I was just reading slashdot and came across this awesome article http://liliputing.com/2012/06/android-is-about-to-get-a-lot-faster-thanks-to-linaro.html about some enterprising individuals who have come up with a way of using the new gcc tool chain and inserting various optimizations into the android kernel and base system that produces a huge positive performance impact. sadly I'm not a programmer (but I am an experienced beta tester and know my way around linux internals) but I think it might be worth having the t3am take a look into patching some of these optimizations into the already AWESOME job they've done with the cm9 p1000 port. Yes, there really is only so much a 2 year old single core 1ghz tablet with 512MB ram can do, but I'm hoping that with some more optimization work such as that brought to us by the linaro team we can squeeze every cpu operation and every bit of ram doing exactly what it should be to make application execution a fast, stable, and hopefully less memory intensive experience.
What do you guys think?
+1
We should try everything to make our tablets perform as if they'd up to date...
Knew someone would post it, and before you say anything, i already did a build with the toolchain yesterday, nothing else, just the linaro toolchain with the CyanogenMod 9.
First thing - it broke kernel build, had to use a prebuilt kernel.
No big improvements, ran a 0xbench and got a small increment in the score, but then most of the other code is still in review, and using it as it is causes breakage with the proprietary libs.
On a sidenote, i'm doing a build for the kernel changes from humberos' tree, with some fixes, do test it and let me know if anything doesn't work as expected (ofc on the other thread), mostly shouldn't, still being safe, shall merge it after that.
EDIT : DONE, http://get.cm/get/2I9
Flash it, and report back on the discussion thread, shall merge it after that
Last edited: