Re:Play Store and Android Apps Coming to Chromebooks
by azakai on Thursday May 19, 2016 @07:40PM
(#11734715)
This does, however mean that
1. This depends on closed-source code (libhoudini is Intel code, and it's not open, last I heard). I guess it's not the first piece of closed-source code in Chromebooks, but it's a crucial piece.
2. This code is still not portable, it just runs on 2 archs, unless someone writes a libhoudini for all other archs as well.
Re:Play Store and Android Apps Coming to Chromebooks
by izacus on Friday May 20, 2016 @04:31AM
(#11736306)
What "other archs"? Android only supports 3 ARM architectures (all of them will load armeabi binaries), x86 and MIPS (with 0 MIPS devices or phones being on the market).
Chromebooks run only ARM or x86.
Again, this is an issue that's relevant to a tiny minority of the applications since vast majority do not ship native binaries.
Re:Play Store and Android Apps Coming to Chromebooks
by azakai on Friday May 20, 2016 @11:16AM
(#11738494)
Even if no official Android devices use MIPS today, that could change tomorrow. And there are lots of Android forks, I would be very surprised if none of them use MIPS.
Re:Play Store and Android Apps Coming to Chromebooks
by lallysingh on Thursday May 19, 2016 @11:37PM
(#11735414)
I don't think it's a major blocker, qemu already let's you emulate any arch that you need. Replacing libhoudini with a qemu based version would likely be straightforward.
2.) For those that don't, libhoudini does an ok job even on slower phones. On faster chromebooks that shouldn't be an issue at all.