Re: [PATCH v9 19/29] x86/insn-eval: Add support to resolve 32-bit address encodings

[Ricardo Neri]

On Fri, Oct 20, 2017 at 08:38:25PM +0200, Borislav Petkov wrote:
> On Fri, Oct 20, 2017 at 11:24:48AM -0700, Ricardo Neri wrote:
> > I will create these helper functions. This change and your suggestion in
> > patch 18 will impact other patches in the series (e.g., the function
> > get_addr_ref_16() in patch 22). Would it make sense to submit a v10 and
> > resume review there?
> > 
> > Also, do you think I am still on-time to make it to v4.15?
> 
> Well, I've been thinking about it: handling huge patchsets is always
> very cumbersome, time-consuming and error prone. So perhaps it would be
> easier - maybe - I'm not saying it will definitely but only maybe - if
> you would split the patchset into, say, two, pieces, or halves, if you
> will.
> 
> And I think the first piece is more or less reviewed and if tip guys
> don't find any booboos, it could go in now. Which would free you to deal
> with the other half later.

Since MPX uses this emulation code and only cares about 64-bit addresses
(given the initial implemention from which I based my code), patches 1-18
need to be pulled together.

Perhaps I can send the v10 of patches 1-18 (or a v1 since is a new
series?). Patches 19-29 would constitute a series of improved emulation
plus UMIP code.

Does it make sense?

Thanks and BR,
Ricardo
> --