Skip to contentSkip to conversation

V4L/DVB bug fixes

PATCH In review [email protected] 3 participants 5 messages May 13, 2006 – May 21, 2006

lore
Share & tools
lore
#0 lore raw

cover V4L/DVB bug fixes

Linus,

Please pull these from master branch at:
kernel.org:/pub/scm/linux/kernel/git/mchehab/v4l-dvb.git

Mostly are just small patches to fix bugs. The most changed driver is cx24123 frontend, since it had several precision loss at math operations, resulting on several digital TV stations not being seen (from my tests, without the patch, only 8 TV stations were available, of a total of 28 ones with the patch).

We are also including some changes at Multimedia Kconfig menu to allow disabling drivers based at V4L1 API. This api were used until kernel 2.4, without providing enough capability to handle all analog TV video/audio standards. The removal of this feature is marked to July/2006. We intend to keep for a while a compatibility layer, already at V4L core, that converts V4L1 calls into V4L2 ones.

This patch series contains the following stuff:

- Fix some errors on bttv_risc_overlay
- Fix mutex in dvb_register_device to work.
- Fix TT budget-ci 1.1 CI slots
- Kbuild: drivers/media/video/bt8xx: remove $(src) from include path
- Saa7134: Fix oops with disable_ir=1
- Fix oops in budget-av with CI
- Set tone/voltage again if the frontend was reinitialised
- Fix some more potential oopses
- Fix a bug at pluto2 Makefile
- Bug fix: Wrong tuner was used pcHDTV HD-3000 card
- Correct buffer size calculations in cx88-core.c
- Pvr350 tv out (saa7127)
- Create V4L1 config options
- Add VIVI Kconfig stuff
- Removed uneeded stuff from pwc Makefile
- Fix compilation with V4L1_COMPAT
- Use after free in drivers/media/video/em28xx/em28xx-video.c
- Kbuild: DVB_BT8XX must select DVB_ZL10353
- Fix for CX24123 & low symbol rates
- Add several debug messages to cx24123 code
- Always wait for diseqc queue to become ready before transmitting a diseqc message
- Various correctness fixes to tuning.
- Tweak bandselect setup fox cx24123
- Add support for TCL M2523_5N_E tuner.
- Cxusb-bluebird: bug-fix: power down corrupts frontend
- Remove broken 'fast firmware load' from cx25840.
- Saa7134: Missing 'break' in Terratec Cinergy 400 TV initialization
- Fix frequency values in the ranges structures of the LG TDVS H06xF tuners
- Get_dvb_firmware: download nxt2002 firmware from new driver location
- Sparc32 vivi fix
- Vivi build fix
- Bt8xx/bttv-cards.c: fix off-by-one errors
- Fix CONFIG_VIDEO_VIVI=y build bug

Cheers,
Mauro.

V4L/DVB development is hosted at http://linuxtv.org
---

 Documentation/dvb/get_dvb_firmware                |    8 
 drivers/media/Kconfig                             |   45 -
 drivers/media/common/Kconfig                      |    1 
 drivers/media/dvb/bt8xx/Kconfig                   |    1 
 drivers/media/dvb/cinergyT2/cinergyT2.c           |    5 
 drivers/media/dvb/dvb-core/dvb_frontend.c         |   12 
 drivers/media/dvb/dvb-core/dvbdev.c               |    4 
 drivers/media/dvb/dvb-usb/cxusb.c                 |   17 
 drivers/media/dvb/frontends/cx24123.c             |  617 +++++++++-----
 drivers/media/dvb/frontends/dvb-pll.c             |    4 
 drivers/media/dvb/pluto2/Makefile                 |    2 
 drivers/media/dvb/ttpci/Kconfig                   |   12 
 drivers/media/dvb/ttpci/budget-av.c               |    6 
 drivers/media/dvb/ttpci/budget-ci.c               |  105 +-
 drivers/media/dvb/ttusb-budget/dvb-ttusb-budget.c |    6 
 drivers/media/radio/Kconfig                       |   30 
 drivers/media/video/Kconfig                       |   81 +
 drivers/media/video/Makefile                      |    7 
 drivers/media/video/bt8xx/Kconfig                 |    2 
 drivers/media/video/bt8xx/Makefile                |    2 
 drivers/media/video/bt8xx/bttv-cards.c            |    4 
 drivers/media/video/bt8xx/bttv-risc.c             |   14 
 drivers/media/video/cx25840/cx25840-firmware.c    |   49 -
 drivers/media/video/cx88/cx88-cards.c             |    2 
 drivers/media/video/cx88/cx88-core.c              |   16 
 drivers/media/video/cx88/cx88-dvb.c               |    2 
 drivers/media/video/cx88/cx88-video.c             |    2 
 drivers/media/video/em28xx/Kconfig                |    2 
 drivers/media/video/em28xx/em28xx-video.c         |   10 
 drivers/media/video/et61x251/Kconfig              |    2 
 drivers/media/video/pwc/Kconfig                   |    2 
 drivers/media/video/pwc/Makefile                  |   17 
 drivers/media/video/saa7127.c                     |    1 
 drivers/media/video/saa7134/saa7134-cards.c       |    1 
 drivers/media/video/saa7134/saa7134-core.c        |    6 
 drivers/media/video/saa7134/saa7134-video.c       |    2 
 drivers/media/video/sn9c102/Kconfig               |    2 
 drivers/media/video/tuner-types.c                 |    4 
 drivers/media/video/tveeprom.c                    |    2 
 drivers/media/video/usbvideo/Kconfig              |    6 
 drivers/media/video/vivi.c                        |    5 
 drivers/media/video/zc0301/Kconfig                |    2 
 include/linux/videodev2.h                         |    5 
 43 files changed, 729 insertions(+), 396 deletions(-)
#1 lore raw
Quoted 4 lines · [email protected] wrote:

