get blocked by a IOMMU. For the first (reserved) page this happens multiple times at every playback, for various synth pages it happens randomly, rarely for PCM playback buffers and the page table memory itself. All these reads seem to follow a similar pattern, observed read offsets beyond the allocation end were 0x00, 0x40, 0x80 and 0xc0 (PCI cache line multiples), so it looks like the device tries to accesses up to 256 extra bytes. As a workaround let's widen these DMA allocations by an extra page if we detect that the device is behind a non-passthrough IOMMU (the DMA memory should be relatively plenty on IOMMU systems). Signed-off-by: Maciej S. Szmigiero <mail(a)maciej.szmigiero.name> --- Changes from v1: Apply this workaround also to PCM playback buffers since it seems they are affected, too. include/sound/emu10k1.h | 1 + sound/pci/emu10k1/emu10k1_main.c | 50 +++++++++++++++++++++++++++++++++++++--- sound/pci/emu10k1/emupcm.c | 9 +++++++- sound/pci/emu10k1/memory.c | 16 ++++++++++--- 4 files changed, 69 insertions(+), 7 deletions(-) diff --git a/include/sound/emu10k1.h b/include/sound/emu10k1.h index db32b7de52e0..ba27abf65408 100644 --- a/include/sound/emu10k1.h +++ b/include/sound/emu10k1.h @@ -1710,6 +1710,7 @@ struct snd_emu10k1 { unsigned int ecard_ctrl; /* ecard control bits */ unsigned int address_mode; /* address mode */ unsigned long dma_mask; /* PCI DMA mask */ + bool iommu_workaround; /* IOMMU workaround needed */ unsigned int delay_pcm_irq; /* in samples */ int max_cache_pages; /* max memory size / PAGE_SIZE */ struct snd_dma_buffer silent_page; /* silent page */ diff --git a/sound/pci/emu10k1/emu10k1_main.c b/sound/pci/emu10k1/emu10k1_main.c index 8decd2a7a404..3638bff26d23 100644 --- a/sound/pci/emu10k1/emu10k1_main.c +++ b/sound/pci/emu10k1/emu10k1_main.c @@ -36,6 +36,7 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/interrupt.h> +#include <linux/iommu.h> #include <linux/pci.h> #include <linux/slab.h> #include <linux/vmalloc.h> @@ -1758,6 +1759,38 @@ static struct snd_emu_chip_details emu_chip_details[] = { { } /* terminator */ }; +/* + * The chip (at least the Audigy 2 CA0102 chip, but most likely others, too) + * has a problem that from time to time it likes to do few DMA reads a bit + * beyond its normal allocation and gets very confused if these reads get + * blocked by a IOMMU. + * + * This behaviour has been observed for the first (reserved) page + * (for which it happens multiple times at every playback), often for various + * synth pages and sometimes for PCM playback buffers and the page table + * memory itself. + * + * As a workaround let's widen these DMA allocations by an extra page if we + * detect that the device is behind a non-passthrough IOMMU. + */ +static void snd_emu10k1_detect_iommu(struct snd_emu10k1 *emu) +{ + struct iommu_domain *domain; + + emu->iommu_workaround = false; + + if (!iommu_present(emu->card->dev->bus)) + return; + + domain = iommu_get_domain_for_dev(emu->card->dev); + if (domain && domain->type == IOMMU_DOMAIN_IDENTITY) + return; + + dev_notice(emu->card->dev, + "non-passthrough IOMMU detected, widening DMA allocations"); + emu->iommu_workaround = true; +} + int snd_emu10k1_create(struct snd_card *card, struct pci_dev *pci, unsigned short extin_mask, @@ -1770,6 +1803,7 @@ int snd_emu10k1_create(struct snd_card *card, struct snd_emu10k1 *emu; int idx, err; int is_audigy; + size_t page_table_size, silent_page_size; unsigned int silent_page; const struct snd_emu_chip_details *c; static struct snd_device_ops ops = { @@ -1867,6 +1901,8 @@ int snd_emu10k1_create(struct snd_card *card, is_audigy = emu->audigy = c->emu10k2_chip; + snd_emu10k1_detect_iommu(emu); + /* set addressing mode */ emu->address_mode = is_audigy ? 