| 1 |
On Monday 29 September 2008, Jon Hardcastle wrote: |
| 2 |
> > > > On Monday 29 September 2008, Jon Hardcastle |
| 3 |
> > |
| 4 |
> > wrote: |
| 5 |
> > > > > Hi, i have spent the weekend trying to get |
| 6 |
> > |
| 7 |
> > my new |
| 8 |
> > |
| 9 |
> > > > Hauppage USB tv stick to |
| 10 |
> > > > |
| 11 |
> > > > > work under linux. |
| 12 |
> > > > > |
| 13 |
> > > > > I have been constantly perplexed by |
| 14 |
> > |
| 15 |
> > references to |
| 16 |
> > |
| 17 |
> > > > compilable kernel modules |
| 18 |
> > > > |
| 19 |
> > > > > I couldn't see! I assumed it was because |
| 20 |
> > |
| 21 |
> > they |
| 22 |
> > |
| 23 |
> > > > referred to old modules or |
| 24 |
> > > > |
| 25 |
> > > > > something. |
| 26 |
> > > > > |
| 27 |
> > > > > But having downloaded and booted into |
| 28 |
> > |
| 29 |
> > Knoppix and run |
| 30 |
> > |
| 31 |
> > > > make menuconfig on |
| 32 |
> > > > |
| 33 |
> > > > > that i have confirmed that kernel seems to |
| 34 |
> > |
| 35 |
> > have |
| 36 |
> > |
| 37 |
> > > > 10's of TV card chipset |
| 38 |
> > > > |
| 39 |
> > > > > drivers availible to compile whereas |
| 40 |
> > > > |
| 41 |
> > > > gentoo-sources/vanilla-sources emerged |
| 42 |
> > > > |
| 43 |
> > > > > on my machine have maybe 2 or 3. |
| 44 |
> > > > > |
| 45 |
> > > > > Can someone offer me some assistance as to |
| 46 |
> > |
| 47 |
> > why this |
| 48 |
> > |
| 49 |
> > > > might be? |
| 50 |
> > > > |
| 51 |
> > > > > Thank you |
| 52 |
> > > > > |
| 53 |
> > > > > (This is my first mailing list post - please |
| 54 |
> > |
| 55 |
> > be |
| 56 |
> > |
| 57 |
> > > > gentle!) |
| 58 |
> > > > |
| 59 |
> > > > |
| 60 |
> > > > because a large part of the dvb/tv drivers are |
| 61 |
> > |
| 62 |
> > developed |
| 63 |
> > |
| 64 |
> > > > outside of the main |
| 65 |
> > > > kernel tree. Also a large part is hidden under |
| 66 |
> > > > experimental. |
| 67 |
> > > > |
| 68 |
> > > > For my dvb-t stick I need to checkout a mercurial |
| 69 |
> > |
| 70 |
> > rep and |
| 71 |
> > |
| 72 |
> > > > do a make, make |
| 73 |
> > > > install in it do get drivers that work. And with |
| 74 |
> > > > 'work' I mean: no sound on |
| 75 |
> > > > first try, but after disconnecting the stick and |
| 76 |
> > > > reconnecting it, it suddenly |
| 77 |
> > > > works. |
| 78 |
> > > |
| 79 |
> > > Hi, I went thru this process but i still dont have the |
| 80 |
> > |
| 81 |
> > option to add |
| 82 |
> > |
| 83 |
> > > support as a kernel module. Should the mercurial add |
| 84 |
> > |
| 85 |
> > steps to the |
| 86 |
> > |
| 87 |
> > > menuconfig? |
| 88 |
> > |
| 89 |
> > no. |
| 90 |
> > it just installs all the drivers in the rep as modules. It |
| 91 |
> > doesn't touch any |
| 92 |
> > configs at all. |
| 93 |
> |
| 94 |
> Hmmm i wonder why i have seen sooo many references to alot more kernel |
| 95 |
> options than i have in either my gentoo-sources or vanilla-sources. |
| 96 |
|
| 97 |
because they patch their kernel. |
| 98 |
|
| 99 |
> Is it |
| 100 |
> possible to 'patch' the kernel source instead of downloading using |
| 101 |
> mercurial. Also how does it 'know' what kernel to make the modules |
| 102 |
> available under? I have 3 sets of kernel code on my machine? |
| 103 |
|
| 104 |
a) /usr/src/linux |
| 105 |
b) uname -r |
| 106 |
c) just make clean && make&& make install for every kernel. |
| 107 |
|
| 108 |
Why patch at all? with the rep and make&& make install you get all the |
| 109 |
drivers. It is not that much 'overhead' compared to the work of patching the |
| 110 |
kernel. And you don't even need to worry about which driver you need - |
| 111 |
autoloading will do that for you ;) |
Archives