May 2021 - Alsa-devel - mailman.alsa-project.org

[PATCH 0/8] ALSA: firewire-lib: check cycle continuity
by Takashi Sakamoto 19 May '21

19 May '21

Hi, Current implementation of ALSA IEC 61883-1/6 packet streaming engine doesn't check whether received packets are exactly per isochronous cycle. This is required to process packets transferred from OXFW970-based devices and devices in RME Fireface series. However, the packet sequence with skipped cycle is inconvenient for media clock recovery. This patchset takes the engine to check cycle continuity at processing packets, including code refactoring. For RME Fireface series, the skipped cycle is handled as receiving an empty packet. For OXFW970-based devices, the skipped cycles are acceptable but media clock recovery is hard. Takashi Sakamoto (8): ALSA: firewire-lib: code refactoring to refer the same frame count per period in domain structure ALSA: firewire-lib: handle the case that empty isochronous packet payload for CIP ALSA: firewire-lib: code refactoring for sequence descriptor' ALSA: firewire-lib: code refactoring for helper function to compute OHCI 1394 cycle ALSA: firewire-lib: code refactoring for parser of IR context header ALSA: firewire-lib: code refactoring for check of CIP header about payload size ALSA: firewire-lib: check cycle continuity ALSA: firewire-lib: insert descriptor for skipped cycle sound/firewire/amdtp-stream.c | 172 ++++++++++++++++++++++------------ sound/firewire/amdtp-stream.h | 10 +- 2 files changed, 119 insertions(+), 63 deletions(-) -- 2.27.0

2 9

[PATCH][next] ALSA: rawmidi: fix incorrect array bounds check on clock_names
by Colin King 19 May '21

19 May '21

From: Colin Ian King <colin.king(a)canonical.com> The array bounds check on clock_names is currently checking the size of the entire array rather than the number of elements in the array leading to a potential array bounds read error. Fix this by using the ARRAY_SIZE macro instead of sizeof. Addresses-Coverity: ("Out-of-bounds read") Fixes: 08fdced60ca0 ("ALSA: rawmidi: Add framing mode") Signed-off-by: Colin Ian King <colin.king(a)canonical.com> --- sound/core/rawmidi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/core/rawmidi.c b/sound/core/rawmidi.c index 4a6534db77d6..6c0a4a67ad2e 100644 --- a/sound/core/rawmidi.c +++ b/sound/core/rawmidi.c @@ -1679,7 +1679,7 @@ static void snd_rawmidi_proc_info_read(struct snd_info_entry *entry, buffer_size, avail, xruns); if (substream->framing == SNDRV_RAWMIDI_MODE_FRAMING_TSTAMP) { clock_type = substream->clock_type >> SNDRV_RAWMIDI_MODE_CLOCK_SHIFT; - if (!snd_BUG_ON(clock_type >= sizeof(clock_names))) + if (!snd_BUG_ON(clock_type >= ARRAY_SIZE(clock_names))) snd_iprintf(buffer, " Framing : tstamp\n" " Clock type : %s\n", -- 2.31.1

2 1

[PATCH 0/2] mfd: wcd934x: add mbhc interrupts
by Srinivas Kandagatla 19 May '21

19 May '21

This patchset adds interrupts required to enable MBHC (Mulit Button Headset Control) support for Jack and headset button detection in WCD938x Codec. Codec side patches are sent in a separate patch set. Tested it on Lenovo Yoga C630. --srini Srinivas Kandagatla (2): mfd: wcd934x: add macro for adding new interrupts mfd: wcd934x: add additional interrupts required for MBHC support drivers/mfd/wcd934x.c | 50 +++++++++++++++++++++---------------------- 1 file changed, 24 insertions(+), 26 deletions(-) -- 2.21.0

2 4

[bug report] ASoC: topology: Add support for multiple kcontrol types to a widget
by Dan Carpenter 19 May '21

