[go: up one dir, main page]
|
|
Log in / Subscribe / Register

"Other operating systems" and "stealing" a bit

"Other operating systems" and "stealing" a bit

Posted Feb 22, 2022 17:00 UTC (Tue) by marcH (subscriber, #57642)
In reply to: "Other operating systems" and "stealing" a bit by pbonzini
Parent article: Shadow stacks for user space

If a PTE bit has been given to software / the OSes, then it is not a "spare" bit anymore. This is not just about Redmond begging, it's about not "stealing" back something that was given.

I think all the confusion about these bits comes from a lack of clarity about _who owns what_ and how. A bit ironic considering this is a feature meant to catch memory corruption.

to post comments

"Other operating systems" and "stealing" a bit

Posted Feb 22, 2022 20:42 UTC (Tue) by nix (subscriber, #2304) [Link]

Both Intel and software on Intel processors has learned this before: in the 286 days lots of software used reserved bits freely, and then the 386 started using them and all hell broke loose. It's literally impossible to do that now, because the processor stops you.

(ref: http://www.os2museum.com/wp/theres-more-to-the-286-xenix-...)

"Other operating systems" and "stealing" a bit

Posted Mar 6, 2022 6:17 UTC (Sun) by oldtomas (guest, #72579) [Link]

"If a PTE bit has been given to software / the OSes..."

The point Paolo is making is that there /is/ a protocol for the software/OS to tell the hardware "go ahead, use this bit for your shadow stack".

It seems that Redmond, though... well, we know that routine :-)


Copyright © 2026, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds