Welcome to GOFASTMOTORSPORTS.com ... ... ... Green Flag - Green Flag - Go Go Go ... ... Brought to you by Ricks Satellite - Where the Big Dish Rules!

Ricks Satellite Wildfeed and Backhaul Forum
Register Latest Topics
 
 
 


Reply
  Author   Comment   Page 1 of 3      1   2   3   Next
jowski

Avatar / Picture

Registered:
Posts: 1,324
Reply with quote  #1 
Does this patch make the s2-liplianin version of stv0900_core.c into one with Updatelee's and Pendragon/Cletus2k's patches?

I know it gives an error "missing header for context diff at line 3 of patch" but it patches.  If what I've attached is a good patch; what do we need to do with it?

Jim

 
Attached Files
txt stv0900_core.c.patch.txt (4.37 KB, 142 views)


__________________
Change the motto for the Office of Homeland Insecurity to
------"Omnia sed Realis Officia"------

majortom

Avatar / Picture

Registered:
Posts: 4,548
Reply with quote  #2 
Sounds like it, though I'm not hip to what your baseline was to create the patch, etc.
I had to go back to s2-liplianin Dec 2010 Prof 7500 IR remote fix, patched that with UDL's patch. Then I took the Pendragon/Cletus2k patch and pasted the entire Ts fifo  routine as posted here by Pendragon. I didn't create a new patch though using diff against the original that i downloaded from mercurial to compare to yours.

Once ya patch stv0900_core.c, recompile s2-liplianin,
make
make install
reboot
test the driver

if ya notice anything flakey, maybe try
make remove
then
make install
reboot 

It's working fine for me here as I described.


__________________
Perfect 10 7.5' mesh Chapparal C/Ku Co-Rotor feed, Norsat 8115 CBAND, Norsat 4106A Ku Thomson Saginaw Actuator, Vbox positioner,
Home Brew Polarotor683 Servo circuit. Openbox S9, Prof7500, v4l-updatelee linux drivers installed.
jowski

Avatar / Picture

Registered:
Posts: 1,324
Reply with quote  #3 
I took both Updatelee's and Cletus2k/Pendragon's work and patched s2-liplianin-0b7d3cc65161.  I'm sure the order of patching doesn't matter since they patch different sections of code; but I did Cletus2k/Pendragon's first.  It 'makes' just fine so I'm sure it'll install.  Actually, I installed a patch against s2-liplianin-4987f8a59d8a and everything went fine with 'make' and 'make install'.  The stv0900_core.c file is the same as the one in s2-liplianin-0b7d3cc65161 so it's not too surprising that it took the patch also.

Right now I don't have an stv0900 device to check it against.  I fried my Prof 7301 over the weekend and the replacement isn't due in until Friday.

Jim


__________________
Change the motto for the Office of Homeland Insecurity to
------"Omnia sed Realis Officia"------
majortom

Avatar / Picture

Registered:
Posts: 4,548
Reply with quote  #4 
bummer that sux. testing afterward I noticed it (prof 7500)  still didn't seem to behave correctly after installing the patch the first time, this go around. That's why I threw in the make remove comment just in case, cause that's what I wound up having to do here. Though I had never had a similar issue in the past re-installing compiled s2-liplianin. The reason I went back to Dec 2010 is cause patching with UDL, & Pendragon/Cletus2k patches against most recent available from repo wasn't working for me. So, I went back to what I knew had worked for me in the past.


__________________
Perfect 10 7.5' mesh Chapparal C/Ku Co-Rotor feed, Norsat 8115 CBAND, Norsat 4106A Ku Thomson Saginaw Actuator, Vbox positioner,
Home Brew Polarotor683 Servo circuit. Openbox S9, Prof7500, v4l-updatelee linux drivers installed.
jowski

Avatar / Picture

Registered:
Posts: 1,324
Reply with quote  #5 
I remade the patch correctly so it looks like other patches and hopefully won't give the context error message.  Here it is.

Maybe we'll hear from the experts and learn more.

