October 2011 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] ASoC: Reduce the number of neigbours we mark dirty when updating power
by Mark Brown 03 Oct '11

03 Oct '11

If two widgets are not currently connected then there is no need to propagate a power state change between them as we mark the affected widgets when we change a connection. Similarly if a neighbour widget is already in the state being set for the current widget then there is no need to recheck. On one system I tested this gave: Power Path Neighbour Before: 114 1066 1327 After: 106 970 1186 which is an improvement, although relatively small. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/soc-dapm.c | 26 ++++++++++++++++++++------ 1 files changed, 20 insertions(+), 6 deletions(-) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index 06d8cfa..e0d9971 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -1215,6 +1215,21 @@ static void dapm_post_sequence_async(void *data, async_cookie_t cookie) } } +static void dapm_widget_set_peer_power(struct snd_soc_dapm_widget *peer, + bool power, bool connect) +{ + /* If a connection is being made or broken then that update + * will have marked the peer dirty, otherwise the widgets are + * not connected and this update has no impact. */ + if (!connect) + return; + + /* If the peer is already in the state we're moving to then we + * won't have an impact on it. */ + if (power != peer->power) + dapm_mark_dirty(peer); +} + static void dapm_widget_set_power(struct snd_soc_dapm_widget *w, bool power, struct list_head *up_list, struct list_head *down_list) @@ -1227,19 +1242,18 @@ static void dapm_widget_set_power(struct snd_soc_dapm_widget *w, bool power, trace_snd_soc_dapm_widget_power(w, power); /* If we changed our power state perhaps our neigbours changed - * also. We're not yet smart enough to update relevant - * neighbours when we change the state of a widget, this acts - * as a proxy for that. It will notify more neighbours than - * is ideal. + * also. */ list_for_each_entry(path, &w->sources, list_sink) { if (path->source) { - dapm_mark_dirty(path->source); + dapm_widget_set_peer_power(path->source, power, + path->connect); } } list_for_each_entry(path, &w->sinks, list_source) { if (path->sink) { - dapm_mark_dirty(path->sink); + dapm_widget_set_peer_power(path->sink, power, + path->connect); } } -- 1.7.6.3

1 0

[alsa-devel] [PATCH] ASoC: Mark headphone, mic, speaker and line widgets as always connected
by Mark Brown 03 Oct '11

03 Oct '11

We're not actually doing any dynamic power management based on connection and output drivers (which are pretty much the same thing) are marked as unconditionally connected already. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/soc-dapm.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index e0d9971..214a709 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -329,7 +329,7 @@ static void dapm_set_path_status(struct snd_soc_dapm_widget *w, } } break; - /* does not effect routing - always connected */ + /* does not affect routing - always connected */ case snd_soc_dapm_pga: case snd_soc_dapm_out_drv: case snd_soc_dapm_output: @@ -341,13 +341,13 @@ static void dapm_set_path_status(struct snd_soc_dapm_widget *w, case snd_soc_dapm_supply: case snd_soc_dapm_aif_in: case snd_soc_dapm_aif_out: - p->connect = 1; - break; - /* does effect routing - dynamically connected */ case snd_soc_dapm_hp: case snd_soc_dapm_mic: case snd_soc_dapm_spk: case snd_soc_dapm_line: + p->connect = 1; + break; + /* does affect routing - dynamically connected */ case snd_soc_dapm_pre: case snd_soc_dapm_post: p->connect = 0; -- 1.7.6.3

1 0

Re: [alsa-devel] [PATCH resend] ALSA: firewire-speakers: fix locking
by Clemens Ladisch 03 Oct '11

03 Oct '11

Stefan Richter wrote: > On Oct 03 Takashi Iwai wrote: >> Stefan Richter wrote: >>> There is a lock inversion between fwspk->mutex and pcm->open_mutex >>> reported by lockdep when fwspk_hw_free is called. >> >> We should put cc to stable kernel, right? > > I am not sure. While I could reproduce the lockdep report reliably at > device unplugs, I never encountered a deadlock. The reported scenario requires that the same PCM device is freed twice at the same time, which is prevented by the file reference counting. Going through the entire dependency chain would require the device being unplugged, being independently close()d, and a ioctl being called with a swapped-out data buffer all happening at the same time. This doesn't look like an actual problem. Regards, Clemens

