Hi,

To pick up this thread once again...


There was one major problem with the patch mentioned earlier. The isochronous setup code is commented out (I guess since it uses the deprecated
interface.)

I have been doing some additional work on the driver the last weeks. Apart from the writing new iso setup code I have added CI descrambling support
and some CI MMI support. I have created a patch at  http://firesat.kurelid.se/firesat-ci-2.6.24.3-v3.patch with my additions. (It is to be applied
after the patch at http://www.kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/patches/bad/ldp/dvb-ad...).


I will gladly lend a hand there as well.

subsystem based firesat driver, the circle of persons who can do a  conversion to the firewire subsystem is bigger.  I for one might be able  to
spend some time on such activities from October onwards or so.
I agree. Let's not "fix it if it ain't broken" yet. I have some time to spare the coming months so if there are no other plans on the driver before


Regards,
Henrik Kurelid


_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Hello

I was eventually to make the ported driver work (please see my posting
from Thu Apr 24 02:24:53 PDT 2008):

"return remap_page_range(vma, vma->vm_start, vma->vm_pgoff <<

PAGE_SHIFT,vma->vm_end - vma->vm_start, vma->vm_page_prot);"
needs to be changed to

"return remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff,
vma->vm_end - vma->vm_start, vma->vm_page_prot);"


rather than

"return remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff << PAGE_SHIFT,
vma->vm_end - vma->vm_start, vma->vm_page_prot);"

remap_pfn_range is the shift operation by PAGE_SHIFT already build in.


Best regards
Tilman
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Hi,

I'd like to help. I'm looking for a "first" project.
Something that may need, for example, only the
cleaning up of sources, correcting typos, reformatting
code to currently accepted standards, etc. I've
recently revised ANSI C and am currently self-training
in ANSI C++.

One possible (!) problem is that I can only access the
web through computers at my local public library.
However, I can usually manage to get a machine,
especially in the evenings. I just thought that I had
better mention this in case it is a real (!) problem.

Cheers,

Stuart



      ___________________________________________________________ 
Want ideas for reducing your carbon footprint? Visit Yahoo! For Good  http://uk.promotions.yahoo.com/forgood/environment.html
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Just a side note:  Try to convert to regular coding style when yo do
functional or merely formatting changes.  One thing that sticks out is


Don't use // comments, except perhaps in hacks that will go away in
another iteration before review requests, let alone merge requests.
Instead use the styles

	/*
	 * canonical multiline
	 * comment style
	 */

	/* single line comment */

#if 0
	temporarily_disabled_code;
	even_more_disabled_code;
#endif

/**
 * kerneldoc comment of subsystem API elements
 */


linux/scripts/checkpatch.pl will also reveal a few other formatting nits

Hmm, wouldn't plain and simple u8 instead of __u8 be appropriate here?
But more importantly, these most certainly should not be bitfields. From
what I have been told, bitfields aren't as strictly specified in the C
language spec as required for on-the-wire data (and for userspace ABIs
for example).

I.e. ctype/cts and suid/sutyp in the above example would need to be
collapsed into a single struct member.  If there are several places in
the code accessing these bifields, then you could supply inline
functions (or second choice: preprocessor macros) as accessors.


These definitions need to be replaced eventually.  Nowadays we typically
define such data structures in the endianess of the hardware (I suppose
big endian), with sparse-annotated types if available, and use accessors
like cpu_to_beXX and so on to write and read such data.

I'm not saying that /you/ should do all these cleanups; but you
definitely should for example clean up comment style at places where you
cleaned up 80 column line lengths and the likes.

Thanks for keeping the ball rolling,
-- 
Stefan Richter
-=====-==--- =--- ----=
http://arcgraph.de/sr/
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

The email address on the last line is certainly Philipp's and possibly
no longer valid.


...in:
drivers/media/dvb/firesat/cmp.c
drivers/media/dvb/firesat/firesat_dvb.c
drivers/media/dvb/firesat/firesat_fe.c
drivers/media/dvb/firesat/firesat.h

Christian and Ben, do you know who holds copyright to those?

Henrik, are you the sole copyright holder of
drivers/media/dvb/firesat/firesat-ci.c?
-- 
Stefan Richter
-=====-==--- =--- ----=
http://arcgraph.de/sr/
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Hi,

I created the initial version of the driver, so the following files  
should also have my copyright:

avc_api.c
avc_api.h
cmp.c
cmp.h
firesat-ci.c
firesat-ci.h
firesat-rc.c
firesat-rc.h
firesat.c
firesat.h

unless any of them has been removed or completely rewritten since then  
(it's hard to follow development, since I don't know if you're using  
any repository, or are just passing patches around).
The version in v4l-dvb mercurial (v4l_experimental/firesat) is the one  
where I stopped development.

andy

_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Philipps e-mail address is not active anymore, this was his e-mail at
As I know most of the code is written by Andreas Monitzer, so I think
his name should be listed here.

Regards,
Christian



_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

I now took the "current" version of dvb-add-firesat-driver.patch from 
http://git.kernel.org/?p=linux/kernel/git/gregkh/staging.git;a=blob;f=dvb-add-firesat-...
(this is the version you used plus device-recognition.patch from 
http://firedtv.bbackx.com/index.php?q=node/5 ), also applied Ben's other 
patch for DVB-S on DVB-S2 devices, and then applied your patch.  I got 
two rejects which looked trivial.  I also removed all trailing 
whitespace from your patch.  The result is temporarily available here:
http://user.in-berlin.de/~s5r6/linux1394/firedtv/

Henrik, could you please test if this version still works?

Also, please look at the changelog of 
04-firesat-update-isochronous-interface-add-ci-support.patch and send me 
a Signed-off-by line for it if everything is OK.  Also remember to add a 
good changelog and your Signed-off-by (with the implications as stated 
in linux/Documentation/SubmittingPatches) when you send additional patches.

Thanks,
-- 
Stefan Richter
-=====-==--- =--- -==--
http://arcgraph.de/sr/
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Hi,

Back from a week of vacation, I have tried the patch with a vanilla 2.6.26 and it seems to work just fine.

Regards,
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]

Great, have you submitted the driver for inclusion in the kernel tree?

thanks,

greg k-h
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
[ message continues ]