Jim

 
Attached Files
txt correctly-made-stv0900_core.c.patch.txt (3.34 KB, 145 views)


__________________
Change the motto for the Office of Homeland Insecurity to
------"Omnia sed Realis Officia"------

signalquest

Registered:
Posts: 416
Reply with quote  #6 
What do these patches do? 

__________________
10' MESH w/ DMX241, and GBOX w/ Von Weise Actuator
1 meter KU w/ SG2100 rotor
PROF 7500 w/ Linux using XBMC, TVHeadend, blindscan-s2, and tune-s2
majortom

Avatar / Picture

Registered:
Posts: 4,548
Reply with quote  #7 
for updatelee's patch, explained here.
http://updatelee.blogspot.com/2010/09/prof-7500-linux-kernel-driver-mods.html

the pendragon/cletus2k patch fixes locking of high speed muxs



__________________
Perfect 10 7.5' mesh Chapparal C/Ku Co-Rotor feed, Norsat 8115 CBAND, Norsat 4106A Ku Thomson Saginaw Actuator, Vbox positioner,
Home Brew Polarotor683 Servo circuit. Openbox S9, Prof7500, v4l-updatelee linux drivers installed.
signalquest

Registered:
Posts: 416
Reply with quote  #8 
Thanks.  What is considered high speed muxes?

__________________
10' MESH w/ DMX241, and GBOX w/ Von Weise Actuator
1 meter KU w/ SG2100 rotor
PROF 7500 w/ Linux using XBMC, TVHeadend, blindscan-s2, and tune-s2
bluzee

Registered:
Posts: 2,171
Reply with quote  #9 
Giving this thread a bump so it's easier to find.  The stock kernel driver won't lock CBS feeds without this patch.  The TSFIFO_MANSPEED lines fix that issue.  Maybe other feeds too, I didn't test the stock driver beyond CBS.  I also remove the "/* Speeds up DVB-S searching */" patch that Igor added some time back.  Lines 115, 116, and 117 in the correctly made patch file. Doesn't seem to be necessary and with them gone you don't have to specify system DVB-S or DVB-S2 when tuning.  Very handy for old dvb apps that have not been updated to the new S2API.  Just give the correct frequency and the Prof card will tune what ever is there no matter the system. It's freaking awesome.

Thanks for the patches guys!

jowski

Avatar / Picture

Registered:
Posts: 1,324
Reply with quote  #10 
Here's my latest combined patch file.  It comments out the "/* Speeds up DVB-S searching */" patch.  That was the only change I made.  It can be found at hxxp://jowski.no-ip.com:3000/correctly-made-stv0900_core.c.patch.1.txt

Enjoy,
Jim

__________________
Change the motto for the Office of Homeland Insecurity to
------"Omnia sed Realis Officia"------
majortom

Avatar / Picture

Registered:
Posts: 4,548
Reply with quote  #11 
I think I finally understand why the s2-liplianin mercurial tree
doesn't compile under a linux 3.x environment. Seems they finally removed support for the old analog tuner cards (v4l). So some headers from the old v4l are missing.

Read this blog post by Devin Heitmueller, one of the LinuxTV developers.

http://www.kernellabs.com/blog/?p=1517

Am I understanding this correctly?




__________________
Perfect 10 7.5' mesh Chapparal C/Ku Co-Rotor feed, Norsat 8115 CBAND, Norsat 4106A Ku Thomson Saginaw Actuator, Vbox positioner,
Home Brew Polarotor683 Servo circuit. Openbox S9, Prof7500, v4l-updatelee linux drivers installed.
jowski

Avatar / Picture

Registered:
Posts: 1,324
Reply with quote  #12 
As a retired software tester (not the "watch DN for free kind" but aircraft and satellite communications) this looks like a classic case of the left hand and right hand not knowing what the other is doing.  One's saying, "hey, watch me" and the other is saying, "get lost I'm busy".  Well, I guess we know how that's working out.

