asterisk/main/cli.c

3068 lines
84 KiB
C
Raw Normal View History

/*
* 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 Standard Command Line Interface
*
* \author Mark Spencer <markster@digium.com>
*/
/*! \li \ref cli.c uses the configuration file \ref cli_permissions.conf
* \addtogroup configuration_file Configuration Files
*/
/*!
* \page cli_permissions.conf cli_permissions.conf
* \verbinclude cli_permissions.conf.sample
*/
/*** MODULEINFO
<support_level>core</support_level>
***/
#include "asterisk.h"
#include "asterisk/_private.h"
#include "asterisk/paths.h" /* use ast_config_AST_MODULE_DIR */
#include <signal.h>
#include <ctype.h>
#include <regex.h>
#include <pwd.h>
#include <grp.h>
#include "asterisk/cli.h"
#include "asterisk/linkedlists.h"
#include "asterisk/module.h"
#include "asterisk/pbx.h"
#include "asterisk/channel.h"
#include "asterisk/utils.h"
#include "asterisk/app.h"
#include "asterisk/lock.h"
Merge team/russell/ast_verbose_threadstorage - instead of defining a free() wrapper in a bunch of files, define it as ast_free() in utils.h and remove the copies from all the files. - centralize and abstract the code used for doing thread storage. The code lives in threadstorage.h, with one function being implemented in utils.c. This new API includes generic thread storage as well as special functions for handling thread local dynamic length string buffers. - update ast_inet_ntoa() to use the new threadstorage API - update ast_state2str() to use the new threadstorage API - update ast_cli() to use the new threadstorage API - Modify manager_event() to use thread storage. Instead of using a buffer of 4096 characters as the workspace for building the manager event, use a thread local dynamic string. Now there is no length limitation on the length of the body of a manager event. - Significantly simplify the handling of ast_verbose() ... - Instead of using a static char buffer and a lock to make sure only one thread can be using ast_verbose() at a time, use a thread local dynamic string as the workspace for preparing the verbose message. Instead of locking around the entire function, the only locking done now is when the message has been built and is being deliviered to the list of registered verbose message handlers. - This function was doing a strdup() on every message passed to it and keeping a queue of the last 200 messages in memory. This has been completely removed. The only place this was used was that if there were any messages in the verbose queue when a verbose handler was registered, all of the messages in the queue would be fed to it. So, I just made sure that the console verbose handler and the network verbose handler (for remote asterisk consoles) were registered before any verbose messages. pbx_gtkconsole and pbx_kdeconsole will now lose a few verbose messages at startup, but I didn't feel the performance hit of this message queue was worth saving the initial verbose output for these very rarely used modules. - I have removed the last three arguments to the verbose handlers, leaving only the string itself because they aren't needed anymore. For example, ast_verbose had some logic for telling the verbose handler to add a newline if the buffer was completely full. Now that the buffer can grow as needed, this doesn't matter anymore. - remove unused function, ast_verbose_dmesg() which was to dispatch the message queue - Convert the list of verbose handlers to use the linked list macros. - add missing newline characters to a few ast_verbose() calls - convert the list of log channels to use the linked list macros in logger.c - fix close_logger() to close all of the files it opened for logging - update ast_log() to use a thread local dynamic string for its workspace for preparing log messages instead of a buffer of size BUFSIZ (8kB on my system) allocated on the stack. The dynamic string in this case is limited to only growing to a maximum size of BUFSIZ. git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39272 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2006-08-08 06:32:04 +00:00
#include "asterisk/threadstorage.h"
#include "asterisk/logger_category.h"
Merged revisions 282047 via svnmerge from https://origsvn.digium.com/svn/asterisk/branches/1.8 ........ r282047 | dvossel | 2010-08-12 15:15:41 -0500 (Thu, 12 Aug 2010) | 35 lines improved translation paths for wideband codecs The problem I'm addressing is that Asterisk's current method of building the least cost translation paths between codecs does not take into account sample rate. For instance, it was possible for siren14 (a 32khz codec), to contain the a translation path to siren7 (a 16khz audio codec) that goes through slin at 8khz. In this case Asterisk takes a 32khz codec, down samples it to 8khz and then up samples it to 16khz which is terrible regardless if it is computationally less expensive. This patch now builds translation paths that give priority to maintaining the best possible sample rate before taking into consideration computational cost. This patch also adds cli commands to expose what translation paths are actually being used. Changes: 1. Translation paths will never contain a step that changes the sample rate unless absolutely necessary. 2. When choosing the best codec to make two channels compatible. Shared codecs with the highest sample rate are given priority. 3. A new cli command to show all translation paths available for a specific codec 'core show translation paths [codec name]' has been added. 4. 'core show translation' which displays the translation matrix now includes the new higher bit audio codecs in the table. 5. 'core show channel [channel name]' now displays the translation paths if translation is used. (closes issue #16841) Reported by: dvossel Review: https://reviewboard.asterisk.org/r/842/ ........ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@282048 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2010-08-12 20:17:17 +00:00
#include "asterisk/translate.h"
#include "asterisk/bridge.h"
#include "asterisk/stasis_channels.h"
#include "asterisk/stasis_bridges.h"
#include "asterisk/vector.h"
#include "asterisk/stream.h"
/*!
* \brief List of restrictions per user.
*/
struct cli_perm {
unsigned int permit:1; /*!< 1=Permit 0=Deny */
char *command; /*!< Command name (to apply restrictions) */
AST_LIST_ENTRY(cli_perm) list;
};
AST_LIST_HEAD_NOLOCK(cli_perm_head, cli_perm);
/*! \brief list of users to apply restrictions. */
struct usergroup_cli_perm {
int uid; /*!< User ID (-1 disabled) */
int gid; /*!< Group ID (-1 disabled) */
struct cli_perm_head *perms; /*!< List of permissions. */
AST_LIST_ENTRY(usergroup_cli_perm) list;/*!< List mechanics */
};
/*! \brief CLI permissions config file. */
static const char perms_config[] = "cli_permissions.conf";
/*! \brief Default permissions value 1=Permit 0=Deny */
static int cli_default_perm = 1;
/*! \brief mutex used to prevent a user from running the 'cli reload permissions' command while
* it is already running. */
AST_MUTEX_DEFINE_STATIC(permsconfiglock);
/*! \brief List of users and permissions. */
static AST_RWLIST_HEAD_STATIC(cli_perms, usergroup_cli_perm);
/*!
* \brief map a debug or verbose level to a module name
*/
struct module_level {
unsigned int level;
AST_RWLIST_ENTRY(module_level) entry;
char module[0];
};
AST_RWLIST_HEAD(module_level_list, module_level);
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
/*! lists of module names and their debug/trace levels */
static struct module_level_list debug_modules = AST_RWLIST_HEAD_INIT_VALUE;
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
static struct module_level_list trace_modules = AST_RWLIST_HEAD_INIT_VALUE;
AST_THREADSTORAGE(ast_cli_buf);
AST_RWLOCK_DEFINE_STATIC(shutdown_commands_lock);
static AST_VECTOR(, struct ast_cli_entry *) shutdown_commands;
/*! \brief Initial buffer size for resulting strings in ast_cli() */
#define AST_CLI_INITLEN 256
void ast_cli(int fd, const char *fmt, ...)
{
int res;
struct ast_str *buf;
va_list ap;
if (!(buf = ast_str_thread_get(&ast_cli_buf, AST_CLI_INITLEN)))
Merge team/russell/ast_verbose_threadstorage - instead of defining a free() wrapper in a bunch of files, define it as ast_free() in utils.h and remove the copies from all the files. - centralize and abstract the code used for doing thread storage. The code lives in threadstorage.h, with one function being implemented in utils.c. This new API includes generic thread storage as well as special functions for handling thread local dynamic length string buffers. - update ast_inet_ntoa() to use the new threadstorage API - update ast_state2str() to use the new threadstorage API - update ast_cli() to use the new threadstorage API - Modify manager_event() to use thread storage. Instead of using a buffer of 4096 characters as the workspace for building the manager event, use a thread local dynamic string. Now there is no length limitation on the length of the body of a manager event. - Significantly simplify the handling of ast_verbose() ... - Instead of using a static char buffer and a lock to make sure only one thread can be using ast_verbose() at a time, use a thread local dynamic string as the workspace for preparing the verbose message. Instead of locking around the entire function, the only locking done now is when the message has been built and is being deliviered to the list of registered verbose message handlers. - This function was doing a strdup() on every message passed to it and keeping a queue of the last 200 messages in memory. This has been completely removed. The only place this was used was that if there were any messages in the verbose queue when a verbose handler was registered, all of the messages in the queue would be fed to it. So, I just made sure that the console verbose handler and the network verbose handler (for remote asterisk consoles) were registered before any verbose messages. pbx_gtkconsole and pbx_kdeconsole will now lose a few verbose messages at startup, but I didn't feel the performance hit of this message queue was worth saving the initial verbose output for these very rarely used modules. - I have removed the last three arguments to the verbose handlers, leaving only the string itself because they aren't needed anymore. For example, ast_verbose had some logic for telling the verbose handler to add a newline if the buffer was completely full. Now that the buffer can grow as needed, this doesn't matter anymore. - remove unused function, ast_verbose_dmesg() which was to dispatch the message queue - Convert the list of verbose handlers to use the linked list macros. - add missing newline characters to a few ast_verbose() calls - convert the list of log channels to use the linked list macros in logger.c - fix close_logger() to close all of the files it opened for logging - update ast_log() to use a thread local dynamic string for its workspace for preparing log messages instead of a buffer of size BUFSIZ (8kB on my system) allocated on the stack. The dynamic string in this case is limited to only growing to a maximum size of BUFSIZ. git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39272 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2006-08-08 06:32:04 +00:00
return;
va_start(ap, fmt);
res = ast_str_set_va(&buf, 0, fmt, ap);
va_end(ap);
if (res != AST_DYNSTR_BUILD_FAILED) {
ast_carefulwrite(fd, ast_str_buffer(buf), ast_str_strlen(buf), 100);
}
}
unsigned int ast_debug_get_by_module(const char *module)
{
struct module_level *ml;
unsigned int res = 0;
AST_RWLIST_RDLOCK(&debug_modules);
AST_LIST_TRAVERSE(&debug_modules, ml, entry) {
if (!strcasecmp(ml->module, module)) {
res = ml->level;
break;
}
}
AST_RWLIST_UNLOCK(&debug_modules);
return res;
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
unsigned int ast_trace_get_by_module(const char *module)
{
struct module_level *ml;
unsigned int res = 0;
AST_RWLIST_RDLOCK(&trace_modules);
AST_LIST_TRAVERSE(&trace_modules, ml, entry) {
if (!strcasecmp(ml->module, module)) {
res = ml->level;
break;
}
}
AST_RWLIST_UNLOCK(&trace_modules);
return res;
}
/*! \internal
* \brief Check if the user with 'uid' and 'gid' is allow to execute 'command',
* if command starts with '_' then not check permissions, just permit
* to run the 'command'.
* if uid == -1 or gid == -1 do not check permissions.
* if uid == -2 and gid == -2 is because rasterisk client didn't send
* the credentials, so the cli_default_perm will be applied.
* \param uid User ID.
* \param gid Group ID.
* \param command Command name to check permissions.
* \retval 1 if has permission
* \retval 0 if it is not allowed.
*/
static int cli_has_permissions(int uid, int gid, const char *command)
{
struct usergroup_cli_perm *user_perm;
struct cli_perm *perm;
/* set to the default permissions general option. */
int isallowg = cli_default_perm, isallowu = -1, ispattern;
regex_t regexbuf;
/* if uid == -1 or gid == -1 do not check permissions.
if uid == -2 and gid == -2 is because rasterisk client didn't send
the credentials, so the cli_default_perm will be applied. */
if ((uid == CLI_NO_PERMS && gid == CLI_NO_PERMS) || command[0] == '_') {
return 1;
}
if (gid < 0 && uid < 0) {
return cli_default_perm;
}
AST_RWLIST_RDLOCK(&cli_perms);
AST_LIST_TRAVERSE(&cli_perms, user_perm, list) {
if (user_perm->gid != gid && user_perm->uid != uid) {
continue;
}
AST_LIST_TRAVERSE(user_perm->perms, perm, list) {
if (strcasecmp(perm->command, "all") && strncasecmp(perm->command, command, strlen(perm->command))) {
/* if the perm->command is a pattern, check it against command. */
ispattern = !regcomp(&regexbuf, perm->command, REG_EXTENDED | REG_NOSUB | REG_ICASE);
if (ispattern && regexec(&regexbuf, command, 0, NULL, 0)) {
regfree(&regexbuf);
continue;
}
if (!ispattern) {
continue;
}
regfree(&regexbuf);
}
if (user_perm->uid == uid) {
/* this is a user definition. */
isallowu = perm->permit;
} else {
/* otherwise is a group definition. */
isallowg = perm->permit;
}
}
}
AST_RWLIST_UNLOCK(&cli_perms);
if (isallowu > -1) {
/* user definition override group definition. */
isallowg = isallowu;
}
return isallowg;
}
static AST_RWLIST_HEAD_STATIC(helpers, ast_cli_entry);
static char *handle_load(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
/* "module load <mod>" */
switch (cmd) {
case CLI_INIT:
e->command = "module load";
e->usage =
"Usage: module load <module name>\n"
" Loads the specified module into Asterisk.\n";
return NULL;
case CLI_GENERATE:
if (a->pos != e->args) {
return NULL;
}
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_LOAD);
}
if (a->argc != e->args + 1) {
return CLI_SHOWUSAGE;
}
if (ast_load_resource(a->argv[e->args])) {
ast_cli(a->fd, "Unable to load module %s\n", a->argv[e->args]);
return CLI_FAILURE;
}
ast_cli(a->fd, "Loaded %s\n", a->argv[e->args]);
return CLI_SUCCESS;
}
static char *handle_reload(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
int x;
switch (cmd) {
case CLI_INIT:
e->command = "module reload";
e->usage =
"Usage: module reload [module ...]\n"
" Reloads configuration files for all listed modules which support\n"
" reloading, or for all supported modules if none are listed.\n";
return NULL;
case CLI_GENERATE:
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_RELOAD);
}
if (a->argc == e->args) {
ast_module_reload(NULL);
return CLI_SUCCESS;
}
for (x = e->args; x < a->argc; x++) {
enum ast_module_reload_result res = ast_module_reload(a->argv[x]);
switch (res) {
case AST_MODULE_RELOAD_NOT_FOUND:
ast_cli(a->fd, "No such module '%s'\n", a->argv[x]);
break;
case AST_MODULE_RELOAD_NOT_IMPLEMENTED:
ast_cli(a->fd, "The module '%s' does not support reloads\n", a->argv[x]);
break;
case AST_MODULE_RELOAD_QUEUED:
ast_cli(a->fd, "Asterisk cannot reload a module yet; request queued\n");
break;
case AST_MODULE_RELOAD_ERROR:
ast_cli(a->fd, "The module '%s' reported a reload failure\n", a->argv[x]);
break;
case AST_MODULE_RELOAD_IN_PROGRESS:
ast_cli(a->fd, "A module reload request is already in progress; please be patient\n");
break;
case AST_MODULE_RELOAD_UNINITIALIZED:
ast_cli(a->fd, "The module '%s' was not properly initialized. Before reloading"
" the module, you must run \"module load %s\" and fix whatever is"
" preventing the module from being initialized.\n", a->argv[x], a->argv[x]);
break;
case AST_MODULE_RELOAD_SUCCESS:
ast_cli(a->fd, "Module '%s' reloaded successfully.\n", a->argv[x]);
break;
}
}
return CLI_SUCCESS;
}
static char *handle_core_reload(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
switch (cmd) {
case CLI_INIT:
e->command = "core reload";
e->usage =
"Usage: core reload\n"
" Execute a global reload.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
if (a->argc != e->args) {
return CLI_SHOWUSAGE;
}
ast_module_reload(NULL);
return CLI_SUCCESS;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
/*!
