patches & bugfixes howto, branch discussion

Beitrag von **obi** » Donnerstag 25. September 2003, 14:12

if you made a patch containing a bugfix, an update or an enhancement for the cross development kit (i$cvsroot/cdk/*), then please post it in this forum.

i don't know if there is much interest in maintaining the rel_1_0_0 cdk, most people i know use HEAD cdk even with apps and drivers from rel_1_0_0. anybody volunteering for maintenance? should HEAD cdk/* be merged into rel_1_0_0 or shall a new rel* branch be created from HEAD cdk and rel_1_0_0 apps & drivers? We have to get rid of old archives on a regular basis because of the user quota on berlios' web server, which always breaks rel_1_0_0 cdk downloads because it is currently kind of unmaintained...

when you create patches then please use the diff option -u. this is important because otherwise patches are really unreadable.

regards,
obi

p.s.: i wrote this in english because most people reporting bugs for rel_1_0_0 don't seem to know german very well. understanding and working with the cdk requires at least basic knowledge of the english language anyway... however if anyone wants to translate it then feel free to do so.

Hi obi,

i don't know if there is much interest in maintaining the rel_1_0_0 cdk, most people i know use HEAD cdk even with apps and drivers from rel_1_0_0. anybody volunteering for maintenance? should HEAD cdk/* be merged into rel_1_0_0 or shall a new rel* branch be created from HEAD cdk and rel_1_0_0 apps & drivers?

allow me to comment on this one. please be so kind and leave at least one stable REL branch available. i cant talk about enigma and lcars, but infact of neutrino there is use for the old drivers, because neutrino head is unable to receive some channels. i am not talking about the satscan, which does not find all infact of wrong NITs by some sat providers (eutelsat, sirius). i am talking about channels, which used to work fine using the rel branch, but give a "kanal nicht verfuegbar" using head, and this on the same box (here, nokia avia600, no problems with SCPC).

to give you an example of what i am talking about (im just going to nail a few things down, creating a services.xml manualy). i did not check channels using levels2 encryption, or some other stuff the d-box2 is not allowed to handle either. i am mainly talking about FTA chanels.

so, at 19.2E everythjing is fine. all FTA and all Betacrypt channels work.

i did not nail down sirius at 5E yet, but stay tuned.

now an easy one, plenty people are able to proove, hotbird at 13E:



the above used to work using REL. and this is not even SCPC.



the above to work with REL as well.

now there are some others, which sometimes also use tid=1 and onid=ffff. if this happened in my research, i commented them out, to make sure there duplicate IDs are not the problem.

so, here are some more, which dont work:









i am unsure now, wich channels worked with REL and wich not, but i am sure TRT INT and EDTV/DUBAI did. at least on my nokia.

also polsat/polsat2 did (i agree the signal is weak in this case, but i had no problems with my 1meter dish).

so as you can see, there still is a real problem with HEAD (the only serious problem left, IMHO). if this problem is infact of zapit or the drivers, i am unable to tell.

so it would just be nice, if a stable REL version (name it whatever you like) would be kept.

thanks,
kerlimann

btw, about SCPC:

<transponder id="0ed9" onid="013e" frequency="11131000" inversion="0" symbol_rate="5632000" fec_inner="3" polarization="1">
<channel service_id="1389" name="BBC PRIME [E]" service_type="01"/>
<channel service_id="138a" name="BBC RUSSIA [R]" service_type="02"/>
<channel service_id="138b" name="BBC ENGLISH [E]" service_type="02"/>
<channel service_id="138c" name="BBC ARABIC [A]" service_type="02"/>
<channel service_id="138d" name="BBC PERSIAN [FA]" service_type="02"/>
</transponder>

the above DOES work, just to make sure, SCPC cant be the problem (IMHO).

Beitrag von **obi** » Donnerstag 9. Oktober 2003, 18:30

you completely misunderstood what i wrote: it's all about cdk/*, not apps/* and not driver/*

obi hat geschrieben:you completely misunderstood what i wrote: it's all about cdk/*, not apps/* and not driver/*

oops, in this case, sorry. the subject of this thread includes "branch discussion" (together with 2 other topics).

one option you gave was, "if HEAD should be merged into REL". the other option was what you describe above now (apps, drivers are kept). i just wanted to point out, that i would prefer the second choice. if this is not what you asked for, sorry.

was ist eigentlich daraus geworden

Beitrag von **woglinde** » Samstag 9. Oktober 2004, 22:44

oehm.

keiner hat mehr so richig zeit und es laeuft alles wie bisher. *g*

Gruss woglinde

woglinde hat geschrieben: keiner hat mehr so richig zeit und es laeuft alles wie bisher. *g*

naja, fuer kernel updates scheint die zeit ja da zu sein. ob das so relevant ist.. <ducking>

tuxbox-devel@lists.berlios.de geht nicht (mehr).

Ciao Micha

Beitrag von **mogway** » Dienstag 24. Mai 2005, 21:23

michaelstaehle hat geschrieben:tuxbox-devel@lists.berlios.de geht nicht (mehr).

Take a look at: http://cvs.tuxbox-cvs.sourceforge.net/

Gruß
mogway

Tuxbox-Board

patches & bugfixes howto, branch discussion

patches & bugfixes howto, branch discussion

Re: patches & bugfixes howto, branch discussion