Ubuntu 11.10 Beta Has No Power Regression Fix

[phoronix]

Phoronix: Ubuntu 11.10 Beta Has No Power Regression Fix

Canonical's Kate Stewart set a milestone for correcting the ASPM power issue by Ubuntu 11.04 Beta 1. Ubuntu 11.04 Beta will be released today, but it will not fix the Linux 2.6.38 power regression that's caused by a change in PCI-E Active State Power Management...

http://www.phoronix.com/vr.php?view=OTg2Mg

[curaga]

They actually do have some useful kernel patches, for example the async initramfs loading one. Can speed up boot a second or so. Been shipping in 6-7 U releases, and no attempt at upstreaming at any point of course

[FireBurn]

<sarcasm>
Since Ubuntu is now consistently going off the reservation, perhaps they should consider developing their own kernel too, you know like Unity, only lower-level.
</sarcasm>

I presume that since this is upstream and difficult to solve, that:

a) it's outside of the Ubuntu development community's expertise to solve without hairy workarounds, and
b) it's affecting virtually every modern Linux distribution?

Didn't Michael have some experimental patches he was working on to solve this, or was it just the results of his profiling correlated against patchsets?

LOL the patch was reverting the code that disabled the feature on devices who's bios declares it isn't supported

[pbryan]

This isn't Ubuntu-only, is it?

<sarcasm>
Since Ubuntu is now consistently going off the reservation, perhaps they should consider developing their own kernel too, you know like Unity, only lower-level.
</sarcasm>

I presume that since this is upstream and difficult to solve, that:

a) it's outside of the Ubuntu development community's expertise to solve without hairy workarounds, and
b) it's affecting virtually every modern Linux distribution?

Didn't Michael have some experimental patches he was working on to solve this, or was it just the results of his profiling correlated against patchsets?

[FireBurn]

HP screwed up the BIOS (InsydeH20 garbage) on my dv6-3210. Forcing ASPM used to work, and after a BIOS update, it doesn't. If I hadn't been in such a hurry to buy a laptop, I would've got something from a dedicated Linux vendor.

woulda shoulda coulda

[DanL]

HP screwed up the BIOS (InsydeH20 garbage) on my dv6-3210. Forcing ASPM used to work, and after a BIOS update, it doesn't. If I hadn't been in such a hurry to buy a laptop, I would've got something from a dedicated Linux vendor.

[FireBurn]

The real fix is for the manufactures to provide. I doubt however most of them will update their bios

I refuse to buy a Samsung laptop to this day due to a bios bug that only showed up when you used 4GB of RAM on a 64bit OS

The bios was only fixed when Windows 7 was released even though I told them about the bug 2 years prior :-(

[RealNC]

Equally people don't want their machines crashing - which was why the code was disabled in the first place for machines that don't declare they support it

Double bad. No crashes = no battery. Screwed double.

[FireBurn]

Well, the last stand of the so called "Linux Desktop", which is on the laptop, is fleeting away. No one will want Linux on their laptops if it means 3 hours less battery time than Windows.

Equally people don't want their machines crashing - which was why the code was disabled in the first place for machines that don't declare they support it

[RealNC]

Well, the last stand of the so called "Linux Desktop", which is on the laptop, is fleeting away. No one will want Linux on their laptops if it means 3 hours less battery time than Windows.