Please pull these from master branch at:
kernel.org:/pub/scm/linux/kernel/git/mchehab/v4l-dvb.git

Linus,

A week has gone by... Do you intend to merge these bug fixes into your tree?

These changesets fix a whole bunch of serious bugs that have been introduced in 2.6.17, and a good many of these bugfixes have been sitting around, waiting to be merged since 2.6.17-rc1, not to mention the fixes that have already been merged much earlier in 2.6.16.y

Please merge these before 2.6.17-rc5

We have already eliminated as many patches as we possibly could. It all comes down to these. Everything else has been held back until the 2.6.18 merge window.

If you are nacking these fixes, then I would have to call 2.6.17 a broken kernel, when it comes to multimedia :-(

Please merge.

Regards,

Michael Krufky

Quoted 64 lines

Mostly are just small patches to fix bugs. The most changed driver is cx24123 frontend, since it had several precision loss at math operations, resulting on several digital TV stations not being seen (from my tests, without the patch, only 8 TV stations were available, of a total of 28 ones with the patch).

We are also including some changes at Multimedia Kconfig menu to allow disabling
drivers based at V4L1 API. This api were used until kernel 2.4, without providing
enough capability to handle all analog TV video/audio standards. The removal of this
feature is marked to July/2006. We intend to keep for a while a compatibility layer, already at V4L core, that converts V4L1 calls into V4L2 ones.

This patch series contains the following stuff:

- Fix some errors on bttv_risc_overlay
- Fix mutex in dvb_register_device to work.
- Fix TT budget-ci 1.1 CI slots
- Kbuild: drivers/media/video/bt8xx: remove $(src) from include path
- Saa7134: Fix oops with disable_ir=1
- Fix oops in budget-av with CI
- Set tone/voltage again if the frontend was reinitialised
- Fix some more potential oopses
- Fix a bug at pluto2 Makefile
- Bug fix: Wrong tuner was used pcHDTV HD-3000 card
- Correct buffer size calculations in cx88-core.c
- fix Pvr350 tv out (saa7127)
- Create V4L1 config options
- Add VIVI Kconfig stuff
- Removed uneeded stuff from pwc Makefile
- Fix compilation with V4L1_COMPAT
- Use after free in drivers/media/video/em28xx/em28xx-video.c
- Kbuild: DVB_BT8XX must select DVB_ZL10353
- Fix for CX24123 & low symbol rates
- Add several debug messages to cx24123 code
- Always wait for diseqc queue to become ready before transmitting a diseqc message
- Various correctness fixes to tuning.
- Tweak bandselect setup fox cx24123
- Add support for TCL M2523_5N_E tuner.
- Cxusb-bluebird: bug-fix: power down corrupts frontend
- Remove broken 'fast firmware load' from cx25840.
- Saa7134: Missing 'break' in Terratec Cinergy 400 TV initialization
- Fix frequency values in the ranges structures of the LG TDVS H06xF tuners
- Get_dvb_firmware: download nxt2002 firmware from new driver location
- Sparc32 vivi fix
- Vivi build fix
- Bt8xx/bttv-cards.c: fix off-by-one errors
- Fix CONFIG_VIDEO_VIVI=y build bug

Cheers,
Mauro.