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
* \brief Find the module level setting
*
* \param module Module name to look for.
* \param mll List to search.
*
* \retval level struct found on success.
* \retval NULL not found.
*/
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
static struct module_level *find_module_level(const char *module, struct module_level_list *mll)
{
struct module_level *ml;
AST_LIST_TRAVERSE(mll, ml, entry) {
if (!strcasecmp(ml->module, module))
return ml;
}
return NULL;
}
static char *complete_number(const char *partial, unsigned int min, unsigned int max, int n)
{
int i, count = 0;
unsigned int prospective[2];
unsigned int part = strtoul(partial, NULL, 10);
char next[13];
if (part < min || part > max) {
return NULL;
}
for (i = 0; i < 21; i++) {
if (i == 0) {
prospective[0] = prospective[1] = part;
} else if (part == 0 && !ast_strlen_zero(partial)) {
break;
} else if (i < 11) {
prospective[0] = prospective[1] = part * 10 + (i - 1);
} else {
prospective[0] = (part * 10 + (i - 11)) * 10;
prospective[1] = prospective[0] + 9;
}
if (i < 11 && (prospective[0] < min || prospective[0] > max)) {
continue;
} else if (prospective[1] < min || prospective[0] > max) {
continue;
}
if (++count > n) {
if (i < 11) {
snprintf(next, sizeof(next), "%u", prospective[0]);
} else {
snprintf(next, sizeof(next), "%u...", prospective[0] / 10);
}
return ast_strdup(next);
}
}
return NULL;
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
#define DEBUG_HANDLER 0
#define TRACE_HANDLER 1
#define VERBOSE_HANDLER 2
static void status_debug_verbose(struct ast_cli_args *a, int handler, int old_val, int cur_val)
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
{
char was_buf[30];
const char *was;
const char *what = "";
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
switch(handler) {
case DEBUG_HANDLER:
what = "Core debug";
break;
case TRACE_HANDLER:
what = "Core trace";
break;
case VERBOSE_HANDLER:
what = "Console verbose";
break;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (old_val) {
snprintf(was_buf, sizeof(was_buf), "%d", old_val);
was = was_buf;
} else {
was = "OFF";
}
if (old_val == cur_val) {
ast_cli(a->fd, "%s is still %s.\n", what, was);
} else {
char now_buf[30];
const char *now;
if (cur_val) {
snprintf(now_buf, sizeof(now_buf), "%d", cur_val);
now = now_buf;
} else {
now = "OFF";
}
ast_cli(a->fd, "%s was %s and is now %s.\n", what, was, now);
}
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
static char *handle_debug_or_trace(int handler, struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
int oldval;
int newlevel;
int atleast = 0;
struct module_level *ml;
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
struct module_level_list *modules;
unsigned int module_option;
int *core_option;
const char *handler_name;
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->argc <= e->args) {
return CLI_SHOWUSAGE;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
if (handler == DEBUG_HANDLER) {
modules = &debug_modules;
module_option = AST_OPT_FLAG_DEBUG_MODULE;
core_option = &option_debug;
handler_name = "debug";
} else {
modules = &trace_modules;
module_option = AST_OPT_FLAG_TRACE_MODULE;
core_option = &option_trace;
handler_name = "trace";
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->argc == e->args + 1 && !strcasecmp(a->argv[e->args], "off")) {
newlevel = 0;
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
} else {
if (!strcasecmp(a->argv[e->args], "atleast")) {
atleast = 1;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->argc != e->args + atleast + 1 && a->argc != e->args + atleast + 2) {
return CLI_SHOWUSAGE;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (sscanf(a->argv[e->args + atleast], "%30d", &newlevel) != 1) {
return CLI_SHOWUSAGE;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->argc == e->args + atleast + 2) {
/* We have specified a module name. */
char *mod = ast_strdupa(a->argv[e->args + atleast + 1]);
int mod_len = strlen(mod);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (3 < mod_len && !strcasecmp(mod + mod_len - 3, ".so")) {
mod[mod_len - 3] = '\0';
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_WRLOCK(modules);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
ml = find_module_level(mod, modules);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (!newlevel) {
if (!ml) {
/* Specified off for a nonexistent entry. */
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_UNLOCK(modules);
ast_cli(a->fd, "Core %s is still 0 for '%s'.\n", handler_name, mod);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return CLI_SUCCESS;
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_REMOVE(modules, ml, entry);
if (AST_RWLIST_EMPTY(modules)) {
ast_clear_flag(&ast_options, module_option);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_UNLOCK(modules);
ast_cli(a->fd, "Core %s was %u and has been set to 0 for '%s'.\n", handler_name,
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
ml->level, mod);
ast_free(ml);
return CLI_SUCCESS;
}
if (ml) {
if ((atleast && newlevel < ml->level) || ml->level == newlevel) {
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
ast_cli(a->fd, "Core %s is still %u for '%s'.\n", handler_name, ml->level, mod);
AST_RWLIST_UNLOCK(modules);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return CLI_SUCCESS;
}
oldval = ml->level;
ml->level = newlevel;
} else {
ml = ast_calloc(1, sizeof(*ml) + strlen(mod) + 1);
if (!ml) {
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_UNLOCK(modules);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return CLI_FAILURE;
}
oldval = ml->level;
ml->level = newlevel;
strcpy(ml->module, mod);
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_INSERT_TAIL(modules, ml, entry);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
ast_set_flag(&ast_options, module_option);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
ast_cli(a->fd, "Core %s was %d and has been set to %u for '%s'.\n", handler_name,
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
oldval, ml->level, ml->module);
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_UNLOCK(modules);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return CLI_SUCCESS;
}
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
/* Update global debug level */
if (!newlevel) {
/* Specified level was 0 or off. */
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_WRLOCK(modules);
while ((ml = AST_RWLIST_REMOVE_HEAD(modules, entry))) {
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
ast_free(ml);
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
ast_clear_flag(&ast_options, AST_OPT_FLAG_DEBUG_MODULE);
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_RWLIST_UNLOCK(modules);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
oldval = *core_option;
if (!atleast || newlevel > *core_option) {
*core_option = newlevel;
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
/* Report debug level status */
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
status_debug_verbose(a, handler, oldval, *core_option);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return CLI_SUCCESS;
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
static char *handle_debug(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
int atleast = 0;
const char *argv3 = a->argv ? S_OR(a->argv[3], "") : "";
switch (cmd) {
case CLI_INIT:
e->command = "core set debug";
e->usage =
"Usage: core set debug [atleast] <level> [module]\n"
" core set debug off\n"
"\n"
" Sets level of debug messages to be displayed or\n"
" sets a module name to display debug messages from.\n"
" 0 or off means no messages should be displayed.\n"
" Equivalent to -d[d[...]] on startup\n";
return NULL;
case CLI_GENERATE:
if (!strcasecmp(argv3, "category")) {
return NULL;
}
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
if (!strcasecmp(argv3, "atleast")) {
atleast = 1;
}
if (a->pos == 3 || (a->pos == 4 && atleast)) {
const char *pos = a->pos == 3 ? argv3 : S_OR(a->argv[4], "");
int numbermatch = (ast_strlen_zero(pos) || strchr("123456789", pos[0])) ? 0 : 21;
if (a->n < 21 && numbermatch == 0) {
return complete_number(pos, 0, 0x7fffffff, a->n);
} else if (pos[0] == '0') {
if (a->n == 0) {
return ast_strdup("0");
}
} else if (a->n == (21 - numbermatch)) {
if (a->pos == 3 && !strncasecmp(argv3, "off", strlen(argv3))) {
return ast_strdup("off");
} else if (a->pos == 3 && !strncasecmp(argv3, "atleast", strlen(argv3))) {
return ast_strdup("atleast");
}
} else if (a->n == (22 - numbermatch) && a->pos == 3 && ast_strlen_zero(argv3)) {
return ast_strdup("atleast");
}
} else if ((a->pos == 4 && !atleast && strcasecmp(argv3, "off") && strcasecmp(argv3, "channel"))
|| (a->pos == 5 && atleast)) {
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_RUNNING);
}
return NULL;
}
/* all the above return, so we proceed with the handler.
* we are guaranteed to be called with argc >= e->args;
*/
return handle_debug_or_trace(DEBUG_HANDLER, e, cmd, a);
}
static char *handle_trace(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
int atleast = 0;
const char *argv3 = a->argv ? S_OR(a->argv[3], "") : "";
switch (cmd) {
case CLI_INIT:
e->command = "core set trace";
e->usage =
"Usage: core set trace [atleast] <level> [module]\n"
" core set trace off\n"
"\n"
" Sets level of trace messages to be displayed or\n"
" sets a module name to display trace messages from.\n"
" 0 or off means no messages should be displayed.\n";
return NULL;
case CLI_GENERATE:
if (!strcasecmp(argv3, "atleast")) {
atleast = 1;
}
if (a->pos == 3 || (a->pos == 4 && atleast)) {
const char *pos = a->pos == 3 ? argv3 : S_OR(a->argv[4], "");
int numbermatch = (ast_strlen_zero(pos) || strchr("123456789", pos[0])) ? 0 : 21;
if (a->n < 21 && numbermatch == 0) {
return complete_number(pos, 0, 0x7fffffff, a->n);
} else if (pos[0] == '0') {
if (a->n == 0) {
return ast_strdup("0");
}
} else if (a->n == (21 - numbermatch)) {
if (a->pos == 3 && !strncasecmp(argv3, "off", strlen(argv3))) {
return ast_strdup("off");
} else if (a->pos == 3 && !strncasecmp(argv3, "atleast", strlen(argv3))) {
return ast_strdup("atleast");
}
} else if (a->n == (22 - numbermatch) && a->pos == 3 && ast_strlen_zero(argv3)) {
return ast_strdup("atleast");
}
} else if ((a->pos == 4 && !atleast && strcasecmp(argv3, "off") && strcasecmp(argv3, "channel"))
|| (a->pos == 5 && atleast)) {
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_RUNNING);
}
return NULL;
}
/* all the above return, so we proceed with the handler.
* we are guaranteed to be called with argc >= e->args;
*/
return handle_debug_or_trace(TRACE_HANDLER, e, cmd, a);
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
static char *handle_verbose(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
int oldval;
int newlevel;
int atleast = 0;
int silent = 0;
const char *argv3 = a->argv ? S_OR(a->argv[3], "") : "";
switch (cmd) {
case CLI_INIT:
e->command = "core set verbose";
e->usage =
"Usage: core set verbose [atleast] <level> [silent]\n"
" core set verbose off\n"
"\n"
" Sets level of verbose messages to be displayed.\n"
" 0 or off means no verbose messages should be displayed.\n"
" The silent option means the command does not report what\n"
" happened to the verbose level.\n"
" Equivalent to -v[v[...]] on startup\n";
return NULL;
case CLI_GENERATE:
if (!strcasecmp(argv3, "atleast")) {
atleast = 1;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->pos == 3 || (a->pos == 4 && atleast)) {
const char *pos = a->pos == 3 ? argv3 : S_OR(a->argv[4], "");
int numbermatch = (ast_strlen_zero(pos) || strchr("123456789", pos[0])) ? 0 : 21;
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->n < 21 && numbermatch == 0) {
return complete_number(pos, 0, 0x7fffffff, a->n);
} else if (pos[0] == '0') {
if (a->n == 0) {
return ast_strdup("0");
}
} else if (a->n == (21 - numbermatch)) {
if (a->pos == 3 && !strncasecmp(argv3, "off", strlen(argv3))) {
return ast_strdup("off");
} else if (a->pos == 3 && !strncasecmp(argv3, "atleast", strlen(argv3))) {
return ast_strdup("atleast");
}
} else if (a->n == (22 - numbermatch) && a->pos == 3 && ast_strlen_zero(argv3)) {
return ast_strdup("atleast");
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
} else if ((a->pos == 4 && !atleast && strcasecmp(argv3, "off"))
|| (a->pos == 5 && atleast)) {
const char *pos = S_OR(a->argv[a->pos], "");
if (a->n == 0 && !strncasecmp(pos, "silent", strlen(pos))) {
return ast_strdup("silent");
}
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return NULL;
}
/* all the above return, so we proceed with the handler.
