Re: [PATCH v3 1/2] x86_64,signal: Fix SS handling for signals delivered to 64-bit programs

[Cyrill Gorcunov <gorcunov@gmail.com>]

On Wed, Mar 18, 2015 at 07:25:22PM +0100, Oleg Nesterov wrote:
> On 03/18, Cyrill Gorcunov wrote:
> >
> > On Wed, Mar 18, 2015 at 11:06:00AM -0700, Andy Lutomirski wrote:
> > > > --- a/arch/x86/crtools.c
> > > > +++ b/arch/x86/crtools.c
> > > > @@ -475,6 +475,7 @@ int restore_gpregs(struct rt_sigframe *f, UserX86RegsEntry *r)
> > > >         CPREG2(rip, ip);
> > > >         CPREG2(eflags, flags);
> > > >         CPREG1(cs);
> > > > +       CPREG1(ss);
> > > >         CPREG1(gs);
> > > >         CPREG1(fs);
> > >
> > > Huh?  Is CRIU actually trying to build an entire sigcontext from
> > > scratch here?  I don't see how this can reliably work across kernel
> > > versions or CPU versions.
> >
> > Yes we are. And why it can't work reliably? As to CPU -- we're
> > testing that cpu features saved in image should match ones
> > provided by the kernel runtime, ie on the machine where we're
> > restoring.
> 
> But, say, __USER_CS can be changed in kernel, and nobody should notice this.

True (and this applies to quotes below). Hopefully it won't be changed frequently
though ;) As to seg registers sure the safe way as Andy pointed is to fetch them
runtime on the machine we're restoring. Thanks, I will update our code!

> 
> But in this case "restore on another machine" or "restore after kernel
> upgrade" can fail.
> 
> So probably restore_gpregs() should only change the general-purpose regs,
> as its name suggests.