AF9015/AF9013 changes

Hello
I just looked latest for 3.3 and there was 3 patch missing I have already PULL requested.

Could you PULL those ASAP from that tree:
http://git.linuxtv.org/anttip/media_tree.git/shortlog/refs/heads/misc

2011-11-19 tda18218: fix 6 MHz default IF frequency 2011-11-19 af9015: limit I2C access to keep FW happy 2011-11-28 af9013: rewrite whole driver

Antti

http://palosaari.fi/

Antti,

Please, don't send pull requests like that. They won't go to my queue, as patchwork won't get it.

For patchwork to get a git pull request, it has to contain the strings bellow:

	The following changes since commit 1a5cd29631a6b75e49e6ad8a770ab9d69cda0fa2:

	[media] tda10021: Add support for DVB-C Annex C (2011-12-20 14:01:08 -0200)

	are available in the git repository at:
	git://linuxtv.org/gliakhovetski/v4l-dvb.git for-3.3

(those are from a real example).

E. g. the pull request should be generated with git request-pull.

This time, I'll get it by hand. I'm assuming that you want me to pull from:
git://linuxtv.org/anttip/media_tree.git misc

based on the URL you've sent me.

Regards,
Mauro

Still, the same problem as before:

drivers/media/dvb/dvb-usb/af9015.c: In function ‘af9015_rc_query’:
drivers/media/dvb/dvb-usb/af9015.c:1091:6: error: ‘adap’ undeclared (first use in this function) drivers/media/dvb/dvb-usb/af9015.c:1091:6: note: each undeclared identifier is reported only once for each function it appears in drivers/media/dvb/dvb-usb/af9015.c:1094:12: error: ‘struct af9015_state’ has no member named ‘sleep’ drivers/media/dvb/dvb-usb/af9015.c:1094:30: error: ‘fe’ undeclared (first use in this function)

As all patches that went to 3.1-rc7 were merged back, this means that there's something else missing there (maybe that patch you asked me to revert).

After digging for a while, it seems that this patch is dependent on the one you asked me to revert:

Author: Antti Palosaari <[email protected]>
Date:   Sat Nov 12 22:33:30 2011 -0300

[media] af9015: limit I2C access to keep FW happy

AF9015 firmware does not like if it gets interrupted by I2C adapter request on some critical phases. During normal operation I2C adapter is used only 2nd demodulator and tuner on dual tuner devices.

Override demodulator callbacks and use mutex for limit access to those "critical" paths to keep AF9015 happy.

Applying it before this patch made it to apply. Please check if my merge conflict didn't break anything.

Regards,
Mauro.