* we are guaranteed to be called with argc >= e->args;
*/
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->argc <= e->args) {
return CLI_SHOWUSAGE;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (a->argc == e->args + 1 && !strcasecmp(a->argv[e->args], "off")) {
newlevel = 0;
} else {
if (!strcasecmp(a->argv[e->args], "atleast")) {
atleast = 1;
}
if (a->argc == e->args + atleast + 2
&& !strcasecmp(a->argv[e->args + atleast + 1], "silent")) {
silent = 1;
}
if (a->argc != e->args + atleast + silent + 1) {
return CLI_SHOWUSAGE;
}
if (sscanf(a->argv[e->args + atleast], "%30d", &newlevel) != 1) {
return CLI_SHOWUSAGE;
}
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
/* Update verbose level */
oldval = ast_verb_console_get();
if (!atleast || newlevel > oldval) {
ast_verb_console_set(newlevel);
} else {
newlevel = oldval;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
if (silent) {
/* Be silent after setting the level. */
return CLI_SUCCESS;
}
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
/* Report verbose level status */
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
status_debug_verbose(a, VERBOSE_HANDLER, oldval, newlevel);
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
return CLI_SUCCESS;
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
static char *handle_logger_mute(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
switch (cmd) {
case CLI_INIT:
e->command = "logger mute";
e->usage =
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
"Usage: logger mute\n"
" Disables logging output to the current console, making it possible to\n"
" gather information without being disturbed by scrolling lines.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
if (a->argc < 2 || a->argc > 3)
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SHOWUSAGE;
if (a->argc == 3 && !strcasecmp(a->argv[2], "silent"))
ast_console_toggle_mute(a->fd, 1);
else
ast_console_toggle_mute(a->fd, 0);
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SUCCESS;
}
static char *handle_refresh(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
/* "module refresh <mod>" */
switch (cmd) {
case CLI_INIT:
e->command = "module refresh";
e->usage =
"Usage: module refresh <module name>\n"
" Unloads and loads the specified module into Asterisk.\n";
return NULL;
case CLI_GENERATE:
if (a->pos != e->args) {
return NULL;
}
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_UNLOAD);
}
if (a->argc != e->args + 1) {
return CLI_SHOWUSAGE;
}
if (ast_unload_resource(a->argv[e->args], AST_FORCE_SOFT)) {
ast_cli(a->fd, "Unable to unload resource %s\n", a->argv[e->args]);
return CLI_FAILURE;
}
if (ast_load_resource(a->argv[e->args])) {
ast_cli(a->fd, "Unable to load module %s\n", a->argv[e->args]);
return CLI_FAILURE;
}
ast_cli(a->fd, "Unloaded and loaded %s\n", a->argv[e->args]);
return CLI_SUCCESS;
}
static char *handle_unload(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
/* "module unload mod_1 [mod_2 .. mod_N]" */
int x;
int force = AST_FORCE_SOFT;
const char *s;
switch (cmd) {
case CLI_INIT:
e->command = "module unload";
e->usage =
"Usage: module unload [-f|-h] <module_1> [<module_2> ... ]\n"
" Unloads the specified module from Asterisk. The -f\n"
" option causes the module to be unloaded even if it is\n"
" in use (may cause a crash) and the -h module causes the\n"
" module to be unloaded even if the module says it cannot, \n"
" which almost always will cause a crash.\n";
return NULL;
case CLI_GENERATE:
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_UNLOAD);
}
if (a->argc < e->args + 1)
return CLI_SHOWUSAGE;
x = e->args; /* first argument */
s = a->argv[x];
if (s[0] == '-') {
if (s[1] == 'f')
force = AST_FORCE_FIRM;
else if (s[1] == 'h')
force = AST_FORCE_HARD;
else
return CLI_SHOWUSAGE;
if (a->argc < e->args + 2) /* need at least one module name */
return CLI_SHOWUSAGE;
x++; /* skip this argument */
}
for (; x < a->argc; x++) {
if (ast_unload_resource(a->argv[x], force)) {
ast_cli(a->fd, "Unable to unload resource %s\n", a->argv[x]);
return CLI_FAILURE;
}
ast_cli(a->fd, "Unloaded %s\n", a->argv[x]);
}
return CLI_SUCCESS;
}
#define MODLIST_FORMAT "%-30s %-40.40s %-10d %-11s %13s\n"
#define MODLIST_FORMAT2 "%-30s %-40.40s %-10s %-11s %13s\n"
AST_MUTEX_DEFINE_STATIC(climodentrylock);
static int climodentryfd = -1;
static int modlist_modentry(const char *module, const char *description,
int usecnt, const char *status, const char *like,
enum ast_module_support_level support_level)
{
/* Comparing the like with the module */
if (strcasestr(module, like) ) {
ast_cli(climodentryfd, MODLIST_FORMAT, module, description, usecnt,
status, ast_module_support_level_to_string(support_level));
return 1;
}
return 0;
}
static void print_uptimestr(int fd, struct timeval timeval, const char *prefix, int printsec)
{
int x; /* the main part - years, weeks, etc. */
struct ast_str *out;
#define SECOND (1)
#define MINUTE (SECOND*60)
#define HOUR (MINUTE*60)
#define DAY (HOUR*24)
#define WEEK (DAY*7)
#define YEAR (DAY*365)
#define NEEDCOMMA(x) ((x) ? ", " : "") /* define if we need a comma */
if (timeval.tv_sec < 0) /* invalid, nothing to show */
return;
if (printsec) { /* plain seconds output */
ast_cli(fd, "%s%lu\n", prefix, (u_long)timeval.tv_sec);
return;
}
out = ast_str_alloca(256);
if (timeval.tv_sec > YEAR) {
x = (timeval.tv_sec / YEAR);
timeval.tv_sec -= (x * YEAR);
ast_str_append(&out, 0, "%d year%s%s", x, ESS(x), NEEDCOMMA(timeval.tv_sec));
}
if (timeval.tv_sec > WEEK) {
x = (timeval.tv_sec / WEEK);
timeval.tv_sec -= (x * WEEK);
ast_str_append(&out, 0, "%d week%s%s", x, ESS(x), NEEDCOMMA(timeval.tv_sec));
}
if (timeval.tv_sec > DAY) {
x = (timeval.tv_sec / DAY);
timeval.tv_sec -= (x * DAY);
ast_str_append(&out, 0, "%d day%s%s", x, ESS(x), NEEDCOMMA(timeval.tv_sec));
}
if (timeval.tv_sec > HOUR) {
x = (timeval.tv_sec / HOUR);
timeval.tv_sec -= (x * HOUR);
ast_str_append(&out, 0, "%d hour%s%s", x, ESS(x), NEEDCOMMA(timeval.tv_sec));
}
if (timeval.tv_sec > MINUTE) {
x = (timeval.tv_sec / MINUTE);
timeval.tv_sec -= (x * MINUTE);
ast_str_append(&out, 0, "%d minute%s%s", x, ESS(x), NEEDCOMMA(timeval.tv_sec));
}
x = timeval.tv_sec;
if (x > 0 || ast_str_strlen(out) == 0) {
/* if there is nothing, print 0 seconds */
ast_str_append(&out, 0, "%d second%s", x, ESS(x));
}
ast_cli(fd, "%s%s\n", prefix, ast_str_buffer(out));
}
static struct ast_cli_entry *cli_next(struct ast_cli_entry *e)
{
if (e) {
return AST_LIST_NEXT(e, list);
} else {
return AST_LIST_FIRST(&helpers);
}
}
static char * handle_showuptime(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
struct timeval curtime = ast_tvnow();
int printsec;
switch (cmd) {
case CLI_INIT:
e->command = "core show uptime [seconds]";
e->usage =
"Usage: core show uptime [seconds]\n"
" Shows Asterisk uptime information.\n"
" The seconds word returns the uptime in seconds only.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
/* regular handler */
if (a->argc == e->args && !strcasecmp(a->argv[e->args-1],"seconds"))
printsec = 1;
else if (a->argc == e->args-1)
printsec = 0;
else
return CLI_SHOWUSAGE;
if (ast_startuptime.tv_sec) {
print_uptimestr(a->fd, ast_tvsub(curtime, ast_startuptime), "System uptime: ", printsec);
}
if (ast_lastreloadtime.tv_sec) {
print_uptimestr(a->fd, ast_tvsub(curtime, ast_lastreloadtime), "Last reload: ", printsec);
}
return CLI_SUCCESS;
}
static char *handle_modlist(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
const char *like;
switch (cmd) {
case CLI_INIT:
e->command = "module show [like]";
e->usage =
"Usage: module show [like keyword]\n"
" Shows Asterisk modules currently in use, and usage statistics.\n";
return NULL;
case CLI_GENERATE:
if (a->pos == e->args) {
return ast_module_helper(a->line, a->word, a->pos, a->n, a->pos, AST_MODULE_HELPER_LOADED);
} else {
return NULL;
}
}
/* all the above return, so we proceed with the handler.
* we are guaranteed to have argc >= e->args
*/
if (a->argc == e->args - 1)
like = "";
else if (a->argc == e->args + 1 && !strcasecmp(a->argv[e->args-1], "like") )
like = a->argv[e->args];
else
return CLI_SHOWUSAGE;
ast_mutex_lock(&climodentrylock);
climodentryfd = a->fd; /* global, protected by climodentrylock */
ast_cli(a->fd, MODLIST_FORMAT2, "Module", "Description", "Use Count", "Status", "Support Level");
ast_cli(a->fd,"%d modules loaded\n", ast_update_module_list(modlist_modentry, like));
climodentryfd = -1;
ast_mutex_unlock(&climodentrylock);
return CLI_SUCCESS;
}
#undef MODLIST_FORMAT
#undef MODLIST_FORMAT2
static char *handle_showcalls(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
static const char * const completions[] = { "seconds", NULL };
struct timeval curtime = ast_tvnow();
int showuptime, printsec;
switch (cmd) {
case CLI_INIT:
e->command = "core show calls [uptime]";
e->usage =
"Usage: core show calls [uptime [seconds]]\n"
" Lists number of currently active calls and total number of calls\n"
" processed through PBX since last restart. If 'uptime' is specified\n"
" the system uptime is also displayed. If 'seconds' is specified in\n"
" addition to 'uptime', the system uptime is displayed in seconds.\n";
return NULL;
case CLI_GENERATE:
if (a->pos != e->args)
return NULL;
return ast_cli_complete(a->word, completions, a->n);
}
/* regular handler */
if (a->argc >= e->args && !strcasecmp(a->argv[e->args-1],"uptime")) {
showuptime = 1;
if (a->argc == e->args+1 && !strcasecmp(a->argv[e->args],"seconds"))
printsec = 1;
else if (a->argc == e->args)
printsec = 0;
else
return CLI_SHOWUSAGE;
} else if (a->argc == e->args-1) {
showuptime = 0;
printsec = 0;
} else
return CLI_SHOWUSAGE;
if (ast_option_maxcalls) {
ast_cli(a->fd, "%d of %d max active call%s (%5.2f%% of capacity)\n",
ast_active_calls(), ast_option_maxcalls, ESS(ast_active_calls()),
((double)ast_active_calls() / (double)ast_option_maxcalls) * 100.0);
} else {
ast_cli(a->fd, "%d active call%s\n", ast_active_calls(), ESS(ast_active_calls()));
}
ast_cli(a->fd, "%d call%s processed\n", ast_processed_calls(), ESS(ast_processed_calls()));
if (ast_startuptime.tv_sec && showuptime) {
print_uptimestr(a->fd, ast_tvsub(curtime, ast_startuptime), "System uptime: ", printsec);
}
return RESULT_SUCCESS;
}
static char *handle_chanlist(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
#define FORMAT_STRING "%-64.64s %-32.32s %-7.7s %-30.30s\n"
#define FORMAT_STRING2 "%-64.64s %-32.32s %-7.7s %-30.30s\n"
#define CONCISE_FORMAT_STRING "%s!%s!%s!%d!%s!%s!%s!%s!%s!%s!%d!%s!%s!%s\n"
#define VERBOSE_FORMAT_STRING "%-80.80s %-24.24s %-24.24s %4d %-7.7s %-12.12s %-25.25s %-15.15s %8.8s %-11.11s %-11.11s %-20.20s\n"
#define VERBOSE_FORMAT_STRING2 "%-80.80s %-24.24s %-24.24s %-4.4s %-7.7s %-12.12s %-25.25s %-15.15s %8.8s %-11.11s %-11.11s %-20.20s\n"
struct ao2_container *channels;
struct ao2_iterator it_chans;
struct ast_channel_snapshot *cs;
int numchans = 0, concise = 0, verbose = 0, count = 0;
switch (cmd) {
case CLI_INIT:
e->command = "core show channels [concise|verbose|count]";
e->usage =
"Usage: core show channels [concise|verbose|count]\n"
" Lists currently defined channels and some information about them. If\n"
" 'concise' is specified, the format is abridged and in a more easily\n"
" machine parsable format. If 'verbose' is specified, the output includes\n"
" more and longer fields. If 'count' is specified only the channel and call\n"
" count is output.\n"
" The 'concise' option is deprecated and will be removed from future versions\n"
" of Asterisk.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
if (a->argc == e->args) {
if (!strcasecmp(a->argv[e->args-1],"concise"))
concise = 1;
else if (!strcasecmp(a->argv[e->args-1],"verbose"))
verbose = 1;
else if (!strcasecmp(a->argv[e->args-1],"count"))
count = 1;
else
return CLI_SHOWUSAGE;
} else if (a->argc != e->args - 1)
return CLI_SHOWUSAGE;
channels = ast_channel_cache_by_name();
if (!count) {
if (!concise && !verbose)
ast_cli(a->fd, FORMAT_STRING2, "Channel", "Location", "State", "Application(Data)");
else if (verbose)
ast_cli(a->fd, VERBOSE_FORMAT_STRING2, "Channel", "Context", "Extension", "Priority", "State", "Application", "Data",
"CallerID", "Duration", "Accountcode", "PeerAccount", "BridgeID");
}
it_chans = ao2_iterator_init(channels, 0);
for (; (cs = ao2_iterator_next(&it_chans)); ao2_ref(cs, -1)) {
char durbuf[16] = "-";
if (!count) {
if ((concise || verbose) && !ast_tvzero(cs->base->creationtime)) {
int duration = (int)(ast_tvdiff_ms(ast_tvnow(), cs->base->creationtime) / 1000);
if (verbose) {
int durh = duration / 3600;
int durm = (duration % 3600) / 60;
int durs = duration % 60;
snprintf(durbuf, sizeof(durbuf), "%02d:%02d:%02d", durh, durm, durs);
} else {
snprintf(durbuf, sizeof(durbuf), "%d", duration);
}
}
if (concise) {
ast_cli(a->fd, CONCISE_FORMAT_STRING, cs->base->name, cs->dialplan->context, cs->dialplan->exten, cs->dialplan->priority, ast_state2str(cs->state),
S_OR(cs->dialplan->appl, "(None)"),
cs->dialplan->data,
cs->caller->number,
cs->base->accountcode,
cs->peer->account,
cs->amaflags,
durbuf,
cs->bridge->id,
cs->base->uniqueid);
} else if (verbose) {
ast_cli(a->fd, VERBOSE_FORMAT_STRING, cs->base->name, cs->dialplan->context, cs->dialplan->exten, cs->dialplan->priority, ast_state2str(cs->state),
S_OR(cs->dialplan->appl, "(None)"),
S_OR(cs->dialplan->data, "(Empty)"),
cs->caller->number,
durbuf,
cs->base->accountcode,
cs->peer->account,
cs->bridge->id);
} else {
char locbuf[40] = "(None)";
char appdata[40] = "(None)";
if (!ast_strlen_zero(cs->dialplan->context) && !ast_strlen_zero(cs->dialplan->exten)) {
snprintf(locbuf, sizeof(locbuf), "%s@%s:%d", cs->dialplan->exten, cs->dialplan->context, cs->dialplan->priority);
}
if (!ast_strlen_zero(cs->dialplan->appl)) {
snprintf(appdata, sizeof(appdata), "%s(%s)", cs->dialplan->appl, S_OR(cs->dialplan->data, ""));
}
ast_cli(a->fd, FORMAT_STRING, cs->base->name, locbuf, ast_state2str(cs->state), appdata);
}
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
ao2_iterator_destroy(&it_chans);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (!concise) {
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
numchans = ast_active_channels();
ast_cli(a->fd, "%d active channel%s\n", numchans, ESS(numchans));
if (ast_option_maxcalls)
ast_cli(a->fd, "%d of %d max active call%s (%5.2f%% of capacity)\n",
ast_active_calls(), ast_option_maxcalls, ESS(ast_active_calls()),
((double)ast_active_calls() / (double)ast_option_maxcalls) * 100.0);
else
ast_cli(a->fd, "%d active call%s\n", ast_active_calls(), ESS(ast_active_calls()));
ast_cli(a->fd, "%d call%s processed\n", ast_processed_calls(), ESS(ast_processed_calls()));
}
ao2_ref(channels, -1);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
return CLI_SUCCESS;
#undef FORMAT_STRING
#undef FORMAT_STRING2
#undef CONCISE_FORMAT_STRING
#undef VERBOSE_FORMAT_STRING
#undef VERBOSE_FORMAT_STRING2
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
static char *handle_softhangup(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
struct ast_channel *c = NULL;
static const char * const completions[] = { "all", NULL };
char *complete;
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
switch (cmd) {
case CLI_INIT:
e->command = "channel request hangup";
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
e->usage =
"Usage: channel request hangup <channel>|<all>\n"
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
" Request that a channel be hung up. The hangup takes effect\n"
" the next time the driver reads or writes from the channel.\n"
" If 'all' is specified instead of a channel name, all channels\n"
" will see the hangup request.\n";
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return NULL;
case CLI_GENERATE:
if (a->pos != e->args) {
return NULL;
}
complete = ast_cli_complete(a->word, completions, a->n);
if (!complete) {
complete = ast_complete_channels(a->line, a->word, a->pos, a->n - 1, e->args);
}
return complete;
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (a->argc != 4) {
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SHOWUSAGE;
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