1 0

[alsa-devel] [PATCH 1/5] ASoC: Factor write of widget power out into a separate function
by Mark Brown 03 Oct '11

03 Oct '11

Split the decision about what the new power should be out from the implementation of that decision. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/soc-dapm.c | 29 ++++++++++++++++++----------- 1 files changed, 18 insertions(+), 11 deletions(-) diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index c277228..dcbd468 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -1197,6 +1197,23 @@ static void dapm_post_sequence_async(void *data, async_cookie_t cookie) } } +static void dapm_widget_set_power(struct snd_soc_dapm_widget *w, bool power, + struct list_head *up_list, + struct list_head *down_list) +{ + if (w->power == power) + return; + + trace_snd_soc_dapm_widget_power(w, power); + + if (power) + dapm_seq_insert(w, up_list, true); + else + dapm_seq_insert(w, down_list, false); + + w->power = power; +} + static void dapm_power_one_widget(struct snd_soc_dapm_widget *w, struct list_head *up_list, struct list_head *down_list) @@ -1241,17 +1258,7 @@ static void dapm_power_one_widget(struct snd_soc_dapm_widget *w, } } - if (w->power == power) - break; - - trace_snd_soc_dapm_widget_power(w, power); - - if (power) - dapm_seq_insert(w, up_list, true); - else - dapm_seq_insert(w, down_list, false); - - w->power = power; + dapm_widget_set_power(w, power, up_list, down_list); break; } } -- 1.7.6.3

1 5

[alsa-devel] [PATCH] ASoC: Fix typo in 24.576MHz rate in WM5100
by Mark Brown 03 Oct '11

03 Oct '11

Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/codecs/wm5100.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/wm5100.c b/sound/soc/codecs/wm5100.c index f88b0a3..d2f260f 100644 --- a/sound/soc/codecs/wm5100.c +++ b/sound/soc/codecs/wm5100.c @@ -1762,7 +1762,7 @@ static int wm5100_set_sysclk(struct snd_soc_codec *codec, int clk_id, fval = 1; break; case 22579200: - case 2457600: + case 24576000: fval = 2; break; default: @@ -1779,7 +1779,7 @@ static int wm5100_set_sysclk(struct snd_soc_codec *codec, int clk_id, case 6144000: case 12288000: - case 2457600: + case 24576000: audio_rate = 48000; break; -- 1.7.6.3

1 0

Re: [alsa-devel] [PATCH] ASoC: samsung: s3c-i2s-v2.c needs module.h
by Mark Brown 03 Oct '11

03 Oct '11

On Mon, Oct 03, 2011 at 12:03:19PM -0400, Paul Gortmaker wrote: > On 11-10-03 06:30 AM, Mark Brown wrote: > > That's exactly what I'm running at the minute with no problem... > Then I'm guessing you don't have the option for it enabled (as I'm pretty sure I'd have noticed if I wasn't building the driver, my systems would be a lot less noisy than it currently is. > Thanks Axel, I'll squish this into the commit that adds module.h > to all the other sound drivers that have it missing. Now that we've worked out what it's for I don't see any reason not to apply Axel's patch right now so I've done that. It would be useful if you could send your patches for module.h upstream, there's no reason to avoid sending them now and it'd be useful to cut down on the noise from small merge issues.

1 0

[alsa-devel] [PATCH 0/6] ASoC fixes from ARM randconfig builds
by Arnd Bergmann 03 Oct '11

03 Oct '11

Hi Mark, As promised, here are the remaining ASoC patches from my randconfig series. I hope I managed to get the submission right this time. I've left out the ones that you already took or NAKed, so this is shorter than I expected. Arnd Arnd Bergmann (6): ASoC: codecs/wm8682: use __devexit_p ASoC: codecs: AK4641 depends on GPIOLIB ASoC: imx: eukrea_tlv320 needs i2c ASoC: sh: use correct __iomem annotations ASoC: samsung: fix Kconfig dependencies ASoC: samsung: add missing __devexit_p() annotations sound/soc/codecs/Kconfig | 2 +- sound/soc/codecs/wm8782.c | 2 +- sound/soc/imx/Kconfig | 1 + sound/soc/samsung/Kconfig | 4 ++++ sound/soc/samsung/ac97.c | 2 +- sound/soc/samsung/i2s.c | 2 +- sound/soc/sh/fsi.c | 10 +++++----- 7 files changed, 14 insertions(+), 9 deletions(-) -- 1.7.5.4

