[lkml]   [2009]   [Aug]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [GIT PULL] Additional x86 fixes for 2.6.31-rc5

On Mon, 3 Aug 2009, Tejun Heo wrote:
> Yeap, this part is strange. I can't understand why it was done that
> way. Interestingly, the segment override doesn't trigger any
> exception, it just gets ignored. Is there a way to retrieve %gs value
> without accessing memory? ie. other than %gs:identity_gs?

No. There's a "load segment limit" instruction, but not a base one. It's
intentional - segments are supposed to be "anonymous" in protected mode
(in real more of vm86 mode, the base is normally the segment register
shifted left by four of course, but even that isn't _strictly_ always

You're very much not supposed to look at the base.


 \ /
  Last update: 2009-10-18 23:28    [W:0.049 / U:11.616 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site