if (!strcasecmp(a->argv[3], "all")) {
struct ast_channel_iterator *iter = NULL;
if (!(iter = ast_channel_iterator_all_new())) {
return CLI_FAILURE;
}
for (; iter && (c = ast_channel_iterator_next(iter)); ast_channel_unref(c)) {
ast_channel_lock(c);
Replace direct access to channel name with accessor functions There are many benefits to making the ast_channel an opaque handle, from increasing maintainability to presenting ways to kill masquerades. This patch kicks things off by taking things a field at a time, renaming the field to '__do_not_use_${fieldname}' and then writing setters/getters and converting the existing code to using them. When all fields are done, we can move ast_channel to a C file from channel.h and lop off the '__do_not_use_'. This patch sets up main/channel_interal_api.c to be the only file that actually accesses the ast_channel's fields directly. The intent would be for any API functions in channel.c to use the accessor functions. No more monkeying around with channel internals. We should use our own APIs. The interesting changes in this patch are the addition of channel_internal_api.c, the moving of the AST_DATA stuff from channel.c to channel_internal_api.c (note: the AST_DATA stuff will have to be reworked to use accessor functions when ast_channel is really opaque), and some re-working of the way channel iterators/callbacks are handled so as to avoid creating fake ast_channels on the stack to pass in matching data by directly accessing fields (since "name" is a stringfield and the fake channel doesn't init the stringfields, you can't use the ast_channel_name_set() function). I went with ast_channel_name(chan) for a getter, and ast_channel_name_set(chan, name) for a setter. The majority of the grunt-work for this change was done by writing a semantic patch using Coccinelle ( http://coccinelle.lip6.fr/ ). Review: https://reviewboard.asterisk.org/r/1655/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@350223 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2012-01-09 22:15:50 +00:00
ast_cli(a->fd, "Requested Hangup on channel '%s'\n", ast_channel_name(c));
ast_softhangup(c, AST_SOFTHANGUP_EXPLICIT);
ast_channel_unlock(c);
}
ast_channel_iterator_destroy(iter);
} else if ((c = ast_channel_get_by_name(a->argv[3]))) {
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
ast_channel_lock(c);
Replace direct access to channel name with accessor functions There are many benefits to making the ast_channel an opaque handle, from increasing maintainability to presenting ways to kill masquerades. This patch kicks things off by taking things a field at a time, renaming the field to '__do_not_use_${fieldname}' and then writing setters/getters and converting the existing code to using them. When all fields are done, we can move ast_channel to a C file from channel.h and lop off the '__do_not_use_'. This patch sets up main/channel_interal_api.c to be the only file that actually accesses the ast_channel's fields directly. The intent would be for any API functions in channel.c to use the accessor functions. No more monkeying around with channel internals. We should use our own APIs. The interesting changes in this patch are the addition of channel_internal_api.c, the moving of the AST_DATA stuff from channel.c to channel_internal_api.c (note: the AST_DATA stuff will have to be reworked to use accessor functions when ast_channel is really opaque), and some re-working of the way channel iterators/callbacks are handled so as to avoid creating fake ast_channels on the stack to pass in matching data by directly accessing fields (since "name" is a stringfield and the fake channel doesn't init the stringfields, you can't use the ast_channel_name_set() function). I went with ast_channel_name(chan) for a getter, and ast_channel_name_set(chan, name) for a setter. The majority of the grunt-work for this change was done by writing a semantic patch using Coccinelle ( http://coccinelle.lip6.fr/ ). Review: https://reviewboard.asterisk.org/r/1655/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@350223 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2012-01-09 22:15:50 +00:00
ast_cli(a->fd, "Requested Hangup on channel '%s'\n", ast_channel_name(c));
ast_softhangup(c, AST_SOFTHANGUP_EXPLICIT);
ast_channel_unlock(c);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
c = ast_channel_unref(c);
} else {
ast_cli(a->fd, "%s is not a known channel\n", a->argv[3]);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SUCCESS;
}
/*! \brief handles CLI command 'cli show permissions' */
static char *handle_cli_show_permissions(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
struct usergroup_cli_perm *cp;
struct cli_perm *perm;
struct passwd *pw = NULL;
struct group *gr = NULL;
switch (cmd) {
case CLI_INIT:
e->command = "cli show permissions";
e->usage =
"Usage: cli show permissions\n"
" Shows CLI configured permissions.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
AST_RWLIST_RDLOCK(&cli_perms);
AST_LIST_TRAVERSE(&cli_perms, cp, list) {
if (cp->uid >= 0) {
pw = getpwuid(cp->uid);
if (pw) {
ast_cli(a->fd, "user: %s [uid=%d]\n", pw->pw_name, cp->uid);
}
} else {
gr = getgrgid(cp->gid);
if (gr) {
ast_cli(a->fd, "group: %s [gid=%d]\n", gr->gr_name, cp->gid);
}
}
ast_cli(a->fd, "Permissions:\n");
if (cp->perms) {
AST_LIST_TRAVERSE(cp->perms, perm, list) {
ast_cli(a->fd, "\t%s -> %s\n", perm->permit ? "permit" : "deny", perm->command);
}
}
ast_cli(a->fd, "\n");
}
AST_RWLIST_UNLOCK(&cli_perms);
return CLI_SUCCESS;
}
/*! \brief handles CLI command 'cli reload permissions' */
static char *handle_cli_reload_permissions(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
switch (cmd) {
case CLI_INIT:
e->command = "cli reload permissions";
e->usage =
"Usage: cli reload permissions\n"
" Reload the 'cli_permissions.conf' file.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
ast_cli_perms_init(1);
return CLI_SUCCESS;
}
/*! \brief handles CLI command 'cli check permissions' */
static char *handle_cli_check_permissions(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
struct passwd *pw = NULL;
struct group *gr;
int gid = -1, uid = -1;
char command[AST_MAX_ARGS] = "";
struct ast_cli_entry *ce = NULL;
int found = 0;
char *group, *tmp;
switch (cmd) {
case CLI_INIT:
e->command = "cli check permissions";
e->usage =
"Usage: cli check permissions {<username>|@<groupname>|<username>@<groupname>} [<command>]\n"
" Check permissions config for a user@group or list the allowed commands for the specified user.\n"
" The username or the groupname may be omitted.\n";
return NULL;
case CLI_GENERATE:
if (a->pos >= 4) {
return ast_cli_generator(a->line + strlen("cli check permissions") + strlen(a->argv[3]) + 1, a->word, a->n);
}
return NULL;
}
if (a->argc < 4) {
return CLI_SHOWUSAGE;
}
tmp = ast_strdupa(a->argv[3]);
group = strchr(tmp, '@');
if (group) {
gr = getgrnam(&group[1]);
if (!gr) {
ast_cli(a->fd, "Unknown group '%s'\n", &group[1]);
return CLI_FAILURE;
}
group[0] = '\0';
gid = gr->gr_gid;
}
if (!group && ast_strlen_zero(tmp)) {
ast_cli(a->fd, "You didn't supply a username\n");
} else if (!ast_strlen_zero(tmp) && !(pw = getpwnam(tmp))) {
ast_cli(a->fd, "Unknown user '%s'\n", tmp);
return CLI_FAILURE;
} else if (pw) {
uid = pw->pw_uid;
}
if (a->argc == 4) {
while ((ce = cli_next(ce))) {
/* Hide commands that start with '_' */
if (ce->_full_cmd[0] == '_') {
continue;
}
if (cli_has_permissions(uid, gid, ce->_full_cmd)) {
ast_cli(a->fd, "%30.30s %s\n", ce->_full_cmd, S_OR(ce->summary, "<no description available>"));
found++;
}
}
if (!found) {
ast_cli(a->fd, "You are not allowed to run any command on Asterisk\n");
}
} else {
ast_join(command, sizeof(command), a->argv + 4);
ast_cli(a->fd, "%s '%s%s%s' is %s to run command: '%s'\n", uid >= 0 ? "User" : "Group", tmp,
group && uid >= 0 ? "@" : "",
group ? &group[1] : "",
cli_has_permissions(uid, gid, command) ? "allowed" : "not allowed", command);
}
return CLI_SUCCESS;
}
static char *__ast_cli_generator(const char *text, const char *word, int state, int lock);
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
static char *handle_commandmatchesarray(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
char *buf, *obuf;
int buflen = 2048;
int len = 0;
char **matches;
int x, matchlen;
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
switch (cmd) {
case CLI_INIT:
e->command = "_command matchesarray";
e->usage =
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
"Usage: _command matchesarray \"<line>\" text \n"
" This function is used internally to help with command completion and should.\n"
" never be called by the user directly.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
if (a->argc != 4)
return CLI_SHOWUSAGE;
if (!(buf = ast_malloc(buflen)))
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_FAILURE;
buf[len] = '\0';
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
matches = ast_cli_completion_matches(a->argv[2], a->argv[3]);
if (matches) {
for (x=0; matches[x]; x++) {
matchlen = strlen(matches[x]) + 1;
if (len + matchlen >= buflen) {
buflen += matchlen * 3;
obuf = buf;
if (!(buf = ast_realloc(obuf, buflen)))
/* Memory allocation failure... Just free old buffer and be done */
ast_free(obuf);
}
if (buf)
len += sprintf( buf + len, "%s ", matches[x]);
ast_free(matches[x]);
matches[x] = NULL;
}
ast_free(matches);
}
if (buf) {
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
ast_cli(a->fd, "%s%s",buf, AST_CLI_COMPLETE_EOF);
ast_free(buf);
} else
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
ast_cli(a->fd, "NULL\n");
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SUCCESS;
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
struct channel_set_debug_args {
int fd;
int is_off;
};
static int channel_set_debug(void *obj, void *arg, void *data, int flags)
{
struct ast_channel *chan = obj;
struct channel_set_debug_args *args = data;
ast_channel_lock(chan);
if (!(ast_channel_fin(chan) & DEBUGCHAN_FLAG) || !(ast_channel_fout(chan) & DEBUGCHAN_FLAG)) {
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (args->is_off) {
ast_channel_fin_set(chan, ast_channel_fin(chan) & ~DEBUGCHAN_FLAG);
ast_channel_fout_set(chan, ast_channel_fout(chan) & ~DEBUGCHAN_FLAG);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
} else {
ast_channel_fin_set(chan, ast_channel_fin(chan) | DEBUGCHAN_FLAG);
ast_channel_fout_set(chan, ast_channel_fout(chan) | DEBUGCHAN_FLAG);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
ast_cli(args->fd, "Debugging %s on channel %s\n", args->is_off ? "disabled" : "enabled",
Replace direct access to channel name with accessor functions There are many benefits to making the ast_channel an opaque handle, from increasing maintainability to presenting ways to kill masquerades. This patch kicks things off by taking things a field at a time, renaming the field to '__do_not_use_${fieldname}' and then writing setters/getters and converting the existing code to using them. When all fields are done, we can move ast_channel to a C file from channel.h and lop off the '__do_not_use_'. This patch sets up main/channel_interal_api.c to be the only file that actually accesses the ast_channel's fields directly. The intent would be for any API functions in channel.c to use the accessor functions. No more monkeying around with channel internals. We should use our own APIs. The interesting changes in this patch are the addition of channel_internal_api.c, the moving of the AST_DATA stuff from channel.c to channel_internal_api.c (note: the AST_DATA stuff will have to be reworked to use accessor functions when ast_channel is really opaque), and some re-working of the way channel iterators/callbacks are handled so as to avoid creating fake ast_channels on the stack to pass in matching data by directly accessing fields (since "name" is a stringfield and the fake channel doesn't init the stringfields, you can't use the ast_channel_name_set() function). I went with ast_channel_name(chan) for a getter, and ast_channel_name_set(chan, name) for a setter. The majority of the grunt-work for this change was done by writing a semantic patch using Coccinelle ( http://coccinelle.lip6.fr/ ). Review: https://reviewboard.asterisk.org/r/1655/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@350223 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2012-01-09 22:15:50 +00:00
ast_channel_name(chan));
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
ast_channel_unlock(chan);
return 0;
}
static char *handle_core_set_debug_channel(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
struct ast_channel *c = NULL;
static const char * const completions_all[] = { "all", NULL };
static const char * const completions_off[] = { "off", NULL };
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
struct channel_set_debug_args args = {
.fd = a->fd,
};
switch (cmd) {
case CLI_INIT:
e->command = "core set debug channel";
e->usage =
"Usage: core set debug channel <all|channel> [off]\n"
" Enables/disables debugging on all or on a specific channel.\n";
return NULL;
case CLI_GENERATE:
if (a->pos == 4) {
char *complete = ast_cli_complete(a->word, completions_all, a->n);
if (!complete) {
complete = ast_complete_channels(a->line, a->word, a->pos, a->n - 1, e->args);
}
return complete;
} else if (a->pos == 5) {
return ast_cli_complete(a->word, completions_off, a->n);
}
return NULL;
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (cmd == (CLI_HANDLER + 1000)) {
/* called from handle_nodebugchan_deprecated */
args.is_off = 1;
} else if (a->argc == e->args + 2) {
/* 'core set debug channel {all|chan_id}' */
if (!strcasecmp(a->argv[e->args + 1], "off"))
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
args.is_off = 1;
else
return CLI_SHOWUSAGE;
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
} else if (a->argc != e->args + 1) {
return CLI_SHOWUSAGE;
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (!strcasecmp("all", a->argv[e->args])) {
if (args.is_off) {
global_fin &= ~DEBUGCHAN_FLAG;
global_fout &= ~DEBUGCHAN_FLAG;
} else {
global_fin |= DEBUGCHAN_FLAG;
global_fout |= DEBUGCHAN_FLAG;
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
ast_channel_callback(channel_set_debug, NULL, &args, OBJ_NODATA | OBJ_MULTIPLE);
} else {
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if ((c = ast_channel_get_by_name(a->argv[e->args]))) {
channel_set_debug(c, NULL, &args, 0);
ast_channel_unref(c);
} else {
ast_cli(a->fd, "No such channel %s\n", a->argv[e->args]);
}
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
ast_cli(a->fd, "Debugging on new channels is %s\n", args.is_off ? "disabled" : "enabled");
return CLI_SUCCESS;
}
static char *handle_debug_category(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
const char *argv4 = a->argv ? S_OR(a->argv[4], "") : "";
int offset = strncasecmp(argv4, "off", strlen(argv4)) ? 0 : 1;
switch (cmd) {
case CLI_INIT:
e->command = "core set debug category";
e->usage =
"Usage: core set debug category <category>[:<sublevel>] [category[:<sublevel] ...]\n"
" core set debug category off [<category> [<category>] ...]\n\n"
" Allows enabling and disabling debug logging categories.\n"
" When a category is enabled all relevant debug messages are logged\n"
" for a given category. However, if a sublevel is specified only\n"
" those categorized messages at or below the coded debug sublevel\n"
" are logged.\n";
return NULL;
case CLI_GENERATE:
if (a->pos < e->args) {
return NULL;
}
if (a->pos == 4 && offset) {
ast_cli_completion_add(ast_strdup("off"));
}
return ast_debug_category_complete(a->argv + 4,
a->pos - e->args, a->word, a->n - 1);
}
if (a->argc <= e->args) {
return CLI_SHOWUSAGE;
}
ast_debug_category_set_sublevels(a->argv + e->args + offset, a->argc - e->args - offset,
offset ? AST_LOG_CATEGORY_DISABLED : AST_LOG_CATEGORY_ENABLED);
return CLI_SUCCESS;
}
static char *handle_nodebugchan_deprecated(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
char *res;
switch (cmd) {
case CLI_INIT:
e->command = "no debug channel";
return NULL;
case CLI_HANDLER:
