Files
asterisk/main/translate.c
Matthew Jordan a2c912e997 media formats: re-architect handling of media for performance improvements
In the old times media formats were represented using a bit field. This was
fast but had a few limitations.
 1. Asterisk was limited in how many formats it could handle.
 2. Formats, being a bit field, could not include any attribute information.
    A format was strictly its type, e.g., "this is ulaw".
This was changed in Asterisk 10 (see
https://wiki.asterisk.org/wiki/display/AST/Media+Architecture+Proposal for
notes on that work) which led to the creation of the ast_format structure.
This structure allowed Asterisk to handle attributes and bundle information
with a format.

Additionally, ast_format_cap was created to act as a container for multiple
formats that, together, formed the capability of some entity. Another
mechanism was added to allow logic to be registered which performed format
attribute negotiation. Everywhere throughout the codebase Asterisk was
changed to use this strategy.

Unfortunately, in software, there is no free lunch. These new capabilities
came at a cost.

Performance analysis and profiling showed that we spend an inordinate
amount of time comparing, copying, and generally manipulating formats and
their related structures. Basic prototyping has shown that a reasonably
large performance improvement could be made in this area. This patch is the
result of that project, which overhauled the media format architecture
and its usage in Asterisk to improve performance.

Generally, the new philosophy for handling formats is as follows:
 * The ast_format structure is reference counted. This removed a large amount
   of the memory allocations and copying that was done in prior versions.
 * In order to prevent race conditions while keeping things performant, the
   ast_format structure is immutable by convention and lock-free. Violate this
   tenet at your peril!
 * Because formats are reference counted, codecs are also reference counted.
   The Asterisk core generally provides built-in codecs and caches the
   ast_format structures created to represent them. Generally, to prevent
   inordinate amounts of module reference bumping, codecs and formats can be
   added at run-time but cannot be removed.
 * All compatibility with the bit field representation of codecs/formats has
   been moved to a compatibility API. The primary user of this representation
   is chan_iax2, which must continue to maintain its bit-field usage of formats
   for interoperability concerns.
 * When a format is negotiated with attributes, or when a format cannot be
   represented by one of the cached formats, a new format object is created or
   cloned from an existing format. That format may have the same codec
   underlying it, but is a different format than a version of the format with
   different attributes or without attributes.
 * While formats are reference counted objects, the reference count maintained
   on the format should be manipulated with care. Formats are generally cached
   and will persist for the lifetime of Asterisk and do not explicitly need
   to have their lifetime modified. An exception to this is when the user of a
   format does not know where the format came from *and* the user may outlive
   the provider of the format. This occurs, for example, when a format is read
   from a channel: the channel may have a format with attributes (hence,
   non-cached) and the user of the format may last longer than the channel (if
   the reference to the channel is released prior to the format's reference).

For more information on this work, see the API design notes:
  https://wiki.asterisk.org/wiki/display/AST/Media+Format+Rewrite

Finally, this work was the culmination of a large number of developer's
efforts. Extra thanks goes to Corey Farrell, who took on a large amount of the
work in the Asterisk core, chan_sip, and was an invaluable resource in peer
reviews throughout this project.

There were a substantial number of patches contributed during this work; the
following issues/patch names simply reflect some of the work (and will cause
the release scripts to give attribution to the individuals who work on them).

Reviews:
 https://reviewboard.asterisk.org/r/3814
 https://reviewboard.asterisk.org/r/3808
 https://reviewboard.asterisk.org/r/3805
 https://reviewboard.asterisk.org/r/3803
 https://reviewboard.asterisk.org/r/3801
 https://reviewboard.asterisk.org/r/3798
 https://reviewboard.asterisk.org/r/3800
 https://reviewboard.asterisk.org/r/3794
 https://reviewboard.asterisk.org/r/3793
 https://reviewboard.asterisk.org/r/3792
 https://reviewboard.asterisk.org/r/3791
 https://reviewboard.asterisk.org/r/3790
 https://reviewboard.asterisk.org/r/3789
 https://reviewboard.asterisk.org/r/3788
 https://reviewboard.asterisk.org/r/3787
 https://reviewboard.asterisk.org/r/3786
 https://reviewboard.asterisk.org/r/3784
 https://reviewboard.asterisk.org/r/3783
 https://reviewboard.asterisk.org/r/3778
 https://reviewboard.asterisk.org/r/3774
 https://reviewboard.asterisk.org/r/3775
 https://reviewboard.asterisk.org/r/3772
 https://reviewboard.asterisk.org/r/3761
 https://reviewboard.asterisk.org/r/3754
 https://reviewboard.asterisk.org/r/3753
 https://reviewboard.asterisk.org/r/3751
 https://reviewboard.asterisk.org/r/3750
 https://reviewboard.asterisk.org/r/3748
 https://reviewboard.asterisk.org/r/3747
 https://reviewboard.asterisk.org/r/3746
 https://reviewboard.asterisk.org/r/3742
 https://reviewboard.asterisk.org/r/3740
 https://reviewboard.asterisk.org/r/3739
 https://reviewboard.asterisk.org/r/3738
 https://reviewboard.asterisk.org/r/3737
 https://reviewboard.asterisk.org/r/3736
 https://reviewboard.asterisk.org/r/3734
 https://reviewboard.asterisk.org/r/3722
 https://reviewboard.asterisk.org/r/3713
 https://reviewboard.asterisk.org/r/3703
 https://reviewboard.asterisk.org/r/3689
 https://reviewboard.asterisk.org/r/3687
 https://reviewboard.asterisk.org/r/3674
 https://reviewboard.asterisk.org/r/3671
 https://reviewboard.asterisk.org/r/3667
 https://reviewboard.asterisk.org/r/3665
 https://reviewboard.asterisk.org/r/3625
 https://reviewboard.asterisk.org/r/3602
 https://reviewboard.asterisk.org/r/3519
 https://reviewboard.asterisk.org/r/3518
 https://reviewboard.asterisk.org/r/3516
 https://reviewboard.asterisk.org/r/3515
 https://reviewboard.asterisk.org/r/3512
 https://reviewboard.asterisk.org/r/3506
 https://reviewboard.asterisk.org/r/3413
 https://reviewboard.asterisk.org/r/3410
 https://reviewboard.asterisk.org/r/3387
 https://reviewboard.asterisk.org/r/3388
 https://reviewboard.asterisk.org/r/3389
 https://reviewboard.asterisk.org/r/3390
 https://reviewboard.asterisk.org/r/3321
 https://reviewboard.asterisk.org/r/3320
 https://reviewboard.asterisk.org/r/3319
 https://reviewboard.asterisk.org/r/3318
 https://reviewboard.asterisk.org/r/3266
 https://reviewboard.asterisk.org/r/3265
 https://reviewboard.asterisk.org/r/3234
 https://reviewboard.asterisk.org/r/3178

