On Mon 2019-06-10 08:47:45, Yu-cheng Yu wrote:
> On Sat, 2019-06-08 at 22:52 +0200, Pavel Machek wrote:
> > Hi!
> > 
> > > > I've no idea what the kernel should do; since you failed to answer the
> > > > question what happens when you point this to garbage.
> > > > 
> > > > Does it then fault or what?
> > > 
> > > Yeah, I think you'll fault with a rather mysterious CR2 value since
> > > you'll go look at the instruction that faulted and not see any
> > > references to the CR2 value.
> > > 
> > > I think this new MSR probably needs to get included in oops output when
> > > CET is enabled.
> > > 
> > > Why don't we require that a VMA be in place for the entire bitmap?
> > > Don't we need a "get" prctl function too in case something like a JIT is
> > > running and needs to find the location of this bitmap to set bits itself?
> > > 
> > > Or, do we just go whole-hog and have the kernel manage the bitmap
> > > itself. Our interface here could be:
> > > 
> > > 	prctl(PR_MARK_CODE_AS_LEGACY, start, size);
> > > 
> > > and then have the kernel allocate and set the bitmap for those code
> > > locations.
> > 
> > For the record, that sounds like a better interface than userspace knowing
> > about the bitmap formats...
> > 									Pavel
> 
> Initially we implemented the bitmap that way.  To manage the bitmap, every time
> the application issues a syscall for a .so it loads, and the kernel does
> copy_from_user() & copy_to_user() (or similar things).  If a system has a few
> legacy .so files and every application does the same, it can take a long time to
> boot up.

Loading .so is already many syscalls, I'd not expect measurable
performance there. Are you sure?
								Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html