/* exit out of switch statement */
break;
default:
return NULL;
}
if (a->argc != e->args + 1)
return CLI_SHOWUSAGE;
/* add a 'magic' value to the CLI_HANDLER command so that
* handle_core_set_debug_channel() will act as if 'off'
* had been specified as part of the command
*/
res = handle_core_set_debug_channel(e, CLI_HANDLER + 1000, a);
return res;
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
static char *handle_showchan(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
struct ast_channel *chan;
struct timeval now;
char cdrtime[256];
struct ast_str *obuf;/*!< Buffer for CDR variables. */
struct ast_str *output;/*!< Accumulation buffer for all output. */
long elapsed_seconds=0;
int hour=0, min=0, sec=0;
struct ast_var_t *var;
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
struct ast_str *write_transpath = ast_str_alloca(256);
struct ast_str *read_transpath = ast_str_alloca(256);
struct ast_str *codec_buf = ast_str_alloca(AST_FORMAT_CAP_NAMES_LEN);
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
struct ast_bridge *bridge;
ast_callid callid;
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
char callid_buf[32];
int stream_num;
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
switch (cmd) {
case CLI_INIT:
e->command = "core show channel";
e->usage =
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
"Usage: core show channel <channel>\n"
" Shows lots of information about the specified channel.\n";
return NULL;
case CLI_GENERATE:
return ast_complete_channels(a->line, a->word, a->pos, a->n, 3);
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (a->argc != 4) {
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SHOWUSAGE;
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
obuf = ast_str_thread_get(&ast_str_thread_global_buf, 16);
if (!obuf) {
return CLI_FAILURE;
}
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
chan = ast_channel_get_by_name(a->argv[3]);
if (!chan) {
ast_cli(a->fd, "%s is not a known channel\n", a->argv[3]);
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
return CLI_SUCCESS;
}
output = ast_str_create(8192);
if (!output) {
ast_channel_unref(chan);
return CLI_FAILURE;
}
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
now = ast_tvnow();
ast_channel_lock(chan);
if (!ast_tvzero(ast_channel_creationtime(chan))) {
elapsed_seconds = now.tv_sec - ast_channel_creationtime(chan).tv_sec;
hour = elapsed_seconds / 3600;
min = (elapsed_seconds % 3600) / 60;
sec = elapsed_seconds % 60;
snprintf(cdrtime, sizeof(cdrtime), "%dh%dm%ds", hour, min, sec);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
} else {
strcpy(cdrtime, "N/A");
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
ast_translate_path_to_str(ast_channel_writetrans(chan), &write_transpath);
ast_translate_path_to_str(ast_channel_readtrans(chan), &read_transpath);
bridge = ast_channel_get_bridge(chan);
callid_buf[0] = '\0';
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
callid = ast_channel_callid(chan);
if (callid) {
ast_callid_strnprint(callid_buf, sizeof(callid_buf), callid);
}
ast_str_append(&output, 0,
" -- General --\n"
" Name: %s\n"
" Type: %s\n"
" UniqueID: %s\n"
" LinkedID: %s\n"
" Caller ID: %s\n"
" Caller ID Name: %s\n"
"Connected Line ID: %s\n"
"Connected Line ID Name: %s\n"
Add private representation of caller, connected and redirecting party ids. This patch adds the feature "Private representation of caller, connected and redirecting party ids", as previously discussed with us (DATUS) and Digium. 1. Feature motivation Until now it is quite difficult to modify a party number or name which can only be seen by exactly one particular instantiated technology channel subscriber. One example where a modified party number or name on one channel is spread over several channels are supplementary services like call transfer or pickup. To implement these features Asterisk internally copies caller and connected ids from one channel to another. Another example are extension subscriptions. The monitoring entities (watchers) are notified of state changes and - if desired - of party numbers or names which represent the involving call parties. One major feature where a private representation of party names is essentially needed, i.e. where a party name shall be exclusively signaled to only one particular user, is a private user-specific name resolution for party numbers. A lookup in a private destination-dependent telephone book shall provide party names which cannot be seen by any other user at any time. 2. Feature Description This feature comes along with the implementation of additional private party id elements for caller id, connected id and redirecting ids inside Asterisk channels. The private party id elements can be read or set by the user using Asterisk dialplan functions. When a technology channel is initiating a call, receives an internal connected-line update event, or receives an internal redirecting update event, it merges the corresponding public id with the private id to create an effective party id. The effective party id is then used for protocol signaling. The channel technologies which initially support the private id representation with this patch are SIP (chan_sip), mISDN (chan_misdn) and PRI (chan_dahdi). Once a private name or number on a channel is set and (implicitly) made valid, it is generally used for any further protocol signaling until it is rewritten or invalidated. To simplify the invalidation of private ids all internally generated connected/redirecting update events and also all connected/redirecting update events which are generated by technology channels -- receiving regarding protocol information - automatically trigger the invalidation of private ids. If not using the private party id representation feature at all, i.e. if using only the 'regular' caller-id, connected and redirecting related functions, the current characteristic of Asterisk is not affected by the new extended functionality. 3. User interface Description To grant access to the private name and number representation from the Asterisk dialplan, the CALLERID, CONNECTEDLINE and REDIRECTING dialplan functions are extended by the following data types. The formats of these data types are equal to the corresponding regular 'non-private' already existing data types: CALLERID: priv-all priv-name priv-name-valid priv-name-charset priv-name-pres priv-num priv-num-valid priv-num-plan priv-num-pres priv-subaddr priv-subaddr-valid priv-subaddr-type priv-subaddr-odd priv-tag CONNECTEDLINE: priv-name priv-name-valid priv-name-pres priv-name-charset priv-num priv-num-valid priv-num-pres priv-num-plan priv-subaddr priv-subaddr-valid priv-subaddr-type priv-subaddr-odd priv-tag REDIRECTING: priv-orig-name priv-orig-name-valid priv-orig-name-pres priv-orig-name-charset priv-orig-num priv-orig-num-valid priv-orig-num-pres priv-orig-num-plan priv-orig-subaddr priv-orig-subaddr-valid priv-orig-subaddr-type priv-orig-subaddr-odd priv-orig-tag priv-from-name priv-from-name-valid priv-from-name-pres priv-from-name-charset priv-from-num priv-from-num-valid priv-from-num-pres priv-from-num-plan priv-from-subaddr priv-from-subaddr-valid priv-from-subaddr-type priv-from-subaddr-odd priv-from-tag priv-to-name priv-to-name-valid priv-to-name-pres priv-to-name-charset priv-to-num priv-to-num-valid priv-to-num-pres priv-to-num-plan priv-to-subaddr priv-to-subaddr-valid priv-to-subaddr-type priv-to-subaddr-odd priv-to-tag Reported by: Thomas Arimont Review: https://reviewboard.asterisk.org/r/2030/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@371120 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2012-08-10 19:54:55 +00:00
"Eff. Connected Line ID: %s\n"
"Eff. Connected Line ID Name: %s\n"
" DNID Digits: %s\n"
" Language: %s\n"
" State: %s (%u)\n"
" NativeFormats: %s\n"
" WriteFormat: %s\n"
" ReadFormat: %s\n"
Merged revisions 282047 via svnmerge from https://origsvn.digium.com/svn/asterisk/branches/1.8 ........ r282047 | dvossel | 2010-08-12 15:15:41 -0500 (Thu, 12 Aug 2010) | 35 lines improved translation paths for wideband codecs The problem I'm addressing is that Asterisk's current method of building the least cost translation paths between codecs does not take into account sample rate. For instance, it was possible for siren14 (a 32khz codec), to contain the a translation path to siren7 (a 16khz audio codec) that goes through slin at 8khz. In this case Asterisk takes a 32khz codec, down samples it to 8khz and then up samples it to 16khz which is terrible regardless if it is computationally less expensive. This patch now builds translation paths that give priority to maintaining the best possible sample rate before taking into consideration computational cost. This patch also adds cli commands to expose what translation paths are actually being used. Changes: 1. Translation paths will never contain a step that changes the sample rate unless absolutely necessary. 2. When choosing the best codec to make two channels compatible. Shared codecs with the highest sample rate are given priority. 3. A new cli command to show all translation paths available for a specific codec 'core show translation paths [codec name]' has been added. 4. 'core show translation' which displays the translation matrix now includes the new higher bit audio codecs in the table. 5. 'core show channel [channel name]' now displays the translation paths if translation is used. (closes issue #16841) Reported by: dvossel Review: https://reviewboard.asterisk.org/r/842/ ........ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@282048 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2010-08-12 20:17:17 +00:00
" WriteTranscode: %s %s\n"
" ReadTranscode: %s %s\n"
" Time to Hangup: %ld\n"
" Elapsed Time: %s\n"
" Bridge ID: %s\n"
" -- PBX --\n"
" Context: %s\n"
" Extension: %s\n"
" Priority: %d\n"
" Call Group: %llu\n"
" Pickup Group: %llu\n"
" Application: %s\n"
" Data: %s\n"
" Call Identifer: %s\n",
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
ast_channel_name(chan),
ast_channel_tech(chan)->type,
ast_channel_uniqueid(chan),
ast_channel_linkedid(chan),
S_COR(ast_channel_caller(chan)->id.number.valid,
ast_channel_caller(chan)->id.number.str, "(N/A)"),
S_COR(ast_channel_caller(chan)->id.name.valid,
ast_channel_caller(chan)->id.name.str, "(N/A)"),
S_COR(ast_channel_connected(chan)->id.number.valid,
ast_channel_connected(chan)->id.number.str, "(N/A)"),
S_COR(ast_channel_connected(chan)->id.name.valid,
ast_channel_connected(chan)->id.name.str, "(N/A)"),
S_COR(ast_channel_connected_effective_id(chan).number.valid,
ast_channel_connected_effective_id(chan).number.str, "(N/A)"),
S_COR(ast_channel_connected_effective_id(chan).name.valid,
ast_channel_connected_effective_id(chan).name.str, "(N/A)"),
S_OR(ast_channel_dialed(chan)->number.str, "(N/A)"),
ast_channel_language(chan),
ast_state2str(ast_channel_state(chan)),
ast_channel_state(chan),
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
ast_format_cap_get_names(ast_channel_nativeformats(chan), &codec_buf),
ast_format_get_name(ast_channel_writeformat(chan)),
ast_format_get_name(ast_channel_readformat(chan)),
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
ast_str_strlen(write_transpath) ? "Yes" : "No",
ast_str_buffer(write_transpath),
ast_str_strlen(read_transpath) ? "Yes" : "No",
ast_str_buffer(read_transpath),
(long)ast_channel_whentohangup(chan)->tv_sec,
cdrtime,
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
bridge ? bridge->uniqueid : "(Not bridged)",
ast_channel_context(chan),
ast_channel_exten(chan),
ast_channel_priority(chan),
ast_channel_callgroup(chan),
ast_channel_pickupgroup(chan),
S_OR(ast_channel_appl(chan), "(N/A)"),
S_OR(ast_channel_data(chan), "(Empty)"),
S_OR(callid_buf, "(None)")
);
ast_str_append(&output, 0, " Variables:\n");
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
AST_LIST_TRAVERSE(ast_channel_varshead(chan), var, entries) {
ast_str_append(&output, 0, "%s=%s\n", ast_var_name(var), ast_var_value(var));
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (!(ast_channel_tech(chan)->properties & AST_CHAN_TP_INTERNAL)
&& ast_cdr_serialize_variables(ast_channel_name(chan), &obuf, '=', '\n')) {
ast_str_append(&output, 0, " CDR Variables:\n%s\n", ast_str_buffer(obuf));
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
ast_str_append(&output, 0, " -- Streams --\n");
for (stream_num = 0; stream_num < ast_stream_topology_get_count(ast_channel_get_stream_topology(chan)); stream_num++) {
struct ast_stream *stream = ast_stream_topology_get_stream(ast_channel_get_stream_topology(chan), stream_num);
struct ast_variable *metadata = ast_stream_get_metadata_list(stream);
ast_str_append(&output, 0,
"Name: %s\n"
" Type: %s\n"
" State: %s\n"
" Group: %d\n"
" Formats: %s\n"
" Metadata:\n",
ast_stream_get_name(stream),
ast_codec_media_type2str(ast_stream_get_type(stream)),
ast_stream_state2str(ast_stream_get_state(stream)),
ast_stream_get_group(stream),
ast_format_cap_get_names(ast_stream_get_formats(stream), &codec_buf)
);
if (metadata) {
struct ast_variable *v;
for(v = metadata; v; v = v->next) {
ast_str_append(&output, 0, " %s: %s\n", v->name, v->value);
}
ast_variables_destroy(metadata);
}
}
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
ast_channel_unlock(chan);
ast_cli(a->fd, "%s", ast_str_buffer(output));
ast_free(output);
stasis: Reduce creation of channel snapshots to improve performance During some performance testing of Asterisk with AGI, ARI, and lots of Local channels, we noticed that there's quite a hit in performance during channel creation and releasing to the dialplan (ARI continue). After investigating the performance spike that occurs during channel creation, we discovered that we create a lot of channel snapshots that are technically unnecessary. This includes creating snapshots during: * AGI execution * Returning objects for ARI commands * During some Local channel operations * During some dialling operations * During variable setting * During some bridging operations And more. This patch does the following: - It removes a number of fields from channel snapshots. These fields were rarely used, were expensive to have on the snapshot, and hurt performance. This included formats, translation paths, Log Call ID, callgroup, pickup group, and all channel variables. As a result, AMI Status, "core show channel", "core show channelvar", and "pjsip show channel" were modified to either hit the live channel or not show certain pieces of data. While this is unfortunate, the performance gain from this patch is worth the loss in behaviour. - It adds a mechanism to publish a cached snapshot + blob. A large number of publications were changed to use this, including: - During Dial begin - During Variable assignment (if no AMI variables are emitted - if AMI variables are set, we have to make snapshots when a variable is changed) - During channel pickup - When a channel is put on hold/unhold - When a DTMF digit is begun/ended - When creating a bridge snapshot - When an AOC event is raised - During Local channel optimization/Local bridging - When endpoint snapshots are generated - All AGI events - All ARI responses that return a channel - Events in the AgentPool, MeetMe, and some in Queue - Additionally, some extraneous channel snapshots were being made that were unnecessary. These were removed. - The result of ast_hashtab_hash_string is now cached in stasis_cache. This reduces a large number of calls to ast_hashtab_hash_string, which reduced the amount of time spent in this function in gprof by around 50%. #ASTERISK-23811 #close Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/3568/ ........ Merged revisions 416211 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@416216 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-06-13 18:24:49 +00:00
ao2_cleanup(bridge);
ast_channel_unref(chan);
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
return CLI_SUCCESS;
}
/*
* helper function to generate CLI matches from a fixed set of values.
* A NULL word is acceptable.