0 : 1; /* set the DMA transfer mask */ @@ -1893,8 +1929,13 @@ int snd_emu10k1_create(struct snd_card *card, emu->port = pci_resource_start(pci, 0); emu->max_cache_pages = max_cache_bytes >> PAGE_SHIFT; + + page_table_size = sizeof(u32) * (emu->address_mode ? MAXPAGES1 : + MAXPAGES0); + if (emu->iommu_workaround) + page_table_size += PAGE_SIZE; if (snd_dma_alloc_pages(SNDRV_DMA_TYPE_DEV, snd_dma_pci_data(pci), - (emu->address_mode ? 32 : 16) * 1024, &emu->ptb_pages) < 0) { + page_table_size, &emu->ptb_pages) < 0) { err = -ENOMEM; goto error; } @@ -1910,8 +1951,11 @@ int snd_emu10k1_create(struct snd_card *card, goto error; } + silent_page_size = EMUPAGESIZE; + if (emu->iommu_workaround) + silent_page_size *= 2; if (snd_dma_alloc_pages(SNDRV_DMA_TYPE_DEV, snd_dma_pci_data(pci), - EMUPAGESIZE, &emu->silent_page) < 0) { + silent_page_size, &emu->silent_page) < 0) { err = -ENOMEM; goto error; } @@ -1995,7 +2039,7 @@ int snd_emu10k1_create(struct snd_card *card, 0x00000000 | SPCS_EMPHASIS_NONE | SPCS_COPYRIGHT; /* Clear silent pages and set up pointers */ - memset(emu->silent_page.area, 0, PAGE_SIZE); + memset(emu->silent_page.area, 0, silent_page_size); silent_page = emu->silent_page.addr << emu->address_mode; for (idx = 0; idx < (emu->address_mode ? MAXPAGES1 : MAXPAGES0); idx++) ((u32 *)emu->ptb_pages.area)[idx] = cpu_to_le32(silent_page | idx); diff --git a/sound/pci/emu10k1/emupcm.c b/sound/pci/emu10k1/emupcm.c index 2683b9717215..80b3279692b8 100644 --- a/sound/pci/emu10k1/emupcm.c +++ b/sound/pci/emu10k1/emupcm.c @@ -411,12 +411,19 @@ static int snd_emu10k1_playback_hw_params(struct snd_pcm_substream *substream, struct snd_emu10k1 *emu = snd_pcm_substream_chip(substream); struct snd_pcm_runtime *runtime = substream->runtime; struct snd_emu10k1_pcm *epcm = runtime->private_data; + size_t alloc_size; int err; if ((err = snd_emu10k1_pcm_channel_alloc(epcm, params_channels(hw_params))) < 0) return err; - if ((err = snd_pcm_lib_malloc_pages(substream, params_buffer_bytes(hw_params))) < 0) + + alloc_size = params_buffer_bytes(hw_params); + if (emu->iommu_workaround) + alloc_size += EMUPAGESIZE; + if ((err = snd_pcm_lib_malloc_pages(substream, alloc_size)) < 0) return err; + if (emu->iommu_workaround && runtime->dma_bytes >= EMUPAGESIZE) + runtime->dma_bytes -= EMUPAGESIZE; if (err > 0) { /* change */ int mapped; if (epcm->memblk != NULL) diff --git a/sound/pci/emu10k1/memory.c b/sound/pci/emu10k1/memory.c index 5cdffe2d31e1..6a5371d10dcf 100644 --- a/sound/pci/emu10k1/memory.c +++ b/sound/pci/emu10k1/memory.c @@ -457,6 +457,16 @@ static void get_single_page_range(struct snd_util_memhdr *hdr, *last_page_ret = last_page; } +static size_t synth_get_alloc_size(struct snd_emu10k1 *emu) +{ + size_t alloc_size = PAGE_SIZE; + + if (emu->iommu_workaround) + alloc_size *= 2; + + return alloc_size; +} + /* * allocate kernel pages */ @@ -471,7 +481,7 @@ static int synth_alloc_pages(struct snd_emu10k1 *emu, struct snd_emu10k1_memblk for (page = first_page; page <= last_page; page++) { if (snd_dma_alloc_pages(SNDRV_DMA_TYPE_DEV, snd_dma_pci_data(emu->pci), - PAGE_SIZE, &dmab) < 0) + synth_get_alloc_size(emu), &dmab) < 0) goto __fail; if (!is_valid_page(emu, dmab.addr)) { snd_dma_free_pages(&dmab); @@ -488,7 +498,7 @@ static int synth_alloc_pages(struct snd_emu10k1 *emu, struct snd_emu10k1_memblk for (page = first_page; page <= last_page; page++) { dmab.area = emu->page_ptr_table[page]; dmab.addr = emu->page_addr_table[page]; - dmab.bytes = PAGE_SIZE; + dmab.bytes = synth_get_alloc_size(emu); snd_dma_free_pages(&dmab); emu->page_addr_table[page] = 0; emu->page_ptr_table[page] = NULL; @@ -513,7 +523,7 @@ static int synth_free_pages(struct snd_emu10k1 *emu, struct snd_emu10k1_memblk * continue; dmab.area = emu->page_ptr_table[page]; dmab.addr = emu->page_addr_table[page]; - dmab.bytes = PAGE_SIZE; + dmab.bytes = synth_get_alloc_size(emu); snd_dma_free_pages(&dmab); emu->page_addr_table[page] = 0; emu->page_ptr_table[page] = NULL;

