Talk Of A "Massive Power Regression" In Linux 3.5

[madjr]

If it doesn't work for you, just don't bloody use it and shut up, or file a bug report and shut up.

haha zealots, you just proved my point once again.

and I have been filling out stupid bugs since 2005 and is always the same regression shit every other release.

like i said, open code, closed zealot minds.

----------

"Insanity: doing the same thing over and over again and expecting different results."
-- Albert Einstein

well thats always been linux motto is free so shut the fuck up if it doesnt work for u :go back to windows.or fixit urself do it urself bla bla. with this kind of atitude and thinking and reacting never in this lifetime linux will win the desktop!
And most users use a pirated copy of windows except US and those preinstalled laptops, so windows is also free. People stick to linux because they like it not because its free, and the power regressions are not even that much of a deal, graphic driver on linux are in a poor shape. I used linux since like 2003 but i dont know why back then it felt much more stable and much more better tested but i also think the hardware might have been much more simple.

yes I agree.

[bug77]

haha zealots, you just proved my point once again.

and I have been filling out stupid bugs since 2005 and is always the same regression shit every other release.

As opposed to which OS that doesn't need frequent patching?
Desides trolling, do you seriously expect the kernel maintainres to have a barn full of computers ranging from Pentium II to Ivy Bridge and Piledriver CPUs, from Voodoo graphics to quad-SLI nvidia 680s and Radeon 6970? Including Atom, various ARM incarnations, chipsets from SiS, ALi and VIA? Even if hardware manufacturers themselves would test each kernel, bugs would still slip through.
The world is what it is. If think the world should should change because it annoys you, maybe you should rethink some stuff. Just sayin'...

[jrch2k8]

As opposed to which OS that doesn't need frequent patching?
Desides trolling, do you seriously expect the kernel maintainres to have a barn full of computers ranging from Pentium II to Ivy Bridge and Piledriver CPUs, from Voodoo graphics to quad-SLI nvidia 680s and Radeon 6970? Including Atom, various ARM incarnations, chipsets from SiS, ALi and VIA? Even if hardware manufacturers themselves would test each kernel, bugs would still slip through.
The world is what it is. If think the world should should change because it annoys you, maybe you should rethink some stuff. Just sayin'...

don't bother anymore is ovbious after few posts he is a cave troll and will bash linux no matter what we say and since his windows 7 [alien edition cuz that thing he describe is not sell here on earth] is so awesome let him enjoy it, no feeds the trolls policy activation notice

[madjr]

As opposed to which OS that doesn't need frequent patching?
Desides trolling, do you seriously expect the kernel maintainres to have a barn full of computers ranging from Pentium II to Ivy Bridge and Piledriver CPUs, from Voodoo graphics to quad-SLI nvidia 680s and Radeon 6970? Including Atom, various ARM incarnations, chipsets from SiS, ALi and VIA? Even if hardware manufacturers themselves would test each kernel, bugs would still slip through.
The world is what it is. If think the world should should change because it annoys you, maybe you should rethink some stuff. Just sayin'...

I don't expect that, but I brought up the topic of better communication, collaboration and coordination and the only reply I got from a few zealots/douches was : Shut The Fuck Up.

I mean If no one wants to collaborate/coordinate and kernel devs are not expected to QA but just release to the wild (often with regressions), and is a big job solely for the distros, then Is not surprising that they could be tempted to just go and fork it.

It seems we are against forking/fragmenting (and we should be, because is mostly duplication) , but when you're talking to deaf ears this is the result.

And no am not a troll, I have strong faith on linux and have been testing hardware, donating, filling out bugs and helping thousands of new users with their technical problems since 2005, but most of the same issues seem to always come up again and again because hardly anyone QAs. Maybe am just very burned out.

Maybe the zealots are right, nothing will change so maybe I (and probably other people who want to see linux succeed) are just wasting their time here.

[Teho]

Maybe the zealots are right, nothing will change so maybe I (and probably other people who want to see linux succeed) are just wasting their time here.

Well considering the fact that Linux is easily the most successful kernel in the existence shows that it's doing pretty fine. However software that is developed by over 300 companies and gets patches from over 3000 people every year is not exactly trivial to manage. Of course sometimes things break but you can't delay the release or even revert the patch as it might cause more unknown problems. Then after the release you can fix the problem in the next bugfix release that might come out in a week or two or in the next major release in 60 days. Even if stuff like this slips through it's not the end of the world. Whining about the lack of "collaboration", "communication" and "coordination" on Phoronix forums simply looks silly to me. It's not like these problems are not acknowledged.

[jrch2k8]

I don't expect that, but I brought up the topic of better communication, collaboration and coordination and the only reply I got from a few zealots/douches was : Shut The Fuck Up.

