Kernel 2.6.18 broke the MotU Fastlane, which uses duplicate endpoint
numbers in a manner that is not only illegal but also confuses the
kernel's endpoint descriptor caching mechanism.  To work around this, we
have to add a separate usb_set_interface() call to guide the USB core to
the correct descriptors.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Reported-and-tested-by: NDavid Fries <david@fries.net>
Cc: <stable@kernel.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Added SNDRV_PCM_INFO_BATCH flag to PCM info field of some drivers that
really don't give the precise pointer value.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Reset the internal period position counter upon stream startup. This
fixes initial aplay underruns and problems related to latency picky
applications such as pulseaudio.

Bumped the version number to 1.3.14.
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Use it to clean up snd_us122l_card_used[].

Without patch unplugging of an US122L soundcard didn't reset the
corresponding element of snd_us122l_card_used[] to 0.
The (SNDRV_CARDS + 1)th plugging in did not result in creating the soundcard
device anymore.
Index values supplied with the modprobe command line were not used correctly
anymore after the first unplugging of an US122L.
Signed-off-by: NKarsten Wiese <fzu@wemgehoertderstaat.de>
Cc: stable@kernel.org
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Within 2.6.30's mergewindow, struct urb's transfer_buffer_length has become
unsigned. This changed an "int > int" comparision to an "unsigned > int" one
in snd_usb_122l.
Fix this by using a local int variable instead of urb->transfer_buffer_length
in comparisions.

Shorten playback_prep_freqn() a bit and tweak error-paths in
usb_stream_prepare_playback().
Signed-off-by: NKarsten Wiese <fzu@wemgehoertderstaat.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Cleanup only, no functional change.
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Remove open-coded memdup_user().
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

To enable periods shorter than 1 ms, we have to make sure that short
periods are only available for alternate settings that have a small
enough data packet interval.  Furthermore, the code that aligns URBs to
USB frames is now superfluous.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The data packet interval needs to be available in the audioformat
structure, together with the other audio format parameters, so that it
can be used to influence ALSA hardware parameters.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This removes the check_hw_params_convention() function because
1) it is not necessary, as the hw_rule_* functions also work correctly
   (i.e., as no-ops) when the device supports all combinations of the
   audio format parameters; and
2) it would become too complex when adding a fourth altsetting-dependent
   hardware parameter, as this would require another three loops to
   check dependecies with rate/channels/format.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When listing the device's sample formats in the stream? proc file, the
sample format number itself is rather obscure, so we better show the
format width, too.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Drop inbound packets that are smaller than expected. This has been
observed at the very beginning of the streaming transaction.

And when the hardware is in panic mode (which can only very rarely
happen in case of massive EMI chaos), mute the input channels.
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Tested-by: NMark Hills <mark@pogo.org.uk>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Limit the number of printed warnings to one in case of streaming errors.
printk() happens to be expensive, especially in code called as often as
here.
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When checking for the maximum queue length, we have to take into account
that MAX_QUEUE is measured in milliseconds (i.e., frames) while the unit
of urb_packs is whatever data packet interval the device uses (possibly
less than one frame when using high speed devices).
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When storing the channel numbers used by a format, and if the device
happens to support 32 channels, the code would try to store 1<<32 in
a 32-bit value.

Since no valid format can have zero channels, we can use 1<<(channels-1)
instead of 1<<channels so that all the channel numbers that we test for
fit into 32 bits.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fix this sparse warning:
  sound/usb/usx2y/usbusx2y.c:231:33: warning: do-while statement is not a compound statement
Signed-off-by: NHannes Eder <hannes@hanneseder.net>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Impact: Move declaration to header file.

Fix this sparse warning:
  sound/usb/usx2y/usx2yhwdeppcm.c:739:5: warning: symbol 'usX2Y_hwdep_pcm_new' was not declared. Should it be static?
Signed-off-by: NHannes Eder <hannes@hanneseder.net>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Remove the MIN_PACKS_URB symbol because other limits can force the
number of packets down to one, regardless of the value of this symbol,
and nobody has ever changed it anyway.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The CM6207 incorrectly advertises its 96 kHz playback setting as 48 kHz
in its USB device descriptor. This patch extends an existing workaround
in usbaudio.c to also cover the CM6207.

