2.6.29.6: modules won't compile

Rerwin reported:
The problem is that a critical file is missing from the kernel source, causing any compile to fail. I first tried compiling dgcmodem, but then tried pctel, which I infer you have compiled, since it is in the sources-k2.6.29.6/3rdparty/modems directory.
And:
error: linux/bounds.h: No such file or directory

That's strange, as far as I can recall, the pctel modem just compiled, no problem.

I'm away from home right now. When I get home I'll check the PC that I compiled it on.

Note, I'm still not 100% settled on what "advanced" kernel will be used for 4.3. The 2.6.29.6 build was an experiment.

Hopefully for 4.3beta1 it will be settled.


Posted on 5 Aug 2009, 20:25


Comments:

Posted on 5 Aug 2009, 23:12 by Jota
4.3 wish-list
My wish-list for 4.3 (by priority):

- upgrade JWM to be able to resize from the top (and maybe with round-corners, why not?)
- kernel 2.6.30 (still faster and less resource intensive than 2.6.29 from pre-beta)
- flashblock in seamonkey
- upgrade to gnumeric 1.8.4
- some plugins in abiword



Posted on 6 Aug 2009, 21:22 by gposil
kernel
I don't know if you've done any more investigation of 2.6.27.x Barry, but the support is ongoing with some very interesting sidelines as regards performance compared to the 29 an 30 lines.


Posted on 7 Aug 2009, 8:10 by BarryK
2.6.27.x
Yes, it is too intriguing. It may be that the 2.6.27.x series is the best balance for Puppy 4.x. One probable advantage, some of the old analog modem drivers will compile (won't on 2.6.29 and later) which will make rerwin happy. It would bypass that pesky hid-* modules problem too!

I'll compile it today!