2 4

[alsa-devel] [PATCH v2 1/2] ASoC: wm9712: Use empty struct initializer
by Fabio Estevam 13 Feb '18

13 Feb '18

From: Fabio Estevam <fabio.estevam(a)nxp.com> { 0 } only clears the first member of the structure. The first member of the snd_soc_dapm_update struct is a pointer, and writing 0 to a pointer results in the following sparse warning: sound/soc/codecs/wm9712.c:229:47: warning: Using plain integer as NULL pointer Use the empty struct initializer that clears all the struct members and fixes the sparse warning. Signed-off-by: Fabio Estevam <fabio.estevam(a)nxp.com> --- Changes since v1: - Fix typos in commit log sound/soc/codecs/wm9712.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/wm9712.c b/sound/soc/codecs/wm9712.c index 7aeeeec..953d94d 100644 --- a/sound/soc/codecs/wm9712.c +++ b/sound/soc/codecs/wm9712.c @@ -226,7 +226,7 @@ static int wm9712_hp_mixer_put(struct snd_kcontrol *kcontrol, struct soc_mixer_control *mc = (struct soc_mixer_control *)kcontrol->private_value; unsigned int mixer, mask, shift, old; - struct snd_soc_dapm_update update = { 0 }; + struct snd_soc_dapm_update update = {}; bool change; mixer = mc->shift >> 8; -- 2.7.4

1 1

[alsa-devel] [PATCH v2 0/2] Add support for AK4458 DAC
by Cosmin-Gabriel Samoila 13 Feb '18

13 Feb '18

We support Normal Mode, TDM Mode and PM. Changes since V1: - support only i2c - i2c probe / remove/suspend/ resume moved into shared code - change enum array to separate variables - aggregate Volume control for Left and Right Channel - remove mux from DAPM - remove unused functions - remove unused controls - remove unused macros - use snd_soc_update_bits instead of soc_read + soc_write - use GPIOD - use SPDX identifier - use dac@10 instgitead of ak4458@10 - use explicit power_on/off functions - add __maybe_unused attribute - use probe_new for i2c_probe Cosmin-Gabriel Samoila (2): ASoC: codecs: Add support for AK4458 DAC driver ASoC: ak4458: Add bindings for AK4458 DAC Documentation/devicetree/bindings/sound/ak4458.txt | 23 + sound/soc/codecs/Kconfig | 6 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/ak4458.c | 684 +++++++++++++++++++++ sound/soc/codecs/ak4458.h | 75 +++ 5 files changed, 790 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/ak4458.txt create mode 100644 sound/soc/codecs/ak4458.c create mode 100644 sound/soc/codecs/ak4458.h -- 2.7.4

2 3

[alsa-devel] [PATCH 0/2] ASoC: missing replace platform/codec to component
by Kuninori Morimoto 13 Feb '18

13 Feb '18

Hi Mark These are missing replace platform/codec patches which are not yet accepted (except ac97 issue) uniphier: aio-dmac arizona -- 1.9.1

3 4

Re: [alsa-devel] [PATCH] ALSA: usb-audio: Fix UAC2 get_ctl request with a RANGE attribute
by Takashi Iwai 13 Feb '18

