Commit Graph

48 Commits

Author SHA1 Message Date
Max Kellermann
bd336e6344 [media] dvb: make DVB frontend *_ops instances "const"
These are immutable.  Making them "const" allows the compiler to move
them to the "rodata" section.

Note that cxd2841er_t_c_ops cannot be made "const", because
cxd2841er_attach() modifies it.  Ouch!

[mchehab@s-opensource.com: fix merge conflicts]
Signed-off-by: Max Kellermann <max.kellermann@gmail.com>

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2016-11-18 15:00:22 -02:00
Nicolas Iooss
19157003ab [media] mb86a20s: always initialize a return value
In mb86a20s_read_status_and_stats(), when mb86a20s_read_status() fails,
the function returns the value in variable rc without initializing it
first. Fix this by propagating the error code from variable status_nr.

This bug has been found using clang and -Wsometimes-uninitialized
warning flag.

Signed-off-by: Nicolas Iooss <nicolas.iooss_linux@m4x.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2016-10-24 17:16:56 -02:00
Mauro Carvalho Chehab
505a0ea706 [media] mb86a20s: fix demod settings
With the current settings, only one channel locks properly.
That's likely because, when this driver was written, Brazil
were still using experimental transmissions.

Change it to reproduce the settings used by the newer drivers.
That makes it lock on other channels.

Tested with both PixelView SBTVD Hybrid (cx231xx-based) and
C3Tech Digital Duo HDTV/SDTV (em28xx-based) devices.

Cc: stable@vger.kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2016-09-05 15:26:30 -03:00
Mauro Carvalho Chehab
dafb65fb98 [media] mb86a20s: fix the locking logic
On this frontend, it takes a while to start output normal
TS data. That only happens on state S9. On S8, the TS output
is enabled, but it is not reliable enough.

However, the zigzag loop is too fast to let it sync.

As, on practical tests, the zigzag software loop doesn't
seem to be helping, but just slowing down the tuning, let's
switch to hardware algorithm, as the tuners used on such
devices are capable of work with frequency drifts without
any help from software.

Cc: stable@vger.kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2016-09-05 15:26:29 -03:00
Colin Ian King
eca2d34b9d [media] mb86a20s: apply mask to val after checking for read failure
Appling the mask 0x0f to the immediate return of the call to
mb86a20s_readreg will always result in a positive value, meaning that the
check of ret < 0 will never work.  Instead, check for a -ve return value
first, and then mask val with 0x0f.

Kudos to Mauro Carvalho Chehab for spotting the mistake in my original fix.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2016-07-15 13:32:58 -03:00
Mauro Carvalho Chehab
a749704984 [media] mb86a20s: get rid of dummy get_frontend()
This is not needed, as the core handles well if get_frontend()
is not present.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2016-02-04 16:27:29 -02:00
Mauro Carvalho Chehab
0df289a209 [media] dvb: Get rid of typedev usage for enums
The DVB API was originally defined using typedefs. This is against
Kernel CodingStyle, and there's no good usage here. While we can't
remove its usage on userspace, we can avoid its usage in Kernelspace.

So, let's do it.

This patch was generated by this shell script:

	for j in $(grep typedef include/uapi/linux/dvb/frontend.h |cut -d' ' -f 3); do for i in $(find drivers/media -name '*.[ch]' -type f) $(find drivers/staging/media -name '*.[ch]' -type f); do sed "s,${j}_t,enum $j," <$i >a && mv a $i; done; done

While here, make CodingStyle fixes on the affected lines.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Acked-by: Stefan Richter <stefanr@s5r6.in-berlin.de> # for drivers/media/firewire/*
2015-06-09 17:47:35 -03:00
Mauro Carvalho Chehab
e32b31ae45 [media] mb86a20s: remove unused debug modprobe parameter
The debug parameter is not used anymore, as this module was
converted already to use dev_dbg().

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
2015-01-26 10:08:29 -02:00
Hans Verkuil
ce08131c42 [media] mb86a16/mb86a20s: fix sparse warnings
drivers/media/dvb-frontends/mb86a16.c:31:14: warning: symbol 'verbose' was not declared. Should it be static?
drivers/media/dvb-frontends/mb86a20s.c:36:4: warning: symbol 'mb86a20s_subchannel' was not declared. Should it be static?
drivers/media/dvb-frontends/mb86a20s.c:1333:24: warning: symbol 'cnr_qpsk_table' was not declared. Should it be static?