19 May '21

Hello Jaska Uimonen, The patch d29d41e28eea: "ASoC: topology: Add support for multiple kcontrol types to a widget" from May 7, 2021, leads to the following static checker warning: sound/soc/soc-topology.c:1559 soc_tplg_dapm_widget_create() error: uninitialized symbol 'kcontrol_type'. sound/soc/soc-topology.c 1418 static int soc_tplg_dapm_widget_create(struct soc_tplg *tplg, 1419 struct snd_soc_tplg_dapm_widget *w) 1420 { 1421 struct snd_soc_dapm_context *dapm = &tplg->comp->dapm; 1422 struct snd_soc_dapm_widget template, *widget; 1423 struct snd_soc_tplg_ctl_hdr *control_hdr; 1424 struct snd_soc_card *card = tplg->comp->card; 1425 unsigned int *kcontrol_type; 1426 struct snd_kcontrol_new *kc; 1427 int mixer_count = 0; 1428 int bytes_count = 0; 1429 int enum_count = 0; 1430 int ret = 0; 1431 int i; 1432 1433 if (strnlen(w->name, SNDRV_CTL_ELEM_ID_NAME_MAXLEN) == 1434 SNDRV_CTL_ELEM_ID_NAME_MAXLEN) 1435 return -EINVAL; 1436 if (strnlen(w->sname, SNDRV_CTL_ELEM_ID_NAME_MAXLEN) == 1437 SNDRV_CTL_ELEM_ID_NAME_MAXLEN) 1438 return -EINVAL; 1439 1440 dev_dbg(tplg->dev, "ASoC: creating DAPM widget %s id %d\n", 1441 w->name, w->id); 1442 1443 memset(&template, 0, sizeof(template)); 1444 1445 /* map user to kernel widget ID */ 1446 template.id = get_widget_id(le32_to_cpu(w->id)); 1447 if ((int)template.id < 0) 1448 return template.id; 1449 1450 /* strings are allocated here, but used and freed by the widget */ 1451 template.name = kstrdup(w->name, GFP_KERNEL); 1452 if (!template.name) 1453 return -ENOMEM; 1454 template.sname = kstrdup(w->sname, GFP_KERNEL); 1455 if (!template.sname) { 1456 ret = -ENOMEM; 1457 goto err; 1458 } 1459 template.reg = le32_to_cpu(w->reg); 1460 template.shift = le32_to_cpu(w->shift); 1461 template.mask = le32_to_cpu(w->mask); 1462 template.subseq = le32_to_cpu(w->subseq); 1463 template.on_val = w->invert ? 0 : 1; 1464 template.off_val = w->invert ? 1 : 0; 1465 template.ignore_suspend = le32_to_cpu(w->ignore_suspend); 1466 template.event_flags = le16_to_cpu(w->event_flags); 1467 template.dobj.index = tplg->index; 1468 1469 tplg->pos += 1470 (sizeof(struct snd_soc_tplg_dapm_widget) + 1471 le32_to_cpu(w->priv.size)); 1472 1473 if (w->num_kcontrols == 0) { 1474 template.num_kcontrols = 0; 1475 goto widget; This patch changed kcontrol_type to a pointer. Originally if we hit this goto kcontrol_type was zero but now it's uninitialized. This will cause a runtime error with UBSan even if it isn't a real bug. 1476 } 1477 1478 control_hdr = (struct snd_soc_tplg_ctl_hdr *)tplg->pos; 1479 dev_dbg(tplg->dev, "ASoC: template %s has %d controls of type %x\n", 1480 w->name, w->num_kcontrols, control_hdr->type); 1481 1482 template.num_kcontrols = le32_to_cpu(w->num_kcontrols); 1483 kc = devm_kcalloc(tplg->dev, le32_to_cpu(w->num_kcontrols), sizeof(*kc), GFP_KERNEL); 1484 if (!kc) 1485 goto err; 1486 1487 kcontrol_type = devm_kcalloc(tplg->dev, le32_to_cpu(w->num_kcontrols), sizeof(unsigned int), 1488 GFP_KERNEL); 1489 if (!kcontrol_type) 1490 goto err; 1491 1492 for (i = 0; i < w->num_kcontrols; i++) { 1493 control_hdr = (struct snd_soc_tplg_ctl_hdr *)tplg->pos; 1494 switch (le32_to_cpu(control_hdr->ops.info)) { 1495 case SND_SOC_TPLG_CTL_VOLSW: 1496 case SND_SOC_TPLG_CTL_STROBE: 1497 case SND_SOC_TPLG_CTL_VOLSW_SX: 1498 case SND_SOC_TPLG_CTL_VOLSW_XR_SX: 1499 case SND_SOC_TPLG_CTL_RANGE: 1500 case SND_SOC_TPLG_DAPM_CTL_VOLSW: 1501 /* volume mixer */ 1502 kc[i].index = mixer_count; 1503 kcontrol_type[i] = SND_SOC_TPLG_TYPE_MIXER; 1504 mixer_count++; 1505 ret = soc_tplg_dapm_widget_dmixer_create(tplg, &kc[i]); 1506 if (ret < 0) 1507 goto hdr_err; 1508 break; 1509 case SND_SOC_TPLG_CTL_ENUM: 1510 case SND_SOC_TPLG_CTL_ENUM_VALUE: 1511 case SND_SOC_TPLG_DAPM_CTL_ENUM_DOUBLE: 1512 case SND_SOC_TPLG_DAPM_CTL_ENUM_VIRT: 1513 case SND_SOC_TPLG_DAPM_CTL_ENUM_VALUE: 1514 /* enumerated mixer */ 1515 kc[i].index = enum_count; 1516 kcontrol_type[i] = SND_SOC_TPLG_TYPE_ENUM; 1517 enum_count++; 1518 ret = soc_tplg_dapm_widget_denum_create(tplg, &kc[i]); 1519 if (ret < 0) 1520 goto hdr_err; 1521 break; 1522 case SND_SOC_TPLG_CTL_BYTES: 1523 /* bytes control */ 1524 kc[i].index = bytes_count; 1525 kcontrol_type[i] = SND_SOC_TPLG_TYPE_BYTES; 1526 bytes_count++; 1527 ret = soc_tplg_dapm_widget_dbytes_create(tplg, &kc[i]); 1528 if (ret < 0) 1529 goto hdr_err; 1530 break; 1531 default: 1532 dev_err(tplg->dev, "ASoC: invalid widget control type %d:%d:%d\n", 1533 control_hdr->ops.get, control_hdr->ops.put, 1534 le32_to_cpu(control_hdr->ops.info)); 1535 ret = -EINVAL; 1536 goto hdr_err; 1537 } 1538 } 1539 1540 template.kcontrol_news = kc; 1541 1542 widget: 1543 ret = soc_tplg_widget_load(tplg, &template, w); 1544 if (ret < 0) 1545 goto hdr_err; 1546 1547 /* card dapm mutex is held by the core if we are loading topology 1548 * data during sound card init. */ 1549 if (card->instantiated) 1550 widget = snd_soc_dapm_new_control(dapm, &template); 1551 else 1552 widget = snd_soc_dapm_new_control_unlocked(dapm, &template); 1553 if (IS_ERR(widget)) { 1554 ret = PTR_ERR(widget); 1555 goto hdr_err; 1556 } 1557 1558 widget->dobj.type = SND_SOC_DOBJ_WIDGET; 1559 widget->dobj.widget.kcontrol_type = kcontrol_type; ^^^^^^^^^^^^^^ 1560 widget->dobj.ops = tplg->ops; 1561 widget->dobj.index = tplg->index; 1562 list_add(&widget->dobj.list, &tplg->comp->dobj_list); 1563 1564 ret = soc_tplg_widget_ready(tplg, widget, w); 1565 if (ret < 0) 1566 goto ready_err; regards, dan carpenter

