Burning Through Power: Linux Regressions Found
For the multiple Linux kernel power regressions that I've talked about on Phoronix now for a number of weeks and have been affecting mobile Linux users en mass, I said I was looking for a better power measuring approach by using an AC power meter / UPS rather than a notebook battery to use in nailing these regressions. Using such a power meter would lead to a fully-automated process by the Phoronix Test Suite as no longer would I need to keep pulling the power plug from a laptop, could use much faster hardware, and allow for some other interesting possibilities. Well, last week I bought a power meter that plays with Linux. So now there's some news to share.
The big "2.6.38 kernel power regression" has now been fully detected using the Phoronix Test Suite stack. The problem / commit has been definitively determined. I also have what should work as a workaround to this power issue that should be easily applicable on the 2.6.38/2.6.39/3.0 kernels on most systems. But with that said, a proper solution for fixing the problem on "out of the box" / clean installations may not be merged until the Linux 3.1 kernel or later.
With the 2.6.38 power regression determined, today (Sunday) I'm also working to nail the nasty Linux 2.6.35 kernel power regression too and there's a few other possible power bugs that my testing software code has also detected. Another one of these is also found in the 2.6.38 merge window. Now with the power meter in hand and improvements being made to the Phoronix Test Suite and related components (Phoromatic, OpenBenchmarking.org, etc), both to the open-source and enterprise versions of the Phoronix products, it should just be easy and closer to hitting and tracking down one regression after another.
This information isn't being released today as I'm waiting to hear back from the respective subsystem maintainer / the one that also signed-off on the patch in question (the name is no stranger to Phoronix) for any further details, but unfortunately he doesn't usually seem to work over the weekends. Additionally, hell yes, to maximize traffic and ad revenues off of this work due to the amount of time and resources I've spent tracking down these Linux kernel power bugs and other problems on top of my already maxed workload; no organization has yet stepped up to sponsor this continued work or any PTS-powered continuous integration on the kernel.
Stay tuned till Monday and I've also been reporting various bits to my personal Twitter feed.
The big "2.6.38 kernel power regression" has now been fully detected using the Phoronix Test Suite stack. The problem / commit has been definitively determined. I also have what should work as a workaround to this power issue that should be easily applicable on the 2.6.38/2.6.39/3.0 kernels on most systems. But with that said, a proper solution for fixing the problem on "out of the box" / clean installations may not be merged until the Linux 3.1 kernel or later.
With the 2.6.38 power regression determined, today (Sunday) I'm also working to nail the nasty Linux 2.6.35 kernel power regression too and there's a few other possible power bugs that my testing software code has also detected. Another one of these is also found in the 2.6.38 merge window. Now with the power meter in hand and improvements being made to the Phoronix Test Suite and related components (Phoromatic, OpenBenchmarking.org, etc), both to the open-source and enterprise versions of the Phoronix products, it should just be easy and closer to hitting and tracking down one regression after another.
This information isn't being released today as I'm waiting to hear back from the respective subsystem maintainer / the one that also signed-off on the patch in question (the name is no stranger to Phoronix) for any further details, but unfortunately he doesn't usually seem to work over the weekends. Additionally, hell yes, to maximize traffic and ad revenues off of this work due to the amount of time and resources I've spent tracking down these Linux kernel power bugs and other problems on top of my already maxed workload; no organization has yet stepped up to sponsor this continued work or any PTS-powered continuous integration on the kernel.
Stay tuned till Monday and I've also been reporting various bits to my personal Twitter feed.
78 Comments