[svn-commits] sruffell: branch linux/mspiceland/dahdi-trunk-adtapiupdatefor117 r5528 - in /...
SVN commits to the Digium repositories
svn-commits at lists.digium.com
Sun Dec 14 16:07:36 CST 2008
Author: sruffell
Date: Sun Dec 14 16:07:35 2008
New Revision: 5528
URL: http://svn.digium.com/view/dahdi?view=rev&rev=5528
Log:
Merged revisions 4618,4624,4628,4630,4636-4637,4641,4643,4648,4651,4653,4655,4657-4659,4662,4664-4665,4667-4669,4671-4673,4675,4677,4679-4682,4684-4686,4688,4690,4692,4694-4695,4697,4699,4702-4703,4705,4709,4712-4713,4717,4720-4721,4724-4725,4728,4742,4758-4761,4770,4773,4776,4781,4784,4788,4791,4794-4795,4798,4801,4804-4805,4817,4823,4828-4830,4834,4837,4840-4841,4847-4849,4856,4861,4864-4865,4868-4870,4894,4900,4905,4909,4917,4921,4930,4956-4957,4971,4979-4980,4986,4990,5002-5003,5007,5017,5021,5025,5029-5030,5034,5045-5046,5051,5056,5060,5064,5068,5084,5090,5097,5105-5107,5111,5115,5118,5121,5124,5127,5130-5133,5138,5148-5150,5156,5162-5163,5167,5171,5175,5179-5180,5186,5190-5191,5195-5196,5203,5207,5211,5230,5232,5236-5237,5249,5257,5269-5270,5275,5288,5303,5310,5315,5320-5321,5335,5340,5345,5350,5355,5360,5367-5369,5374,5383-5384,5397,5402-5403,5413,5420,5433-5434,5444,5453,5459-5460,5481-5486,5492,5497-5498,5503-5504,5509,5523 via svnmerge from
https://origsvn.digium.com/svn/dahdi/linux/trunk
........
r4618 | sruffell | 2008-07-21 10:39:35 -0500 (Mon, 21 Jul 2008) | 4 lines
Fixed typo where chan is cleared out accidentally.
(Related to issue #0013112)
........
r4624 | kpfleming | 2008-07-22 12:37:36 -0500 (Tue, 22 Jul 2008) | 2 lines
from r4429 of zaptel... make battery loss to red alarm translation actually work
........
r4628 | kpfleming | 2008-07-29 17:03:52 -0500 (Tue, 29 Jul 2008) | 2 lines
make an initial split of dahdi/kernel.h into userspace- and kernelspace-specific parts
........
r4630 | kpfleming | 2008-07-29 17:11:25 -0500 (Tue, 29 Jul 2008) | 2 lines
missed one definition needed by Asterisk 1.6/trunk
........
r4636 | kpfleming | 2008-07-30 10:14:58 -0500 (Wed, 30 Jul 2008) | 2 lines
remove extraneous file from .PHONY
........
r4637 | kpfleming | 2008-07-30 10:18:58 -0500 (Wed, 30 Jul 2008) | 2 lines
use a better technique to ensure that version.h gets updated when necessary
........
r4641 | kpfleming | 2008-07-31 11:21:27 -0500 (Thu, 31 Jul 2008) | 2 lines
would help if we didn't copy a NULL pointer and then try to dereference it...
........
r4643 | kpfleming | 2008-08-01 16:18:59 -0500 (Fri, 01 Aug 2008) | 8 lines
fix problem accepting non-linearized skbs from NIC drivers
(closes issue #13204)
Reported by: biohumanoid
Patches:
ztd-eth.c.patch uploaded by biohumanoid (license 459) (modified by me to actually apply to DAHDI)
........
r4648 | kpfleming | 2008-08-01 17:07:28 -0500 (Fri, 01 Aug 2008) | 2 lines
make skb_linearize() work for kernels <= 2.6.17 too
........
r4651 | kpfleming | 2008-08-01 17:46:02 -0500 (Fri, 01 Aug 2008) | 6 lines
fix a couple of places where the driver assumed 4 spans were present on the card
(closes issue #13141)
Reported by: pabelanger
........
r4653 | kpfleming | 2008-08-01 18:07:28 -0500 (Fri, 01 Aug 2008) | 6 lines
use correct logic to handle a slave channel being removed from a master channel, so if the master channel itself is not reconfigured it will still end up properly linking to any remaining slaves
(closes issue #11611)
Reported by: richvdh
Tested by: tzafrir
........
r4655 | russell | 2008-08-01 23:17:00 -0500 (Fri, 01 Aug 2008) | 2 lines
Space out the definition of an enum, and remove an unused typedef
........
r4657 | russell | 2008-08-01 23:38:56 -0500 (Fri, 01 Aug 2008) | 2 lines
convert a list of defines to an enum
........
r4658 | russell | 2008-08-01 23:39:28 -0500 (Fri, 01 Aug 2008) | 2 lines
Convert most of the existing comments in kernel.h to doxygen format
........
r4659 | russell | 2008-08-01 23:41:11 -0500 (Fri, 01 Aug 2008) | 1 line
ignore tags
........
r4662 | russell | 2008-08-02 08:59:43 -0500 (Sat, 02 Aug 2008) | 3 lines
- use kzalloc instead of kmalloc+memset
- reformat dahdi_q_sig to match proper coding guidelines
........
r4664 | russell | 2008-08-02 09:56:38 -0500 (Sat, 02 Aug 2008) | 5 lines
- align case with switch
- constify return of sigstr()
- rework implementation of fill_alarm_string to only use one call to snprintf()
and avoid more complicated string building logic
........
r4665 | russell | 2008-08-02 10:15:50 -0500 (Sat, 02 Aug 2008) | 2 lines
kzalloc was introduced in later versions of 2.6, so use kcalloc instead
........
r4667 | russell | 2008-08-02 14:59:37 -0500 (Sat, 02 Aug 2008) | 4 lines
- Define kzalloc() for kernel versions that don't have it
- Use kzalloc instead of kmalloc+memset
- formatting tweaks to better conform to coding guidelines
........
r4668 | russell | 2008-08-02 15:07:24 -0500 (Sat, 02 Aug 2008) | 2 lines
Update copyright, and fix a typo in some HDLC code that I wasn't building
........
r4669 | russell | 2008-08-02 15:17:43 -0500 (Sat, 02 Aug 2008) | 3 lines
- simplify some code using ARRAY_SIZE
- don't put multiple statements on the same line
........
r4671 | russell | 2008-08-02 15:22:13 -0500 (Sat, 02 Aug 2008) | 2 lines
Use C99 instead of GNU style initializers
........
r4672 | russell | 2008-08-02 15:33:18 -0500 (Sat, 02 Aug 2008) | 5 lines
- use C99 instead of GNU initializers
- adding spaces and braces according to guidelines
- use for loop instead of a while loop for open coded list traversal
- reduce some indentation
........
r4673 | russell | 2008-08-02 15:42:54 -0500 (Sat, 02 Aug 2008) | 6 lines
Rework free_tone_zone a bit to avoid a potential memory leak in a weird scenario.
It was theoretically possible that a zone was registered after we removed a zone,
but before we tried to put it back after determining that it was busy. If this
occurred, then when putting the zone back, we would overwrite the reference to
the newly created zone, leaking the associated memory.
........
r4675 | russell | 2008-08-02 17:23:42 -0500 (Sat, 02 Aug 2008) | 2 lines
Convert dahdi timers to a kernel list instead of an open coded list
........
r4677 | russell | 2008-08-03 09:31:03 -0500 (Sun, 03 Aug 2008) | 3 lines
- Formatting changes to better conform to guidelines
- tweak hw_echocan_on to reduce indentation and use an initializer instead of memset
........
r4679 | russell | 2008-08-03 10:42:46 -0500 (Sun, 03 Aug 2008) | 5 lines
Rework the table of signalling bits for signalling types and hook states
in dahdi_rbs_sethook() to a much more readable and less error prone format.
In passing, get rid of the dahdi_txsig_t typedef, and fix uses of it
throughout the tree.
........
r4680 | russell | 2008-08-03 10:44:24 -0500 (Sun, 03 Aug 2008) | 2 lines
Fix a couple last references to dahdi_txsig_t.