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
2014-09-03 09:37:47 -03:00
Mauro Carvalho Chehab
277bfd2f4d [media] mb86a20s: Fix the code that estimates the measurement interval
Instead of looking at the guard interval field, it was using
the interval length, with is wrong. Fix it.

Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
2014-07-22 21:34:03 -03:00
Mauro Carvalho Chehab
4247368bf7 [media] mb86a20s: Fix Interleaving
Interleaving code was wrong at mb86a20s: instead, it was looking
at the Guard Interval. Fix it.

Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
2014-07-22 21:33:41 -03:00
Mauro Carvalho Chehab
276dfc4b99 [media] mb86a20s: fix ISDB-T mode handling
The driver was reporting an incorrect mode, when mode 2
is selected.

While testing it, noticed that neither mode 1 or guard
interval 1/32 is supported by this device. Document it,
and ensure that it will report _AUTO when it doesn't lock,
in order to not report a wrong detection to userspace.

Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
2014-07-22 21:33:25 -03:00
Mauro Carvalho Chehab
37e59f876b [media, edac] Change my email address
There are several left overs with my old email address.
Remove their occurrences and add myself at CREDITS, to
allow people to be able to reach me on my new addresses.

Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
2014-02-07 08:03:07 -02:00
Mauro Carvalho Chehab
9d32069faa [media] mb86a20s: Fix TS parallel mode
changeset 768e6dadd7 caused a regression on using mb86a20s
in parallel mode, as the parallel mode selection got
overriden by mb86a20s_init2.

Cc: stable@vger.kernel.org
Signed-off-by: Mauro Carvalho Chehab <m.chehab@samsung.com>
2013-08-22 12:06:29 -03:00
Mauro Carvalho Chehab
0562aef296 [media] mb86a20s: better name temp vars at mb86a20s_layer_bitrate()
Using 'i' for the guard interval temporary var is a bad idea, as
'i' is generally used by "anonymous" indexes.
Let's rename modulation, fec and guard interval temp vars with
a meaningful name, as that makes easier to understand the code
and avoids cut-and-paste types of error.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-04-08 07:12:19 -03:00
Mauro Carvalho Chehab
b1f8933119 [media] mb86a20s: Fix estimate_rate setting
As reported by Dan Carpenter <dan.carpenter@oracle.com>:
        Smatch warnings:
        drivers/media/dvb-frontends/mb86a20s.c:644 mb86a20s_layer_bitrate() error: buffer overflow 'state->estimated_rate' 3 <= 3
What happens there is that estimate_rate index should be the layer
number, and not the guard interval.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-04-08 07:11:06 -03:00
Mauro Carvalho Chehab
5cb88ca84a [media] mb86a20s: Use 'layer' instead of 'i' on all places
We're using the anonymous 'i' to indicate the layer number
on several places on the driver. That's not good, as some
cut-and-paste type of change might be doing the wrong thing.
So, call it as "layer" everywhere.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-04-08 07:10:53 -03:00
Mauro Carvalho Chehab
41c6e9dd09 [media] mb86a20s: fix audio sub-channel check
As reported by Dan Carpenter <dan.carpenter@oracle.com>
FYI, there are new smatch warnings show up in:
	tree:   git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next master
	head:   da17d7bda957ae4697b6abc0793f74fb9b50b58f
	commit: 04fa725e7b [media] mb86a20s: Implement set_frontend cache logic
	New smatch warnings:
	drivers/media/dvb-frontends/mb86a20s.c:1897 mb86a20s_set_frontend() error: buffer overflow 'mb86a20s_subchannel' 8 <= 8
	04fa725e Mauro Carvalho Chehab 2013-03-04  1894  		if (c->isdbt_sb_subchannel > ARRAY_SIZE(mb86a20s_subchannel))
	04fa725e Mauro Carvalho Chehab 2013-03-04  1895  			c->isdbt_sb_subchannel = 0;
	04fa725e Mauro Carvalho Chehab 2013-03-04  1896
	04fa725e Mauro Carvalho Chehab 2013-03-04 @1897  		state->subchannel = mb86a20s_subchannel[c->isdbt_sb_subchannel];