4 31

[alsa-devel] [PATCH] ALSA: HDA: Fix naming of input jacks for IDT parser
by David Henningsson 03 Oct '11

03 Oct '11

Actually, it's just Sigmatel that wants to make things its own way, making life harder for us in userspace. Make the driver more uniform by removing this code. -- David Henningsson, Canonical Ltd. http://launchpad.net/~diwic

2 1

[PATCH] ALSA: HDA: Fix naming of input jacks for IDT parser
by David Henningsson 03 Oct '11

03 Oct '11

The Sigmatel/IDT parser should have the same naming convention for input jacks as the other codecs have. BugLink: http://bugs.launchpad.net/bugs/859704 Signed-off-by: David Henningsson <david.henningsson(a)canonical.com> --- sound/pci/hda/patch_sigmatel.c | 14 +++----------- 1 files changed, 3 insertions(+), 11 deletions(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 987e3cf..dd977b6 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4130,22 +4130,14 @@ static int stac92xx_add_jack(struct hda_codec *codec, #ifdef CONFIG_SND_HDA_INPUT_JACK int def_conf = snd_hda_codec_get_pincfg(codec, nid); int connectivity = get_defcfg_connect(def_conf); - char name[32]; - int err; if (connectivity && connectivity != AC_JACK_PORT_FIXED) return 0; - snprintf(name, sizeof(name), "%s at %s %s Jack", - snd_hda_get_jack_type(def_conf), - snd_hda_get_jack_connectivity(def_conf), - snd_hda_get_jack_location(def_conf)); - - err = snd_hda_input_jack_add(codec, nid, type, name); - if (err < 0) - return err; -#endif /* CONFIG_SND_HDA_INPUT_JACK */ + return snd_hda_input_jack_add(codec, nid, type, NULL); +#else return 0; +#endif /* CONFIG_SND_HDA_INPUT_JACK */ } static int stac_add_event(struct sigmatel_spec *spec, hda_nid_t nid, -- 1.7.5.4 --------------040603080804090909080905--

1 0

[alsa-devel] [PATCH v4] ALSA: hdmi: expose ELD control
by Pierre-Louis Bossart 03 Oct '11

