kernel suspend2 and CK2 patching

[MikeCindi]

I have really enjoyed the perfomance boost that the 2.6.21-ck2 kernel has given VL-SOHO on my laptop. I then finally got around to patching the "default" kernel for SOHO (2.6.20.3) with suspend2 and all went well. But I was unable to patch the CK2 patched kernel with suspend2. Several "hunks" gave errors. Then I tried a custom kernel for my laptop using the latest 2.6.22.1 and all went well but was unable to get suspend2 to work with it (the patch applied okay but it was never functional after that). Thus I wonder if what I'm trying to do is reasonable. Any suggestions?
Mike

[Vxt]

Hi,  M-C

I have never tried the suspend patch - (no laptop)
Have you considered the newest (I.Molnar)  "Fair-Schedulng"  version ?

Suggestion:

Use alternate kernel versions (in loader) to assess B4 committing.

Best of luck



 

[MikeCindi]

I recently reworked the 2.6.22.1 kernel to use the "built-in" suspend option instead of TuxOnIce (suspend2). That works very well for me. It does not allow for "hibernation" (suspend to disk) or at least I've not figured that one out yet. Under WinXP I use both (suspend to RAM and disk) often. Thus currently I have 3 kernels available to me the "vanilla" 2.6.20.7 (updated 2.6.20.3 that comes default with 5.8 SOHO), the 2.6.21-ck2 enabled kernel (the improvement in responsiveness on my laptop is amazing over the other 2), and my "customized" 2.6.22.1.

In the near future I will try to apply the ck2 patch to my 2.6.22.1 kernel as well as attempting the suspend2 patch again. Most of the other "restraints" from using VL all the time have been alleviated by running Windows in either VirtualBox or VMware. (My biggest reason for not wanting to dual boot anymore is the significantly better performance of Audacity under VL vs. XP. a 60 minute speech file exported to mp3 takes about 7 minutes in VL and 17 minutes in XP. Often I don't have that 10 minutes to wait due to battery issues...leading to the desire to have suspend capabilities. I just know that everyone else wanted to know this. )

[MikeCindi]

I started the patch process on my 2.6.22.1 kernel and got the following:

Hunk #1 FAILED at 1.
1 out of 1 hunk FAILED -- saving rejects to file Makefile.rej

This is the content of Makefile.rej

Code:

***************
*** 1,8 ****
  VERSION = 2
  PATCHLEVEL = 6
  SUBLEVEL = 22
- EXTRAVERSION =
- NAME = Holy Dancing Manatees, Batman!
 
  # *DOCUMENTATION*
  # To see a list of typical targets execute "make help"
--- 1,9 ----
  VERSION = 2
  PATCHLEVEL = 6
  SUBLEVEL = 22
+ EXTRAVERSION = -ck1
+ NAME = So long, and thanks for all the fish
+ JANAME = �よ��ら�����魚を�り���
 
  # *DOCUMENTATION*
  # To see a list of typical targets execute "make help"

Is this their way of saying "goodbye"?
I went ahead and re-built the kernel.

BTW: the patch seems to have worked well but I'm unsure of how to determine this for sure. Subjectively it seems to be "improved" and the "built in" suspend to RAM seems to work.

EDIT: And finally, I tried to patch the kernel with the TuxOnIce (suspend2) but there were some "failed hunks" that would require editing one of the kernel sources.  After looking that over it seemed to be more complicated than my abilities at the moment. I then tried applying the TOI patch to a "vanilla" 2.6.22.1 kernel which worked well but several hunks failed when applying the CK2 patch. Thus the choice for me is: do I want suspend to disk ability without the enhancement of the CK2 patch or the performance boost?