ASTERISK-23114 #close
Reported by: mjordan
  media_formats_translation_core.diff uploaded by kharwell (License 6464)
  rb3506.diff uploaded by mjordan (License 6283)
  media_format_app_file.diff uploaded by kharwell (License 6464) 
  misc-2.diff uploaded by file (License 5000)
  chan_mild-3.diff uploaded by file (License 5000) 
  chan_obscure.diff uploaded by file (License 5000) 
  jingle.diff uploaded by file (License 5000) 
  funcs.diff uploaded by file (License 5000) 
  formats.diff uploaded by file (License 5000) 
  core.diff uploaded by file (License 5000) 
  bridges.diff uploaded by file (License 5000) 
  mf-codecs-2.diff uploaded by file (License 5000) 
  mf-app_fax.diff uploaded by file (License 5000) 
  mf-apps-3.diff uploaded by file (License 5000) 
  media-formats-3.diff uploaded by file (License 5000) 

ASTERISK-23715
  rb3713.patch uploaded by coreyfarrell (License 5909)
  rb3689.patch uploaded by mjordan (License 6283)
  
ASTERISK-23957
  rb3722.patch uploaded by mjordan (License 6283) 
  mf-attributes-3.diff uploaded by file (License 5000) 

ASTERISK-23958
Tested by: jrose
  rb3822.patch uploaded by coreyfarrell (License 5909) 
  rb3800.patch uploaded by jrose (License 6182)
  chan_sip.diff uploaded by mjordan (License 6283) 
  rb3747.patch uploaded by jrose (License 6182)

ASTERISK-23959 #close
Tested by: sgriepentrog, mjordan, coreyfarrell
  sip_cleanup.diff uploaded by opticron (License 6273)
  chan_sip_caps.diff uploaded by mjordan (License 6283) 
  rb3751.patch uploaded by coreyfarrell (License 5909) 
  chan_sip-3.diff uploaded by file (License 5000) 

ASTERISK-23960 #close
Tested by: opticron
  direct_media.diff uploaded by opticron (License 6273) 
  pjsip-direct-media.diff uploaded by file (License 5000) 
  format_cap_remove.diff uploaded by opticron (License 6273) 
  media_format_fixes.diff uploaded by opticron (License 6273) 
  chan_pjsip-2.diff uploaded by file (License 5000) 

ASTERISK-23966 #close
Tested by: rmudgett
  rb3803.patch uploaded by rmudgetti (License 5621)
  chan_dahdi.diff uploaded by file (License 5000) 
  
ASTERISK-24064 #close
Tested by: coreyfarrell, mjordan, opticron, file, rmudgett, sgriepentrog, jrose
  rb3814.patch uploaded by rmudgett (License 5621) 
  moh_cleanup.diff uploaded by opticron (License 6273) 
  bridge_leak.diff uploaded by opticron (License 6273) 
  translate.diff uploaded by file (License 5000) 
  rb3795.patch uploaded by rmudgett (License 5621) 
  tls_fix.diff uploaded by mjordan (License 6283) 
  fax-mf-fix-2.diff uploaded by file (License 5000) 
  rtp_transfer_stuff uploaded by mjordan (License 6283) 
  rb3787.patch uploaded by rmudgett (License 5621) 
  media-formats-explicit-translate-format-3.diff uploaded by file (License 5000) 
  format_cache_case_fix.diff uploaded by opticron (License 6273) 
  rb3774.patch uploaded by rmudgett (License 5621) 
  rb3775.patch uploaded by rmudgett (License 5621) 
  rtp_engine_fix.diff uploaded by opticron (License 6273) 
  rtp_crash_fix.diff uploaded by opticron (License 6273) 
  rb3753.patch uploaded by mjordan (License 6283) 
  rb3750.patch uploaded by mjordan (License 6283) 
  rb3748.patch uploaded by rmudgett (License 5621) 
  media_format_fixes.diff uploaded by opticron (License 6273) 
  rb3740.patch uploaded by mjordan (License 6283) 
  rb3739.patch uploaded by mjordan (License 6283) 
  rb3734.patch uploaded by mjordan (License 6283) 
  rb3689.patch uploaded by mjordan (License 6283) 
  rb3674.patch uploaded by coreyfarrell (License 5909) 
  rb3671.patch uploaded by coreyfarrell (License 5909) 
  rb3667.patch uploaded by coreyfarrell (License 5909) 
  rb3665.patch uploaded by mjordan (License 6283) 
  rb3625.patch uploaded by coreyfarrell (License 5909) 
  rb3602.patch uploaded by coreyfarrell (License 5909) 
  format_compatibility-2.diff uploaded by file (License 5000) 
  core.diff uploaded by file (License 5000) 
  


git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@419044 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-07-20 22:06:33 +00:00

