Inner this library, layouts of all structures are public. At a compilation
time, each size of the structures can be calculated. It means that we can
use automatic variable instead of calling alloca(3) to program this
library because in both ways storages are kept on stack frame of process
VMA. Besides, the usage of automatic variables requires less instructions
than calls of alloca(3). Furthermore, alloca(3) is not described in any
C language standards.
This commit replaces calls of alloca(3) just for structures with automatic
variables, for control features.
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When a hint description has only either device_input or device_output,
we shouldn't handle it as a full duplex but rather a single
direction. In that way, we can avoid to list up a playback stream
like dmix or surround51 as a capture stream in the namehint.
Reported-by: Trent Reed <treed0803@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Due to the uninitialized field before the error path, passing an
invalid iface argument may result in a bad free() call. Initialize
the fields properly beforehand.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The current code of snd_device_name_hint() has a bug when listing up
devices without the device index (e.g. ctl). Because it assigns the
default device index 0 unconditionally and it has a check at the later
point to filter entries with dev >= 0, it ended up with empty
outputs.
The fix is simply to remove the bogus assignment of dev = 0.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
snd_device_name_hint() tries to free the allocated list at the error
path via snd_device_name_free_hint(). But snd_device_name_free_hint()
expects a list terminated by NULL while snd_device_name_hint() doesn't
add it. Adding it may again result in an error and thus isn't
guaranteed to work. Hence we can't add NULL at the error path.
Instead, now the code always allocates one entry more, and zero-clears
the newly allocated beforehand to guarantee the NULL termination.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Sometimes a hook manipulates the config tree, which makes currently
running iterators point to freed memory. As a workaround, make two
copies, one for the iterators and another for the hooks.
BugLink: https://bugs.launchpad.net/bugs/1008600
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This commit and its parent make the function reentrant.
Signed-off-by: Jerome Forissier <jerome@taodyne.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This is the first step towards making this function reentrant.
Signed-off-by: Jerome Forissier <jerome@taodyne.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This tries to fix the issue when logical device indexes does not match the
hardware device indexes (like hdmi -> 0:3, 1:7, 2:8, 3:9).
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Get_card_name() can be called more than once on the same list, so it
must free the previous list->cardname before replacing it.
Signed-off-by: John Lindgren <john.lindgren@tds.net>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Card-independent devices such as "null" or "pulse" should only be
added once, not once for each card.
Signed-off-by: John Lindgren <john.lindgren@tds.net>\
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
list->card is wrongly assumed to be initialized, but the previous
initialization is within a conditional that is false when only
card-independent devices are found. (This is the case when searching
for mixers on my system; the end result is that the "pulse" mixer is
listed three times.)
Signed-off-by: John Lindgren <john.lindgren@tds.net>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
snd_device_name_hint() corrupts the config name space after its call.
This results in the error from the suceeding calls of snd_pcm_open()
after snd_device_name_hint().
The bug is in try_config() in namehint.c; it calls snd_config_delete(res)
but res can be two different objects in the function. One is the object
obtained via snd_config_search_definition(), and another is the one from
snd_config_search_alias_hooks(). The former is the expanded objects,
thus it should be freed. But, the latter is a reference, and must not be
freed.
This patch adds the check to free or not.
Reported-by: John Lindgren <john.lindgren@tds.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This makes it possible to write them to .data.rel.ro or to .rodata if
there is no relocation involved (arrays of character arrays).
Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
get_dev_name() checks wrongly the device_output, and it doesn't check
properly the case when only device is set and device_input and output
are unset. This resulted in listing of non-existing HDMI device, for
example. The patch fixes both issues.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
- add long card name to device description
- create empty PCM plugin to allow right hint description parsing
- reorder devices in alsa.conf
- make namehint more configurable (using default.namehint.showall switch)
- add two levels basic and exteded for hints to default configuration files
- do not show direct device aliases
- removed all known memory leaks
- add snd_device_name_hint() and snd_device_name_free_hint() functions
- add snd_ctl_iface_conf_name() functions
- do not accept parameters for the plugin definition without @args section
- add defaults.pcm.dmix.card/device and dsnoop.card/device definitions
- add hints for HDA-Intel.conf, pcm/dmix.conf, pcm/dsnoop.conf and alsa.conf
- add test/namehint test utility
- doxygen related cleanups