13 Feb '18

On Thu, 08 Feb 2018 07:21:36 +0100, Kirill Marinushkin wrote: > > On 02/07/18 06:45, Takashi Iwai wrote: > > On Mon, 29 Jan 2018 06:37:55 +0100, > > Kirill Marinushkin wrote: > >> The layout of the UAC2 Control request and response varies depending on > >> the request type. With the current implementation, only the Layout 2 > >> Parameter Block (with the 2-byte sized RANGE attribute) is handled > >> properly. For the Control requests with the 1-byte sized RANGE attribute > >> (Bass Control, Mid Control, Tremble Control), the response is parsed > >> incorrectly. > >> > >> This commit: > >> * fixes the wLength field value in the request > >> * fixes parsing the range values from the response > >> > >> Fixes: 23caaf19b11e ("ALSA: usb-mixer: Add support for Audio Class v2.0") > >> Signed-off-by: Kirill Marinushkin <k.marinushkin(a)gmail.com> > >> Cc: Jaroslav Kysela <perex(a)perex.cz> > >> Cc: Takashi Iwai <tiwai(a)suse.com> > >> Cc: Jaejoong Kim <climbbb.kim(a)gmail.com> > >> Cc: Bhumika Goyal <bhumirks(a)gmail.com> > >> Cc: Stephen Barber <smbarber(a)chromium.org> > >> Cc: Julian Scheel <julian(a)jusst.de> > >> Cc: alsa-devel(a)alsa-project.org > >> Cc: linux-kernel(a)vger.kernel.org > > Sorry for the late reply, as I've been (and still) off. > > > > Does this bug actually hit on any real devices, or is it only a > > logical error so far? In the former case, a Cc to stable is > > mandatory. > > > > In anyway, I'll review and merge it properly once after I back to > > work. > > > > > > thanks, > > > > Takashi > > Hello Takashi, > > Thank you for your answer. I will wait until you are back to work, don't > worry about the late replies. > > I did not hit the issue on a real device. > > During my UAC2 experiments, I reproduced this issue on the development > board, and then tested my solution on it. OK, if it happened on a development system, it's a real issue, and we should put Cc to stable. Now I applied the patch. Thanks! Takashi

2 3

[alsa-devel] [PATCH v3 0/3] add UniPhier audio nodes
by Katsuhiro Suzuki 13 Feb '18

13 Feb '18

This series adds device-tree nodes for Socionext audio system for UniPhier LD11/LD20 SoCs. This one enables I2S output for Line-In, Line-Out, S/PDIF output and compress audio output. Katsuhiro Suzuki (3): arm64: dts: uniphier: add sound node for UniPhier arm64: dts: uniphier: add speaker out for UniPhier LD11/LD20 boards arm64: dts: uniphier: add compress audio out for UniPhier LD11/LD20 .../boot/dts/socionext/uniphier-ld11-global.dts | 62 ++++++++++++++ arch/arm64/boot/dts/socionext/uniphier-ld11.dtsi | 92 +++++++++++++++++++++ .../boot/dts/socionext/uniphier-ld20-global.dts | 62 ++++++++++++++ arch/arm64/boot/dts/socionext/uniphier-ld20.dtsi | 96 ++++++++++++++++++++++ 4 files changed, 312 insertions(+) -- 2.15.1

2 4

[alsa-devel] [PATCH] ASoC: tlv320aic31xx: Fix master mode clock I2S bus clocks
by Peter Ujfalusi 13 Feb '18

13 Feb '18

In the reset state of the codec we do not have complete playback or capture routes. The audio playback/capture will not work due to missing clock signals on the I2S bus if PLL, MDAC/NDAC/DAC MADC/NADC/ADC is powered down. To make sure that even if all output/input is disconnected the codec is generating clocks, we need to have valid DAPM route in every case to power up the must needed parts of the codec. I have verified that switching DAC (during playback) or ADC (during capture) will stop the I2S clocks, so the only solution is to connect the 'Off' routes as well to output/input. Tested on am43x-epos-evm with aic3111 codec in master mode. Signed-off-by: Peter Ujfalusi <peter.ujfalusi(a)ti.com> --- sound/soc/codecs/tlv320aic31xx.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/sound/soc/codecs/tlv320aic31xx.c b/sound/soc/codecs/tlv320aic31xx.c index 858cb8be445f..d42a23eb916e 100644 --- a/sound/soc/codecs/tlv320aic31xx.c +++ b/sound/soc/codecs/tlv320aic31xx.c @@ -586,9 +586,11 @@ common31xx_audio_map[] = { {"DAC Left Input", "Left Data", "DAC IN"}, {"DAC Left Input", "Right Data", "DAC IN"}, {"DAC Left Input", "Mono", "DAC IN"}, + {"DAC Left Input", "Off", "DAC IN"}, {"DAC Right Input", "Left Data", "DAC IN"}, {"DAC Right Input", "Right Data", "DAC IN"}, {"DAC Right Input", "Mono", "DAC IN"}, + {"DAC Right Input", "Off", "DAC IN"}, {"DAC Left", NULL, "DAC Left Input"}, {"DAC Right", NULL, "DAC Right Input"}, @@ -601,6 +603,9 @@ common31xx_audio_map[] = { {"HP Right", "Switch", "Output Right"}, {"HPR Driver", NULL, "HP Right"}, {"HPR", NULL, "HPR Driver"}, + + {"HPL", NULL, "DAC Left"}, + {"HPR", NULL, "DAC Right"}, }; static const struct snd_soc_dapm_route @@ -621,16 +626,20 @@ aic31xx_audio_map[] = { {"MIC1LP P-Terminal", "FFR 10 Ohm", "MIC1LP"}, {"MIC1LP P-Terminal", "FFR 20 Ohm", "MIC1LP"}, {"MIC1LP P-Terminal", "FFR 40 Ohm", "MIC1LP"}, + {"MIC1LP P-Terminal", "Off", "MIC1LP"}, {"MIC1RP P-Terminal", "FFR 10 Ohm", "MIC1RP"}, {"MIC1RP P-Terminal", "FFR 20 Ohm", "MIC1RP"}, {"MIC1RP P-Terminal", "FFR 40 Ohm", "MIC1RP"}, + {"MIC1RP P-Terminal", "Off", "MIC1RP"}, {"MIC1LM P-Terminal", "FFR 10 Ohm", "MIC1LM"}, {"MIC1LM P-Terminal", "FFR 20 Ohm", "MIC1LM"}, {"MIC1LM P-Terminal", "FFR 40 Ohm", "MIC1LM"}, + {"MIC1LM P-Terminal", "Off", "MIC1LM"}, {"MIC1LM M-Terminal", "FFR 10 Ohm", "MIC1LM"}, {"MIC1LM M-Terminal", "FFR 20 Ohm", "MIC1LM"}, {"MIC1LM M-Terminal", "FFR 40 Ohm", "MIC1LM"}, + {"MIC1LM M-Terminal", "Off", "MIC1LM"}, {"MIC_GAIN_CTL", NULL, "MIC1LP P-Terminal"}, {"MIC_GAIN_CTL", NULL, "MIC1RP P-Terminal"}, -- Peter Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki

4 7

[alsa-devel] ASoC: rsnd: status debug message and its suppress
by Kuninori Morimoto 13 Feb '18

13 Feb '18

Hi Mark Renesas sound driver want to know status when debugging, but sometimes it is too much information. These patches indicate status information when debugging, but we can suppress it by special define Kuninori Morimoto (2): ASoC: rsnd: indicate IRQ error status for debug ASoC: rsnd: suppress rsnd_dai_call() debug message sound/soc/sh/rcar/core.c | 11 ++++++++++- sound/soc/sh/rcar/rsnd.h | 20 ++++++++++++++++++++ sound/soc/sh/rcar/src.c | 22 ++++++++++++++++++++-- sound/soc/sh/rcar/ssi.c | 16 +++++++++++++++- 4 files changed, 65 insertions(+), 4 deletions(-) Best regards --- Kuninori Morimoto

2 4

[alsa-devel] [asoc:topic/uniphier 3/7] sound/soc/uniphier/aio-core.c:907:5: sparse: symbol 'aiodma_rb_get_rp' was not declared. Should it be static?
by kbuild test robot 13 Feb '18

13 Feb '18

tree: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git topic/uniphier head: db4cb3d04eba91db48da0032c44b83ed7185973c commit: f37fe2f9987b0460f25a87b1380f8e97a5959121 [3/7] ASoC: uniphier: add support for UniPhier AIO common driver reproduce: # apt-get install sparse git checkout f37fe2f9987b0460f25a87b1380f8e97a5959121 make ARCH=x86_64 allmodconfig make C=1 CF=-D__CHECK_ENDIAN__ sparse warnings: (new ones prefixed by >>) >> sound/soc/uniphier/aio-core.c:907:5: sparse: symbol 'aiodma_rb_get_rp' was not declared. Should it be Please review and possibly fold the followup patch. --- 0-DAY kernel test infrastructure Open Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation

2 2

[alsa-devel] [asoc:topic/rt5659 1/3] sound/soc/codecs/rt5659.c:4048:6: sparse: symbol 'rt5659_intel_hd_header_probe_setup' was not declared. Should it be static?
by kbuild test robot 13 Feb '18

13 Feb '18

tree: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git topic/rt5659 head: de86a33253d07675abbb1fa294ffb4f3df6f219e commit: 041e74b71491acadad74b275e8b05add165d92b4 [1/3] ASoC: rt5659: Add the support of Intel HDA Header reproduce: # apt-get install sparse git checkout 041e74b71491acadad74b275e8b05add165d92b4 make ARCH=x86_64 allmodconfig make C=1 CF=-D__CHECK_ENDIAN__ sparse warnings: (new ones prefixed by >>) >> sound/soc/codecs/rt5659.c:4048:6: sparse: symbol 'rt5659_intel_hd_header_probe_setup' was not declared. Should it be Please review and possibly fold the followup patch. --- 0-DAY kernel test infrastructure Open Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation

2 2