1495 lines
40 KiB
C

/*
* Asterisk -- An open source telephony toolkit.
*
* Copyright (C) 1999 - 2006, Digium, Inc.
*
* Mark Spencer <markster@digium.com>
*
* See http://www.asterisk.org for more information about
* the Asterisk project. Please do not directly contact
* any of the maintainers of this project for assistance;
* the project provides a web site, mailing lists and IRC
* channels for your use.
*
* This program is free software, distributed under the terms of
* the GNU General Public License Version 2. See the LICENSE file
* at the top of the source tree.
*/
/*! \file
*
* \brief Translate via the use of pseudo channels
*
* \author Mark Spencer <markster@digium.com>
*/
/*** MODULEINFO
<support_level>core</support_level>
***/
#include "asterisk.h"
ASTERISK_FILE_VERSION(__FILE__, "$Revision$")
#include <sys/time.h>
#include <sys/resource.h>
#include <math.h>
#include "asterisk/lock.h"
#include "asterisk/channel.h"
#include "asterisk/translate.h"
#include "asterisk/module.h"
#include "asterisk/frame.h"
#include "asterisk/sched.h"
#include "asterisk/cli.h"
#include "asterisk/term.h"
#include "asterisk/format.h"
/*! \todo
* TODO: sample frames for each supported input format.
* We build this on the fly, by taking an SLIN frame and using
* the existing converter to play with it.
*/
/*! max sample recalc */
#define MAX_RECALC 1000
/*! \brief the list of translators */
static AST_RWLIST_HEAD_STATIC(translators, ast_translator);
struct translator_path {
struct ast_translator *step; /*!< Next step translator */
uint32_t table_cost; /*!< Complete table cost to destination */
uint8_t multistep; /*!< Multiple conversions required for this translation */
};
/*!
* \brief a matrix that, for any pair of supported formats,
* indicates the total cost of translation and the first step.
* The full path can be reconstricted iterating on the matrix
* until step->dstfmt == desired_format.
*
* Array indexes are 'src' and 'dest', in that order.
*
* Note: the lock in the 'translators' list is also used to protect
* this structure.
*/
static struct translator_path **__matrix;
/*!
* \brief table for converting index to format values.
*
* \note this table is protected by the table_lock.
*/
static unsigned int *__indextable;
/*! protects the __indextable for resizing */
static ast_rwlock_t tablelock;
/* index size starts at this*/
#define INIT_INDEX 32
/* index size grows by this as necessary */
#define GROW_INDEX 16
/*! the current largest index used by the __matrix and __indextable arrays*/
static int cur_max_index;
/*! the largest index that can be used in either the __indextable or __matrix before resize must occur */
static int index_size;
static void matrix_rebuild(int samples);
/*!
* \internal
* \brief converts codec id to index value.
*/
static int codec_to_index(unsigned int id)
{
int x;
ast_rwlock_rdlock(&tablelock);
for (x = 0; x < cur_max_index; x++) {
if (__indextable[x] == id) {
/* format already exists in index2format table */
ast_rwlock_unlock(&tablelock);
return x;
}
}
ast_rwlock_unlock(&tablelock);
return -1; /* not found */
}
/*!
* \internal
* \brief converts codec to index value.
*/
static int codec2index(struct ast_codec *codec)
{
return codec_to_index(codec->id);
}
/*!
* \internal
* \brief converts format to codec index value.
*/
static int format2index(struct ast_format *format)
{
return codec_to_index(ast_format_get_codec_id(format));
}
/*!
* \internal
* \brief add a new codec to the matrix and index table structures.
*
* \note it is perfectly safe to call this on codecs already indexed.
*
* \retval 0, success
* \retval -1, matrix and index table need to be resized
*/
static int add_codec2index(struct ast_codec *codec)
{
if (codec2index(codec) != -1) {
/* format is already already indexed */
return 0;
}
ast_rwlock_wrlock(&tablelock);
if (cur_max_index == (index_size)) {
ast_rwlock_unlock(&tablelock);
return -1; /* hit max length */
}
__indextable[cur_max_index] = codec->id;
cur_max_index++;
ast_rwlock_unlock(&tablelock);
return 0;
}
/*!
* \internal
* \brief converts index value back to codec
*/
static struct ast_codec *index2codec(int index)
{
struct ast_codec *codec;
if (index >= cur_max_index) {
return 0;
}
ast_rwlock_rdlock(&tablelock);
codec = ast_codec_get_by_id(__indextable[index]);
ast_rwlock_unlock(&tablelock);
return codec;
}
/*!
* \internal
* \brief resize both the matrix and index table so they can represent
* more translators
*
* \note _NO_ locks can be held prior to calling this function
*
* \retval 0, success
* \retval -1, failure. Old matrix and index table can still be used though
*/
static int matrix_resize(int init)
{
struct translator_path **tmp_matrix = NULL;
unsigned int *tmp_table = NULL;
int old_index;
int x;
AST_RWLIST_WRLOCK(&translators);
ast_rwlock_wrlock(&tablelock);
old_index = index_size;
if (init) {
index_size += INIT_INDEX;
} else {
index_size += GROW_INDEX;
}
/* make new 2d array of translator_path structures */
if (!(tmp_matrix = ast_calloc(1, sizeof(struct translator_path *) * (index_size)))) {
goto resize_cleanup;
}
for (x = 0; x < index_size; x++) {
if (!(tmp_matrix[x] = ast_calloc(1, sizeof(struct translator_path) * (index_size)))) {
goto resize_cleanup;
}
}
/* make new index table */
if (!(tmp_table = ast_calloc(1, sizeof(unsigned int) * index_size))) {
goto resize_cleanup;
}
/* if everything went well this far, free the old and use the new */
if (!init) {
for (x = 0; x < old_index; x++) {
ast_free(__matrix[x]);
}
ast_free(__matrix);
memcpy(tmp_table, __indextable, sizeof(unsigned int) * old_index);
ast_free(__indextable);
}
/* now copy them over */
__matrix = tmp_matrix;
__indextable = tmp_table;
matrix_rebuild(0);
ast_rwlock_unlock(&tablelock);
AST_RWLIST_UNLOCK(&translators);
return 0;
resize_cleanup:
ast_rwlock_unlock(&tablelock);
AST_RWLIST_UNLOCK(&translators);
if (tmp_matrix) {
for (x = 0; x < index_size; x++) {
ast_free(tmp_matrix[x]);
}
ast_free(tmp_matrix);
}
ast_free(tmp_table);
return -1;
}
/*!
* \internal
* \brief reinitialize the __matrix during matrix rebuild
*
* \note must be protected by the translators list lock
*/
static void matrix_clear(void)
{
int x;
for (x = 0; x < index_size; x++) {
memset(__matrix[x], '\0', sizeof(struct translator_path) * (index_size));
}
}
/*!
* \internal
* \brief get a matrix entry
*
