Re: [RFC v18 00/15] i386 cleanup PART 2

[Claudio Fontana]

On 2/15/21 12:37 PM, Alex Bennée wrote:
> 
> Claudio Fontana <cfontana@suse.de> writes:
> 
> <snip>
>> Looking forward to your comments on this proposal,
>>
> <snip>
> 
> So I've reviewed as much as I'm comfortable with - I'm going to defer to
> the x86 experts on the split of stuff for x86. However from my point of
> view I think it's a nice step in improving modularity and reducing the
> maze of #ifdefs in the code.
> 

Thanks a lot Alex for your review.

If I can leverage your TCG knowledge a bit more, as I forgot a lot about it,

have you noticed in "i386: split tcg btp_helper into softmmu and user parts"
and "i386: split smm helper (softmmu)"

those new

#ifndef CONFIG_USER_ONLY?

Basically we remove some preamble stuff that would end up calling empty stubs 
for user mode,
but do you think there could be some other consequence I am not seeing?

Maybe there is even more that could be removed in the code immediately 
preceding those CONFIG_USER_ONLY?

In particular I am referring to patches:

  i386: split tcg btp_helper into softmmu and user parts
  i386: split smm helper (softmmu)

I am commenting those patches now inline, so that it is easier to see what I am 
talking about.

Again, thanks a lot!

Claudio