*/
char *ast_cli_complete(const char *word, const char * const choices[], int state)
{
int i, which = 0, len;
len = ast_strlen_zero(word) ? 0 : strlen(word);
for (i = 0; choices[i]; i++) {
if ((!len || !strncasecmp(word, choices[i], len)) && ++which > state) {
if (state != -1) {
return ast_strdup(choices[i]);
}
if (ast_cli_completion_add(ast_strdup(choices[i]))) {
return NULL;
}
}
}
return NULL;
}
char *ast_complete_channels(const char *line, const char *word, int pos, int state, int rpos)
{
int wordlen = strlen(word), which = 0;
struct ao2_container *cached_channels;
char *ret = NULL;
struct ao2_iterator iter;
struct ast_channel_snapshot *snapshot;
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
if (pos != rpos) {
return NULL;
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
cached_channels = ast_channel_cache_all();
iter = ao2_iterator_init(cached_channels, 0);
for (; (snapshot = ao2_iterator_next(&iter)); ao2_ref(snapshot, -1)) {
if (!strncasecmp(word, snapshot->base->name, wordlen) && (++which > state)) {
if (state != -1) {
ret = ast_strdup(snapshot->base->name);
ao2_ref(snapshot, -1);
break;
}
if (ast_cli_completion_add(ast_strdup(snapshot->base->name))) {
ao2_ref(snapshot, -1);
break;
}
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
}
}
ao2_iterator_destroy(&iter);
ao2_ref(cached_channels, -1);
Convert the ast_channel data structure over to the astobj2 framework. There is a lot that could be said about this, but the patch is a big improvement for performance, stability, code maintainability, and ease of future code development. The channel list is no longer an unsorted linked list. The main container for channels is an astobj2 hash table. All of the code related to searching for channels or iterating active channels has been rewritten. Let n be the number of active channels. Iterating the channel list has gone from O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1). Searching for a channel by extension is still O(n), but uses a new method for doing so, which is more efficient. The ast_channel object is now a reference counted object. The benefits here are plentiful. Some benefits directly related to issues in the previous code include: 1) When threads other than the channel thread owning a channel wanted access to a channel, it had to hold the lock on it to ensure that it didn't go away. This is no longer a requirement. Holding a reference is sufficient. 2) There are places that now require less dealing with channel locks. 3) There are places where channel locks are held for much shorter periods of time. 4) There are places where dealing with more than one channel at a time becomes _MUCH_ easier. ChanSpy is a great example of this. Writing code in the future that deals with multiple channels will be much easier. Some additional information regarding channel locking and reference count handling can be found in channel.h, where a new section has been added that discusses some of the rules associated with it. Mark Michelson also assisted with the development of this patch. He did the conversion of ChanSpy and introduced a new API, ast_autochan, which makes it much easier to deal with holding on to a channel pointer for an extended period of time and having it get automatically updated if the channel gets masqueraded. Mark was also a huge help in the code review process. Thanks to David Vossel for his assistance with this branch, as well. David did the conversion of the DAHDIScan application by making it become a wrapper for ChanSpy internally. The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch. Review: http://reviewboard.digium.com/r/203/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@190423 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2009-04-24 14:04:26 +00:00
return ret;
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
static char *group_show_channels(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
#define FORMAT_STRING "%-25s %-20s %-20s\n"
struct ast_group_info *gi = NULL;
int numchans = 0;
regex_t regexbuf;
int havepattern = 0;
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
switch (cmd) {
case CLI_INIT:
e->command = "group show channels";
e->usage =
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
"Usage: group show channels [pattern]\n"
" Lists all currently active channels with channel group(s) specified.\n"
" Optional regular expression pattern is matched to group names for each\n"
" channel.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
if (a->argc < 3 || a->argc > 4)
return CLI_SHOWUSAGE;
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
if (a->argc == 4) {
if (regcomp(&regexbuf, a->argv[3], REG_EXTENDED | REG_NOSUB))
return CLI_SHOWUSAGE;
havepattern = 1;
}
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
ast_cli(a->fd, FORMAT_STRING, "Channel", "Group", "Category");
ast_app_group_list_rdlock();
gi = ast_app_group_list_head();
while (gi) {
if (!havepattern || !regexec(&regexbuf, gi->group, 0, NULL, 0)) {
Replace direct access to channel name with accessor functions There are many benefits to making the ast_channel an opaque handle, from increasing maintainability to presenting ways to kill masquerades. This patch kicks things off by taking things a field at a time, renaming the field to '__do_not_use_${fieldname}' and then writing setters/getters and converting the existing code to using them. When all fields are done, we can move ast_channel to a C file from channel.h and lop off the '__do_not_use_'. This patch sets up main/channel_interal_api.c to be the only file that actually accesses the ast_channel's fields directly. The intent would be for any API functions in channel.c to use the accessor functions. No more monkeying around with channel internals. We should use our own APIs. The interesting changes in this patch are the addition of channel_internal_api.c, the moving of the AST_DATA stuff from channel.c to channel_internal_api.c (note: the AST_DATA stuff will have to be reworked to use accessor functions when ast_channel is really opaque), and some re-working of the way channel iterators/callbacks are handled so as to avoid creating fake ast_channels on the stack to pass in matching data by directly accessing fields (since "name" is a stringfield and the fake channel doesn't init the stringfields, you can't use the ast_channel_name_set() function). I went with ast_channel_name(chan) for a getter, and ast_channel_name_set(chan, name) for a setter. The majority of the grunt-work for this change was done by writing a semantic patch using Coccinelle ( http://coccinelle.lip6.fr/ ). Review: https://reviewboard.asterisk.org/r/1655/ git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@350223 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2012-01-09 22:15:50 +00:00
ast_cli(a->fd, FORMAT_STRING, ast_channel_name(gi->chan), gi->group, (ast_strlen_zero(gi->category) ? "(default)" : gi->category));
numchans++;
}
gi = AST_LIST_NEXT(gi, group_list);
}
ast_app_group_list_unlock();
if (havepattern)
regfree(&regexbuf);
Merge a ton of NEW_CLI conversions. Thanks to everyone that helped out! :) (closes issue #10724) Reported by: eliel Patches: chan_skinny.c.patch uploaded by eliel (license 64) chan_oss.c.patch uploaded by eliel (license 64) chan_mgcp.c.patch2 uploaded by eliel (license 64) pbx_config.c.patch uploaded by seanbright (license 71) iax2-provision.c.patch uploaded by eliel (license 64) chan_gtalk.c.patch uploaded by eliel (license 64) pbx_ael.c.patch uploaded by seanbright (license 71) file.c.patch uploaded by seanbright (license 71) image.c.patch uploaded by seanbright (license 71) cli.c.patch uploaded by moy (license 222) astobj2.c.patch uploaded by moy (license 222) asterisk.c.patch uploaded by moy (license 222) res_limit.c.patch uploaded by seanbright (license 71) res_convert.c.patch uploaded by seanbright (license 71) res_crypto.c.patch uploaded by seanbright (license 71) app_osplookup.c.patch uploaded by seanbright (license 71) app_rpt.c.patch uploaded by seanbright (license 71) app_mixmonitor.c.patch uploaded by seanbright (license 71) channel.c.patch uploaded by seanbright (license 71) translate.c.patch uploaded by seanbright (license 71) udptl.c.patch uploaded by seanbright (license 71) threadstorage.c.patch uploaded by seanbright (license 71) db.c.patch uploaded by seanbright (license 71) cdr.c.patch uploaded by moy (license 222) pbd_dundi.c.patch uploaded by moy (license 222) app_osplookup-rev83558.patch uploaded by moy (license 222) res_clioriginate.c.patch uploaded by moy (license 222) git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@85460 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2007-10-11 19:03:06 +00:00
ast_cli(a->fd, "%d active channel%s\n", numchans, ESS(numchans));
return CLI_SUCCESS;
#undef FORMAT_STRING
}
static char *handle_cli_wait_fullybooted(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
switch (cmd) {
case CLI_INIT:
e->command = "core waitfullybooted";
e->usage =
"Usage: core waitfullybooted\n"
" Wait until Asterisk has fully booted.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
while (!ast_test_flag(&ast_options, AST_OPT_FLAG_FULLY_BOOTED)) {
usleep(100);
}
ast_cli(a->fd, "Asterisk has fully booted.\n");
return CLI_SUCCESS;
}
#ifdef HAVE_MALLOC_TRIM
/*!
* \internal
* \brief Attempt to reclaim unused heap memory.
*
* Users have reported that asterisk will sometimes be killed because it can't allocate
* more than around 3G of memory on a 32 bit system.
*
* Using malloc_trim() will help us to determine if it's because there's a leak or because
* the heap is so fragmented that there isn't enough contiguous memory available.
*
* \note malloc_trim() is a GNU extension and is therefore not portable.
*/
static char *handle_cli_malloc_trim(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
extern int malloc_trim(size_t __pad) __THROW;
switch (cmd) {
case CLI_INIT:
e->command = "malloc trim";
e->usage =
"Usage: malloc trim\n"
" Try to give excess memory back to the OS.\n";
return NULL;
case CLI_GENERATE:
return NULL;
}
if (malloc_trim(0)) {
ast_cli(a->fd, "Returned some memory to the OS.\n");
} else {
ast_cli(a->fd, "No memory returned to the OS.\n");
}
return CLI_SUCCESS;
}
#endif
static char *handle_help(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a);
static struct ast_cli_entry cli_cli[] = {
AST_CLI_DEFINE(handle_commandmatchesarray, "Returns command matches array"),
AST_CLI_DEFINE(handle_debug_category, "Enable/disable debugging categories"),
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
AST_CLI_DEFINE(handle_debug, "Set level of debug chattiness"),
Scope Tracing: A new facility for tracing scope enter/exit What's wrong with ast_debug? ast_debug is fine for general purpose debug output but it's not really geared for scope tracing since it doesn't present its output in a way that makes capturing and analyzing flow through Asterisk easy. How is scope tracing better? Scope tracing uses the same "cleanup" attribute that RAII_VAR uses to print messages to a separate "trace" log level. Even better, the messages are indented and unindented based on a thread-local call depth counter. When output to a separate log file, the output is uncluttered and easy to follow. Here's an example of the output. The leading timestamps and thread ids are removed and the output cut off at 68 columns for commit message restrictions but you get the idea. --> res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 --> res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 --> res_pjsip_session.c:3669 handle_incoming_response PJSIP/ --> chan_pjsip.c:3265 chan_pjsip_incoming_response_after --> chan_pjsip.c:3194 chan_pjsip_incoming_response P chan_pjsip.c:3245 chan_pjsip_incoming_respon <-- chan_pjsip.c:3194 chan_pjsip_incoming_response P <-- chan_pjsip.c:3265 chan_pjsip_incoming_response_after <-- res_pjsip_session.c:3669 handle_incoming_response PJSIP/ <-- res_pjsip_session.c:3661 handle_incoming_response PJSIP/1173 <-- res_pjsip_session.c:3680 handle_incoming PJSIP/1173-00000001 The messages with the "-->" or "<--" were produced by including the following at the top of each function: SCOPE_TRACE(1, "%s\n", ast_sip_session_get_name(session)); Scope isn't limited to functions any more than RAII_VAR is. You can also see entry and exit from "if", "for", "while", etc blocks. There is also an ast_trace() macro that doesn't track entry or exit but simply outputs a message to the trace log using the current indent level. The deepest message in the sample (chan_pjsip.c:3245) was used to indicate which "case" in a "select" was executed. How do you use it? More documentation is available in logger.h but here's an overview: * Configure with --enable-dev-mode. Like debug, scope tracing is #ifdef'd out if devmode isn't enabled. * Add a SCOPE_TRACE() call to the top of your function. * Set a logger channel in logger.conf to output the "trace" level. * Use the CLI (or cli.conf) to set a trace level similar to setting debug level... CLI> core set trace 2 res_pjsip.so Summary Of Changes: * Added LOG_TRACE logger level. Actually it occupies the slot formerly occupied by the now defunct "event" level. * Added core asterisk option "trace" similar to debug. Includes ability to specify global trace level in asterisk.conf and CLI commands to turn on/off and set levels. Levels can be set globally (probably not a good idea), or by module/source file. * Updated sample asterisk.conf and logger.conf. Tracing is disabled by default in both. * Added __ast_trace() to logger.c which keeps track of the indent level using TLS. It's #ifdef'd out if devmode isn't enabled. * Added ast_trace() and SCOPE_TRACE() macros to logger.h. These are all #ifdef'd out if devmode isn't enabled. Why not use gcc's -finstrument-functions capability? gcc's facility doesn't allow access to local data and doesn't operate on non-function scopes. Known Issues: The only know issue is that we currently don't know the line number where the scope exited. It's reported as the same place the scope was entered. There's probably a way to get around it but it might involve looking at the stack and doing an 'addr2line' to get the line number. Kind of like ast_backtrace() does. Not sure if it's worth it. Change-Id: Ic5ebb859883f9c10a08c5630802de33500cad027
2020-05-14 18:24:19 +00:00
AST_CLI_DEFINE(handle_trace, "Set level of trace chattiness"),
verbosity: Fix performance of console verbose messages. The per console verbose level feature as previously implemented caused a large performance penalty. The fix required some minor incompatibilities if the new rasterisk is used to connect to an earlier version. If the new rasterisk connects to an older Asterisk version then the root console verbose level is always affected by the "core set verbose" command of the remote console even though it may appear to only affect the current console. If an older version of rasterisk connects to the new version then the "core set verbose" command will have no effect. * Fixed the verbose performance by not generating a verbose message if nothing is going to use it and then filtered any generated verbose messages before actually sending them to the remote consoles. * Split the "core set debug" and "core set verbose" CLI commands to remove the per module verbose support that cannot work with the per console verbose level. * Added a silent option to the "core set verbose" command. * Fixed "core set debug off" tab completion. * Made "core show settings" list the current console verbosity in addition to the root console verbosity. * Changed the default verbose level of the 'verbose' setting in the logger.conf [logfiles] section. The default is now to once again follow the current root console level. As a result, using the AMI Command action with "core set verbose" could again set the root console verbose level and affect the verbose level logged. (closes issue AST-1252) Reported by: Guenther Kelleter Review: https://reviewboard.asterisk.org/r/3114/ ........ Merged revisions 405431 from http://svn.asterisk.org/svn/asterisk/branches/11 ........ Merged revisions 405432 from http://svn.asterisk.org/svn/asterisk/branches/12 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@405436 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2014-01-14 18:14:02 +00:00
AST_CLI_DEFINE(handle_verbose, "Set level of verbose chattiness"),
AST_CLI_DEFINE(handle_help, "Display help list, or specific help on a command"),
AST_CLI_DEFINE(handle_logger_mute, "Toggle logging output to a console"),
AST_CLI_DEFINE(handle_modlist, "List modules and info"),
AST_CLI_DEFINE(handle_load, "Load a module by name"),
AST_CLI_DEFINE(handle_reload, "Reload configuration for a module"),
AST_CLI_DEFINE(handle_core_reload, "Global reload"),
AST_CLI_DEFINE(handle_unload, "Unload a module by name"),
AST_CLI_DEFINE(handle_refresh, "Completely unloads and loads a module by name"),
AST_CLI_DEFINE(handle_showuptime, "Show uptime information"),
AST_CLI_DEFINE(handle_cli_reload_permissions, "Reload CLI permissions config"),
AST_CLI_DEFINE(handle_cli_show_permissions, "Show CLI permissions"),
AST_CLI_DEFINE(handle_cli_check_permissions, "Try a permissions config for a user"),
AST_CLI_DEFINE(handle_cli_wait_fullybooted, "Wait for Asterisk to be fully booted"),
#ifdef HAVE_MALLOC_TRIM
AST_CLI_DEFINE(handle_cli_malloc_trim, "Return excess memory to the OS"),
#endif
};
static struct ast_cli_entry cli_channels_cli[] = {
AST_CLI_DEFINE(handle_nodebugchan_deprecated, "Disable debugging on channel(s)"),
AST_CLI_DEFINE(handle_chanlist, "Display information on channels"),
AST_CLI_DEFINE(handle_showcalls, "Display information on calls"),
AST_CLI_DEFINE(handle_showchan, "Display information on a specific channel"),
AST_CLI_DEFINE(handle_core_set_debug_channel, "Enable/disable debugging on a channel"),
AST_CLI_DEFINE(group_show_channels, "Display active channels with group(s)"),
AST_CLI_DEFINE(handle_softhangup, "Request a hangup on a given channel"),
};
/*!
* Some regexp characters in cli arguments are reserved and used as separators.
*/
static const char cli_rsvd[] = "[]{}|*%";
/*!
* initialize the _full_cmd string and related parameters,
* return 0 on success, -1 on error.
