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

[alsa-devel] Stuttering 100mA and 500mA MaxPower usb audio investigation on bus powered and external powererd usb hubs
by Jelle de Jong 10 Nov '08

10 Nov '08

Hello everybody, I have been testing several scenarios to find some glues why some usb audio devices did not fully work and gave stuttering sounds I found the following: usb devices define a max power attribute during there enumeration process. I have some usb devices that have this at 500mA and some that have this on 100mA. I also have some usb hubs that have external extra power supply and some that are only a hub with bus power. Here it comes: Make sure the 4 port usb hub is full with for example: - usb keyboard (100mA) - usb mouse (100mA) - usb audio (100mA or 500mA) - usb stick (200mA/500mA) - usb hub with external power (500mA MaxPower) (AC adapter delivers 1A) - usb hub without external power (100mA MaxPower) Make sure to use audio with several tones low tones (most dvds will do) When connecting a 100mA usb audio devices on a hub with external power it will give stuttering sound!! (most usb audo devices are 100mA and most hubs are external powered due to more then one usb stick) When connecting a 500mA usb audio device on a hub with external power it will work perfectly. When connecting a 100mA or a 500mA usb audio device on a hub without external power it will work perfectly. I have the idea there is something wrong with the power distribution system when using external powered usb hubs... Any ideas how to correctly address this problem/bug? Best regards, Jelle

2 2

[alsa-devel] Backported sbxfi driver (UNTESTED!)
by Romain 10 Nov '08

10 Nov '08

Hi, I've tested the sbxfi driver and it works great on my setup. Had little problems before getting it running : i had to comment out the snd_printk stuff, otherwise the module wouldn't load (snd_printk* symbols not found, in dmesg) This was probably not the cleanest way to deal but it's just debug messages and i'm new wrt kernel driver hacking. I also had to disable my onboard audio in BIOS, otherwise the X-Fi would not get picked up. i'm using Fedora 9 with Linux 2.6.26.6-79.fc9.i686 kernel (standard supply from fedora repos),and pulseaudio I'm using with standard 2.1 jack output, i do not yet own home cinema speakers so i couldn't test all the cabling options. Thanks for this great work, Best Regards, Romain.

1 0

[alsa-devel] [PATCH] ALSA: hda - Add another HP model (6730s) for AD1884A.
by Michel Marti 09 Nov '08

09 Nov '08

Signed-off-by: Michel Marti <mma(a)objectxp.com> --- sound/pci/hda/patch_analog.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index d3fd432..ee62e31 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -3861,6 +3861,7 @@ static const char *ad1884a_models[AD1884A_MODELS] = { static struct snd_pci_quirk ad1884a_cfg_tbl[] = { SND_PCI_QUIRK(0x103c, 0x3030, "HP", AD1884A_MOBILE), SND_PCI_QUIRK(0x103c, 0x3056, "HP", AD1884A_MOBILE), + SND_PCI_QUIRK(0x103c, 0x3614, "HP 6730s", AD1884A_MOBILE), SND_PCI_QUIRK(0x17aa, 0x20ac, "Thinkpad X300", AD1884A_THINKPAD), {} }; -- 1.5.6.5

2 2

[alsa-devel] [PATCH] ALSA: hdsp: check for iobox and upload firmware during ioctl
by Tim Blechmann 09 Nov '08

09 Nov '08

currently, the error message when trying to run hdspmixer or hdspconf if the breakout box is not connected is somehow misleading, since it asks the user to upload the firmware. this patch adds a test, whether the breakout box is connected and tries to upload the firmware in the case, that it is not present, e.g. because of power failures of the breakout box. Signed-off-by: Tim Blechmann <tim(a)klingt.org> --- sound/pci/rme9652/hdsp.c | 17 +++++++++++++---- 1 files changed, 13 insertions(+), 4 deletions(-) diff --git a/sound/pci/rme9652/hdsp.c b/sound/pci/rme9652/hdsp.c index cf9eb41..5bd0be6 100644 --- a/sound/pci/rme9652/hdsp.c +++ b/sound/pci/rme9652/hdsp.c @@ -4548,11 +4548,18 @@ static int snd_hdsp_hwdep_ioctl(struct snd_hwdep *hw, struct file *file, unsigne { struct hdsp *hdsp = (struct hdsp *)hw->private_data; void __user *argp = (void __user *)arg; + int err; switch (cmd) { case SNDRV_HDSP_IOCTL_GET_PEAK_RMS: { struct hdsp_peak_rms __user *peak_rms = (struct hdsp_peak_rms __user *)arg; + if ((err = hdsp_check_for_iobox (hdsp)) < 0) + return err; + + if ((err = hdsp_check_for_firmware(hdsp, 1)) < 0) + return err; + if (!(hdsp->state & HDSP_FirmwareLoaded)) { snd_printk(KERN_ERR "Hammerfall-DSP: firmware needs to be uploaded to the card.\n"); return -EINVAL; @@ -4572,10 +4579,12 @@ static int snd_hdsp_hwdep_ioctl(struct snd_hwdep *hw, struct file *file, unsigne unsigned long flags; int i; - if (!(hdsp->state & HDSP_FirmwareLoaded)) { - snd_printk(KERN_ERR "Hammerfall-DSP: Firmware needs to be uploaded to the card.\n"); - return -EINVAL; - } + if ((err = hdsp_check_for_iobox (hdsp)) < 0) + return err; + + if ((err = hdsp_check_for_firmware(hdsp, 1)) < 0) + return err; + spin_lock_irqsave(&hdsp->lock, flags); info.pref_sync_ref = (unsigned char)hdsp_pref_sync_ref(hdsp); info.wordclock_sync_check = (unsigned char)hdsp_wc_sync_check(hdsp); -- 1.5.6.3