V4L/DVB development is hosted at http://linuxtv.org
---

Documentation/dvb/get_dvb_firmware                |    8 drivers/media/Kconfig                             |   45 -
drivers/media/common/Kconfig                      |    1 drivers/media/dvb/bt8xx/Kconfig                   |    1 drivers/media/dvb/cinergyT2/cinergyT2.c           |    5 drivers/media/dvb/dvb-core/dvb_frontend.c         |   12 drivers/media/dvb/dvb-core/dvbdev.c               |    4 drivers/media/dvb/dvb-usb/cxusb.c                 |   17 drivers/media/dvb/frontends/cx24123.c             |  617 +++++++++-----
drivers/media/dvb/frontends/dvb-pll.c             |    4 drivers/media/dvb/pluto2/Makefile                 |    2 drivers/media/dvb/ttpci/Kconfig                   |   12 drivers/media/dvb/ttpci/budget-av.c               |    6 drivers/media/dvb/ttpci/budget-ci.c               |  105 +-
drivers/media/dvb/ttusb-budget/dvb-ttusb-budget.c |    6 drivers/media/radio/Kconfig                       |   30 drivers/media/video/Kconfig                       |   81 +
drivers/media/video/Makefile                      |    7 drivers/media/video/bt8xx/Kconfig                 |    2 drivers/media/video/bt8xx/Makefile                |    2 drivers/media/video/bt8xx/bttv-cards.c            |    4 drivers/media/video/bt8xx/bttv-risc.c             |   14 drivers/media/video/cx25840/cx25840-firmware.c    |   49 -
drivers/media/video/cx88/cx88-cards.c             |    2 drivers/media/video/cx88/cx88-core.c              |   16 drivers/media/video/cx88/cx88-dvb.c               |    2 drivers/media/video/cx88/cx88-video.c             |    2 drivers/media/video/em28xx/Kconfig                |    2 drivers/media/video/em28xx/em28xx-video.c         |   10 drivers/media/video/et61x251/Kconfig              |    2 drivers/media/video/pwc/Kconfig                   |    2 drivers/media/video/pwc/Makefile                  |   17 drivers/media/video/saa7127.c                     |    1 drivers/media/video/saa7134/saa7134-cards.c       |    1 drivers/media/video/saa7134/saa7134-core.c        |    6 drivers/media/video/saa7134/saa7134-video.c       |    2 drivers/media/video/sn9c102/Kconfig               |    2 drivers/media/video/tuner-types.c                 |    4 drivers/media/video/tveeprom.c                    |    2 drivers/media/video/usbvideo/Kconfig              |    6 drivers/media/video/vivi.c                        |    5 drivers/media/video/zc0301/Kconfig                |    2 include/linux/videodev2.h                         |    5 43 files changed, 729 insertions(+), 396 deletions(-)

_______________________________________________
v4l-dvb-maintainer mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/v4l-dvb-maintainer

#2 lore raw

Am Samstag, den 20.05.2006, 15:34 -0400 schrieb Michael Krufky:

Quoted 21 lines
Quoted 4 lines · [email protected] wrote:

Please pull these from master branch at:
kernel.org:/pub/scm/linux/kernel/git/mchehab/v4l-dvb.git

Linus,

A week has gone by... Do you intend to merge these bug fixes into your tree?

These changesets fix a whole bunch of serious bugs that have been introduced in 2.6.17, and a good many of these bugfixes have been sitting around, waiting to be merged since 2.6.17-rc1, not to mention the fixes that have already been merged much earlier in 2.6.16.y

Please merge these before 2.6.17-rc5

We have already eliminated as many patches as we possibly could. It all comes down to these. Everything else has been held back until the 2.6.18 merge window.

Quoted 9 lines

If you are nacking these fixes, then I would have to call 2.6.17 a broken kernel, when it comes to multimedia :-(

Please merge.

Regards,

Michael Krufky

Hi Mike,

I can see you are concerned and you are right.

OTOH we have some load currently and I wouldn't mind to go through what Andrew worked out for it so far. Especially as we never had much/any useful from those testers, but might change if it is employed long enough ;)

At least, and thanks again for calming me down once, it is not about some white space only now.

Cheers,
Hermann

Quoted 99 lines
Quoted 99 lines

Mostly are just small patches to fix bugs. The most changed driver is cx24123 frontend, since it had several precision loss at math operations, resulting on several digital TV stations not being seen (from my tests, without the patch, only 8 TV stations were available, of a total of 28 ones with the patch).

