For those experimenting / using 3.17 or 3.18 and experiencing freezes: There seems to be a lockup problem in the kernel, currently discussed in lkml that I was experiencing as well. Although my problems were not caused by high load (I useVAAPI to decode HD H.264 content) all my problems went away after I downgraded to 3.16.7. MythTV + ddbridge are now solid as a rock!
Hope this may help anyone looking at inexplainable freezes while watching TV (MythTV in my case).
I'm seeing I2C timeouts in ddbridge module (and linked tuners) with latest media_build_experimental:
[49327.084218] DDBridge I2C timeout, card 0, port 0
[49327.084233] ddbridge 0000:01:00.0: DDBridge IRS 00000301
[49327.084237] cxd2843: i2c_read error
[49327.084376] tda18212: i2c_write error
[49537.146442] INFO: task kdvb-ad-0-fe-0:1895 blocked for more than 120 seconds.
[49537.146451] Tainted: G O 3.17.2 #10
[49537.146455] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[49537.146459] kdvb-ad-0-fe-0 D ffff8800ab76cd98 0 1895 2 0x00000000
[49537.146469] ffff8800ab727c60 0000000000000002 ffff8800ab76c890 ffff8800ab727fd8
[49537.146476] 0000000000013180 0000000000013180 ffffffff81c17480 ffff8800ab76c890
[49537.146482] ffff8800b7041d20 ffff8800b7041d24 ffff8800ab76c890 00000000ffffffff
[49537.146490] Call Trace:
[49537.146504] [<ffffffff8161a409>] schedule_preempt_disabled+0x29/0x70
[49537.146511] [<ffffffff8161bf72>] __mutex_lock_slowpath+0x152/0x1e0
[49537.146518] [<ffffffff8161c01f>] mutex_lock+0x1f/0x2f
[49537.146537] [<ffffffffa01c6dab>] gate_ctrl+0x2b/0x280 [cxd2843]
[49537.146550] [<ffffffffa00adc0e>] locked_gate_ctrl+0x3e/0x90 [ddbridge]
[49537.146559] [<ffffffffa01d861d>] set_params+0x3cd/0x910 [tda18212dd]
[49537.146567] [<ffffffff8109e3df>] ? try_to_del_timer_sync+0x4f/0x70
[49537.146577] [<ffffffffa01c57dc>] set_parameters+0x5c/0x1030 [cxd2843]
[49537.146583] [<ffffffff8109cc10>] ? internal_add_timer+0xb0/0xb0
[49537.146593] [<ffffffffa01c6879>] tune+0x49/0x70 [cxd2843]
[49537.146608] [<ffffffffa007e366>] dvb_frontend_thread+0x3e6/0x710 [dvb_core]
[49537.146617] [<ffffffff81083520>] ? prepare_to_wait_event+0x110/0x110
[49537.146630] [<ffffffffa007df80>] ? dvb_frontend_release+0x110/0x110 [dvb_core]
[49537.146638] [<ffffffff81066742>] kthread+0xd2/0xf0
[49537.146645] [<ffffffff81066670>] ? kthread_create_on_node+0x170/0x170
[49537.146652] [<ffffffff8161e16c>] ret_from_fork+0x7c/0xb0
[49537.146659] [<ffffffff81066670>] ? kthread_create_on_node+0x170/0x170
I successfully built last media_build_experimental against 3.17.2 this morning and have ddbridge DVB + dvbloopback reception. Nice to be at cutting edge again 
Hi Oliver,
Thx for the updated media_build_experimental! One question though: why do you remove the mutex locks/unlocks in dvbdev.c:dvb_device_open? And, as someone else pointed out: we dvbloopback driver users need the extra unlock/lock around err = file->f_op->open(inode,file); to prevent tuner lockups. I agree with you this should be solved in the offending module, but I'm no kernel/module hacker and it seems noone wants to, or can pick up the task to fix this long standing bug in dvbloopback/ffdecsawrapper.
Anyway, it's not a big problem to uncomment the (un)locks and add a couple of lines myself. I was just wondering why you deviate from the original code by commenting the existing lock/unlocks?
Martin
I managed to merge 0.9.15a into media_build_experimental and seem to have overcome long-term instabilities of my CINE C2/T2 (v7). My tuners were structurally failing after a couple of days resulting in 0B recordings and this now seems to be resolved although I admit I need a little longer testing to be sure.
The patches didn't apply neatly to the latest media_build_experimental but I managed to apply them manually. I also applied the last patch to v4l/ddbridge-core.c since that had errors as well but still no luck. See attachment.
Compilation goes a lot further, but ultimately fails on 3.17. See output in attachment.
No problem, the system perfoms quite well against the default ubuntu 14.04 kernel now. I was just looking into completely home built solution. Is there any chance the cine c2/t2 v7 (0006) will ever be supported in the mainline kernel?
Excuse my English. My German is not very good and this is the only place I found with very recent activity around the ddbridge module for linux...
I tried to compile media_build_experimental against vanilla 3.17.0 this morning and it failed.
I renamed media_build_experimental to media_build_experimental-3.17 because I didn't want to break my working tree.