*/
static int set_full_cmd(struct ast_cli_entry *e)
{
int i;
char buf[80];
ast_join(buf, sizeof(buf), e->cmda);
e->_full_cmd = ast_strdup(buf);
if (!e->_full_cmd) {
ast_log(LOG_WARNING, "-- cannot allocate <%s>\n", buf);
return -1;
}
e->cmdlen = strcspn(e->_full_cmd, cli_rsvd);
for (i = 0; e->cmda[i]; i++)
;
e->args = i;
return 0;
}
/*! \brief cleanup (free) cli_perms linkedlist. */
static void destroy_user_perms(void)
{
struct cli_perm *perm;
struct usergroup_cli_perm *user_perm;
AST_RWLIST_WRLOCK(&cli_perms);
while ((user_perm = AST_LIST_REMOVE_HEAD(&cli_perms, list))) {
while ((perm = AST_LIST_REMOVE_HEAD(user_perm->perms, list))) {
ast_free(perm->command);
ast_free(perm);
}
ast_free(user_perm);
}
AST_RWLIST_UNLOCK(&cli_perms);
}
int ast_cli_perms_init(int reload)
{
struct ast_flags config_flags = { reload ? CONFIG_FLAG_FILEUNCHANGED : 0 };
struct ast_config *cfg;
char *cat = NULL;
struct ast_variable *v;
struct usergroup_cli_perm *user_group, *cp_entry;
struct cli_perm *perm = NULL;
struct passwd *pw;
struct group *gr;
if (ast_mutex_trylock(&permsconfiglock)) {
ast_log(LOG_NOTICE, "You must wait until last 'cli reload permissions' command finish\n");
return 1;
}
cfg = ast_config_load2(perms_config, "" /* core, can't reload */, config_flags);
if (!cfg) {
ast_mutex_unlock(&permsconfiglock);
return 1;
} else if (cfg == CONFIG_STATUS_FILEUNCHANGED) {
ast_mutex_unlock(&permsconfiglock);
return 0;
}
/* free current structures. */
destroy_user_perms();
while ((cat = ast_category_browse(cfg, cat))) {
if (!strcasecmp(cat, "general")) {
/* General options */
for (v = ast_variable_browse(cfg, cat); v; v = v->next) {
if (!strcasecmp(v->name, "default_perm")) {
cli_default_perm = (!strcasecmp(v->value, "permit")) ? 1: 0;
}
}
continue;
}
/* users or groups */
gr = NULL, pw = NULL;
if (cat[0] == '@') {
/* This is a group */
gr = getgrnam(&cat[1]);
if (!gr) {
ast_log (LOG_WARNING, "Unknown group '%s'\n", &cat[1]);
continue;
}
} else {
/* This is a user */
pw = getpwnam(cat);
if (!pw) {
ast_log (LOG_WARNING, "Unknown user '%s'\n", cat);
continue;
}
}
user_group = NULL;
/* Check for duplicates */
AST_RWLIST_WRLOCK(&cli_perms);
AST_LIST_TRAVERSE(&cli_perms, cp_entry, list) {
if ((pw && cp_entry->uid == pw->pw_uid) || (gr && cp_entry->gid == gr->gr_gid)) {
/* if it is duplicated, just added this new settings, to
the current list. */
user_group = cp_entry;
break;
}
}
AST_RWLIST_UNLOCK(&cli_perms);
if (!user_group) {
/* alloc space for the new user config. */
user_group = ast_calloc(1, sizeof(*user_group));
if (!user_group) {
continue;
}
user_group->uid = (pw ? pw->pw_uid : -1);
user_group->gid = (gr ? gr->gr_gid : -1);
user_group->perms = ast_calloc(1, sizeof(*user_group->perms));
if (!user_group->perms) {
ast_free(user_group);
continue;
}
}
for (v = ast_variable_browse(cfg, cat); v; v = v->next) {
if (ast_strlen_zero(v->value)) {
/* we need to check this condition cause it could break security. */
ast_log(LOG_WARNING, "Empty permit/deny option in user '%s'\n", cat);
continue;
}
if (!strcasecmp(v->name, "permit")) {
perm = ast_calloc(1, sizeof(*perm));
if (perm) {
perm->permit = 1;
perm->command = ast_strdup(v->value);
}
} else if (!strcasecmp(v->name, "deny")) {
perm = ast_calloc(1, sizeof(*perm));
if (perm) {
perm->permit = 0;
perm->command = ast_strdup(v->value);
}
} else {
/* up to now, only 'permit' and 'deny' are possible values. */
ast_log(LOG_WARNING, "Unknown '%s' option\n", v->name);
continue;
}
if (perm) {
/* Added the permission to the user's list. */
AST_LIST_INSERT_TAIL(user_group->perms, perm, list);
perm = NULL;
}
}
AST_RWLIST_WRLOCK(&cli_perms);
AST_RWLIST_INSERT_TAIL(&cli_perms, user_group, list);
AST_RWLIST_UNLOCK(&cli_perms);
}
ast_config_destroy(cfg);
ast_mutex_unlock(&permsconfiglock);
return 0;
}
static void cli_shutdown(void)
{
ast_cli_unregister_multiple(cli_cli, ARRAY_LEN(cli_cli));
}
static void cli_channels_shutdown(void)
{
ast_cli_unregister_multiple(cli_channels_cli, ARRAY_LEN(cli_channels_cli));
}
/*! \brief initialize the _full_cmd string in * each of the builtins. */
void ast_builtins_init(void)
{
AST_VECTOR_INIT(&shutdown_commands, 0);
ast_cli_register_multiple(cli_cli, ARRAY_LEN(cli_cli));
ast_register_cleanup(cli_shutdown);
}
void ast_cli_channels_init(void)
{
ast_cli_register_multiple(cli_channels_cli, ARRAY_LEN(cli_channels_cli));
ast_register_cleanup(cli_channels_shutdown);
}
/*!
* match a word in the CLI entry.
* returns -1 on mismatch, 0 on match of an optional word,
* 1 on match of a full word.
*
* The pattern can be
* any_word match for equal
* [foo|bar|baz] optionally, one of these words
* {foo|bar|baz} exactly, one of these words
* % any word
*/
static int word_match(const char *cmd, const char *cli_word)
{
int l;
char *pos;
if (ast_strlen_zero(cmd) || ast_strlen_zero(cli_word))
return -1;
if (!strchr(cli_rsvd, cli_word[0])) /* normal match */
return (strcasecmp(cmd, cli_word) == 0) ? 1 : -1;
l = strlen(cmd);
/* wildcard match - will extend in the future */
if (l > 0 && cli_word[0] == '%') {
return 1; /* wildcard */
}
/* Start a search for the command entered against the cli word in question */
pos = strcasestr(cli_word, cmd);
while (pos) {
/*
*Check if the word matched with is surrounded by reserved characters on both sides
* and isn't at the beginning of the cli_word since that would make it check in a location we shouldn't know about.
* If it is surrounded by reserved chars and isn't at the beginning, it's a match.
*/
if (pos != cli_word && strchr(cli_rsvd, pos[-1]) && strchr(cli_rsvd, pos[l])) {
return 1; /* valid match */
}
/* Ok, that one didn't match, strcasestr to the next appearance of the command and start over.*/
pos = strcasestr(pos + 1, cmd);
}
/* If no matches were found over the course of the while loop, we hit the end of the string. It's a mismatch. */
return -1;
}
/*! \brief if word is a valid prefix for token, returns the pos-th
* match as a malloced string, or NULL otherwise.
* Always tell in *actual how many matches we got.
*/
static char *is_prefix(const char *word, const char *token,
int pos, int *actual)
{
int lw;
char *s, *t1;
*actual = 0;
if (ast_strlen_zero(token))
return NULL;
if (ast_strlen_zero(word))
word = ""; /* dummy */
lw = strlen(word);
if (strcspn(word, cli_rsvd) != lw)
return NULL; /* no match if word has reserved chars */
if (strchr(cli_rsvd, token[0]) == NULL) { /* regular match */
if (strncasecmp(token, word, lw)) /* no match */
return NULL;
*actual = 1;
return (pos != 0) ? NULL : ast_strdup(token);
}
/* now handle regexp match */
/* Wildcard always matches, so we never do is_prefix on them */
t1 = ast_strdupa(token + 1); /* copy, skipping first char */
while (pos >= 0 && (s = strsep(&t1, cli_rsvd)) && *s) {
if (*s == '%') /* wildcard */
continue;
if (strncasecmp(s, word, lw)) /* no match */
continue;
(*actual)++;
if (pos-- == 0)
return ast_strdup(s);
}
return NULL;
}
/*!
* \internal
* \brief locate a cli command in the 'helpers' list (which must be locked).
* The search compares word by word taking care of regexps in e->cmda
* This function will return NULL when nothing is matched, or the ast_cli_entry that matched.
* \param cmds
* \param match_type has 3 possible values:
* 0 returns if the search key is equal or longer than the entry.
* note that trailing optional arguments are skipped.
* -1 true if the mismatch is on the last word XXX not true!
* 1 true only on complete, exact match.
*
*/
static struct ast_cli_entry *find_cli(const char * const cmds[], int match_type)
{
int matchlen = -1; /* length of longest match so far */
struct ast_cli_entry *cand = NULL, *e=NULL;
while ( (e = cli_next(e)) ) {
/* word-by word regexp comparison */
const char * const *src = cmds;
const char * const *dst = e->cmda;
int n = 0;
for (;; dst++, src += n) {
n = word_match(*src, *dst);
if (n < 0)
break;
}
if (ast_strlen_zero(*dst) || ((*dst)[0] == '[' && ast_strlen_zero(dst[1]))) {
/* no more words in 'e' */
if (ast_strlen_zero(*src)) /* exact match, cannot do better */
break;
/* Here, cmds has more words than the entry 'e' */
if (match_type != 0) /* but we look for almost exact match... */
continue; /* so we skip this one. */
/* otherwise we like it (case 0) */
} else { /* still words in 'e' */
if (ast_strlen_zero(*src))
continue; /* cmds is shorter than 'e', not good */
/* Here we have leftover words in cmds and 'e',
* but there is a mismatch. We only accept this one if match_type == -1
* and this is the last word for both.
*/
if (match_type != -1 || !ast_strlen_zero(src[1]) ||
!ast_strlen_zero(dst[1])) /* not the one we look for */
continue;
/* good, we are in case match_type == -1 and mismatch on last word */
}
if (src - cmds > matchlen) { /* remember the candidate */
matchlen = src - cmds;
cand = e;
}
}
return e ? e : cand;
}
static char *find_best(const char *argv[])
{
static char cmdline[80];
int x;
/* See how close we get, then print the candidate */
const char *myargv[AST_MAX_CMD_LEN] = { NULL, };
AST_RWLIST_RDLOCK(&helpers);
for (x = 0; argv[x]; x++) {
myargv[x] = argv[x];
if (!find_cli(myargv, -1))
break;
}
AST_RWLIST_UNLOCK(&helpers);
ast_join(cmdline, sizeof(cmdline), myargv);
return cmdline;
}
static int cli_is_registered(struct ast_cli_entry *e)
{
struct ast_cli_entry *cur = NULL;
while ((cur = cli_next(cur))) {
if (cur == e) {
return 1;
}
}
return 0;
}
static void remove_shutdown_command(struct ast_cli_entry *e)
{
ast_rwlock_wrlock(&shutdown_commands_lock);
AST_VECTOR_REMOVE_ELEM_UNORDERED(&shutdown_commands, e, AST_VECTOR_ELEM_CLEANUP_NOOP);
ast_rwlock_unlock(&shutdown_commands_lock);
}
int ast_cli_unregister(struct ast_cli_entry *e)
{
if (e->inuse) {
ast_log(LOG_WARNING, "Can't remove command that is in use\n");
} else {
AST_RWLIST_WRLOCK(&helpers);
AST_RWLIST_REMOVE(&helpers, e, list);
AST_RWLIST_UNLOCK(&helpers);
remove_shutdown_command(e);
ast_free(e->_full_cmd);
e->_full_cmd = NULL;
if (e->handler) {
/* this is a new-style entry. Reset fields and free memory. */
char *cmda = (char *) e->cmda;
memset(cmda, '\0', sizeof(e->cmda));
ast_free(e->command);
e->command = NULL;
e->usage = NULL;
}
}
return 0;
}
int __ast_cli_register(struct ast_cli_entry *e, struct ast_module *module)
{
struct ast_cli_entry *cur;
int i, lf, ret = -1;
struct ast_cli_args a; /* fake argument */
char **dst = (char **)e->cmda; /* need to cast as the entry is readonly */
char *s;
AST_RWLIST_WRLOCK(&helpers);
if (cli_is_registered(e)) {
ast_log(LOG_WARNING, "Command '%s' already registered (the same ast_cli_entry)\n",
S_OR(e->_full_cmd, e->command));
ret = 0; /* report success */
goto done;
}
memset(&a, '\0', sizeof(a));
e->module = module;
/* No module reference needed here, the module called us. */
e->handler(e, CLI_INIT, &a);
/* XXX check that usage and command are filled up */
s = ast_skip_blanks(e->command);
s = e->command = ast_strdup(s);
for (i=0; !ast_strlen_zero(s) && i < AST_MAX_CMD_LEN-1; i++) {
*dst++ = s; /* store string */
s = ast_skip_nonblanks(s);
if (*s == '\0') /* we are done */
break;
*s++ = '\0';
s = ast_skip_blanks(s);
}
*dst++ = NULL;
if (find_cli(e->cmda, 1)) {
ast_log(LOG_WARNING, "Command '%s' already registered (or something close enough)\n",
S_OR(e->_full_cmd, e->command));
goto done;
}
if (set_full_cmd(e)) {
ast_log(LOG_WARNING, "Error registering CLI Command '%s'\n",
S_OR(e->_full_cmd, e->command));
goto done;
}
lf = e->cmdlen;
AST_RWLIST_TRAVERSE_SAFE_BEGIN(&helpers, cur, list) {
int len = cur->cmdlen;
if (lf < len)
len = lf;
if (strncasecmp(e->_full_cmd, cur->_full_cmd, len) < 0) {
AST_RWLIST_INSERT_BEFORE_CURRENT(e, list);
break;
}
}
AST_RWLIST_TRAVERSE_SAFE_END;
if (!cur)
AST_RWLIST_INSERT_TAIL(&helpers, e, list);
ret = 0; /* success */
done:
AST_RWLIST_UNLOCK(&helpers);
if (ret) {
ast_free(e->command);
e->command = NULL;
}
return ret;
}
/*
* register/unregister an array of entries.