1 0

[PATCH 0/7] ALSA: Prep work for PCI rescan support
by Takashi Iwai 18 May '21

18 May '21

Hi, this is a patch set as a prep work for the future-planned PCI rescan support(*). The essential part is the patch to track in-flight tasks, and the rest are cleanups and fixes in the core code. With those changes, the driver can perform a sort of software suspend on the device without the hardware intervention, which is required during the PCI BAR movement, for example. Takashi (*) https://lore.kernel.org/alsa-devel/e25017c6-e5e4-7a24-e793-14a2e70a434e@amd… === Takashi Iwai (7): ALSA: core: Use READ_ONCE() / WRITE_ONCE() for power state change ALSA: control: Add power state check commonly for ioctl handlers ALSA: control: Drop superfluous snd_power_wait() calls ALSA: control: Minor optimization for SNDRV_CTL_IOCTL_POWER_STATE ALSA: control: Track in-flight control read/write/tlv accesses ALSA: pcm: Block the release until the system resume finishes ALSA: Drop superfluous argument from snd_power_wait() include/sound/core.h | 39 +++++++++++++++++++++--- sound/core/control.c | 41 +++++++++++++------------- sound/core/control_compat.c | 17 +++++------ sound/core/init.c | 59 ++++++++++++++++++++++++++++++++----- sound/core/pcm_native.c | 6 +++- sound/soc/soc-core.c | 2 +- sound/usb/usx2y/us122l.c | 2 +- 7 files changed, 121 insertions(+), 45 deletions(-) -- 2.26.2