2 1

[alsa-devel] Intel G45 HDMI audio - missing samples
by Shane W 08 Nov '08

08 Nov '08

Hi, One issue with Intel audio over HDMI on a DG45-ID motherboard. When playing a short sample via HDMI, it misses the first quarter second or so of the audio. This is particularly noticable when using text to speech as the first word is lost. I can get around this by having a seperate program hold the audio device open while using text to speech, that seems to get the whole sample but it's a cludgy workaround. The same issue does not exist on the optical port. Best, Shane

1 0

[alsa-devel] Intel HDA Headphones vs. Speaker Mute
by Richard Spindler 07 Nov '08

07 Nov '08

Hi, the builtin sound card in my laptop does not behave as it should, when running it with the ALSA system. While there are separate mixer controls for headphones and speakers, the Master control levels both, speakers and headphones simultaneously, while the headphone control does not act at all. Also in my opinion the speakers should mute automatically as soon as I plug in the headphones, this does not happen too. I am running Ubuntu 8.04.1: $ uname -a Linux monostatos 2.6.24-19-generic #1 SMP Wed Aug 20 22:56:21 UTC 2008 i686 GNU/Linux $ lspci -nn |grep Audio 00:1b.0 Audio device [0403]: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller [8086:27d8] (rev 02) If you need any more information about that machine, or if there is anything I can do to help fix that problem, do not hesitate to ask. Cheers -Richard PS.: I also published a lengthy rant, and some ideas about that problem at http://propirate.net/oracle/archives/2008/11/05/alsa-headaches-erm-headphon…

2 1

[alsa-devel] Hints for usb audio patch
by LCID Fire 07 Nov '08

07 Nov '08

Hi. I have a patch lying around for some month now and would need some hints how to correctly implement it. 1. What's the correct way to save the state of a special feature? There is a switch which turns on a certain feature in the usb mixer. Looking at the struct usb_mixer_interface I implemented it similar to "audigy2nx_leds" - by adding a new u8 (and the necessary functions). 2. Should one rather set ignore_ctl_error to 1 (which I currently do) or sort the errors out by adding a handling for a specific device? 3. For the switch to active I need an interrupt message for which I implemented snd_usb_interrupt_trans since I did not find any other convenience function to do this.

2 1

[alsa-devel] [PATCH 0/4] ASoC updates
by Mark Brown 07 Nov '08

07 Nov '08

The following changes since commit 0ee4663617fb0f78cec4cc6558a096ccbd8c3ffc: Takashi Iwai (1): ALSA: ASoC - Remove unnecessary inclusion of linux/version.h are available in the git repository at: git://opensource.wolfsonmicro.com/linux-2.6-asoc for-tiwai David Anders (1): ASoC: Add new parameter to s3c24xx_pcm_enqueue Marek Vasut (1): ASoC: Add Palm/PXA27x unified ASoC audio driver Mark Brown (1): ASoC: Remove core version number Troy Kisky (1): ASoC: TLV320AIC23B Support more sample rates arch/arm/mach-pxa/include/mach/palmasoc.h | 13 ++ include/sound/soc.h | 2 - sound/soc/codecs/tlv320aic23.c | 227 +++++++++++++++++++----- sound/soc/pxa/Kconfig | 9 + sound/soc/pxa/Makefile | 2 + sound/soc/pxa/palm27x.c | 269 +++++++++++++++++++++++++++++ sound/soc/s3c24xx/s3c24xx-pcm.c | 12 +- sound/soc/soc-core.c | 1 - 8 files changed, 483 insertions(+), 52 deletions(-) create mode 100644 arch/arm/mach-pxa/include/mach/palmasoc.h create mode 100644 sound/soc/pxa/palm27x.c

5 14

[alsa-devel] snd_ctl compiler warnings
by Alan Horstmann 07 Nov '08

07 Nov '08

The code I am working with at present includes unremarkable lines such as: snd_ctl_elem_value_t *line_in_vol; snd_ctl_elem_value_alloca(&line_in_vol); Building this with a cross compiler for arm based on gcc 4.2.1 gives this warning for the 2nd line: : warning: the address of 'line_in_vol' will always evaluate as 'true' but an older compiler on the desktop does not show this. Would this warning be expected in these circumstances, or is the compiler being over zealous? Am I missing something or is there a straightforward way to avoid such warnings? (Code runs OK). Thanks Alan

2 4

[alsa-devel] [PATCH] ALSA: have snd_BUG_ON() always refer to arguments
by Mike Frysinger 07 Nov '08

07 Nov '08

The snd_BUG_ON() macro should always expand its argument even if we're forcing it to false. This kills off unused warnings that did not exist before the snd_assert() -> snd_BUG_ON() conversion. Signed-off-by: Mike Frysinger <vapier(a)gentoo.org> --- include/sound/core.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/include/sound/core.h b/include/sound/core.h index 35424a9..87de9fc 100644 --- a/include/sound/core.h +++ b/include/sound/core.h @@ -387,7 +387,7 @@ void snd_verbose_printd(const char *file, int line, const char *format, ...) #define snd_printd(fmt, args...) /* nothing */ #define snd_BUG() /* nothing */ -#define snd_BUG_ON(cond) ({/*(void)(cond);*/ 0;}) /* always false */ +#define snd_BUG_ON(cond) ((cond) && 0) /* always false */ #endif /* CONFIG_SND_DEBUG */ -- 1.6.0.3

3 14