* \note This function must be protected by the translators list lock
*/
static struct translator_path *matrix_get(unsigned int x, unsigned int y)
{
return __matrix[x] + y;
}
/*
* wrappers around the translator routines.
*/
static void destroy(struct ast_trans_pvt *pvt)
{
struct ast_translator *t = pvt->t;
if (t->destroy) {
t->destroy(pvt);
}
ao2_cleanup(pvt->f.subclass.format);
ast_free(pvt);
ast_module_unref(t->module);
}
/*!
* \brief Allocate the descriptor, required outbuf space,
* and possibly desc.
*/
static struct ast_trans_pvt *newpvt(struct ast_translator *t)
{
struct ast_trans_pvt *pvt;
int len;
char *ofs;
/*
* compute the required size adding private descriptor,
* buffer, AST_FRIENDLY_OFFSET.
*/
len = sizeof(*pvt) + t->desc_size;
if (t->buf_size)
len += AST_FRIENDLY_OFFSET + t->buf_size;
pvt = ast_calloc(1, len);
if (!pvt) {
return NULL;
}
pvt->t = t;
ofs = (char *)(pvt + 1); /* pointer to data space */
if (t->desc_size) { /* first comes the descriptor */
pvt->pvt = ofs;
ofs += t->desc_size;
}
if (t->buf_size) {/* finally buffer and header */
pvt->outbuf.c = ofs + AST_FRIENDLY_OFFSET;
}
ast_module_ref(t->module);
/* call local init routine, if present */
if (t->newpvt && t->newpvt(pvt)) {
ast_free(pvt);
ast_module_unref(t->module);
return NULL;
}
/* Setup normal static translation frame. */
pvt->f.frametype = AST_FRAME_VOICE;
pvt->f.mallocd = 0;
pvt->f.offset = AST_FRIENDLY_OFFSET;
pvt->f.src = pvt->t->name;
pvt->f.data.ptr = pvt->outbuf.c;
/* if the translator has not provided a format find one in the cache or create one */
if (!pvt->f.subclass.format) {
if (!ast_strlen_zero(pvt->t->format)) {
pvt->f.subclass.format = ast_format_cache_get(pvt->t->format);
}
if (!pvt->f.subclass.format) {
struct ast_codec *codec = ast_codec_get(t->dst_codec.name,
t->dst_codec.type, t->dst_codec.sample_rate);
if (!codec) {
ast_log(LOG_ERROR, "Unable to get destination codec\n");
destroy(pvt);
return NULL;
}
pvt->f.subclass.format = ast_format_create(codec);
ao2_ref(codec, -1);
}
if (!pvt->f.subclass.format) {
ast_log(LOG_ERROR, "Unable to create format\n");
destroy(pvt);
return NULL;
}
}
return pvt;
}
/*! \brief framein wrapper, deals with bound checks. */
static int framein(struct ast_trans_pvt *pvt, struct ast_frame *f)
{
int ret;
int samples = pvt->samples; /* initial value */
/* Copy the last in jb timing info to the pvt */
ast_copy_flags(&pvt->f, f, AST_FRFLAG_HAS_TIMING_INFO);
pvt->f.ts = f->ts;
pvt->f.len = f->len;
pvt->f.seqno = f->seqno;
if (f->samples == 0) {
ast_log(LOG_WARNING, "no samples for %s\n", pvt->t->name);
}
if (pvt->t->buffer_samples) { /* do not pass empty frames to callback */
if (f->datalen == 0) { /* perform native PLC if available */
/* If the codec has native PLC, then do that */
if (!pvt->t->native_plc)
return 0;
}
if (pvt->samples + f->samples > pvt->t->buffer_samples) {
ast_log(LOG_WARNING, "Out of buffer space\n");
return -1;
}
}
/* we require a framein routine, wouldn't know how to do
* it otherwise.
*/
ret = pvt->t->framein(pvt, f);
/* diagnostic ... */
if (pvt->samples == samples)
ast_log(LOG_WARNING, "%s did not update samples %d\n",
pvt->t->name, pvt->samples);
return ret;
}
/*! \brief generic frameout routine.
* If samples and datalen are 0, take whatever is in pvt
* and reset them, otherwise take the values in the caller and
* leave alone the pvt values.
*/
struct ast_frame *ast_trans_frameout(struct ast_trans_pvt *pvt,
int datalen, int samples)
{
struct ast_frame *f = &pvt->f;
if (samples) {
f->samples = samples;
} else {
if (pvt->samples == 0) {
return NULL;
}
f->samples = pvt->samples;
pvt->samples = 0;
}
if (datalen) {
f->datalen = datalen;
} else {
f->datalen = pvt->datalen;
pvt->datalen = 0;
}
return ast_frisolate(f);
}
static struct ast_frame *default_frameout(struct ast_trans_pvt *pvt)
{
return ast_trans_frameout(pvt, 0, 0);
}
/* end of callback wrappers and helpers */
void ast_translator_free_path(struct ast_trans_pvt *p)
{
struct ast_trans_pvt *pn = p;
while ( (p = pn) ) {
pn = p->next;
destroy(p);
}
}
/*! \brief Build a chain of translators based upon the given source and dest formats */
struct ast_trans_pvt *ast_translator_build_path(struct ast_format *dst, struct ast_format *src)
{
struct ast_trans_pvt *head = NULL, *tail = NULL;
int src_index, dst_index;
src_index = format2index(src);
dst_index = format2index(dst);
if (src_index < 0 || dst_index < 0) {
ast_log(LOG_WARNING, "No translator path: (%s codec is not valid)\n", src_index < 0 ? "starting" : "ending");
return NULL;
}
AST_RWLIST_RDLOCK(&translators);
while (src_index != dst_index) {
struct ast_trans_pvt *cur;
struct ast_translator *t = matrix_get(src_index, dst_index)->step;
if (!t) {
ast_log(LOG_WARNING, "No translator path from %s to %s\n",
ast_format_get_name(src), ast_format_get_name(dst));
AST_RWLIST_UNLOCK(&translators);
return NULL;
}
if (!(cur = newpvt(t))) {
ast_log(LOG_WARNING, "Failed to build translator step from %s to %s\n",
ast_format_get_name(src), ast_format_get_name(dst));
if (head) {
ast_translator_free_path(head);
}
AST_RWLIST_UNLOCK(&translators);
return NULL;
}
if (!head) {
head = cur;
} else {
tail->next = cur;
}
tail = cur;
cur->nextin = cur->nextout = ast_tv(0, 0);
/* Keep going if this isn't the final destination */
src_index = cur->t->dst_fmt_index;
}
AST_RWLIST_UNLOCK(&translators);
return head;
}
/*! \brief do the actual translation */
struct ast_frame *ast_translate(struct ast_trans_pvt *path, struct ast_frame *f, int consume)
{
struct ast_trans_pvt *p = path;
struct ast_frame *out;
struct timeval delivery;
int has_timing_info;
long ts;
long len;
int seqno;
has_timing_info = ast_test_flag(f, AST_FRFLAG_HAS_TIMING_INFO);
ts = f->ts;
len = f->len;
seqno = f->seqno;
if (!ast_tvzero(f->delivery)) {
if (!ast_tvzero(path->nextin)) {
/* Make sure this is in line with what we were expecting */
if (!ast_tveq(path->nextin, f->delivery)) {
/* The time has changed between what we expected and this
most recent time on the new packet. If we have a
valid prediction adjust our output time appropriately */
if (!ast_tvzero(path->nextout)) {
path->nextout = ast_tvadd(path->nextout,
ast_tvsub(f->delivery, path->nextin));
}
path->nextin = f->delivery;
}
} else {
/* This is our first pass. Make sure the timing looks good */
path->nextin = f->delivery;
path->nextout = f->delivery;