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-04-08 07:10:33 -03:00
Mauro Carvalho Chehab
4f62a20dfb [media] mb86a20s: Use a macro for the number of layers
Instead of using the magic number "3", use NUM_LAYERS macro
on all places that are related to the ISDB-T layers.
This makes the source code a little more readable.
No functional changes.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-04-08 07:10:17 -03:00
Mauro Carvalho Chehab
0e4bbedd63 [media] mb86a20s: Don't assume a 32.57142MHz clock
Now that some devices initialize register 0x2a with different
values, add the calculus formula, instead of hardcoding it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:37:07 -03:00
Mauro Carvalho Chehab
04fa725e7b [media] mb86a20s: Implement set_frontend cache logic
Up to now, the driver was simply assuming TV mode, 13 segs.
Implement the logic to control the ISDB operational mode.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:36:40 -03:00
Mauro Carvalho Chehab
3a2e47519e [media] mb86a20s: cleanup the status at set_frontend()
As the device got re-initialized, the stats should vanish
until the device gets lock again.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:35:16 -03:00
Mauro Carvalho Chehab
8b8e444a27 [media] mb86a20s: Don't reset strength with the other stats
Signal strength is always available. There's no reason to reset
it, as it has its own logic to reset it already.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:34:52 -03:00
Mauro Carvalho Chehab
a78b41d5fd [media] mb86a20s: Always reset the frontend with set_frontend
Always init the frontend when set_frontend is called. The rationale
is: it was noticed that, on some devices, it fails to lock with a
different channel. It seems that some other registers need to be
restored to its initial state, when the channel changes.
As it is better to reset everything, even wasting a few more
miliseconds than to loose channel lock, let's change the logic
to always reset.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:34:31 -03:00
Mauro Carvalho Chehab
17e67d4c7f [media] mb86a20s: change AGC tuning parameters
Use the AGC settings present on a newer device.
The initial settings were taken from one of the first devices with
mb86a20s, and there are several reports that this is not working
properly on some places.
So, instead of keeping using it, get the parameters taken from a
newer device. Tests are welcomed.
Tested also with cx231xx PixelView SBTVD Hybrid with no regressions
noticed so far.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:34:05 -03:00
Mauro Carvalho Chehab
0921ecfdc2 [media] mb86a20s: don't allow updating signal strength too fast
Getting signal strength requires some loop poking with I2C.
Don't let it happen too fast.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:33:43 -03:00
Mauro Carvalho Chehab
dad78c5662 [media] mb86a20s: Fix signal strength calculus
A register typo made the calculation to not work. Fix it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:33:20 -03:00
Mauro Carvalho Chehab
15b1c5a068 [media] mb86a20s: provide CNR stats before FE_HAS_SYNC
State 9 means TS started to be output, and it should be
associated with FE_HAS_SYNC.
The mb86a20scan get CNR statistics at state 7, when frame sync
is obtained.
As CNR may help to adjust the antenna, provide it earlier.
A latter patch could eventually start outputing MER measures
earlier, but that would require a bigger change, and probably
won't be better than the current way, as the time between
changing from state 8 to 9 is generally lower than the time
to get the stats collected.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:32:55 -03:00
Mauro Carvalho Chehab
768e6dadd7 [media] mb86a20s: adjust IF based on what's set on the tuner
Instead of hardcoding a fixed IF frequency of 3.3 MHz, use
the IF frequency provided by the tuner driver.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:32:33 -03:00
Mauro Carvalho Chehab
d56e326f7b [media] mb86a20s: don't pollute dmesg with debug messages
There are a few debug tests that are shown with dev_err() or
dev_info(). Replace them by dev_dbg().

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-03-04 16:32:08 -03:00
Mauro Carvalho Chehab
f67102c49a [media] mb86a20s: remove global BER/PER counters if per-layer counters vanish
If, for any reason, all per-layers counters stop, remove the
corresponding global counter.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-24 14:28:32 -02:00
Mauro Carvalho Chehab
d9b6f08a7f [media] mb86a20s: add a logic for post-BER measurement
The logic here is similar to the preBER.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-24 14:28:14 -02:00
Mauro Carvalho Chehab
313cf4efa4 [media] mb86a20s: fix the PER reset logic
The logic that resets the device is wrong. It should be resetting
just the layer that got read. Also, stop is needed before updating
the counters.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-24 14:27:16 -02:00
Mauro Carvalho Chehab
ad0abbf118 [media] mb86a20s: some fixes at preBER logic
The logic that resets the device is wrong. It should be resetting
just the layer that got read. Also, stop is needed before updating
the counters.
While there, rename it, as we'll soon introduce a postBER logic
there.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-24 14:26:55 -02:00
Mauro Carvalho Chehab
593ae89a3f [media] mb86a20s: add block count measures (PER/UCB)
Add both per-layer and global block error count and block count,
for PER and UCB measurements.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-24 14:24:59 -02:00
Mauro Carvalho Chehab
25188bd0e6 [media] mb86a20s: add CNR measurement
Add Signal/Noise ratio measurement. On this device, a global measure
is taken by the demod. It also provides per-layer CNR measurements,
based on Modulation Error measures (MER).