We are also including some changes at Multimedia Kconfig menu to allow disabling drivers based at V4L1 API. This api were used until kernel 2.4, without providing enough capability to handle all analog TV video/audio standards. The removal of this feature is marked to July/2006. We intend to keep for a while a compatibility layer, already at V4L core, that converts V4L1 calls into V4L2 ones.

This patch series contains the following stuff:

- Fix some errors on bttv_risc_overlay
- Fix mutex in dvb_register_device to work.
- Fix TT budget-ci 1.1 CI slots
- Kbuild: drivers/media/video/bt8xx: remove $(src) from include path
- Saa7134: Fix oops with disable_ir=1
- Fix oops in budget-av with CI
- Set tone/voltage again if the frontend was reinitialised
- Fix some more potential oopses
- Fix a bug at pluto2 Makefile
- Bug fix: Wrong tuner was used pcHDTV HD-3000 card
- Correct buffer size calculations in cx88-core.c
- fix Pvr350 tv out (saa7127)
- Create V4L1 config options
- Add VIVI Kconfig stuff
- Removed uneeded stuff from pwc Makefile
- Fix compilation with V4L1_COMPAT
- Use after free in drivers/media/video/em28xx/em28xx-video.c
- Kbuild: DVB_BT8XX must select DVB_ZL10353
- Fix for CX24123 & low symbol rates
- Add several debug messages to cx24123 code
- Always wait for diseqc queue to become ready before transmitting a diseqc message
- Various correctness fixes to tuning.
- Tweak bandselect setup fox cx24123
- Add support for TCL M2523_5N_E tuner.
- Cxusb-bluebird: bug-fix: power down corrupts frontend
- Remove broken 'fast firmware load' from cx25840.
- Saa7134: Missing 'break' in Terratec Cinergy 400 TV initialization
- Fix frequency values in the ranges structures of the LG TDVS H06xF tuners
- Get_dvb_firmware: download nxt2002 firmware from new driver location
- Sparc32 vivi fix
- Vivi build fix
- Bt8xx/bttv-cards.c: fix off-by-one errors
- Fix CONFIG_VIDEO_VIVI=y build bug

Cheers,
Mauro.

V4L/DVB development is hosted at http://linuxtv.org
---

Documentation/dvb/get_dvb_firmware                |    8 
drivers/media/Kconfig                             |   45 -
drivers/media/common/Kconfig                      |    1 
drivers/media/dvb/bt8xx/Kconfig                   |    1 
drivers/media/dvb/cinergyT2/cinergyT2.c           |    5 
drivers/media/dvb/dvb-core/dvb_frontend.c         |   12 
drivers/media/dvb/dvb-core/dvbdev.c               |    4 
drivers/media/dvb/dvb-usb/cxusb.c                 |   17 
drivers/media/dvb/frontends/cx24123.c             |  617 +++++++++-----
drivers/media/dvb/frontends/dvb-pll.c             |    4 
drivers/media/dvb/pluto2/Makefile                 |    2 
drivers/media/dvb/ttpci/Kconfig                   |   12 
drivers/media/dvb/ttpci/budget-av.c               |    6 
drivers/media/dvb/ttpci/budget-ci.c               |  105 +-
drivers/media/dvb/ttusb-budget/dvb-ttusb-budget.c |    6 
drivers/media/radio/Kconfig                       |   30 
drivers/media/video/Kconfig                       |   81 +
drivers/media/video/Makefile                      |    7 
drivers/media/video/bt8xx/Kconfig                 |    2 
drivers/media/video/bt8xx/Makefile                |    2 
drivers/media/video/bt8xx/bttv-cards.c            |    4 
drivers/media/video/bt8xx/bttv-risc.c             |   14 
drivers/media/video/cx25840/cx25840-firmware.c    |   49 -
drivers/media/video/cx88/cx88-cards.c             |    2 
drivers/media/video/cx88/cx88-core.c              |   16 
drivers/media/video/cx88/cx88-dvb.c               |    2 
drivers/media/video/cx88/cx88-video.c             |    2 
drivers/media/video/em28xx/Kconfig                |    2 
drivers/media/video/em28xx/em28xx-video.c         |   10 
drivers/media/video/et61x251/Kconfig              |    2 
drivers/media/video/pwc/Kconfig                   |    2 
drivers/media/video/pwc/Makefile                  |   17 
drivers/media/video/saa7127.c                     |    1 
drivers/media/video/saa7134/saa7134-cards.c       |    1 
drivers/media/video/saa7134/saa7134-core.c        |    6 
drivers/media/video/saa7134/saa7134-video.c       |    2 
drivers/media/video/sn9c102/Kconfig               |    2 
drivers/media/video/tuner-types.c                 |    4 
drivers/media/video/tveeprom.c                    |    2 
drivers/media/video/usbvideo/Kconfig              |    6 
drivers/media/video/vivi.c                        |    5 
drivers/media/video/zc0301/Kconfig                |    2 
include/linux/videodev2.h                         |    5 
43 files changed, 729 insertions(+), 396 deletions(-)
#3 lore raw
Quoted 18 lines · hermann pitton

