Release v1.2.3

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

dlsym: fix the snd_dlopen(), incorrect variable use

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: USB-Audio: Add HP Thunderbolt Dock in the IEC958 blacklist

Both headset and speaker on HP Thunderbolt Dock don't support IEC958
interface.

Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

dlsym: fix again snd_dlpath()

There is double lock in 8d866e4cd886991f605399af6bd6cf4ac6eea3e2 .

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: Fix memory leak at snd_pcm_new when THREAD_SAVE_API is defined

The pthread_mutexattr_t object should be destroyed by calling
pthread_mutexattr_destroy(), otherwise it may cause a potential
memory leak due to the different implement of pthread_mutexattr_init()

Signed-off-by: chunxu.li <chunxuxiao@gmail.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix the wrong return value in uc_mgr_open_ctl()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: remove strict: in ${OpenName} substitution

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix the error path in parse_toplevel_path()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix the wrong uc_mgr_open_ctl() call in if_eval_control_exists()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

dlsym: fix the reference to the origin lib directory

There is reference to the variable on heap in
commit 33089f30a316060094dfa95a2949fd479299c7 .

Cache the origin directory better.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

use-case.h: define Linked variable

We have a requirement to "hide" some hardware drivers, because
other (main) UCM configuration will refer them.

BugLink: https://github.com/alsa-project/alsa-ucm-conf/issues/30
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

tlv: implement nearest rounding in snd_tlv_convert_from_dB() for xdir == 0

We should round-up values to nearest to get more precise results.

BugLink: https://github.com/alsa-project/alsa-utils/issues/35
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

control: use more precise rounding in snd_ctl_ascii_value_parse()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: remove debug printf() in snd_config_hook_load()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: clarify -ENODATA description (recovery, event)

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: fix spelling in documentation for -EBADFD

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>

pcm: add documentation for -ENODATA error code

This error code can be used e.g. with echo reference PCM devices
(the SND_USE_CASE_MOD_ECHO_REF UCM token).

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>

snd_dlopen: implement the relocatable version for glibc

BugLink: https://github.com/alsa-project/alsa-lib/issues/34
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

snd_dlopen: do not use absolute plugin path for snd_dlopen() calls

In commit b906db19ef9f4fb7886003072e550391cc1ac651, the snd_dlopen()
implements the automatic lookup to the ALSA_PLUGIN_DIR directory.
It is not necessary to add the absolute paths in callers now.

The plugin names are also searched in ld.so.conf paths as the fallback now,
but it should not be a big problem.

BugLink: https://github.com/alsa-project/alsa-lib/issues/34
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: add support for XDG_CONFIG_HOME search path

Include $XDG_CONFIG_HOME/alsa/asoundrc _or_ ~/.config/alsa/asoundrc as
the last user-defined configuration.

BugLink: https://github.com/alsa-project/alsa-lib/issues/49
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: load function (hook) - add OR support

While I was trying to work on XDG_CONFIG_HOME support, I though
that it may be nice to support OR for the file specification.

But then I found that we can already do this via the refer hook.

I commit this change anyway, because the snd_config_hook_load()
function is more clean now.

The OR is implemented with three | characters like:

  "~/.asoundrc|||~/.asoundrc2|||/opt/some/other/path"

(first file found wins)

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: USB-Audio.conf - add another card name for Asus Xonar 5

BugLink: https://github.com/alsa-project/alsa-lib/issues/48
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: return immediately when the state is SETUP in snd_pcm_drain()

We are already in the target state. Do not call the plugin callback.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: fix snd_pcm_drain() excluding SETUP state from valid states

once draining is done, the pcm enters the SETUP state, which ought to
be valid for snd_pcm_drain()

signed-off-by: Sylvain BERTRAND <sylvain.bertrand@legeek.net>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: implement CardIdByName substitution

The syntax is ${CardIdByName:CARDNAME[#INDEX]}.

The CARDNAME is the ALSA's soundcard name (short form).
The INDEX is the instance (0 = first, 1 = second etc.).

Example: ${CardIdByName:HDA Intel PCH}
(which is identical to ${CardIdByName:HDA Intel PCH#0})

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: shuffle code in compound_merge()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

github actions: ubuntu - fix the non-interactive build

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute arguments in sequences only for syntax 3+

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute RenameDevice and DeleteDevice lists

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute Comment in Transition and Device

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute verb name and file field

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute component sequence string

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute device strings in the device lists

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute device / modifier names, too

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: implement the toplevel ucm configuration file parser

There is a big issue to validate all possible configuration paths.
Let create ucm2/ucm.conf file which describe the lookups. It may
be also customized later to follow the kernel-side development.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute ${CardNumber}

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute ${OpenName}

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: Define/DefineRegex is supported in Syntax 3

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: String condition - implement Empty

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: handle strict: prefix correctly for the UCM card name

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute also value strings

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: allow to specify the toplevel directory using '/' as first character in File path

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: config - change the in-place include evaluation order

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: allow syntax version 3

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: substitute arguments in sequences

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: configuration - add DefineRegex

Syntax:

  DefineRegex.VariableName {
    String "something to match"
    Regex "(me)thi(ng)"
  }

Result:

- configuration variable "VariableName" is equal to "mething"
- configuration variable "VariableName1" is equal to "me"
- configuration variable "VariableName2" is equal to "ng"

Notes:

The "Flags" string in the DefineRegex compound may specify
the compilation flags (e = extended, i = ignore case,
s = no substitutes, n = newline). See 'man 3 regex' for
more details.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: configuration - allow to define the configuration variables

It may be useful for the library files to use the runtime configuration
variables.

