The dlopen() function might fail also for another reason than
a missing file, thus return the error string from dlerror().
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Check the availability of PTHREAD_MUTEX_RECURSIVE in configure script
and use it only when possible. A fairly old version of glibc still
seems working, but just to be sure.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Currently the top-level config directory is specified only via
configure script option, and is fixed after that. It's inconvenient
when the library is moved to another base directory, or if you want to
use a library code (e.g. with $LD_PRELOAD) with the incompatible
config setups.
This patch allows user to override the top-level config path via the
environment varialbe, $ALSA_CONFIG_DIR. For that, a new helper
function, snd_config_topdir(), was introduced, and the codes referring
to the top config dir have been modified to handle it dynamically.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Users can include file by giving a relative path or just a file name via
alsaconf syntax <xxx>. ALSA conf will search the file in top configuration
directory and additional configuration directories defined by users via
alsaconf syntax <searchdir:xxx>.
ALSA conf will search and open an included file in the following order
of priority:
1. directly open the file by its name;
2. search for the file name in top config direcotry "/usr/share/alsa/";
3. search for the file name in additional configuration directories
specified by users, via alsaconf syntax
<searchdir:relative-path/to/user/share/alsa>;
These directories will be searched in the same order as their
definition.
The configuration directories defined by a file will only be used to
search the files included by this file.
Signed-off-by: Mengdong Lin <mengdong.lin@linux.intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Most of open functions in alsa-lib have the call pattern:
snd_config_update();
return snd_xxx_open(x, snd_config, ...);
This means that the toplevel config gets updated, and passed to a
local open function. Although snd_config_update() itself has a
pthread mutex to be thread safe, the whole procedure above isn't
thread safe. Namely, the global snd_config tree may be deleted and
recreated at any time while the open function is being processed.
This may lead to a data corruption and crash of the program.
For avoiding the corruption, this patch introduces a refcount to
config tree object. A few new helper functions are introduced as
well:
- snd_config_update_ref() does update and take the refcount of the
toplevel tree. The obtained config tree has to be freed via
snd_config_unref() below.
- snd_config_ref() and snd_config_unref() manage the refcount of the
config object. The latter eventually deletes the object when all
references are gone.
Along with these additions, the caller of snd_config_update() and
snd_config global tree in alsa-lib are replaced with the new helpers.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
If you build with --disable-pcm, the rawmidi code fails to link because
it uses the page_size helper declared in local.h, but the page_* funcs
are in the pcm mmap module. Move these funcs into conf.c so that they
are always available to the rest of the system.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The configure script blindly adds -D_GNU_SOURCE to all build settings,
even on non-GNU systems. This isn't too much of a big deal (even if
it uses the wrong variable -- CFLAGS instead of CPPFLAGS), except that
the alsa-lib source itself determines whether to use GNU features when
this is defined (such as versionsort). So when we build on non-glibc
systems, we get build failures like:
src/ucm/parser.c:1268:18: error: 'versionsort' undeclared (first use in this function)
#define SORTFUNC versionsort
^
src/ucm/parser.c:1272:54: note: in expansion of macro 'SORTFUNC'
err = scandir(filename, &namelist, filename_filter, SORTFUNC);
^
The correct way to add these flags is to use the autoconf helper
AC_USE_SYSTEM_EXTENSIONS. Unfortunately, that triggers some more
bugs in the alsa build. This macro adds defines to config.h and
not directly to CPPFLAGS, so it relies on files correctly including
config.h before anything else. A number of alsa files do not do
this leading to build failures. The fix there is to shuffle the
includes around so that the local ones come first.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Otherwise, they get misinterpreted as argument separators
in USB-Audio PCM definitions, and thus prevent SPDIF blacklist entries
from working.
While at it, add my Logitec C910 webcam to the SPDIF blacklist.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
PTHREAD_RECURSIVE_MUTEX_INITIALIZER_NP is not in POSIX, as _NP
(non-portable) suggests.
exposing such a symbol in musl libc would lock in the ABI for all
times and makes it impossible to do future changes to the under-
lying struct without hideous symbol versioning hacks.
use the portable way instead: pthread_once was designed for such
cases.
Signed-off-by: Timo Teräs <timo.teras@iki.fi>
Tested-by: John Spencer <maillist-alsa@barfooze.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Fixes an issue where a variable is used undeclared, which can cause seg
faults on some systems if the configuration file is not formatted
properly.