2 11

[PATCH -next] ASoC: hisilicon: fix missing clk_disable_unprepare() on error in hi6210_i2s_startup()
by Yang Yingliang 18 May '21

18 May '21

After calling clk_prepare_enable(), clk_disable_unprepare() need be called when calling clk_set_rate() failed. Fixes: 0bf750f4cbe1 ("ASoC: hisilicon: Add hi6210 i2s audio driver") Reported-by: Hulk Robot <hulkci(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- sound/soc/hisilicon/hi6210-i2s.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/sound/soc/hisilicon/hi6210-i2s.c b/sound/soc/hisilicon/hi6210-i2s.c index 907f5f1f7b44..ff05b9779e4b 100644 --- a/sound/soc/hisilicon/hi6210-i2s.c +++ b/sound/soc/hisilicon/hi6210-i2s.c @@ -102,18 +102,15 @@ static int hi6210_i2s_startup(struct snd_pcm_substream *substream, for (n = 0; n < i2s->clocks; n++) { ret = clk_prepare_enable(i2s->clk[n]); - if (ret) { - while (n--) - clk_disable_unprepare(i2s->clk[n]); - return ret; - } + if (ret) + goto err_unprepare_clk; } ret = clk_set_rate(i2s->clk[CLK_I2S_BASE], 49152000); if (ret) { dev_err(i2s->dev, "%s: setting 49.152MHz base rate failed %d\n", __func__, ret); - return ret; + goto err_unprepare_clk; } /* enable clock before frequency division */ @@ -165,6 +162,11 @@ static int hi6210_i2s_startup(struct snd_pcm_substream *substream, hi6210_write_reg(i2s, HII2S_SW_RST_N, val); return 0; + +err_unprepare_clk: + while (n--) + clk_disable_unprepare(i2s->clk[n]); + return ret; } static void hi6210_i2s_shutdown(struct snd_pcm_substream *substream, -- 2.25.1

2 1

[PATCH] ASoC: fsl: fix SND_SOC_IMX_RPMSG dependency
by Arnd Bergmann 18 May '21

18 May '21

From: Arnd Bergmann <arnd(a)arndb.de> Kconfig produces a warning with SND_SOC_FSL_RPMSG=y and SND_IMX_SOC=m: WARNING: unmet direct dependencies detected for SND_SOC_IMX_RPMSG Depends on [m]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_IMX_SOC [=m] && RPMSG [=y] Selected by [y]: - SND_SOC_FSL_RPMSG [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && COMMON_CLK [=y] && RPMSG [=y] && SND_IMX_SOC [=m]!=n Add a dependency to prevent this configuration. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- sound/soc/fsl/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/fsl/Kconfig b/sound/soc/fsl/Kconfig index 0917d65d6921..556c284f49dd 100644 --- a/sound/soc/fsl/Kconfig +++ b/sound/soc/fsl/Kconfig @@ -119,6 +119,7 @@ config SND_SOC_FSL_RPMSG tristate "NXP Audio Base On RPMSG support" depends on COMMON_CLK depends on RPMSG + depends on SND_IMX_SOC || SND_IMX_SOC = n select SND_SOC_IMX_RPMSG if SND_IMX_SOC != n help Say Y if you want to add rpmsg audio support for the Freescale CPUs. -- 2.29.2

3 2

[PATCH -next] ASoC: soc-core: use DEVICE_ATTR_RO macro
by YueHaibing 18 May '21

18 May '21

Use DEVICE_ATTR_RO helper instead of plain DEVICE_ATTR, which makes the code a bit shorter and easier to read. Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> --- sound/soc/soc-core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 1c0904acb935..2d969e31f37d 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -75,9 +75,9 @@ static ssize_t pmdown_time_show(struct device *dev, return sprintf(buf, "%ld\n", rtd->pmdown_time); } -static ssize_t pmdown_time_set(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t pmdown_time_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) { struct snd_soc_pcm_runtime *rtd = dev_get_drvdata(dev); int ret; @@ -89,7 +89,7 @@ static ssize_t pmdown_time_set(struct device *dev, return count; } -static DEVICE_ATTR(pmdown_time, 0644, pmdown_time_show, pmdown_time_set); +static DEVICE_ATTR_RW(pmdown_time); static struct attribute *soc_dev_attrs[] = { &dev_attr_pmdown_time.attr, -- 2.17.1