Example:

 Define.Var1 "hw:${CardId},2"
 Value.PlaybackPCM "${var:Var1}"

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: config - substitute File string to allow variables in include

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: configuration - substitute ${ConfDir} and ${ConfTopDir}

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: configuration - implement in-place Include

An example:

Include {
File "Inc.conf"
Before.SectionDevice "Mic"
}

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: rename _once command to _boot command

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: rename SectionOnce to BootSequence

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix compilation error in set_defaults_user()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix SectionOnce comment

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: initialize mgr->once_list

BugLink: https://github.com/alsa-project/alsa-lib/pull/50
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: handle 'set _defaults' command

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: handle 'set _once' command

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: execute SectionDefaults lately (when the first verb is set)

We need a way to set the SectionOnce at first before SectionDefaults.
It also does not make sense to run SectionDefaults when the API is
used only to query for values or so.

Execute the SectionDefaults commands only when the first verb
is set (before the verb specific commands).

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: parse SectionOnce section in the master UCM configuration file

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

topology: Add missing ATTRIBUTE_UNUSED

... to shut up the compiler warnings.

Fixes: b6c9afb4f59b ("topology: implement snd_tplg_decode")
Signed-off-by: Takashi Iwai <tiwai@suse.de>

pcm: rate: Fix uninitialized variable warning

The recent gcc warning indicates the uninitialized variable commit_err:
  pcm_rate.c:1104:6: warning: ‘commit_err’ may be used uninitialized in this function [-Wmaybe-uninitialized]
     if (commit_err < 0)
         ^

Add a proper initialization to commit_err.

Fixes: 29041c522071 ("fix infinite draining of the rate plugin in SND_PCM_NONBLOCK mode")
Signed-off-by: Takashi Iwai <tiwai@suse.de>

pcm: rate: Fix compile warning wrt bit ops and comparison

We've got a gcc warning:
  pcm_rate.c: In function ‘snd_pcm_rate_drain’:
  pcm_rate.c:1090:19: warning: suggest parentheses around comparison in operand of ‘&’ [-Wparentheses]
     if (pcm->mode & SND_PCM_NONBLOCK != 0) {
                   ^

Drop the zero comparison for fixing the warning and for simplicity.

Fixes: 29041c522071 ("fix infinite draining of the rate plugin in SND_PCM_NONBLOCK mode")
Signed-off-by: Takashi Iwai <tiwai@suse.de>

ucm: merge the array items from the condition blocks

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

topology: use snd_config_is_array() function

... instead the own array test.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: add snd_config_is_array() function

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: use bool type for the join member in the config structure

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: change namehint for pcm.front

It seems that we have regressions for pcm.front (some
applications are using this device also for the default
raw input). Change the description and allow both
directions.

Fixes: https://github.com/alsa-project/alsa-lib/issues/39
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

header: Fix build error with old kernel headers

Old development environment doesn't necessarily contain the headers
defining __kernel_long_t that is recently used for y2038 timespec
conditional.  Define it explicitly in such a case.

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

pcm: dmix: assembly: give more freedom over the constraints

Minor changes that could help the compiler to produce a
better (smaller/faster) code.

Signed-off-by: Frédéric Recoules <frederic.recoules@orange.fr>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

pcm: dmix: assembly: refactor the PIC mode

Newer versions of GCC (> 5.0) accept that ebx is declared
in the clobber list even in PIC mode. Meanwhile, even
unlikely, the compiler may use ebx as base address of
one of the memory entry, making subsequent access to
them unreliable ('size', 'dst_step', 'src_step', 'sum_step').

Adding ebx in the clobber solves the problem.
By the way, the entry 'old_ebx' is no longer required.

Signed-off-by: Frédéric Recoules <frederic.recoules@orange.fr>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

pcm: dmix: assembly: fix the interface for safety (mmx)

- add mm0 in the clobber list if the compiler
  is aware of the mmx technology;
- otherwise, add the mmx aliased x87 floating point
  registers in the clobbers;
- the configure now checks if the compiler is aware of
  the MMX technology.

  The compiler assumes none of the mmx or x87 registers are used
  by the function. If it chooses to store some data in them, they
  will be overwritten by the chunk.
  Recall that any mmx instruction invalidate the whole set of
  x87 floating point registers.

  Note: currently does not impact the binary output.

Signed-off-by: Frédéric Recoules <frederic.recoules@orange.fr>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

pcm: dmix: assembly: fix the interface for safety

- move 'size' and 'old_ebx' in the output section
  since they are clobbered;
- add the "memory" clobber since input pointers
  are accessed;
- (minor) add the "cc" clobber since flags are
  clobbered.