Signed-off-by: Allan Wirth <allan@allanwirth.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
versionsort() is a GNU-ism and can't be relied on for non-GNU systems.
[modified to define SORTFUNC instead of copying lines by tiwai]
Signed-off-by: Arun Raghavan <arun.raghavan@collabora.co.uk>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Per POSIX:
L Specifies that a following a, A, e, E, f, F, g, or G conversion
specifier applies to a long double argument.
L is only intended to be used with long doubles, not long long ints.
the proper way is to use "ll" instead.
Signed-off-by: John Spencer <maillist-alsa@barfooze.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When this is done, *.conf files can be placed in that directory and they
will be processed by as if they were included directly.
A directory (typically /usr/share/alsa/alsa.conf.d/) has been
added into the distribution.
v2: Used existing conf syntax rather than processing via autotools
v3: Split file loading into separate function and made error handling
more consistent.
Signed-off-by: Colin Guthrie <colin@mageia.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Coverity Static Analysis helps developers find hard-to-spot,
yet potentially crash-causing defects early in the development phase,
reducing the cost,time, and risk of software errors.
This patch has fix for situations where variable can be NULL
but not been checked beforehand
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Avoid configuration file processing races when multiple threads call
the *open() functions together (for example using alsaloop with
multiple -T jobs can reproduce this issue).
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
setlocale() is not thread-safe. It can actually trigger a crash if
another thread uses locale informations at the same time in the process.
Library code should use POSIX newlocale/duplocale/uselocale/freelocale
instead. Those functions only change the locale data for the calling
thread.
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
This reverts commit 96da0c842d.
This way of fix brokes card-specific configuration loading.
See http://bugzilla.redhat.com bug#521988 for details.
Appropriate way to handle this problem is to fix the dmix configuration file.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Reported by Kevin Goodsell.
Summary: load_for_all_cards fails with existing configuration elements
In snd_config_hook_load_for_all_cards, the first call to
snd_config_search attempts to locate an existing configuration node with
the name of the driver. Typically none is found, and everything is
good. However, if such a node is located, the next line assumes it is a
leaf node with type 'string' and calls snd_config_get_string to fetch
the string value. If this fails, the entire hook is abandoned.
Because of this, setting something like the following in asoundrc:
cards.<driver name>.foo 0
is sufficient to disable the entire card-specific configuration.
As a concrete example, I have a HDA-Intel sound card. dmix.conf includes
a way to set period_size, period_time, and periods by using
configuration elements of the form cards.<driver name>.pcm.dmix.<var>.
In ~/.asoundrc I add
cards.HDA-Intel.pcm.dmix.period_size 1024
This will cause HDA-Intel.conf to fail to load, and the pcm defined in
default.conf will fail to find the device-specific pcm
cards.HDA-Intel.pcm.default, and fall back on the default pcm using
plughw. By attempting to configure dmix, I have disabled it.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When adding a configuration node to another, check that the child node
does not already have a parent. Otherwise, the old parent's children
list would become corrupted.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Make sure that we do not crash when encountering configuration nodes
with a NULL string value, or that at least we run into an assert().
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
snd_config_add() checks for duplicate ids, but it was possible to create
duplicates by adding a note and changing the id afterwards with
snd_config_set_id(); so we have to add a check there, too.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Make sure that we do not crash when encountering configuration nodes
with a NULL id. Furthermore, since we cannot avoid having NULL ids
anyway, allow the id of a top-level node to be reset to NULL.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
The variable $ALSA_CONFIG_PATH specifies the config path, but the current
code accepts the empty string and results in a mysterious error because
no config file is found.
This patch fixes the check of the variable and takes the default value
if the string is empty.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Added --with-configdir and --with-plugindir options to configure
which specify the directories for config files and plugin objects
respectively. The default paths when these options are not
specified are unchanged.
- 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
safe_strtoll() now accepts numbers in any base. It formerly assumed that
its input was a decimal number, which had the consequence that
hexadecimal or octal numbers would be parsed as strings when occurring
outside of parameter lists.
This obsoletes some workarounds in the file permission parsing code that
relied on this bug.
From: Pierre Ossman <drzeus-list@drzeus.cx>
Freeing the global config update structure when it had been created
without any configs present caused an assertion to trigger. Since this
is a valid scenario and the assertion didn't really protect against
anything, it should simply be removed. Attached patch does exactly that.