I mean If no one wants to collaborate/coordinate and kernel devs are not expected to QA but just release to the wild (often with regressions), and is a big job solely for the distros, then Is not surprising that they could be tempted to just go and fork it.

It seems we are against forking/fragmenting (and we should be, because is mostly duplication) , but when you're talking to deaf ears this is the result.

And no am not a troll, I have strong faith on linux and have been testing hardware, donating, filling out bugs and helping thousands of new users with their technical problems since 2005, but most of the same issues seem to always come up again and again because hardly anyone QAs. Maybe am just very burned out.

Maybe the zealots are right, nothing will change so maybe I (and probably other people who want to see linux succeed) are just wasting their time here.

1.) only from a couple of guys [quite low for phoronix btw]
2.) no one will fork linux that is just insane[technically speaking] [maybe they'll choose bsd[flavor]]
2a.) you seem to mistake what a regression is but a regression is not a kernel dev making a booboo and then puting an .i. in the C header file asking the user to fix it nor is a pseudo commie dev from the URSS hiding from the KGB wishing to screw users and the rest of lazy devs didn't crosscheck the code. a regression[depending the level] is a massively circular complex issues that could affect an entire subsystem without any reason or explanation due to a random combination of factors[hardware 99% of the time] that must likely can only be studied or detected when this random set events coincide in a particular system[99% of the time ACPI is the guilty bastard]
2b.) ACPI is an ISO standard that is awesome to print and use it to clean your ass from the bios/motherboard vendors POV, i mean none of them respect or follow the acpi standard at all and to be worse they introduce some variants outside the standard in the more hackish possible way in every series/plataform they can, in fact sometimes motherboard introduce alien acpi code between revisions or more creepy through bios updates of the motherboards and to put the tip of the iceberg they try to detect the running OS and depending of the OS the deliver an alternate table spec[most likely 100% broken or pointing to false registers] and guess which table is the right one?? yes the windows[latest] and regularly the linux table is totally broken[note that microsfot and many vendors have actual litigations on this] and if you add the fact most laptop vendor modify this uber bugged bios and quite commonly broke them more[i have seen tons of hp/dell laptops that can't be upgraded to win7 from vista cuz the bios don't allow it and if you call support they tell you to reinstall vista from restore cuz win7 is not supported or buy the model XXXY rev XY and i have many other from HP that broke if you install vista oem(aka no suspend or low power setting or backlight and you have to go back to the bundled crapware infested hp vista)], so if you actually check the acpi code detection in the kernel you will cry cuz is probably the most insanely complex code ever written[but is very good in finding most of the good settings inside that storm of bugs]
3.) if you ever ask why apple make their own hardware see 2b [probably today they have enough money to litigate this but not sure if this would be of worth to them since their stuff sells quite good]
4.) kernel ppl are not zealots they just get angry when you loop asking for stuff you don't bother to investigate or understand especially in this areas where everyhting is done with reverse engeniiering and witchcraft cuz vendors do everything possible to stop them [ami/phoenix bios and some OEM are well known to do this cuz microsoft force them <-- litigations open too], so yes is easy to blame the devs cuz they couldnt bypass 100% of the traps in that mess of bios your mobo has from the vendor you choose wich do everything it can to make linux unusable or draw power like a GTO
5.) check how hardware PM works in wikipedia and you will see why ACPI is so important[and hence why microsoft has such a tight grip on bios vendors] then come and whine.
6.) you wanna help linux? you wanna see it mainstream? you want an awesome out of the box experience? you wanna see a wonderful stable kernel? then educate yourself buy hardware only from good know bios providers[or openboot supporter suppliers] and ask your current politicians to force bios company to follow the freaking ISO standard like they should ask very vocally to your vendor to follow the standard[boycott them in forums] or if you are rich sue them to death or of you study laws try a class action suit against those vendors cuz if they provide the ISO standard like they should linux problem/regressions/[you name it] will drastically get close to 0[linux support ISO ACPI and as good as they are allowed the frankestein version of your favorite bios vendor].

zealot: [technical savvy that bother in research the deep reason of the issue disscussed] tend to get angry and vocal like some of the posters cuz some guy come to whine and demand about a thing he don't even have a freaking clue about cuz the troll[<-- by the book definition] feels the zealot should for some reason tell him he is right and find a miracle solution out of his ass cuz the troll assume the zealot is a lazy ass that only have to press a big red button to fix all evil in universe but he just like to make the troll suffer but the troll in his arrogance assume he is the very first troll ever in ask the same idiocy but never contemplate the fact that millions of troll do it before so the zealot is about to get a heart attack from the 3 million troll asking the same crap over and over again knowing that the troll [against any reason] will never accept it and ever ever ever will pull a finger or let the discussion go in a way that can provide useful information/practice/testing/idea to fix the issue at hand