Am Samstag, den 20.05.2006, 15:34 -0400 schrieb Michael Krufky:

Quoted 7 lines
Quoted 3 lines · [email protected] wrote:

Please pull these from master branch at:
kernel.org:/pub/scm/linux/kernel/git/mchehab/v4l-dvb.git

Linus,

Quoted 8 lines

A week has gone by... Do you intend to merge these bug fixes into your tree?

These changesets fix a whole bunch of serious bugs that have been introduced in 2.6.17, and a good many of these bugfixes have been sitting around, waiting to be merged since 2.6.17-rc1, not to mention the fixes that have already been merged much earlier in 2.6.16.y

Please merge these before 2.6.17-rc5

We have already eliminated as many patches as we possibly could. It all comes down to these. Everything else has been held back until the 2.6.18 merge window.

Quoted 10 lines

Hi Mike,

I can see you are concerned and you are right.

OTOH we have some load currently and I wouldn't mind to go through what Andrew worked out for it so far. Especially as we never had much/any useful from those testers, but might change if it is employed long enough ;)

Hermann,

The bugfixes that I am concerned about do not include the refactoring changes from Andrew -- those are not even bugfixes at all, and they are only accessible through our mercurial tree. We do not plan to send those upstream until 2.6.18 -- This has nothing at all to do with Mauro's latest git-pull-request, or my email to Linus.

Quoted 3 lines

At least, and thanks again for calming me down once, it is not about some white space only now.

I'm not sure what you mean. Sure, we do have some drastic changes coming up for 2.6.18 ... but this email is about bugfixes that HAVE in fact been thoroughly tested.

It sounds to me as if you may be getting the pending changesets in Mauro's git tree confused with current changes in our mercurial development repository.

Just take a look at v4l-dvb.git ... (or at the diffstat earlier in this thread) .. You will find that the drastic changes that you have mentioned aren't there yet. Those do need some more testing before being sent upstream... I think they'll be ready in time for the 2.6.18 merge window, but this is not the issue right now.

Thanks for your concern,

Michael Krufky

#4 lore raw

Am Sonntag, den 21.05.2006, 13:33 -0400 schrieb Michael Krufky:

Quoted 6 lines
Quoted 4 lines · hermann pitton

Am Samstag, den 20.05.2006, 15:34 -0400 schrieb Michael Krufky:

Quoted 1 line

[email protected] wrote:

[...]

Quoted 21 lines
Quoted 2 lines

Hermann,

The bugfixes that I am concerned about do not include the refactoring changes from Andrew -- those are not even bugfixes at all, and they are only accessible through our mercurial tree. We do not plan to send those upstream until 2.6.18 -- This has nothing at all to do with Mauro's latest git-pull-request, or my email to Linus.

Quoted 3 lines

At least, and thanks again for calming me down once, it is not about some white space only now.

I'm not sure what you mean. Sure, we do have some drastic changes coming up for 2.6.18 ... but this email is about bugfixes that HAVE in fact been thoroughly tested.

It sounds to me as if you may be getting the pending changesets in Mauro's git tree confused with current changes in our mercurial development repository.

Quoted 7 lines

Just take a look at v4l-dvb.git ... (or at the diffstat earlier in
this thread) .. You will find that the drastic changes that you have mentioned aren't there yet. Those do need some more testing before being sent upstream... I think they'll be ready in time for the 2.6.18 merge window, but this is not the issue right now.

Thanks for your concern,

;)

Mike,

for some users it will really look like broken multimedia on 2.6.17 like you said. I don't worry to send them to something else then, but it would really be nice to get the obvious fixes in or a hint, why it is too late or so.

The mixture of a for some broken 2.6.17 and a upcoming 2.6.18 with huge changes and a maybe too small testing base could become funny.

But yes, as you said, it are only concerns for now.

Cheers,
Hermann