Re: [Xen-devel] [PATCH v5 09/16] x86: add multiboot2 protocol support

[Daniel Kiper <daniel.kiper@xxxxxxxxxx>]

On Tue, Aug 30, 2016 at 09:14:28AM -0600, Jan Beulich wrote:
> >>> On 30.08.16 at 16:41, <daniel.kiper@xxxxxxxxxx> wrote:
> > On Thu, Aug 25, 2016 at 05:50:04AM -0600, Jan Beulich wrote:
> >> >>> On 20.08.16 at 00:43, <daniel.kiper@xxxxxxxxxx> wrote:
> >> > +        case MULTIBOOT2_TAG_TYPE_MMAP:
> >> > +            if ( get_mb2_data(tag, mmap, entry_size) < 
> >> > sizeof(*mmap_src) )
> >> > +                break;
> >> > +
> >> > +            mbi_out->flags |= MBI_MEMMAP;
> >> > +            mbi_out->mmap_length = get_mb2_data(tag, mmap, size);
> >> > +            mbi_out->mmap_length -= sizeof(multiboot2_tag_mmap_t);
> >> > +            mbi_out->mmap_length /= get_mb2_data(tag, mmap, entry_size);
> >> > +            mbi_out->mmap_length *= sizeof(memory_map_t);
> >> > +
> >> > +            mbi_out->mmap_addr = alloc_mem(mbi_out->mmap_length);
> >> > +
> >> > +            mmap_src = get_mb2_data(tag, mmap, entries);
> >> > +            mmap_dst = (memory_map_t *)mbi_out->mmap_addr;
> >> > +
> >> > +            for ( i = 0; i < mbi_out->mmap_length / 
> >> > sizeof(memory_map_t); i++ )
> >> > +            {
> >> > +                /* Init size member properly. */
> >> > +                mmap_dst[i].size = sizeof(memory_map_t);
> >> > +                mmap_dst[i].size -= sizeof(((memory_map_t){0}).size);
> >> > +                /* Now copy a given region data. */
> >> > +                mmap_src = (void *)mmap_src + i * get_mb2_data(tag, 
> >> > mmap, entry_size);
> >>
> >> Why do you multiply by i here? The way you've written it you want to
> >> increment mmap_src by entry_size _at the end_ of each iteration. Or
> >
> > This is another option. If you wish I can do that.
>
> This is not just another option - afaict the code as is won't work for
> more than one map entry.

Ugh... Sorry, you are right. Stupid mistake. I will change this.

Daniel

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel