Age | Commit message (Collapse) | Author |
|
Resolve two open issues. One of them that accidentally wasn't closed.
(closes issue #15446)
(closes issue #16447)
Reported by: lpistone
Tested by: okrief
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8421 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Wouldn't compile on kernels >= 2.6.20.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8416 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
So just push all the maintenance mode processing off to the workqueue.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8415 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Since t1xxp_maint can be called from interrupt context with the
DAHDI_MAINT_LOOPSTOP cmd, push the processing of that command to a workqueue
since it may sleep in the t1_getreg call. DAHDI-560.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8410 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
If we timeout a command, don't free it right away unless we are the one who
removed it from whatever list it was on. Also, increase the timeout (2 seconds
wasn't enough when the firmware for the VPMOCT was being loaded on a system with
4K stacks) and check the return value from t1_getreg in case there were timeouts.
DAHDI-560.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8400 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
* If the receive packet isn't the correct size, we can't just drop it, we have
to resubmit it to the receive queue immediately so it will get cleaned up
properly.
* It isn't safe to leave the buffers on any lists while calling
the function to resubmit it to the descriptor ring, since it may be freed
which potentially results in a corrupted list.
* If the card is held up in a state where it is waiting for receive data, there
is a chance that when the transmit process is started up again it could DMA
data into a buffer that has already been freed. By using our own freelist,
there isn't a chance that the board will DMA data into memory that has been
recycled for another purpose.
* Disable the tasklet when processing hard underruns in order to not corrupt the
descriptor rings.
* Make sure voicebus_stop and voicebus_release aren't running at the same time
as the hard underrun handler. This too can result in corrupted descriptor
lists.
DAHDI-560.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8379 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Oddities in the receive processor state make this completion not so useful
anymore. It is more straightforward to simply poll the state on shutdown.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8378 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Since the locking will always take place in a bottom half now (either a timer or
tasklet), we do not need the special locking macros anymore.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8377 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8376 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8375 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8370 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Fixed a minor issue where stateless DAHDI_MAINT ioctl commands, such
as error insertion and clearing, were indicating the span going into
loopback alarm state when it really was not.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8298 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
* Fixed the bug where the card could not be brought out of local
loopback in E1 mode.
* Fixed a bunch of issues where the drivers didn't report unsupported
maintenance functions correctly.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8291 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8279 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Looping modes are now mutually exclusive. If two looping modes are enabled
simultaneously it tends to hose up our framer chip. Now, all looping modes are
cleared in the driver before any are set.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8274 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8255 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8250 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
E1 selection. DAHDI-553
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8245 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8220 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Brings in any backward compatibility definitions.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8204 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
In testing this option didn't appear to function properly. Just mark it debug
only so it's not part of the "official" interface in case there are some users
of it out there that I'm not aware of. DAHDI-267.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8203 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Setting the maximum latency can be useful if you have a system event that
normally causes a latency increase, but you would rather have a break in the
audio or frame slip, then let the latency grow to the current default maximum
which is 25ms. DAHDI-278.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8198 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8189 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
The vpmsupport module parameter (and a few other ones) do not do anything if
they are changed after the driver is first loaded. Therefore, let's make these
module parameters read only at runtime.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8188 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Since we're calling list_move_tail, we need to use the safe version.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8183 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Also, check for good frames as the wctdm24xxp driver does now.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8178 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8177 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Some systems do not like the port I/O and this change allows it to work.
DAHDI-515.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8176 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Basically, fetch the Hx8 firmware like many of the other firmwares.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8171 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Define CONFIG_DAHDI_NO_ECHOCAN_DISABLE in include/dahdi/dahdi_config.h in
order to prevent the tone detector from disabling software echocan.
I slightly modified the noechodisable.patch.
(closes issue #16805)
Reported by: sles
Patches:
noechodisable.patch uploaded by sles (license 991)
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8162 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
handle_transmit was clearing out the wrong size value and checking for timeouts
wasn't handled correctly.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8156 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
compatibility.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8147 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
atomic_inc_return was added in 2.6.10, and synchronize_rcu() was added in
2.6.12.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8142 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Needs work to make backward compatible with the range of supported kernels.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8141 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8140 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Reported by: seanbright
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8135 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
include/dahdi/kernel.h"
Reverts r8117. The wctdm24xxp driver can include <stdbool.h> directly.
Reported by: seanbright
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8134 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8124 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
The Hx8 series cards support BRI modules in addition to analog modules.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8123 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
This description needs to be filled in more.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8121 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8120 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8119 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
This interface is only used to facilitate debugging.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8118 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
I intend to use the bool definition in the wctdm24xxp driver and don't want to
duplicate the definition.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8117 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Request_module_nowait is only useful when asynchronous initialization is used
(committed to trunk in r8094), and that can only happen in kernel version after
2.6.30. Some earlier kernels do not have the request_module_nowait interface.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8111 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8110 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Just removing the errors only. I left the warnings for now.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8104 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Add TDMoE Multi-Frame support as described in the article at the following URL:
http://www.thrallingpenguin.com/articles/tdmoe-mf.htm
TDMoE-MF is known to be implemented in hardware solutions from Redfone
Communications.
This patch additionally implements RCU within dahdi_dynamic to decrease lock
contention, latency, and context switching. Because of the use of RCU locking,
all prior known issues with loading and unloading of the modules are resolved,
providing the spans are shutdown with "dahdi_cfg -s".
It also contains an attempt, which works, at fixing a kernel change with
skb_linearize(). The use of kernel version number does not work with SuSE SLES
10, as it appears they have backported the 2.6.18 change in to their 2.6.16
version.
This merges in the work Jbenden did at:
http://svn.digium.com/svn/dahdi/team/jbenden/tdmoe-mf@8102
(issue #13483)
Patch by: JBenden
Reported by: JBenden
Tested by: JBenden
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8103 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
Saves about 10us each interrupt on 3.06 GHz Xeon.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8097 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|
|
On a 2.6.32.7 debug kernel with a TDM2400, this can save up to 30ms every second
of processing time without any appreciable impact on runtime performance since
most of the commands to the modules during normal operation are reads, and
therefore only one would be in each SFRAME anyway. The impact of this change is
less for TDM410 and TDM800 since they already limit the number of modules that
they look for queued commands for.
There is still more room for optimizing module command handling.
git-svn-id: http://svn.asterisk.org/svn/dahdi/linux/trunk@8096 a0bf4364-ded3-4de4-8d8a-66a801d63aff
|