[plug] 2.4.24 mremap root exploit

[William Kenworthy]

This is the second 2.6 Ive tried, and the slowest yet (2.6.2 with mm
patches, other was vanilla).  It very much depends on the type of
loading, but for jobs that take a week or more at 100% cpu, an extra
couple of days is a problem.  But if you have a desktop based system, it
does seem nicer if you leave such a job churning away in the
background.  This is workstation type processing, not so much server or
desktop type use though, so I am not sure how this would translate to a
database or webserver which is why I am interested in how Craig finds
it.

BillK

On Wed, 2004-02-18 at 23:01, Craig Ringer wrote:
> On Wed, 2004-02-18 at 22:33, William Kenworthy wrote:
> > Interesting to see how it goes.  Subjectively nicer on the desktop, but
> > ~1/3 slower using transcode which fits in with others have said on the
> > gentoo lists (The first guy who brought it up (posted the numbers) 
> > started a mini-flamewar for suggesting 2.6 could be slower!)
> 
> Heh... of course it's slower for some tasks - isn't every major OS
> upgrade, esp when developed for scalability? Who cares though - really?
> If it's more solid and consistent in performance, and more responsive,
> I'd consider that a worthwhile trade-off. I must admit surprise at how
> /much/ slower he reported - I'd expect no more than a max slowdown of 5
> - 10% in a worst case to have been allowed. Driver issue perhaps?
>