  The compiler is missing some dataflow information
  about the execution of the assembly chunks.
  For instance, it assumes that 'size' remains unchanged
  and that no input pointer is accessed.
  The compiler optimizer may take advantage of these
  assumption and produce a wrong code.

Note: currently produces the same binary output.

Signed-off-by: Frédéric Recoules <frederic.recoules@orange.fr>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

pcm: dmix: assembly: change the token by symbolic names

It eases the refactoring of assembly chunk since we can now
add/remove/move entries without worrying about maintaining
the token numbering in the template.

Note: does not impact the binary output.

Signed-off-by: Frédéric Recoules <frederic.recoules@orange.fr>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

test: pcm_min - add snd_pcm_drain() call and indentation fixes

Fixes: https://github.com/alsa-project/alsa-lib/issues/46
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

fix infinite draining of the rate plugin in SND_PCM_NONBLOCK mode

The draining function of the rate plugin does not handle properly the
SND_PCM_NONBLOCK case. It can write data to the slave plugin each time the
function is called, but does not update its internal state in order to
reach a stopping condition. Use a last_commit_ptr workaround to reach such
condition.

Signed-off-by: Sylvain BERTRAND <sylvain.bertrand@legeek.net>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

topology: Use bool parser to parse boolean value

It should be safe to change this as correct values with int would be 0
or 1 and bool parser handles them correctly.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

use-case.h: add USB as allowed device name

Some USB devices are really generic where the
PCM stream can be routed to different paths.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: rate - fix the remaining size calculation in snd_pcm_rate_drain()

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: USB-Audio: Fix S/PDIF output of ASUS Xonar AE

ASUS Xonar AE is a PCI-Express card containing USB controller:

USB controller [0c03]: ASMedia Technology Inc. ASM1042A USB 3.0 Host Controller [1b21:1142]

and the actual USB sound card:

ID 0b05:180f ASUSTek Computer, Inc. XONAR SOUND CARD

As other Xonar USB sound cards, it uses second device for digital
output.

Signed-off-by: Ivan Mironov <mironov.ivan@gmail.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: dmix: fix sw_params handling of timestamp types in direct plugins

In pcms using direct plugins (dmix/dsnoop/dshare), the timestamp type could
be different from the terminating hw plugin, then the kernel driver.

Be sure such pcms have plugins using consistently the same timestamp type.

signed-off-by: Sylvain Bertrand <sylvain.bertrand@legeek.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

ucm: fix copy-n-paste typo (RemoveDevice list)

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix the device remove operation

- allow to remove device only from conflicting lists
- unify the error codes when the device is not found (ENODEV)

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

usecase: allow indexes also for modifier names

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

pcm: add chmap option to route plugin

Add a config definition "chmap" to override the channel maps in the same
way as in the hw and null plugins.

Signed-off-by: Jonas Holmberg <jonashg@axis.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

conf: fix namehint for pcm.front and pcm.iec958

As suggested in:

Fixes: https://github.com/alsa-project/alsa-lib/issues/39
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

namehint: remember the direction from the upper levels

The current code resets the direction info for each level.
Simply remove this code.

Fixes: https://github.com/alsa-project/alsa-lib/issues/39
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: fix uc_mgr_scan_master_configs()

Skip the lookup, if the top level filename does not exist.

Fixes: https://github.com/alsa-project/alsa-ucm-conf/issues/16
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

ucm: clarify the index syntax for the device names

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

topology: add back asrc to widget_map in dapm.c

Fixes: aa1bac2d04bd ("topology: add snd_tplg_save()")
BugLink: https://github.com/thesofproject/sof/issues/2543
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: USB-Audio: Add C-Media USB Headphone Set to the IEC958 blacklist

Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/317
Signed-off-by: Tanu Kaskinen <tanuk@iki.fi>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

configure: add --embed for python3-config (python 3.8)

Fixes: https://github.com/alsa-project/alsa-lib/issues/33
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

Change PCM device number of Asus Xonar U5

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: namehint - add omit_noargs to the hint section

Do not list simple surround devices in the namehint function by default.

Fixes: https://github.com/alsa-project/alsa-lib/issues/27
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

conf: change the order of PCM devices in alsa.conf

Appearently, some applications use the first device for probe or so.

Fixes: https://github.com/alsa-project/alsa-lib/issues/27
Signed-off-by: Jaroslav Kysela <perex@perex.cz>

Release v1.2.2

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

github build.yml: cosmetic changes

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

alsa.m4: another fix to not link all binaries with -latopology

Preserve LIBS when the -latopology is tested. It's added later.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

README.md: add build status badge

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

github: create test workflow for fedora and ubuntu

Signed-off-by: Jaroslav Kysela <perex@perex.cz>

do not set close-on-exec flag on descriptor if it was already set

There is no need to set this again if O_CLOEXEC is supported.

Signed-off-by: Rolf Eike Beer <eb@emlix.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>