Jim


__________________
Change the motto for the Office of Homeland Insecurity to
------"Omnia sed Realis Officia"------
bluzee

Registered:
Posts: 2,171
Reply with quote  #13 
They are saying that v4l was removed in 2.6.38 however s2-liplianin still compiles in 2.6.38 so I don't think that is the issue with it and the 3.0 kernels.  I'm not sure what the issue with the 3.0 kernel and s2-liplianin is.  I do know that stv0900_core.c is the same in s2-liplianin and the current linux-media tar ball so it can be patched the same way. Linux-media compiles fine for 3.0 kernels. 

It would still be nice if the authors  would send the patches here ... http://patchwork.linuxtv.org/.  Hopefully they would get added to the kernel and we could stop patching.
cletus2k

Avatar / Picture

Registered:
Posts: 277
Reply with quote  #14 
Quote:
Originally Posted by signalquest
Thanks.  What is considered high speed muxes?


Just going from memory here but I feel like anything over somewhere around 60Mbit wouldn't lock..  Anything below that the problem didn't manifest itself..

Easy test would be CBS on Galaxy28.. DVB-S2 8PSK 9/10 FEC has to be close to 80Mbit..

pendragon

Registered:
Posts: 982
Reply with quote  #15 
I've supported open-source projects for well over 30 years. There's a lot of good and a lot of bad. There are a lot of well-meaning people and a lot of despots. Unfortunately the chaos that breeds the positive aspects of open-source also makes it a complete backwater to well-accepted principles of modern software development. 

We have branches like s2-liplianin because the main Linux tree is so choked it's often not possible to keep it current. This leads to uncontrolled baseline merges where patches disappear and new bugs are created. This then breaks the branches. The development of the DVB APIs have been a cesspool of back stabbing and slapped together, backward-looking design concepts. It's hardly surprising what has resulted, and this makes Microsoft look like a well-machined operation.

As an author I have shepherded open-source fixes through gauntlets of politics, ignorance, plain stupidity and poor baseline control. I used to think this was a necessary evil, but now the effort to improve a piece of code is entirely overwhelmed by the effort to get it "blessed". It's not worth it to me anymore. For what it's worth, a number of people asked me if they could submit the high data rate patch described here. As is my policy, I always say yes. 

The problem with compiling s2-liplianin on Linux 3.x  is not API deprecation. It is death by a thousand cuts and a branch maintainer falling behind. A few months ago I started fixing each each little compile problem as it came up. Unfortunately this uncovered a number of design choices that have to be made, and as I am not one of the despots, anything I would do would be a wasted effort.

For the moment I would recommend building the DVB drivers as part of the main 3.x kernel tree. For the most part the tuners commonly used by this community are not well supported in either the main tree or s2-liplianin so you won't be missing much either way. Both have their share of unique and ugly bugs that may never be sorted out.

With regard to high data rate transponders, the patch discussed here is an absolute must for any of the STV0903-based tuners (e.g. Prof 7301, 7500 and 8000). However I have noticed several cases where transport stream blips happened even with it. So far this is only anecdotal, but within the last weak I scrubbed three simultaneous 80 Mbit/s recordings of the same TP using two Linux and one Windows tuner. Over several hours of recordings, the relatively small number of TS errors were not always at the same places. What bothers me is the CNR of the this TP was a solid 15 dB and did not appear to be varying. I can't find anything obviously wrong with the driver/application, so for the moment I'm mostly doing information gathering. If I find anything I'll let everyone know.
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Fellow Members, your posts are welcome here! Do not worry about posting everything perfect. Different receivers and LNB's will give you different Frequencies and Symbol Rates. Some set top boxes, PCI cards and USB receivers, Do Not Require all of the same information that others may need. It is not Required to post everything that others may need to tune in a feed. It is just most important to share the find. We can always adjust the Frequency and Symbol Rates and try the various Modulations and FEC's on our own receivers until we get a lock and then give a polite reply with what works for your receiver, as that information might help others as well. We all appreciate the efforts and energy of the Posters!

Thank You for Visiting GOFASTMOTORSPORTS.com - Keep Your Eyes on the Sky and the Track!