}
/* Predict next incoming sample */
path->nextin = ast_tvadd(path->nextin, ast_samp2tv(
f->samples, ast_format_get_sample_rate(f->subclass.format)));
}
delivery = f->delivery;
for (out = f; out && p ; p = p->next) {
framein(p, out);
if (out != f) {
ast_frfree(out);
}
out = p->t->frameout(p);
}
if (out) {
/* we have a frame, play with times */
if (!ast_tvzero(delivery)) {
/* Regenerate prediction after a discontinuity */
if (ast_tvzero(path->nextout)) {
path->nextout = ast_tvnow();
}
/* Use next predicted outgoing timestamp */
out->delivery = path->nextout;
/* Predict next outgoing timestamp from samples in this
frame. */
path->nextout = ast_tvadd(path->nextout, ast_samp2tv(
out->samples, ast_format_get_sample_rate(out->subclass.format)));
if (f->samples != out->samples && ast_test_flag(out, AST_FRFLAG_HAS_TIMING_INFO)) {
ast_debug(4, "Sample size different %d vs %d\n", f->samples, out->samples);
ast_clear_flag(out, AST_FRFLAG_HAS_TIMING_INFO);
}
} else {
out->delivery = ast_tv(0, 0);
ast_set2_flag(out, has_timing_info, AST_FRFLAG_HAS_TIMING_INFO);
if (has_timing_info) {
out->ts = ts;
out->len = len;
out->seqno = seqno;
}
}
/* Invalidate prediction if we're entering a silence period */
if (out->frametype == AST_FRAME_CNG) {
path->nextout = ast_tv(0, 0);
}
}
if (consume) {
ast_frfree(f);
}
return out;
}
/*!
* \internal
* \brief Compute the computational cost of a single translation step.
*
* \note This function is only used to decide which translation path to
* use between two translators with identical src and dst formats. Computational
* cost acts only as a tie breaker. This is done so hardware translators
* can naturally have precedence over software translators.
*/
static void generate_computational_cost(struct ast_translator *t, int seconds)
{
int num_samples = 0;
struct ast_trans_pvt *pvt;
struct rusage start;
struct rusage end;
int cost;
int out_rate = t->dst_codec.sample_rate;
if (!seconds) {
seconds = 1;
}
/* If they don't make samples, give them a terrible score */
if (!t->sample) {
ast_debug(3, "Translator '%s' does not produce sample frames.\n", t->name);
t->comp_cost = 999999;
return;
}
pvt = newpvt(t);
if (!pvt) {
ast_log(LOG_WARNING, "Translator '%s' appears to be broken and will probably fail.\n", t->name);
t->comp_cost = 999999;
return;
}
getrusage(RUSAGE_SELF, &start);
/* Call the encoder until we've processed the required number of samples */
while (num_samples < seconds * out_rate) {
struct ast_frame *f = t->sample();
if (!f) {
ast_log(LOG_WARNING, "Translator '%s' failed to produce a sample frame.\n", t->name);
destroy(pvt);
t->comp_cost = 999999;
return;
}
framein(pvt, f);
ast_frfree(f);
while ((f = t->frameout(pvt))) {
num_samples += f->samples;
ast_frfree(f);
}
}
getrusage(RUSAGE_SELF, &end);
cost = ((end.ru_utime.tv_sec - start.ru_utime.tv_sec) * 1000000) + end.ru_utime.tv_usec - start.ru_utime.tv_usec;
cost += ((end.ru_stime.tv_sec - start.ru_stime.tv_sec) * 1000000) + end.ru_stime.tv_usec - start.ru_stime.tv_usec;
destroy(pvt);
t->comp_cost = cost / seconds;
if (!t->comp_cost) {
t->comp_cost = 1;
}
}
/*!
* \internal
*
* \brief If no table cost value was pre set by the translator. An attempt is made to
* automatically generate that cost value from the cost table based on our src and
* dst formats.
*
* \note This function allows older translators built before the translation cost
* changed away from using onely computational time to continue to be registered
* correctly. It is expected that translators built after the introduction of this
* function will manually assign their own table cost value.
*
* \note This function is safe to use on any audio formats that used to be defined in the
* first 64 bits of the old bit field codec representation.
*
* \retval Table Cost value greater than 0.
* \retval 0 on error.
*/
static int generate_table_cost(struct ast_codec *src, struct ast_codec *dst)
{
int src_rate = src->sample_rate;
int src_ll = 0;
int dst_rate = dst->sample_rate;
int dst_ll = 0;
if ((src->type != AST_MEDIA_TYPE_AUDIO) ||
(dst->type != AST_MEDIA_TYPE_AUDIO)) {
/* This method of generating table cost is limited to audio.
* Translators for media other than audio must manually set their
* table cost. */
return 0;
}
src_ll = !strcmp(src->name, "slin");
dst_ll = !strcmp(dst->name, "slin");
if (src_ll) {
if (dst_ll && (src_rate == dst_rate)) {
return AST_TRANS_COST_LL_LL_ORIGSAMP;
} else if (!dst_ll && (src_rate == dst_rate)) {
return AST_TRANS_COST_LL_LY_ORIGSAMP;
} else if (dst_ll && (src_rate < dst_rate)) {
return AST_TRANS_COST_LL_LL_UPSAMP;
} else if (!dst_ll && (src_rate < dst_rate)) {
return AST_TRANS_COST_LL_LY_UPSAMP;
} else if (dst_ll && (src_rate > dst_rate)) {
return AST_TRANS_COST_LL_LL_DOWNSAMP;
} else if (!dst_ll && (src_rate > dst_rate)) {
return AST_TRANS_COST_LL_LY_DOWNSAMP;
} else {
return AST_TRANS_COST_LL_UNKNOWN;
}
} else {
if (dst_ll && (src_rate == dst_rate)) {
return AST_TRANS_COST_LY_LL_ORIGSAMP;
} else if (!dst_ll && (src_rate == dst_rate)) {
return AST_TRANS_COST_LY_LY_ORIGSAMP;
} else if (dst_ll && (src_rate < dst_rate)) {
return AST_TRANS_COST_LY_LL_UPSAMP;
} else if (!dst_ll && (src_rate < dst_rate)) {
return AST_TRANS_COST_LY_LY_UPSAMP;
} else if (dst_ll && (src_rate > dst_rate)) {
return AST_TRANS_COST_LY_LL_DOWNSAMP;
} else if (!dst_ll && (src_rate > dst_rate)) {
return AST_TRANS_COST_LY_LY_DOWNSAMP;
} else {
return AST_TRANS_COST_LY_UNKNOWN;
}
}
}
/*!
* \brief rebuild a translation matrix.
* \note This function expects the list of translators to be locked
*/
static void matrix_rebuild(int samples)
{
struct ast_translator *t;
int newtablecost;
int x; /* source format index */
int y; /* intermediate format index */
int z; /* destination format index */
ast_debug(1, "Resetting translation matrix\n");
matrix_clear();
/* first, compute all direct costs */
AST_RWLIST_TRAVERSE(&translators, t, list) {
if (!t->active) {
continue;
}
x = t->src_fmt_index;
z = t->dst_fmt_index;
if (samples) {
generate_computational_cost(t, samples);
}
/* This new translator is the best choice if any of the below are true.
* 1. no translation path is set between x and z yet.
* 2. the new table cost is less.
* 3. the new computational cost is less. Computational cost is only used
* to break a tie between two identical translation paths.
*/
