From a few days ago, grsecurity has a different opinion on the new security work in the kernel, calling into question the implementation of both the new write-only protection and KASLR from 3.14: https://news.ycombinator.com/item?id=11698381

Although GKH's pithy response on G+ hits at their incentives: "it's hard watching your business model slip away, I feel bad :(" https://plus.google.com/+gregkroahhartman/posts/2rM46nj9RsL

Separately, KernelNewbies is my go-to place for high(ish) level summaries of kernel work across the entire kernel (not just headline features). http://kernelnewbies.org/Linux_4.6

The grsecurity post is technically correct .. but that's how Linux development is always done. First lay the foundation, then fold over subsystems one at a time in piecemeal increments.

Everytime grsecurity comes up someone asks why it is not in the main kernel yet. The short answer is that the developers couldn't or didn't want to work that way. Kes Cook is actually nibbling away at that work right now, doing the thankless work of merging one part of grsecurity functionality, one piece at a time. That the first commit isn't complete or even very useful is to be expected.

That is a very very important piece of information that I didn't see (or understand). Organic integration is going. Thanks to all them, be it the grsec team or Mr Cook.

Kernel security is good, and more kernel security is better. I'm still hoping that Linux gets W^X (Write XOR eXecute) protection like some BSDs have - it eliminates several classes of vulnerabilities at a single stroke.

Has anyone figured out why Linux devs seem to really dig Google+?

And also Google groups sadly... I hate this website so much...

ah ah ah zdnet the pro windows that will now speak of the Linux kernel. I think I will go to Hurd!