# make
make -C /usr/src/media_build_experimental-3.17/v4l
make[1]: Entering directory `/usr/src/media_build_experimental-3.17/v4l'
scripts/make_makefile.pl
make[2]: Entering directory `/usr/src/media_build_experimental-3.17/linux'
Patches for 3.17.0 already applied.
make[2]: Leaving directory `/usr/src/media_build_experimental-3.17/linux'
./scripts/make_kconfig.pl /lib/modules/3.17.0/build /lib/modules/3.17.0/source
Preparing to compile for kernel version 3.17.0
WARNING: This is the V4L/DVB backport tree, with experimental drivers
backported to run on legacy kernels from the development tree at:
http://git.linuxtv.org/media-tree.git.
It is generally safe to use it for testing a new driver or
feature, but its usage on production environments is risky.
Don't use it in production. You've been warned.
../experimental/update_kconfig.pl
./scripts/make_myconfig.pl
make[1]: Leaving directory `/usr/src/media_build_experimental-3.17/v4l'
make[1]: Entering directory `/usr/src/media_build_experimental-3.17/v4l'
perl scripts/make_config_compat.pl /lib/modules/3.17.0/source ./.myconfig ./config-compat.h
creating symbolic links...
make -C firmware prep
make[2]: Entering directory `/usr/src/media_build_experimental-3.17/v4l/firmware'
make[2]: Leaving directory `/usr/src/media_build_experimental-3.17/v4l/firmware'
make -C firmware
make[2]: Entering directory `/usr/src/media_build_experimental-3.17/v4l/firmware'
CC ihex2fw
Generating vicam/firmware.fw
Generating ttusb-budget/dspbootcode.bin
Generating cpia2/stv0672_vp4.bin
Generating av7110/bootcode.bin
make[2]: Leaving directory `/usr/src/media_build_experimental-3.17/v4l/firmware'
Kernel build directory is /lib/modules/3.17.0/build
make -C ../linux apply_patches
make[2]: Entering directory `/usr/src/media_build_experimental-3.17/linux'
Patches for 3.17.0 already applied.
make[2]: Leaving directory `/usr/src/media_build_experimental-3.17/linux'
make -C /lib/modules/3.17.0/build SUBDIRS=/usr/src/media_build_experimental-3.17/v4l modules
make[2]: Entering directory `/usr/src/kernel/linux-3.17'
CC [M] /usr/src/media_build_experimental-3.17/v4l/mt2131.o
In file included from <command-line>:0:0:
/usr/src/media_build_experimental-3.17/v4l/compat.h: In function 'snd_card_create':
/usr/src/media_build_experimental-3.17/v4l/compat.h:686:2: warning: passing argument 1 of 'snd_card_new' makes pointer from integer without a cast [enabled by default]
*card = snd_card_new(idx, id, module, extra_size);
^
In file included from /usr/src/media_build_experimental-3.17/v4l/compat.h:246:0,
from <command-line>:0:
include/sound/core.h:283:5: note: expected 'struct device *' but argument is of type 'int'
int snd_card_new(struct device *parent, int idx, const char *xid,
^
In file included from <command-line>:0:0:
/usr/src/media_build_experimental-3.17/v4l/compat.h:686:2: warning: passing argument 2 of 'snd_card_new' makes integer from pointer without a cast [enabled by default]
*card = snd_card_new(idx, id, module, extra_size);
^
In file included from /usr/src/media_build_experimental-3.17/v4l/compat.h:246:0,
from <command-line>:0:
include/sound/core.h:283:5: note: expected 'int' but argument is of type 'const char *'
int snd_card_new(struct device *parent, int idx, const char *xid,
^
In file included from <command-line>:0:0:
/usr/src/media_build_experimental-3.17/v4l/compat.h:686:2: warning: passing argument 3 of 'snd_card_new' from incompatible pointer type [enabled by default]
*card = snd_card_new(idx, id, module, extra_size);
^
In file included from /usr/src/media_build_experimental-3.17/v4l/compat.h:246:0,
from <command-line>:0:
include/sound/core.h:283:5: note: expected 'const char *' but argument is of type 'struct module *'
int snd_card_new(struct device *parent, int idx, const char *xid,
^
In file included from <command-line>:0:0:
/usr/src/media_build_experimental-3.17/v4l/compat.h:686:2: warning: passing argument 4 of 'snd_card_new' makes pointer from integer without a cast [enabled by default]
*card = snd_card_new(idx, id, module, extra_size);
^
In file included from /usr/src/media_build_experimental-3.17/v4l/compat.h:246:0,
from <command-line>:0:
include/sound/core.h:283:5: note: expected 'struct module *' but argument is of type 'int'
int snd_card_new(struct device *parent, int idx, const char *xid,
^
In file included from <command-line>:0:0:
/usr/src/media_build_experimental-3.17/v4l/compat.h:686:2: error: too few arguments to function 'snd_card_new'
*card = snd_card_new(idx, id, module, extra_size);
^
In file included from /usr/src/media_build_experimental-3.17/v4l/compat.h:246:0,
from <command-line>:0:
include/sound/core.h:283:5: note: declared here
int snd_card_new(struct device *parent, int idx, const char *xid,
^
make[3]: *** [/usr/src/media_build_experimental-3.17/v4l/mt2131.o] Error 1
make[2]: *** [_module_/usr/src/media_build_experimental-3.17/v4l] Error 2
make[2]: Leaving directory `/usr/src/kernel/linux-3.17'
make[1]: *** [default] Error 2
make[1]: Leaving directory `/usr/src/media_build_experimental-3.17/v4l'
make: *** [all] Error 2