if (!matrix_get(x, z)->step ||
(t->table_cost < matrix_get(x, z)->step->table_cost) ||
(t->comp_cost < matrix_get(x, z)->step->comp_cost)) {
matrix_get(x, z)->step = t;
matrix_get(x, z)->table_cost = t->table_cost;
}
}
/*
* For each triple x, y, z of distinct formats, check if there is
* a path from x to z through y which is cheaper than what is
* currently known, and in case, update the matrix.
* Repeat until the matrix is stable.
*/
for (;;) {
int changed = 0;
for (x = 0; x < cur_max_index; x++) { /* source format */
for (y = 0; y < cur_max_index; y++) { /* intermediate format */
if (x == y) { /* skip ourselves */
continue;
}
for (z = 0; z < cur_max_index; z++) { /* dst format */
if ((z == x || z == y) || /* skip null conversions */
!matrix_get(x, y)->step || /* no path from x to y */
!matrix_get(y, z)->step) { /* no path from y to z */
continue;
}
/* calculate table cost from x->y->z */
newtablecost = matrix_get(x, y)->table_cost + matrix_get(y, z)->table_cost;
/* if no step already exists between x and z OR the new cost of using the intermediate
* step is cheaper, use this step. */
if (!matrix_get(x, z)->step || (newtablecost < matrix_get(x, z)->table_cost)) {
matrix_get(x, z)->step = matrix_get(x, y)->step;
matrix_get(x, z)->table_cost = newtablecost;
matrix_get(x, z)->multistep = 1;
changed++;
if (DEBUG_ATLEAST(10)) {
struct ast_codec *x_codec = index2codec(x);
struct ast_codec *y_codec = index2codec(y);
struct ast_codec *z_codec = index2codec(z);
ast_log(LOG_DEBUG,
"Discovered %u cost path from %s to %s, via %s\n",
matrix_get(x, z)->table_cost, x_codec->name,
y_codec->name, z_codec->name);
ao2_ref(x_codec, -1);
ao2_ref(y_codec, -1);
ao2_ref(z_codec, -1);
}
}
}
}
}
if (!changed) {
break;
}
}
}
static void codec_append_name(const struct ast_codec *codec, struct ast_str **buf)
{
if (codec) {
ast_str_append(buf, 0, "(%s@%u)", codec->name, codec->sample_rate);
} else {
ast_str_append(buf, 0, "(nothing)");
}
}
const char *ast_translate_path_to_str(struct ast_trans_pvt *p, struct ast_str **str)
{
if (!p || !p->t) {
return "";
}
codec_append_name(&p->t->src_codec, str);
while (p) {
ast_str_append(str, 0, "->");
codec_append_name(&p->t->dst_codec, str);
p = p->next;
}
return ast_str_buffer(*str);
}
static char *complete_trans_path_choice(const char *line, const char *word, int pos, int state)
{
int i = 1, which = 0;
int wordlen = strlen(word);
struct ast_codec *codec;
while ((codec = ast_codec_get_by_id(i))) {
++i;
if (codec->type != AST_MEDIA_TYPE_AUDIO) {
ao2_ref(codec, -1);
continue;
}
if (!strncasecmp(word, codec->name, wordlen) && ++which > state) {
char *res = ast_strdup(codec->name);
ao2_ref(codec, -1);
return res;
}
ao2_ref(codec, -1);
}
return NULL;
}
static void handle_cli_recalc(struct ast_cli_args *a)
{
int time = a->argv[4] ? atoi(a->argv[4]) : 1;
if (time <= 0) {
ast_cli(a->fd, " Recalc must be greater than 0. Defaulting to 1.\n");
time = 1;
}
if (time > MAX_RECALC) {
ast_cli(a->fd, " Maximum limit of recalc exceeded by %d, truncating value to %d\n", time - MAX_RECALC, MAX_RECALC);
time = MAX_RECALC;
}
ast_cli(a->fd, " Recalculating Codec Translation (number of sample seconds: %d)\n\n", time);
AST_RWLIST_WRLOCK(&translators);
matrix_rebuild(time);
AST_RWLIST_UNLOCK(&translators);
}
static char *handle_show_translation_table(struct ast_cli_args *a)
{
int x, y, i, k;
int longest = 0, num_codecs = 0, curlen = 0;
struct ast_str *out = ast_str_create(1024);
struct ast_codec *codec;
/* Get the length of the longest (usable?) codec name,
so we know how wide the left side should be */
for (i = 1; (codec = ast_codec_get_by_id(i)); ao2_ref(codec, -1), ++i) {
++num_codecs;
if (codec->type != AST_MEDIA_TYPE_AUDIO) {
continue;
}
curlen = strlen(codec->name);
if (curlen > longest) {
longest = curlen;
}
}
AST_RWLIST_RDLOCK(&translators);
ast_cli(a->fd, " Translation times between formats (in microseconds) for one second of data\n");
ast_cli(a->fd, " Source Format (Rows) Destination Format (Columns)\n\n");
for (i = 0; i < num_codecs; i++) {
struct ast_codec *row = i ? ast_codec_get_by_id(i) : NULL;
x = -1;
if ((i > 0) && (row->type != AST_MEDIA_TYPE_AUDIO)) {
ao2_ref(row, -1);
continue;
}
if ((i > 0) && (x = codec2index(row)) == -1) {
ao2_ref(row, -1);
continue;
}
ast_str_set(&out, 0, " ");
for (k = 0; k < num_codecs; k++) {
struct ast_codec *col = k ? ast_codec_get_by_id(k) : NULL;
y = -1;
if ((k > 0) && (col->type != AST_MEDIA_TYPE_AUDIO)) {
ao2_ref(col, -1);
continue;
}
if ((k > 0) && (y = codec2index(col)) == -1) {
ao2_ref(col, -1);
continue;
}
if (k > 0) {
curlen = strlen(col->name);
}
if (curlen < 5) {
curlen = 5;
}
if (x >= 0 && y >= 0 && matrix_get(x, y)->step) {
/* Actual codec output */
ast_str_append(&out, 0, "%*u", curlen + 1, (matrix_get(x, y)->table_cost/100));
} else if (i == 0 && k > 0) {
/* Top row - use a dynamic size */
ast_str_append(&out, 0, "%*s", curlen + 1, col->name);
} else if (k == 0 && i > 0) {
/* Left column - use a static size. */
ast_str_append(&out, 0, "%*s", longest, row->name);
} else if (x >= 0 && y >= 0) {
/* Codec not supported */
ast_str_append(&out, 0, "%*s", curlen + 1, "-");
} else {
/* Upper left hand corner */
ast_str_append(&out, 0, "%*s", longest, "");
}
ao2_cleanup(col);
}
ast_str_append(&out, 0, "\n");
ast_cli(a->fd, "%s", ast_str_buffer(out));
ao2_cleanup(row);
}
ast_free(out);
AST_RWLIST_UNLOCK(&translators);
return CLI_SUCCESS;
}
static char *handle_show_translation_path(struct ast_cli_args *a, const char *codec_name, unsigned int sample_rate)
{
int i = 1;
struct ast_str *str = ast_str_alloca(1024);
struct ast_translator *step;
struct ast_codec *dst_codec;
struct ast_codec *src_codec = ast_codec_get(codec_name, AST_MEDIA_TYPE_AUDIO, sample_rate);
if (!src_codec) {
ast_cli(a->fd, "Source codec \"%s\" is not found.\n", codec_name);
return CLI_FAILURE;
}
AST_RWLIST_RDLOCK(&translators);
ast_cli(a->fd, "--- Translation paths SRC Codec \"%s\" sample rate %u ---\n",
codec_name, src_codec->sample_rate);
while ((dst_codec = ast_codec_get_by_id(i))) {
int src, dst;
char src_buffer[64];
char dst_buffer[64];
++i;
if (src_codec == dst_codec ||
dst_codec->type != AST_MEDIA_TYPE_AUDIO) {
ao2_ref(dst_codec, -1);
continue;
}
dst = codec2index(dst_codec);
src = codec2index(src_codec);
if (src < 0 || dst < 0) {
ast_str_set(&str, 0, "No Translation Path");