03 Oct '11

Applications may want to read ELD information to understand what codecs are supported on the HDMI receiver and handle the a-v delay for better lip-sync. ELD information is exposed in a device-specific IFACE_PCM kcontrol. Tested both with amixer and PulseAudio; with a corresponding patch passthrough modes are enabled automagically. ELD control size is set to zero in case of errors or wrong configurations. No notifications are implemented for now, it is expected that jack detection is used to reconfigure the audio outputs. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> --- sound/pci/hda/hda_eld.c | 13 +++++--- sound/pci/hda/hda_local.h | 2 + sound/pci/hda/patch_hdmi.c | 68 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 78 insertions(+), 5 deletions(-) diff --git a/sound/pci/hda/hda_eld.c b/sound/pci/hda/hda_eld.c index c34f730..f1c621d 100644 --- a/sound/pci/hda/hda_eld.c +++ b/sound/pci/hda/hda_eld.c @@ -318,6 +318,11 @@ int snd_hdmi_get_eld(struct hdmi_eld *eld, int size; unsigned char *buf; + /* + * ELD size is initialized to zero in caller function. If no errors and + * ELD is valid, actual eld_size is assigned in hdmi_update_eld() + */ + if (!eld->eld_valid) return -ENOENT; @@ -327,14 +332,13 @@ int snd_hdmi_get_eld(struct hdmi_eld *eld, snd_printd(KERN_INFO "HDMI: ELD buf size is 0, force 128\n"); size = 128; } - if (size < ELD_FIXED_BYTES || size > PAGE_SIZE) { + if (size < ELD_FIXED_BYTES || size > ELD_MAX_SIZE) { snd_printd(KERN_INFO "HDMI: invalid ELD buf size %d\n", size); return -ERANGE; } - buf = kmalloc(size, GFP_KERNEL); - if (!buf) - return -ENOMEM; + /* set ELD buffer */ + buf = eld->eld_buffer; for (i = 0; i < size; i++) { unsigned int val = hdmi_get_eld_data(codec, nid, i); @@ -356,7 +360,6 @@ int snd_hdmi_get_eld(struct hdmi_eld *eld, ret = hdmi_update_eld(eld, buf, size); error: - kfree(buf); return ret; } diff --git a/sound/pci/hda/hda_local.h b/sound/pci/hda/hda_local.h index aaefa7c..04d730f 100644 --- a/sound/pci/hda/hda_local.h +++ b/sound/pci/hda/hda_local.h @@ -621,6 +621,7 @@ struct cea_sad { }; #define ELD_FIXED_BYTES 20 +#define ELD_MAX_SIZE 256 #define ELD_MAX_MNL 16 #define ELD_MAX_SAD 16 @@ -645,6 +646,7 @@ struct hdmi_eld { int spk_alloc; int sad_count; struct cea_sad sad[ELD_MAX_SAD]; + char eld_buffer[ELD_MAX_SIZE]; #ifdef CONFIG_PROC_FS struct snd_info_entry *proc_entry; #endif diff --git a/sound/pci/hda/patch_hdmi.c b/sound/pci/hda/patch_hdmi.c index 3f1f6ac..3425401 100644 --- a/sound/pci/hda/patch_hdmi.c +++ b/sound/pci/hda/patch_hdmi.c @@ -324,6 +324,66 @@ static int cvt_nid_to_cvt_index(struct hdmi_spec *spec, hda_nid_t cvt_nid) return -EINVAL; } +static int hdmi_eld_ctl_info(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct hdmi_spec *spec; + int pin_idx; + + spec = codec->spec; + uinfo->type = SNDRV_CTL_ELEM_TYPE_BYTES; + + pin_idx = kcontrol->private_value; + uinfo->count = spec->pins[pin_idx].sink_eld.eld_size; + + return 0; +} + +static int hdmi_eld_ctl_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct hdmi_spec *spec; + int pin_idx; + + spec = codec->spec; + pin_idx = kcontrol->private_value; + + memcpy(ucontrol->value.bytes.data, + spec->pins[pin_idx].sink_eld.eld_buffer, ELD_MAX_SIZE); + + return 0; +} + +static struct snd_kcontrol_new eld_bytes_ctl = { + .access = SNDRV_CTL_ELEM_ACCESS_READ | SNDRV_CTL_ELEM_ACCESS_VOLATILE, + .iface = SNDRV_CTL_ELEM_IFACE_PCM, + .name = "ELD", + .info = hdmi_eld_ctl_info, + .get = hdmi_eld_ctl_get, +}; + +static int hdmi_create_eld_ctl(struct hda_codec *codec, int pin_idx, + int device) +{ + struct snd_kcontrol *kctl; + struct hdmi_spec *spec = codec->spec; + int err; + + kctl = snd_ctl_new1(&eld_bytes_ctl, codec); + if (!kctl) + return -ENOMEM; + kctl->private_value = pin_idx; + kctl->id.device = device; + + err = snd_hda_ctl_add(codec, spec->pins[pin_idx].pin_nid, kctl); + if (err < 0) + return err; + + return 0; +} + #ifdef BE_PARANOID static void hdmi_get_dip_index(struct hda_codec *codec, hda_nid_t pin_nid, int *packet_index, int *byte_index) @@ -1193,6 +1253,14 @@ static int generic_hdmi_build_controls(struct hda_codec *codec) if (err < 0) return err; snd_hda_spdif_ctls_unassign(codec, pin_idx); + + /* add control for ELD Bytes */ + err = hdmi_create_eld_ctl(codec, + pin_idx, + spec->pcm_rec[pin_idx].device); + + if (err < 0) + return err; } return 0; -- 1.7.6.2

2 1