2 3

[PATCH -next] ASoC: rk3328: fix missing clk_disable_unprepare() on error in rk3328_platform_probe()
by Yang Yingliang 18 May '21

18 May '21

Fix the missing clk_disable_unprepare() before return from rk3328_platform_probe() in the error handling case. Fixes: c32759035ad2 ("ASoC: rockchip: support ACODEC for rk3328") Reported-by: Hulk Robot <hulkci(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- sound/soc/codecs/rk3328_codec.c | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) diff --git a/sound/soc/codecs/rk3328_codec.c b/sound/soc/codecs/rk3328_codec.c index bfefefcc76d8..758d439e8c7a 100644 --- a/sound/soc/codecs/rk3328_codec.c +++ b/sound/soc/codecs/rk3328_codec.c @@ -474,7 +474,8 @@ static int rk3328_platform_probe(struct platform_device *pdev) rk3328->pclk = devm_clk_get(&pdev->dev, "pclk"); if (IS_ERR(rk3328->pclk)) { dev_err(&pdev->dev, "can't get acodec pclk\n"); - return PTR_ERR(rk3328->pclk); + ret = PTR_ERR(rk3328->pclk); + goto err_unprepare_mclk; } ret = clk_prepare_enable(rk3328->pclk); @@ -484,19 +485,34 @@ static int rk3328_platform_probe(struct platform_device *pdev) } base = devm_platform_ioremap_resource(pdev, 0); - if (IS_ERR(base)) - return PTR_ERR(base); + if (IS_ERR(base)) { + ret = PTR_ERR(base); + goto err_unprepare_pclk; + } rk3328->regmap = devm_regmap_init_mmio(&pdev->dev, base, &rk3328_codec_regmap_config); - if (IS_ERR(rk3328->regmap)) - return PTR_ERR(rk3328->regmap); + if (IS_ERR(rk3328->regmap)) { + ret = PTR_ERR(rk3328->regmap); + goto err_unprepare_pclk; + } platform_set_drvdata(pdev, rk3328); - return devm_snd_soc_register_component(&pdev->dev, &soc_codec_rk3328, + ret = devm_snd_soc_register_component(&pdev->dev, &soc_codec_rk3328, rk3328_dai, ARRAY_SIZE(rk3328_dai)); + if (ret) + goto err_unprepare_pclk; + + return 0; + +err_unprepare_pclk: + clk_disable_unprepare(rk3328->pclk); + +err_unprepare_mclk: + clk_disable_unprepare(rk3328->mclk); + return ret; } static const struct of_device_id rk3328_codec_of_match[] __maybe_unused = { -- 2.25.1

2 1

[PATCH alsa-ucm-conf 0/3] byt/cht: Add support for controlling mute LEDs using the new snd_ctl_led kmod
by Hans de Goede 18 May '21

18 May '21

Hi All, This series adds support for controlling spk/mic mute LEDs one some Bay- and Cherry-Trail devices using the new snd_ctl_led kmod functionality. This requires the latest alsa-lib and alsa-utils git code and when running on Fedora it also requires putting selinux in permissive mode for now. I've filed a bug against the selinux-policy to get this fixed: https://bugzilla.redhat.com/show_bug.cgi?id=1958210 Note the first patch is mostly unrelated, it is a resend of an earlier patch which seems to have fallen through the cracks. Regards, Hans Hans de Goede (3): cht-bsw-rt5672: Add support for the components string cht-bsw-rt5672: Add support for controlling speaker- and mic-mute LEDs bytcr-rt5640: Add support for controlling a speaker-mute LED ucm2/bytcr-rt5640/bytcr-rt5640.conf | 6 ++++ ucm2/cht-bsw-rt5672/HiFi.conf | 37 +++++++++++++++++++++++-- ucm2/cht-bsw-rt5672/cht-bsw-rt5672.conf | 6 ++++ 3 files changed, 47 insertions(+), 2 deletions(-) -- 2.31.1

2 4