This resolves issue 0004249 in the ALSA bug tracker.
Signed-off-by: NJoris van Rantwijk <jorispubl@xs4all.nl>
Cc: <stable@kernel.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The detection of non-continuous rates (given via rate tables) isn't
processed properly (e.g. for type II).

This patch fixes and simplifies the detection code.
Tested-by: NJoris van Rantwijk <jorispubl@xs4all.nl>
Cc: <stable@kernel.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fix the snd_usbmidi_create_endpoints_midiman() function, which forgot to
set the out_interval member of the endpoint info structure for Midiman/
M-Audio devices.  Since kernel 2.6.24, any non-zero value makes the
driver use interrupt transfers instead of bulk transfers.  With EHCI
controllers, these random interval values result in unbearably large
latencies for output MIDI transfers.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Reported-by: NDavid <devurandom@foobox.com>
Tested-by: NDavid <devurandom@foobox.com>
Cc: <stable@kernel.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

For audio devices that do not have proper audio descriptors (e.g.,
Edirol UA-20), we use hardcoded parameters from our quirks list.
However, we must still read the maximum packet size from the standard
endpoint descriptor; otherwise, we might use packets that are too big
and therefore rejected by the USB core.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Cc: <stable@kernel.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Indicates fixes affecting control messages and switching of input mode
on Audio 8 DJ and Audio 4 DJ.
Signed-off-by: NMark Hills <mark@pogo.org.uk>
Acked-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Remove a duplicate control which causes an error when it is registered,
and causes later controls to not be registered. The device does not have
a fourth ground lift control.
Signed-off-by: NMark Hills <mark@pogo.org.uk>
Acked-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

In the context of the Audio 4 DJ (when compared to Audio 8 DJ), hardware
input mode 0 is not used. Expose modes 1 (line) and 2 (phono) to the user
as modes 0 and 1 respectively.
Signed-off-by: NMark Hills <mark@pogo.org.uk>
Acked-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Do not start the device with input mode undefined. Mimic the behaviour of
the Audio 8 DJ and start in phono input mode.
Signed-off-by: NMark Hills <mark@pogo.org.uk>
Acked-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Fixes a bug where an incorrect command was sent which had no effect on the
device.
Signed-off-by: NMark Hills <mark@pogo.org.uk>
Acked-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Remove NOP hwdep ops in sound drivers.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Rather use printf format instead of hardcoding prefix like 0x.
A next step would be to predefine certain formats.
Signed-off-by: NAndreas Bergmeier <lcid-fire@gmx.net>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Distribute the packets evenly among the URBs, instead of making all URBs
except the last one to have the maximum size.  This makes the timing of
pointer updates more regular and removes some special cases from the
code.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Once our URBs contain enough packets, queueing more URBs does not give
us any additional underrun protection (as we use double-buffering) but
just increases latency unnecessarily.  Therefore, we try to limit the
queue length to some reasonable value.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

When sending a silence URB (before playback has started, or when it is
paused), use the number of frames that would be normally sent instead of
a single frame so that the rate at which completion interrupts arrive is
consistent.
Signed-off-by: NClemens Ladisch <clemens@ladisch.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Ignore errors (wrong usb interface data) found when using the serato
scratch live box with alsa

Thus the alsa controls can be accessed (beware: they don't work though -
but at least it's one ugly error message less)
Signed-off-by: NAndreas Bergmeier <lcid-fire@gmx.net>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

- Added support for two new audio devices from Native Instuments,
  'Audio4DJ' and 'GuitarRig mobile'
- Add missing statement about 'Session IO' in Kconfig help text
- Version number bumped to 1.3.11
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Cache mixer values in usb-audio driver to reduce too excessive
accesses to the hardware.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The Logitech QuickCam Pro 9000 does not appear to any product identification
strings in its USB device descriptor. Therefore it receives a device name of
"USB Device 0x46d:0x990". Th e attached patch below adds a USB quirk to
provide a more friendly name.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Signed-off-by: NAndrea Borgia <andrea@borgia.bo.it>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>