Re: Activity!

Hi.

On 15/12/11 09:23, Harald Judt wrote:
> Am 14.12.2011 22:15, schrieb Piotr Karbowski:
>> On 14.12.2011 13:34, Nigel Cunningham wrote:
>>> Hi all.
>>>
>>> Some of you may have already noticed that I've pushed an update to
>>> git.tuxonice.net. Please be weary of using it at the moment - I'm still
>>> testing and debugging, and there are a couple of changes I want to look
>>> at more closely. I hope to have more time to work on this tomorrow
>>> night. Failing that, it should happen the day after because I'm back to
>>> 4 days a week at work now (glad of that - I can get other things done
>>> again!).
>>>
>>> Regards,
>>>
>>> Nigel
>>
>> Glad to hear that. I am really happy that tuxonice is not dead project!
>> I tried the in-kernel suspend and it was horrible, random freezing
>> system, sometimes it frozen while suspending, sometimes on resuming.
>> With Tuxonice I don't remember one single freeze since a year or more!
>>
>> -- Piotr.
> 
> Yeah, great news! I am a long-time tuxonice user. However, I've been
> using in-kernel suspend since 3.2-rc releases now, in lack of updates
> and because I didn't have the time to make tuxonice work. Here are my
> experiences with in-kernel suspend:
> 
> * Does not work on an older pentium4 with only 3gib ram at all, maybe
> because of additional hardware connected/installed. That was some time
> ago, I could retry and see how good it got in the meanwhile. Tuxonice
> works fine though.
> 
> * Works without issues on up-to-date hardware with plenty of ram
> (16gib), not failing once.
> 
> * Sometimes does not hibernate on a laptop with 4gib ram. Could be
> related to driver or rc status though.
> 
> * It seems to me that it needs longer to start writing the image, but
> writing the image is quite fast.
> 
> * No progress display during hibernation. Output at resuming is ok
> though, not as nice as tuxonice-userui, but good enough for me.
> 
> * Image size and extra pages can be configured now. This is a necessity,
> especially for the machine with 16gib ram. I don't want to reserve
> 6-8gib swap which is mostly unused except for hibernating.
> 
> * Not an option for machines that don't have a lot of memory. there it
> starts to swap to free enough memory which sucks a lot.
> 
> * When the machine crashes during hibernation, all you see is a black
> screen. Really great for debugging.
> 
> 
> Here are some drawbacks of tuxonice:
> 
> * Cannot apply Con Koliva's bfs patch, they break compression. This
> really sucks. I've applied bfs and bfq patches to mainline head, and
> they are quite an improvement, especially on the laptop. The only
> solution would be to not use image compression which is not an option.

I'll see what I can do about that.

> * Laptop special keys (thinkvantage button, fn keys) stopped working
> after hibernating and resuming (lenovo t400). They work fine with
> in-kernel suspend, in fact they start working again when suspending (not
> hibernating) and resuming. (Could have been fixed by newer kernel, not a
> tuxonice issue? On the other hand there have been no changes to the
> module for quite some time IIRC.)

Does current git do that? It sounds like I'm not doing something I
should be doing, or doing it in the wrong order.
> 
> Advantages of tuxonice:
> 
> * Always stable and reliable wherever I tested it. When it didn't,
> in-kernel suspend would not work either.
> 
> * Can write images greater than 1/2 of total memory.
> 
> * Fast and configurable.
> 
> * Debugging output.
> 
> * Nice progress display.

Broken for me at the moment - another thing to look at :)

> * I may decide to reboot while writing the image. What a convenience! I
> use that very often when dual-booting.
> 
> * Hybrid mode (hibernation and suspend).
> 
> * You could use fbsplash and get a really nice tuxonice theme. I did
> never spend the time to set that up.
> 
> * Great community!
> 
> * Great support!
> 
> 
> All in all, the kernel devs did a great job with in-kernel suspend, but
> it's still not quite on a par with tuxonice in some important aspects.
> 
> I wish you fun with hacking and thanks for your great support so far,
> Nigel.

Happy to be of service!

Nigel
-- 
Evolution (n): A hypothetical process whereby improbable
events occur with alarming frequency, order arises from chaos, and
no one is given credit.