} else {
step = matrix_get(src, dst)->step;
if (step) {
codec_append_name(&step->src_codec, &str);
while (src != dst) {
src = step->dst_fmt_index;
step = matrix_get(src, dst)->step;
if (!step) {
ast_str_append(&str, 0, "->");
codec_append_name(dst_codec, &str);
break;
}
ast_str_append(&str, 0, "->");
codec_append_name(&step->src_codec, &str);
}
}
}
snprintf(src_buffer, sizeof(src_buffer), "%s:%u", src_codec->name, src_codec->sample_rate);
snprintf(dst_buffer, sizeof(dst_buffer), "%s:%u", dst_codec->name, dst_codec->sample_rate);
ast_cli(a->fd, "\t%-16.16s To %-16.16s: %-60.60s\n",
src_buffer, dst_buffer, ast_str_buffer(str));
ast_str_reset(str);
ao2_ref(dst_codec, -1);
}
AST_RWLIST_UNLOCK(&translators);
ao2_ref(src_codec, -1);
return CLI_SUCCESS;
}
static char *handle_cli_core_show_translation(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
static const char * const option[] = { "recalc", "paths", NULL };
switch (cmd) {
case CLI_INIT:
e->command = "core show translation";
e->usage =
"Usage: 'core show translation' can be used in two ways.\n"
" 1. 'core show translation [recalc [<recalc seconds>]]\n"
" Displays known codec translators and the cost associated\n"
" with each conversion. If the argument 'recalc' is supplied along\n"
" with optional number of seconds to test a new test will be performed\n"
" as the chart is being displayed.\n"
" 2. 'core show translation paths [codec [sample_rate]]'\n"
" This will display all the translation paths associated with a codec.\n"
" If a codec has multiple sample rates, the sample rate must be\n"
" provided as well.\n";
return NULL;
case CLI_GENERATE:
if (a->pos == 3) {
return ast_cli_complete(a->word, option, a->n);
}
if (a->pos == 4 && !strcasecmp(a->argv[3], option[1])) {
return complete_trans_path_choice(a->line, a->word, a->pos, a->n);
}
/* BUGBUG - add tab completion for sample rates */
return NULL;
}
if (a->argc > 6)
return CLI_SHOWUSAGE;
if (a->argv[3] && !strcasecmp(a->argv[3], option[1]) && a->argc == 5) { /* show paths */
return handle_show_translation_path(a, a->argv[4], 0);
} else if (a->argv[3] && !strcasecmp(a->argv[3], option[1]) && a->argc == 6) {
unsigned int sample_rate;
if (sscanf(a->argv[5], "%30u", &sample_rate) != 1) {
ast_cli(a->fd, "Invalid sample rate: %s.\n", a->argv[5]);
return CLI_FAILURE;
}
return handle_show_translation_path(a, a->argv[4], sample_rate);
} else if (a->argv[3] && !strcasecmp(a->argv[3], option[0])) { /* recalc and then fall through to show table */
handle_cli_recalc(a);
} else if (a->argc > 3) { /* wrong input */
return CLI_SHOWUSAGE;
}
return handle_show_translation_table(a);
}
static struct ast_cli_entry cli_translate[] = {
AST_CLI_DEFINE(handle_cli_core_show_translation, "Display translation matrix")
};
/*! \brief register codec translator */
int __ast_register_translator(struct ast_translator *t, struct ast_module *mod)
{
struct ast_translator *u;
char tmp[80];
RAII_VAR(struct ast_codec *, src_codec, NULL, ao2_cleanup);
RAII_VAR(struct ast_codec *, dst_codec, NULL, ao2_cleanup);
src_codec = ast_codec_get(t->src_codec.name, t->src_codec.type, t->src_codec.sample_rate);
if (!src_codec) {
ast_assert(0);
ast_log(LOG_WARNING, "Failed to register translator: unknown source codec %s\n", t->src_codec.name);
return -1;
}
dst_codec = ast_codec_get(t->dst_codec.name, t->dst_codec.type, t->dst_codec.sample_rate);
if (!dst_codec) {
ast_log(LOG_WARNING, "Failed to register translator: unknown destination codec %s\n", t->dst_codec.name);
return -1;
}
if (add_codec2index(src_codec) || add_codec2index(dst_codec)) {
if (matrix_resize(0)) {
ast_log(LOG_WARNING, "Translator matrix can not represent any more translators. Out of resources.\n");
return -1;
}
add_codec2index(src_codec);
add_codec2index(dst_codec);
}
if (!mod) {
ast_log(LOG_WARNING, "Missing module pointer, you need to supply one\n");
return -1;
}
if (!t->buf_size) {
ast_log(LOG_WARNING, "empty buf size, you need to supply one\n");
return -1;
}
if (!t->table_cost && !(t->table_cost = generate_table_cost(src_codec, dst_codec))) {
ast_log(LOG_WARNING, "Table cost could not be generated for %s, "
"Please set table_cost variable on translator.\n", t->name);
return -1;
}
t->module = mod;
t->src_fmt_index = codec2index(src_codec);
t->dst_fmt_index = codec2index(dst_codec);
t->active = 1;
if (t->src_fmt_index < 0 || t->dst_fmt_index < 0) {
ast_log(LOG_WARNING, "Invalid translator path: (%s codec is not valid)\n", t->src_fmt_index < 0 ? "starting" : "ending");
return -1;
}
if (t->src_fmt_index >= cur_max_index) {
ast_log(LOG_WARNING, "Source codec %s is larger than cur_max_index\n", t->src_codec.name);
return -1;
}
if (t->dst_fmt_index >= cur_max_index) {
ast_log(LOG_WARNING, "Destination codec %s is larger than cur_max_index\n", t->dst_codec.name);
return -1;
}
if (t->buf_size) {
/*
* Align buf_size properly, rounding up to the machine-specific
* alignment for pointers.
*/
struct _test_align { void *a, *b; } p;
int align = (char *)&p.b - (char *)&p.a;
t->buf_size = ((t->buf_size + align - 1) / align) * align;
}
if (t->frameout == NULL) {
t->frameout = default_frameout;
}
generate_computational_cost(t, 1);
ast_verb(2, "Registered translator '%s' from codec %s to %s, table cost, %d, computational cost %d\n",
term_color(tmp, t->name, COLOR_MAGENTA, COLOR_BLACK, sizeof(tmp)),
t->src_codec.name, t->dst_codec.name, t->table_cost, t->comp_cost);
AST_RWLIST_WRLOCK(&translators);
/* find any existing translators that provide this same srcfmt/dstfmt,
and put this one in order based on computational cost */
AST_RWLIST_TRAVERSE_SAFE_BEGIN(&translators, u, list) {
if ((u->src_fmt_index == t->src_fmt_index) &&
(u->dst_fmt_index == t->dst_fmt_index) &&
(u->comp_cost > t->comp_cost)) {
AST_RWLIST_INSERT_BEFORE_CURRENT(t, list);
t = NULL;
break;
}
}
AST_RWLIST_TRAVERSE_SAFE_END;
/* if no existing translator was found for this codec combination,
add it to the beginning of the list */
if (t) {
AST_RWLIST_INSERT_HEAD(&translators, t, list);
}
matrix_rebuild(0);
AST_RWLIST_UNLOCK(&translators);
return 0;
}
/*! \brief unregister codec translator */
int ast_unregister_translator(struct ast_translator *t)
{
char tmp[80];
struct ast_translator *u;
int found = 0;
AST_RWLIST_WRLOCK(&translators);
AST_RWLIST_TRAVERSE_SAFE_BEGIN(&translators, u, list) {
if (u == t) {