Reviewed-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-23 19:10:49 -02:00
Mauro Carvalho Chehab
d01a8ee37a [media] mb86a20s: improve bit error count for BER
Do a better job on setting the bit error counters, in order to
have all layer measures to happen in a little less than one
second.

Reviewed-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-23 19:10:42 -02:00
Mauro Carvalho Chehab
149d518ad0 [media] mb86a20s: add BER measurement
Add the methods to read bit error/bit count measurements from
mb86a20s. On ISDB-T devices, those reads are done per layer.
However, as userspace applications may not be aware of that,
add a global measure that will sum the bit errors and bit
counts for each layer, storing them into a global value.

Reviewed-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-23 19:10:35 -02:00
Mauro Carvalho Chehab
09b6d21e10 [media] mb86a20s: calculate statistics at .read_status()
Instead of providing separate callbacks to read the several FE
stats properties, the better seems to use just one method that will:
    - Read lock status;
    - Read signal strength;
    - if locked, get TMCC data;
    - if locked, get DVB statistics.
As the DVB frontend thread will call this read_status callback
on every 3 seconds, and userspace can even call it earlier,
all stats data and layers layout will be updated together if
available, with is a good thing.

Reviewed-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-23 19:10:27 -02:00
Mauro Carvalho Chehab
f167e302c6 [media] mb86a20s: don't use state before initializing it
As reported by Feng's kbuild test:
	From: kbuild test robot <fengguang.wu@intel.com>
	Subject: drivers/media/dvb-frontends/mb86a20s.c:706 mb86a20s_attach() error: potential null dereference 'state'.  (kzalloc returns null)
	Date: Wed, 23 Jan 2013 19:30:43 +0800

	commit: f66d81b54d [media] mb86a20s: convert it to use dev_info/dev_err/dev_dbg

	drivers/media/dvb-frontends/mb86a20s.c:706 mb86a20s_attach() error: potential null dereference 'state'.  (kzalloc returns null)
	drivers/media/dvb-frontends/mb86a20s.c:706 mb86a20s_attach() error: we previously assumed 'state' could be null (see line 705)

As, at mb86a20s_attach(), we have an i2c pointer, use it for all printk
messages there, instead of state->i2c.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-23 13:22:22 -02:00
Mauro Carvalho Chehab
f66d81b54d [media] mb86a20s: convert it to use dev_info/dev_err/dev_dbg
Instead of having its own set of macros, use the Kernel default
ones for debug, error and info.

While here, do some cleanup on the debug printk's.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 09:13:08 -02:00
Mauro Carvalho Chehab
dd4493ef34 [media] mb86a20s: Function reorder
Reorder functions to have everything related to stats/status read
close. That will make the file more organized as other stats
routines will be added.

No functional changes.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 08:53:11 -02:00
Mauro Carvalho Chehab
d36e418a7b [media] mb86a20s: Split status read logic from DVB callback
Split the logic that reads the status from the DVB callback. That
helps to properly return an error code, if status read fails.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 08:49:39 -02:00
Mauro Carvalho Chehab
04585921ac [media] mb86a20s: fix interleaving and FEC retrival
Get the proper bits from the TMCC table registers.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 08:17:38 -02:00
Mauro Carvalho Chehab
ce77d120ed [media] mb86a20s: make AGC work better
It is recommented to change register 0x0440 value to 0, in order
to fix some AGC bug.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 08:14:03 -02:00
Mauro Carvalho Chehab
fd53744efe [media] mb86a20s: Fix i2c gate on error
If an error happens, restore tuner I2C gate to the right
value.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 08:12:31 -02:00
Mauro Carvalho Chehab
a77cfcac79 [media] mb86a20s: improve error handling at get_frontend
The read/write errors are not handled well on get_frontend. Fix it,
by letting the frontend cached values to represent the DVB properties
that were successfully retrieved.
While here, use "c" for dtv_frontend_properties cache, instead of
"p", as this is more common.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2013-01-22 08:10:52 -02:00
Mauro Carvalho Chehab
9a0bf528b4 [media] move the dvb/frontends to drivers/media/dvb-frontends
Raise the DVB frontends one level up, as the intention is to remove
the drivers/media/dvb directory.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-08-13 23:13:41 -03:00