*/
int __ast_cli_register_multiple(struct ast_cli_entry *e, int len, struct ast_module *module)
{
int i, res = 0;
for (i = 0; i < len; i++) {
res |= __ast_cli_register(e + i, module);
}
return res;
}
int ast_cli_unregister_multiple(struct ast_cli_entry *e, int len)
{
int i, res = 0;
for (i = 0; i < len; i++)
res |= ast_cli_unregister(e + i);
return res;
}
/*! \brief helper for final part of handle_help
* if locked = 1, assume the list is already locked
*/
static char *help1(int fd, const char * const match[], int locked)
{
char matchstr[80] = "";
struct ast_cli_entry *e = NULL;
int len = 0;
int found = 0;
if (match) {
ast_join(matchstr, sizeof(matchstr), match);
len = strlen(matchstr);
}
if (!locked)
AST_RWLIST_RDLOCK(&helpers);
while ( (e = cli_next(e)) ) {
/* Hide commands that start with '_' */
if (e->_full_cmd[0] == '_')
continue;
if (match && strncasecmp(matchstr, e->_full_cmd, len))
continue;
ast_cli(fd, "%-30s -- %s\n", e->_full_cmd,
S_OR(e->summary, "<no description available>"));
found++;
}
if (!locked)
AST_RWLIST_UNLOCK(&helpers);
if (!found && matchstr[0])
ast_cli(fd, "No such command '%s'.\n", matchstr);
return CLI_SUCCESS;
}
static char *handle_help(struct ast_cli_entry *e, int cmd, struct ast_cli_args *a)
{
char fullcmd[80];
struct ast_cli_entry *my_e;
char *res = CLI_SUCCESS;
if (cmd == CLI_INIT) {
e->command = "core show help";
e->usage =
"Usage: core show help [topic]\n"
" When called with a topic as an argument, displays usage\n"
" information on the given command. If called without a\n"
" topic, it provides a list of commands.\n";
return NULL;
} else if (cmd == CLI_GENERATE) {
/* skip first 14 or 15 chars, "core show help " */
int l = strlen(a->line);
if (l > 15) {
l = 15;
}
/* XXX watch out, should stop to the non-generator parts */
return __ast_cli_generator(a->line + l, a->word, a->n, 0);
}
if (a->argc == e->args) {
return help1(a->fd, NULL, 0);
}
AST_RWLIST_RDLOCK(&helpers);
my_e = find_cli(a->argv + 3, 1); /* try exact match first */
if (!my_e) {
res = help1(a->fd, a->argv + 3, 1 /* locked */);
AST_RWLIST_UNLOCK(&helpers);
return res;
}
if (my_e->usage)
ast_cli(a->fd, "%s", my_e->usage);
else {
ast_join(fullcmd, sizeof(fullcmd), a->argv + 3);
ast_cli(a->fd, "No help text available for '%s'.\n", fullcmd);
}
AST_RWLIST_UNLOCK(&helpers);
return res;
}
static char *parse_args(const char *s, int *argc, const char *argv[], int max, int *trailingwhitespace)
{
char *duplicate, *cur;
int x = 0;
int quoted = 0;
int escaped = 0;
int whitespace = 1;
int dummy = 0;
if (trailingwhitespace == NULL)
trailingwhitespace = &dummy;
*trailingwhitespace = 0;
if (s == NULL) /* invalid, though! */
return NULL;
/* make a copy to store the parsed string */
if (!(duplicate = ast_strdup(s)))
return NULL;
cur = duplicate;
/* Remove leading spaces from the command */
while (isspace(*s)) {
cur++;
s++;
}
/* scan the original string copying into cur when needed */
for (; *s ; s++) {
if (x >= max - 1) {
ast_log(LOG_WARNING, "Too many arguments, truncating at %s\n", s);
break;
}
if (*s == '"' && !escaped) {
quoted = !quoted;
if (quoted && whitespace) {
/* start a quoted string from previous whitespace: new argument */
argv[x++] = cur;
whitespace = 0;
}
} else if ((*s == ' ' || *s == '\t') && !(quoted || escaped)) {
/* If we are not already in whitespace, and not in a quoted string or
processing an escape sequence, and just entered whitespace, then
finalize the previous argument and remember that we are in whitespace
*/
if (!whitespace) {
*cur++ = '\0';
whitespace = 1;
}
} else if (*s == '\\' && !escaped) {
escaped = 1;
} else {
if (whitespace) {
/* we leave whitespace, and are not quoted. So it's a new argument */
argv[x++] = cur;
whitespace = 0;
}
*cur++ = *s;
escaped = 0;
}
}
/* Null terminate */
*cur++ = '\0';
/* XXX put a NULL in the last argument, because some functions that take
* the array may want a null-terminated array.
* argc still reflects the number of non-NULL entries.
*/
argv[x] = NULL;
*argc = x;
*trailingwhitespace = whitespace;
return duplicate;
}
char **ast_cli_completion_matches(const char *text, const char *word)
{
struct ast_vector_string *vec = ast_cli_completion_vector(text, word);
char **match_list;
if (!vec) {
return NULL;
}
if (AST_VECTOR_APPEND(vec, NULL)) {
/* We failed to NULL terminate the elements */
AST_VECTOR_CALLBACK_VOID(vec, ast_free);
AST_VECTOR_PTR_FREE(vec);
return NULL;
}
match_list = AST_VECTOR_STEAL_ELEMENTS(vec);
AST_VECTOR_PTR_FREE(vec);
return match_list;
}
AST_THREADSTORAGE_RAW(completion_storage);
/*!
* \internal
* \brief Add a value to the vector.
*
* \param vec Vector to add \a value to. Must be from threadstorage.
* \param value The value to add.
*
* \retval 0 Success
* \retval -1 Failure
*/
static int cli_completion_vector_add(struct ast_vector_string *vec, char *value)
{
if (!value) {
return 0;
}
if (!vec || AST_VECTOR_ADD_SORTED(vec, value, strcasecmp)) {
if (vec) {
ast_threadstorage_set_ptr(&completion_storage, NULL);
AST_VECTOR_CALLBACK_VOID(vec, ast_free);
AST_VECTOR_FREE(vec);
}
ast_free(value);
return -1;
}
return 0;
}
int ast_cli_completion_add(char *value)
{
return cli_completion_vector_add(ast_threadstorage_get_ptr(&completion_storage), value);
}
struct ast_vector_string *ast_cli_completion_vector(const char *text, const char *word)
{
char *retstr, *prevstr;
size_t max_equal;
size_t which = 0;
struct ast_vector_string *vec = ast_calloc(1, sizeof(*vec));
/* Recursion into this function is a coding error. */
ast_assert(!ast_threadstorage_get_ptr(&completion_storage));
if (!vec) {
return NULL;
}
if (ast_threadstorage_set_ptr(&completion_storage, vec)) {
ast_log(LOG_ERROR, "Failed to initialize threadstorage for completion.\n");
ast_free(vec);
return NULL;
}
while ((retstr = ast_cli_generator(text, word, which)) != NULL) {
if (cli_completion_vector_add(vec, retstr)) {
ast_threadstorage_set_ptr(&completion_storage, NULL);
goto vector_cleanup;
}
++which;
}
ast_threadstorage_set_ptr(&completion_storage, NULL);
if (!AST_VECTOR_SIZE(vec)) {
AST_VECTOR_PTR_FREE(vec);
return NULL;
}
prevstr = AST_VECTOR_GET(vec, 0);
max_equal = strlen(prevstr);
which = 1;
/* Find the longest substring that is common to all results
* (it is a candidate for completion), and store a copy in entry 0.
*/
while (which < AST_VECTOR_SIZE(vec)) {
size_t i = 0;
retstr = AST_VECTOR_GET(vec, which);
/* Check for and remove duplicate strings. */
if (!strcasecmp(prevstr, retstr)) {
AST_VECTOR_REMOVE(vec, which, 1);
ast_free(retstr);
continue;
}
while (i < max_equal && toupper(prevstr[i]) == toupper(retstr[i])) {
i++;
}
max_equal = i;
prevstr = retstr;
++which;
}
/* Insert longest match to position 0. */
retstr = ast_strndup(AST_VECTOR_GET(vec, 0), max_equal);
if (!retstr || AST_VECTOR_INSERT_AT(vec, 0, retstr)) {
ast_free(retstr);
goto vector_cleanup;
Fix a variety of memory leaks This patch addresses a number of memory leaks in a variety of modules that were found by a static analysis tool. A brief summary of the changes: * app_minivm: free ast_str objects on off nominal paths * app_page: free the ast_dial object if the requested channel technology cannot be appended to the dialing structure * app_queue: if a penalty rule failed to match any existing rule list names, the created rule would not be inserted and its memory would be leaked * app_read: dispose of the created silence detector in the presence of off nominal circumstances * app_voicemail: dispose of an allocated unique ID field for MWI event un-subscribe requests in off nominal paths; dispose of configuration objects when using the secret.conf option * chan_dahdi: dispose of the allocated frame produced by ast_dsp_process * chan_iax2: properly unref peer in CLI command "iax2 unregister" * chan_sip: dispose of the allocated frame produced by sip_rtp_read's call of ast_dsp_process; free memory in parse unit tests * func_dialgroup: properly deref ao2 object grhead in nominal path of dialgroup_read * func_odbc: free resultset in off nominal paths of odbc_read * cli: free match_list in off nominal paths of CLI match completion * config: free comment_buffer/list_buffer when configuration file load is unchanged; free the same buffers any time they were created and config files were processed * data: free XML nodes in various places * enum: free context buffer in off nominal paths * features: free ast_call_feature in off nominal paths of applicationmap config processing * netsock2: users of ast_sockaddr_resolve pass in an ast_sockaddr struct that is allocated by the method. Failures in ast_sockaddr_resolve could result in the users of the method not knowing whether or not the buffer was allocated. The method will now not allocate the ast_sockaddr struct if it will return failure. * pbx: cleanup hash table traversals in off nominal paths; free ignore pattern buffer if it already exists for the specified context * xmldoc: cleanup various nodes when we no longer need them * main/editline: various cleanup of pointers not being freed before being assigned to other memory, cleanup along off nominal paths * menuselect/mxml: cleanup of value buffer for an attribute when that attribute did not specify a value * res_calendar*: responses are allocated via the various *_request method returns and should not be allocated in the various write_event methods; ensure attendee buffer is freed if no data exists in the parsed node; ensure that calendar objects are de-ref'd appropriately * res_jabber: free buffer in off nominal path * res_musiconhold: close the DIR* object in off nominal paths * res_rtp_asterisk: if we run out of ports, close the rtp socket object and free the rtp object * res_srtp: if we fail to create the session in libsrtp, destroy the temporary ast_srtp object (issue ASTERISK-19665) Reported by: Matt Jordan Review: https://reviewboard.asterisk.org/r/1922 ........ Merged revisions 366880 from http://svn.asterisk.org/svn/asterisk/branches/1.8 ........ Merged revisions 366881 from http://svn.asterisk.org/svn/asterisk/branches/10 git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@366917 65c4cc65-6c06-0410-ace0-fbb531ad65f3
2012-05-18 14:43:44 +00:00
}
return vec;
vector_cleanup:
AST_VECTOR_CALLBACK_VOID(vec, ast_free);
AST_VECTOR_PTR_FREE(vec);
return NULL;
}
/*! \brief returns true if there are more words to match */
static int more_words (const char * const *dst)
{
int i;
for (i = 0; dst[i]; i++) {
if (dst[i][0] != '[')
return -1;
}
return 0;
}
/*
* generate the entry at position 'state'
*/
static char *__ast_cli_generator(const char *text, const char *word, int state, int lock)
{
const char *argv[AST_MAX_ARGS];
struct ast_cli_entry *e = NULL;
int x = 0, argindex, matchlen;
int matchnum=0;
char *ret = NULL;
char matchstr[80] = "";
int tws = 0;
/* Split the argument into an array of words */
char *duplicate = parse_args(text, &x, argv, ARRAY_LEN(argv), &tws);
if (!duplicate) /* malloc error */
return NULL;
/* Compute the index of the last argument (could be an empty string) */
argindex = (!ast_strlen_zero(word) && x>0) ? x-1 : x;
/* rebuild the command, ignore terminating white space and flatten space */
ast_join(matchstr, sizeof(matchstr)-1, argv);
matchlen = strlen(matchstr);
if (tws) {
strcat(matchstr, " "); /* XXX */
if (matchlen)
matchlen++;
}
if (lock)
AST_RWLIST_RDLOCK(&helpers);
while ( (e = cli_next(e)) ) {
/* XXX repeated code */
int src = 0, dst = 0, n = 0;
if (e->command[0] == '_')
continue;
/*
* Try to match words, up to and excluding the last word, which
* is either a blank or something that we want to extend.
*/
for (;src < argindex; dst++, src += n) {
n = word_match(argv[src], e->cmda[dst]);
if (n < 0)
break;
}
if (src != argindex && more_words(e->cmda + dst)) /* not a match */
continue;
ret = is_prefix(argv[src], e->cmda[dst], state - matchnum, &n);
matchnum += n; /* this many matches here */
if (ret) {
/*
* argv[src] is a valid prefix of the next word in this
* command. If this is also the correct entry, return it.
*/
if (matchnum > state)
break;
ast_free(ret);
ret = NULL;
} else if (ast_strlen_zero(e->cmda[dst])) {
/*
* This entry is a prefix of the command string entered
* (only one entry in the list should have this property).
* Run the generator if one is available. In any case we are done.
*/
if (e->handler) { /* new style command */
struct ast_cli_args a = {
.line = matchstr, .word = word,
.pos = argindex,
.n = state - matchnum,
.argv = argv,
.argc = x};
/* If the command is in a module it must be running. */
if (!e->module || ast_module_running_ref(e->module)) {
ret = e->handler(e, CLI_GENERATE, &a);
ast_module_unref(e->module);
}
}
if (ret)
break;
}
}
if (lock)
AST_RWLIST_UNLOCK(&helpers);
ast_free(duplicate);
return ret;
}
char *ast_cli_generator(const char *text, const char *word, int state)
{
return __ast_cli_generator(text, word, state, 1);
}
static int allowed_on_shutdown(struct ast_cli_entry *e)
{
int found = 0;
int i;
ast_rwlock_rdlock(&shutdown_commands_lock);
for (i = 0; i < AST_VECTOR_SIZE(&shutdown_commands); ++i) {
if (e == AST_VECTOR_GET(&shutdown_commands, i)) {
found = 1;
break;
}
}
ast_rwlock_unlock(&shutdown_commands_lock);
return found;
}
int ast_cli_command_full(int uid, int gid, int fd, const char *s)
{
const char *args[AST_MAX_ARGS + 1];
struct ast_cli_entry *e = NULL;
int x;
char *duplicate = parse_args(s, &x, args + 1, AST_MAX_ARGS, NULL);
char tmp[AST_MAX_ARGS + 1];
char *retval = CLI_FAILURE;
struct ast_cli_args a = {
.fd = fd, .argc = x, .argv = args+1 };
if (duplicate == NULL)
return RESULT_FAILURE;
if (x < 1) /* We need at least one entry, otherwise ignore */
goto done;
AST_RWLIST_RDLOCK(&helpers);
e = find_cli(args + 1, 0);
if (e)
ast_atomic_fetchadd_int(&e->inuse, 1);
AST_RWLIST_UNLOCK(&helpers);
if (e == NULL) {
ast_cli(fd, "No such command '%s' (type 'core show help %s' for other possible commands)\n", s, find_best(args + 1));
goto done;
}
if (ast_shutting_down() && !allowed_on_shutdown(e)) {
ast_cli(fd, "Command '%s' cannot be run during shutdown\n", s);
goto done;
}
ast_join(tmp, sizeof(tmp), args + 1);
/* Check if the user has rights to run this command. */
if (!cli_has_permissions(uid, gid, tmp)) {
ast_cli(fd, "You don't have permissions to run '%s' command\n", tmp);
goto done;
}
/*
* Within the handler, argv[-1] contains a pointer to the ast_cli_entry.
* Remember that the array returned by parse_args is NULL-terminated.
*/
args[0] = (char *)e;
/* If the command is in a module it must be running. */
if (!e->module || ast_module_running_ref(e->module)) {
retval = e->handler(e, CLI_HANDLER, &a);
ast_module_unref(e->module);
}
if (retval == CLI_SHOWUSAGE) {
ast_cli(fd, "%s", S_OR(e->usage, "Invalid usage, but no usage information available.\n"));
} else if (retval == CLI_FAILURE) {
ast_cli(fd, "Command '%s' failed.\n", s);
}
done:
if (e) {
ast_atomic_fetchadd_int(&e->inuse, -1);
}
ast_free(duplicate);
return retval == CLI_SUCCESS ? RESULT_SUCCESS : RESULT_FAILURE;
}
int ast_cli_command_multiple_full(int uid, int gid, int fd, size_t size, const char *s)
{
char cmd[512];
int x, y = 0, count = 0;
for (x = 0; x < size; x++) {
cmd[y] = s[x];
y++;
if (s[x] == '\0') {
ast_cli_command_full(uid, gid, fd, cmd);
y = 0;
count++;
}
}
return count;
}
void ast_cli_print_timestr_fromseconds(int fd, int seconds, const char *prefix)
{
print_uptimestr(fd, ast_tv(seconds, 0), prefix, 0);
}
int ast_cli_allow_at_shutdown(struct ast_cli_entry *e)
{
int res;
ast_rwlock_wrlock(&shutdown_commands_lock);
res = AST_VECTOR_APPEND(&shutdown_commands, e);
ast_rwlock_unlock(&shutdown_commands_lock);
return res;
}