........
r4681 | russell | 2008-08-03 10:48:43 -0500 (Sun, 03 Aug 2008) | 2 lines
Get rid of another typedef, sf_detect_state_t, and fix references
........
r4682 | russell | 2008-08-03 11:25:15 -0500 (Sun, 03 Aug 2008) | 2 lines
Get rid of dahdi_rxsig_t typedef
........
r4684 | russell | 2008-08-03 11:38:30 -0500 (Sun, 03 Aug 2008) | 2 lines
reduce indentation and tweak formatting of a for loop in proc_read
........
r4685 | russell | 2008-08-03 11:47:46 -0500 (Sun, 03 Aug 2008) | 7 lines
Change all instances of printk in DAHDI to include a priority
(closes issue #11504)
Reported by: tzafrir
Patches:
20080717__issue11504_dahdi_printk_without_priority.diff uploaded by bbryant (license 36)
........
r4686 | russell | 2008-08-03 11:49:33 -0500 (Sun, 03 Aug 2008) | 2 lines
Strip trailing whitespace
........
r4688 | tzafrir | 2008-08-03 13:06:15 -0500 (Sun, 03 Aug 2008) | 5 lines
Do use printk-s without a priority when continuing previous printk.
Also remove a small duplication in dahdi-base.c, as I'm too lazy to
fix in two separate places.
........
r4690 | russell | 2008-08-03 18:04:23 -0500 (Sun, 03 Aug 2008) | 7 lines
Re-work a for loop in proc_read. This loop traversed the array of all DAHDI
channels and determined whether they belonged to the span being printed or
not. This has been simplified by simply traversing the array of channels on
the span structure, itself.
Thanks to tzafrir for the idea in #asterisk-dev.
........
r4692 | russell | 2008-08-03 19:27:47 -0500 (Sun, 03 Aug 2008) | 2 lines
Tweak spacing and reduce indentation
........
r4694 | tzafrir | 2008-08-04 08:43:18 -0500 (Mon, 04 Aug 2008) | 3 lines
Fix small regression from r4690: do use channo for printing channel
numbers.
........
r4695 | tzafrir | 2008-08-04 09:02:10 -0500 (Mon, 04 Aug 2008) | 7 lines
Firmware update: 5959 (FTMF and BRI ID)
* Fix some cases of double digits.
* Fix some rare cases of a BRI unit providing incorrect directionality.
* Extra debugging code in card_fxs.c to trace future DTMF issues.
* Properly note Zaptel (as opposed to DAHDI) ioctls.
........
r4697 | russell | 2008-08-04 16:43:45 -0500 (Mon, 04 Aug 2008) | 4 lines
- Tweak some formatting to match coding guidelines
- Localize a couple of variables
- Use snprintf instead of sprintf
........
r4699 | russell | 2008-08-04 19:54:49 -0500 (Mon, 04 Aug 2008) | 4 lines
In the function start_digit(), the same logic was replicated 3 times; one time
for dtmf, one for mfr2_fwd, and one for mfr2_rev. So, I created a new function
which handles all three cases in a way that is (IMO) easier to read.
........
r4702 | russell | 2008-08-05 10:55:32 -0500 (Tue, 05 Aug 2008) | 2 lines
Copy tools README over to the linux package
........
r4703 | russell | 2008-08-05 10:58:35 -0500 (Tue, 05 Aug 2008) | 2 lines
Fix alignment issues caused by different tab sizes
........
r4705 | tzafrir | 2008-08-05 13:34:16 -0500 (Tue, 05 Aug 2008) | 2 lines
Fix a small regression from r4667: Please don't NULL a good pointer.
........
r4709 | kpfleming | 2008-08-05 18:56:00 -0500 (Tue, 05 Aug 2008) | 3 lines
first pass at licensing cleanup here... put dahdi/user.h under LGPL instead of GPL
........
r4712 | kpfleming | 2008-08-06 10:05:25 -0500 (Wed, 06 Aug 2008) | 1 line
more license header updates
........
r4713 | kpfleming | 2008-08-06 10:18:45 -0500 (Wed, 06 Aug 2008) | 3 lines
change some ioctls that don't actually use a parameter to be defined that way
........
r4717 | sruffell | 2008-08-06 13:09:58 -0500 (Wed, 06 Aug 2008) | 32 lines
A significant change to the DAHDI transcoder interface and the wctc4xxp
driver which breaks backwards compatibility.
Basically, I've replaced the memory mapped interface with a read/write
interface that allows codec_dahdi to communicate with the hardware transcoder
the exact size of the packet to be transcoded. This eliminates issues with
remote devices that send G729.B CNG packets even though asterisk does not
support them.
More information about the svn-commits
mailing list