Re: [Qemu-devel] [PATCH 2/2] postcopy: Require RAMBlocks that are whole pages

[Peter Xu <peterx@redhat.com>]

On Thu, May 18, 2017 at 09:08:34AM +0100, Dr. David Alan Gilbert wrote:
> * Juan Quintela (quintela@redhat.com) wrote:
> > "Dr. David Alan Gilbert (git)" <dgilbert@redhat.com> wrote:
> > > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> > >
> > > It turns out that it's legal to create a VM with RAMBlocks that aren't
> > > a multiple of the pagesize in use; e.g. a 1025M main memory using
> > > 2M host pages.  That breaks postcopy's atomic placement of pages,
> > > so disallow it.
> > >
> > > Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> > 
> > Reviewed-by: Juan Quintela <quintela@redhat.com>
> 
> Thanks
> 
> > >      }
> > >  
> > >      /* We don't support postcopy with shared RAM yet */
> > > -    if (qemu_ram_foreach_block(test_range_shared, NULL)) {
> > > +    if (qemu_ram_foreach_block(test_ramblock_postcopiable, NULL)) {
> > 
> > When I was looking at this code, I still don't know why
> > qemu_ram_foreach_block() don't pass the block directly.  It needs it
> > almost all callers.
> > 
> > When I saw it I was about to change it, but got sidetracked on other
> > things :-p
> 
> I think originally it passed very little information at all, and
> that RAMBlocks were these mystical things no one outside exec.c
> was really supposed to know about.

(Yeah I got the same question before. That's why I got
 RAMBLOCK_FOREACH() but didn't use qemu_ram_foreach_block() since I
 need at least page size info for the block...)

-- 
Peter Xu