Re: [Qemu-devel] Regression caused by d2f39ad "exec.c: Ensure right alignment also for file backed ram"

[Haozhong Zhang]

On 10/24/16 13:44 +0200, Igor Mammedov wrote:
> On Fri, 21 Oct 2016 16:28:01 +0800
> Haozhong Zhang <address@hidden> wrote:
>
> > Hi,
> >
> > Commit d2f39ad "exec.c: Ensure right alignment also for file backed
> > ram" added an additional alignment requirement besides to the previous
> > page size for the size of the backend file. On x86, the alignment
> > requirement for the size of the backend file is changed from 4KB in
> > QEMU 2.6 to 2MB in QEMU 2.7.
> >
> > This change immediately breaks the usage of "-object 
> > memory-backend-file,...,size=$SIZE"
> > on x86, where $SIZE is multiple of 4KB but not 2MB. It works on QEMU 2.6.
> > Could this be considered as a regression?
> >
> > The commit message shows it's for s390. I'm wondering whether the same
> > regression happens on s390 and ARM. If not, could I fix the regression
> > on x86 by keeping using the original alignment only on x86, e.g.
> I don't see a nice way to work around this regression, so perhaps
> we should tree 2.7 as broken (wrt migration when *dimm devices with file
> backends with pagesize < 2Mb are used) and unbreak 2.8 and onwards.
>
> Could you send a proper patch to list with commit message
> explaining what is broken and how + CLI example.

Sure, I'll send a patch later.

> > modified   exec.c
> > @@ -1254,7 +1254,11 @@ static void *file_ram_alloc(RAMBlock *block,
> >      }
> >
> >      block->page_size = qemu_fd_getpagesize(fd);
> > +#if defined(__x86_64__) || defined(__i386__)
> > +    block->mr->align = block->page_size;
> > +#else
> >      block->mr->align = MAX(block->page_size, QEMU_VMALLOC_ALIGN);
> maybe it's worth to do other way around i.e. for special s390 case
> do above and for the rest leave page_size alignment?
>
>
> > +#endif
> >
> >      if (memory < block->page_size) {
> >          error_setg(errp, "memory size 0x" RAM_ADDR_FMT " must be equal to "
> >
> >
> > Thanks,
> > Haozhong