AST_RWLIST_REMOVE_CURRENT(list);
ast_verb(2, "Unregistered translator '%s' from codec %s to %s\n",
term_color(tmp, t->name, COLOR_MAGENTA, COLOR_BLACK, sizeof(tmp)),
t->src_codec.name, t->dst_codec.name);
found = 1;
break;
}
}
AST_RWLIST_TRAVERSE_SAFE_END;
if (found) {
matrix_rebuild(0);
}
AST_RWLIST_UNLOCK(&translators);
return (u ? 0 : -1);
}
void ast_translator_activate(struct ast_translator *t)
{
AST_RWLIST_WRLOCK(&translators);
t->active = 1;
matrix_rebuild(0);
AST_RWLIST_UNLOCK(&translators);
}
void ast_translator_deactivate(struct ast_translator *t)
{
AST_RWLIST_WRLOCK(&translators);
t->active = 0;
matrix_rebuild(0);
AST_RWLIST_UNLOCK(&translators);
}
/*! \brief Calculate our best translator source format, given costs, and a desired destination */
int ast_translator_best_choice(struct ast_format_cap *dst_cap,
struct ast_format_cap *src_cap,
struct ast_format **dst_fmt_out,
struct ast_format **src_fmt_out)
{
unsigned int besttablecost = INT_MAX;
unsigned int beststeps = INT_MAX;
RAII_VAR(struct ast_format *, best, NULL, ao2_cleanup);
RAII_VAR(struct ast_format *, bestdst, NULL, ao2_cleanup);
RAII_VAR(struct ast_format_cap *, joint_cap, NULL, ao2_cleanup);
int i;
int j;
if (!(joint_cap = ast_format_cap_alloc(AST_FORMAT_CAP_FLAG_DEFAULT))) {
return -1;
}
ast_format_cap_get_compatible(dst_cap, src_cap, joint_cap);
for (i = 0; i < ast_format_cap_count(joint_cap); ++i) {
struct ast_format *fmt =
ast_format_cap_get_format(joint_cap, i);
if (!fmt) {
continue;
}
if (!best) {
/* No ao2_ref operations needed, we're done with fmt */
best = fmt;
continue;
}
if (ast_format_get_sample_rate(best) <
ast_format_get_sample_rate(fmt)) {
ao2_replace(best, fmt);
}
ao2_ref(fmt, -1);
}
if (best) {
ao2_replace(*dst_fmt_out, best);
ao2_replace(*src_fmt_out, best);
return 0;
}
/* need to translate */
AST_RWLIST_RDLOCK(&translators);
for (i = 0; i < ast_format_cap_count(dst_cap); ++i) {
struct ast_format *dst =
ast_format_cap_get_format(dst_cap, i);
if (!dst) {
continue;
}
for (j = 0; j < ast_format_cap_count(src_cap); ++j) {
struct ast_format *src =
ast_format_cap_get_format(src_cap, j);
int x, y;
if (!src) {
continue;
}
x = format2index(src);
y = format2index(dst);
if (x < 0 || y < 0) {
ao2_ref(src, -1);
continue;
}
if (!matrix_get(x, y) || !(matrix_get(x, y)->step)) {
ao2_ref(src, -1);
continue;
}
if (((matrix_get(x, y)->table_cost < besttablecost) ||
(matrix_get(x, y)->multistep < beststeps))) {
/* better than what we have so far */
ao2_replace(best, src);
ao2_replace(bestdst, dst);
besttablecost = matrix_get(x, y)->table_cost;
beststeps = matrix_get(x, y)->multistep;
}
ao2_ref(src, -1);
}
ao2_ref(dst, -1);
}
AST_RWLIST_UNLOCK(&translators);
if (!best) {
return -1;
}
ao2_replace(*dst_fmt_out, bestdst);
ao2_replace(*src_fmt_out, best);
return 0;
}
unsigned int ast_translate_path_steps(struct ast_format *dst_format, struct ast_format *src_format)
{
unsigned int res = -1;
/* convert bitwise format numbers into array indices */
int src = format2index(src_format);
int dest = format2index(dst_format);
if (src < 0 || dest < 0) {
ast_log(LOG_WARNING, "No translator path: (%s codec is not valid)\n", src < 0 ? "starting" : "ending");
return -1;
}
AST_RWLIST_RDLOCK(&translators);
if (matrix_get(src, dest)->step) {
res = matrix_get(src, dest)->multistep + 1;
}
AST_RWLIST_UNLOCK(&translators);
return res;
}
static void check_translation_path(
struct ast_format_cap *dest, struct ast_format_cap *src,
struct ast_format_cap *result, struct ast_format *src_fmt,
enum ast_media_type type)
{
int index, src_index = format2index(src_fmt);
/* For a given source format, traverse the list of
known formats to determine whether there exists
a translation path from the source format to the
destination format. */
for (index = 0; (src_index >= 0) && index < cur_max_index; index++) {
struct ast_codec *codec = index2codec(index);
RAII_VAR(struct ast_format *, fmt, ast_format_create(codec), ao2_cleanup);
ao2_ref(codec, -1);
if (ast_format_get_type(fmt) != type) {
continue;
}
/* if this is not a desired format, nothing to do */
if (ast_format_cap_iscompatible_format(dest, fmt) == AST_FORMAT_CMP_NOT_EQUAL) {
continue;
}
/* if the source is supplying this format, then
we can leave it in the result */
if (ast_format_cap_iscompatible_format(src, fmt) == AST_FORMAT_CMP_EQUAL) {
continue;
}
/* if we don't have a translation path from the src
to this format, remove it from the result */
if (!matrix_get(src_index, index)->step) {
ast_format_cap_remove(result, fmt);
continue;
}
/* now check the opposite direction */
if (!matrix_get(index, src_index)->step) {
ast_format_cap_remove(result, fmt);
}
}
}
void ast_translate_available_formats(struct ast_format_cap *dest, struct ast_format_cap *src, struct ast_format_cap *result)
{
struct ast_format *cur_dest, *cur_src;
int index;
for (index = 0; index < ast_format_cap_count(dest); ++index) {
if (!(cur_dest = ast_format_cap_get_format(dest, index))) {
continue;
}
/* We give preference to a joint format structure if possible */
if ((cur_src = ast_format_cap_get_compatible_format(src, cur_dest))) {
ast_format_cap_append(result, cur_src, 0);
ao2_ref(cur_src, -1);
} else {
/* Otherwise we just use the destination format */
ast_format_cap_append(result, cur_dest, 0);
}
ao2_ref(cur_dest, -1);
}
/* if we don't have a source format, we just have to try all
possible destination formats */
if (!src) {
return;
}
for (index = 0; index < ast_format_cap_count(src); ++index) {
if (!(cur_src = ast_format_cap_get_format(src, index))) {
continue;
}
AST_RWLIST_RDLOCK(&translators);
check_translation_path(dest, src, result,
cur_src, AST_MEDIA_TYPE_AUDIO);
check_translation_path(dest, src, result,
cur_src, AST_MEDIA_TYPE_VIDEO);
AST_RWLIST_UNLOCK(&translators);
ao2_ref(cur_src, -1);
}
}
static void translate_shutdown(void)
{
int x;
ast_cli_unregister_multiple(cli_translate, ARRAY_LEN(cli_translate));
ast_rwlock_wrlock(&tablelock);
for (x = 0; x < index_size; x++) {
ast_free(__matrix[x]);
}
ast_free(__matrix);
ast_free(__indextable);
ast_rwlock_unlock(&tablelock);
ast_rwlock_destroy(&tablelock);
}
int ast_translate_init(void)
{
int res = 0;
ast_rwlock_init(&tablelock);
res = matrix_resize(1);
res |= ast_cli_register_multiple(cli_translate, ARRAY_LEN(cli_translate));
ast_register_atexit(translate_shutdown);
return res;
}