[adriankx]

wanna hear something funny? my cool thinkpad edge13 amd box has a 2% battery capacity so i cant really notice power regression, and i also think those cocksuckers from lenovo suck at making amd machines. My laptops has same bugs since day 1 (broken ACPI_PSS and needs DSTS override)and never been fixed even after posting and mailing their support site. I was told i run an unssuported operating system. How ever funny suspend and resume works better on linux with OSS drivers that on windows where my system randomly awakes without being touched. So yes all software closed source is crap, but my hopes where that OSS developers listen and we are all a big happy family communicating all toghether and so on. But it seems linux is developed by a few server minded-orientated companies (redhat novell and others) those guys dont take suggestions. Linux sells u this things where u can report bugs get involved bla bla but it rarely ever happens, all they need is a free beta testers userbase for they entreprise editions.
It might have started using the right ideeas but in the end corporations use the comunity to make profit.

What we nuub users want beutiful 2d accel tear free desktops great open and closed source graphic drivers will never happen unless all mighty corps i mentioned think they can get a profit from desktop market. In the end all mighty dollar rullz everything. We can just full ourselfs with freedom there is no such thing and it will never be.

[bug77]

I don't expect that, but I brought up the topic of better communication, collaboration and coordination ...

Why? Is any of the persons than can answer/act upon that watching these forums?
I think you got exactly what you usually get for being off-topic.

[madjr]

zealot: [technical savvy that bother in research the deep reason of the issue disscussed] tend to get angry and vocal like some of the posters cuz some guy come to whine and demand about a thing he don't even have a freaking clue about cuz the troll[<-- by the book definition] feels the zealot should for some reason tell him he is right and find a miracle solution out of his ass cuz the troll assume the zealot is a lazy ass that only have to press a big red button to fix all evil in universe but he just like to make the troll suffer but the troll in his arrogance assume he is the very first troll ever in ask the same idiocy but never contemplate the fact that millions of troll do it before so the zealot is about to get a heart attack from the 3 million troll asking the same crap over and over again knowing that the troll [against any reason] will never accept it and ever ever ever will pull a finger or let the discussion go in a way that can provide useful information/practice/testing/idea to fix the issue at hand

sorry your zealot definition is wrong and no I didn't call the kernel devs zealots, but the douches around here.

anyway I appreciate the other more technical part of your response.

Well considering the fact that Linux is easily the most successful kernel in the existence shows that it's doing pretty fine. However software that is developed by over 300 companies and gets patches from over 3000 people every year is not exactly trivial to manage. Of course sometimes things break but you can't delay the release or even revert the patch as it might cause more unknown problems. Then after the release you can fix the problem in the next bugfix release that might come out in a week or two or in the next major release in 60 days. Even if stuff like this slips through it's not the end of the world. Whining about the lack of "collaboration", "communication" and "coordination" on Phoronix forums simply looks silly to me. It's not like these problems are not acknowledged.

Sorry I was talking more about the "linux-desktop" as a whole, more than the linux kernel.

Anyway thanks for the mature response, you're one of the few people who don't insult others around here.

[blackiwid]

wanna hear something funny? my cool thinkpad edge13 amd box has a 2% battery capacity

What we nuub users want beutiful 2d accel tear free desktops great open and closed source graphic drivers will never happen unless all mighty corps i mentioned think they can get a profit from desktop market. In the end all mighty dollar rullz everything. We can just full ourselfs with freedom there is no such thing and it will never be.

I have a thinkpad too and yes lenovo does not make really good work... but the other companies do not better so whatever ^^

Amd does much to make more mainboards compatible with coreboot... so yes without coreboot we will become crap always on the other side it does not hurt me much that accu time could be better because I seldom are many hours without power...

But I dont understand why somebody would be so pissed about such a bug why:

1. in the end the distributors are responsible what version of kernel they give to you, so if something like that happens they will likely fix it with some patches
2. in ubuntu and next fedora not kernel 3.5 but kernel 3.6 will be shipped so maybe this problem will be gone with that.

And another point, the kernel has a kind of mixed release modell. so they release kernels mainly as timebased releases. So they have to make compromises about what they will include if its good enough or brings more or makes more kaput.
And there is a third point the "stable" kernels gets way more users that test it than some stuff thats in some branches, so if they are nearly at the point that their change as example replacing a driver infrastructur or something like that are better than the old thing, they will include it maybe to get feedback...

so where is the problem in doing that when you want more than that just use a distribution and you will not even see this patch... developers do not make that out of fun because they want to torture you they think they are doing the right to as example fix this problem faster.

yes its cool to have a test for anything but as somebody else sain in this case you need much much test-hardware thats harder as to write a small software-test for each peace...

and freedom is not about that you get all what you want for free, freedom is to have the choice to wait, to pay somebody to fix it, to fix it by yourself. And to not use it...

To have somebody to do what he not wants to do or not wants it to do like you want him to do it, is not freedom its slavery...