Module: sip-router
Branch: tmp/dlgnewref
Commit: 9ba4ee25e32368a9c17c3f86784a52a97b748b93
URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=9ba4ee2…
Author: Daniel-Constantin Mierla <miconda(a)gmail.com>
Committer: Daniel-Constantin Mierla <miconda(a)gmail.com>
Date: Mon Nov 21 23:52:43 2011 +0100
dialog(k): refactored dlg ref counting
- dlg pointer is no longer stored in tm callbacks parameter and ref
counting for these kind of needs was removed
- the internal id (hash entry, dlg id) is given now to tm callbacks. The
callbacks will do dlg lookup to get the structure
- there are longer-time persistent ref counting for:
* storing dlg structure in dialog hash table
* keeping dlg in timer list
* waiting for the reply of BYE to destroy dialog structure
- all the other ref countings are for short time, only for the period of
doing specific processing
- the new ref counting is safer in relation with tm callbacks and avoids
tight dependecy which may have been easily broken by future tm changes
---
modules_k/dialog/dialog.c | 102 +++++++----
modules_k/dialog/dlg_db_handler.c | 8 +-
modules_k/dialog/dlg_handlers.c | 387 +++++++++++++++++++------------------
modules_k/dialog/dlg_handlers.h | 23 ++-
modules_k/dialog/dlg_hash.c | 60 +++++--
modules_k/dialog/dlg_hash.h | 73 +++++---
modules_k/dialog/dlg_profile.c | 145 ++++++---------
modules_k/dialog/dlg_profile.h | 44 ++---
modules_k/dialog/dlg_req_within.c | 12 +-
modules_k/dialog/dlg_timer.h | 8 +-
modules_k/dialog/dlg_var.c | 338 ++++++++++++++++++++++----------
modules_k/dialog/dlg_var.h | 34 ++--
12 files changed, 716 insertions(+), 518 deletions(-)
Diff: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commitdiff;h=9ba…
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#179 - segfault in tm/timers?
User who did this - Bayan Towfiq (btowfiq)
----------
Hi Ovi,
I'm not sure--what would you like me to check in gdb. Please give me any commands you'd like to run on the backtrace.
Bayan
----------
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=179#comment377
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#179 - segfault in tm/timers?
User who did this - Ovidiu Sas (osas)
----------
That's "s-3level".
Maybe something is miss-aligned or the pointer is not properly set to point to the right field inside a dialog structure.
It would be interesting to see what's before and after that memory block.
----------
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=179#comment373
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#173 - Double Free -- Crash/Coredump and possible security vulnerability
User who did this - Daniel-Constantin Mierla (miconda)
----------
Try with attached patch -- an workaround solution for now discussed with Timo, but I guess he had no time to check and commit it yet.
----------
One or more files have been attached.
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=173#comment376
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#173 - Double Free -- Crash/Coredump and possible security vulnerability
User who did this - Brandon Armstead (CRYY2010)
----------
Timo,
This bug is turning into a nightmare! :(. It is totally weird since it seems to be present in 3.1 and 3.2 --- we've been running the 3.1 build since you worked out the last dialog issue. However it appears to be present in all recent builds (origin/3.1, origin/3.2, and several dialog commits downwards from HEAD of 3.1 & 3.2).
I really really really appreciate any time and effort you can put into resolving this matter.
Thanks for your help as always!
----------
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=173#comment375
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#173 - Double Free -- Crash/Coredump and possible security vulnerability
User who did this - Timo Reimann (tr)
----------
I know of quite a few people who use dlg_manage() regularly and repeatedly for subsequent requests, so I wouldn't consider it to be the #1 suspect. I'll take a closer look at it though, just to be sure.
Thanks!
----------
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=173#comment374
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#179 - segfault in tm/timers?
User who did this - Bayan Towfiq (btowfiq)
----------
(gdb) p *cbp
Cannot access memory at address 0x732d336c6576656c
:-\
----------
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=179#comment372
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#179 - segfault in tm/timers?
User who did this - Daniel-Constantin Mierla (miconda)
----------
It does not look related to timer at all, but to callbacks execution, which looks very strange considering the line where happens.
Can you print the content of cbp in frame 0:
gdb> p *cbp
I see in the logs there are some errors related to some private development you did to flatstore module. Check also there not to have some buffer overflow or inadequate write to the parameters received from tm callback.
----------
More information can be found at the following URL:
http://sip-router.org/tracker/index.php?do=details&task_id=179#comment371
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.