Patchwork Fix wrong region_offset when overlaying a page with another

login
register
mail settings
Submitter Avi Kivity
Date Jan. 1, 2012, 5:23 p.m.
Message ID <1325438591-7355-1-git-send-email-avi@redhat.com>
Download mbox | patch
Permalink /patch/133761/
State New
Headers show

Comments

Avi Kivity - Jan. 1, 2012, 5:23 p.m.
cpu_register_physical_memory_log() does not update region_offset
if a page was previously registered for the same address.  This
could cause mmio accesses going to the wrong place, by using the
old region_offset.

Signed-off-by: Avi Kivity <avi@redhat.com>
---

Once qemu-test is merged, remind me to post a testlet for this.

 exec.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)
Andreas Färber - Jan. 2, 2012, 5:49 p.m.
Am 01.01.2012 18:23, schrieb Avi Kivity:
> cpu_register_physical_memory_log() does not update region_offset
> if a page was previously registered for the same address.  This
> could cause mmio accesses going to the wrong place, by using the
> old region_offset.
> 
> Signed-off-by: Avi Kivity <avi@redhat.com>

Line numbers don't match master ;) but applies cleanly and looks good.

Acked-by: Andreas Färber <afaerber@suse.de>

Andreas

>  exec.c |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/exec.c b/exec.c
> index 8a3f621..c366835 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -2542,6 +2542,7 @@ void cpu_register_physical_memory_log(target_phys_addr_t start_addr,
>                  p->region_offset = 0;
>              } else {
>                  p->phys_offset = phys_offset;
> +                p->region_offset = region_offset;
>                  if ((phys_offset & ~TARGET_PAGE_MASK) <= IO_MEM_ROM ||
>                      (phys_offset & IO_MEM_ROMD))
>                      phys_offset += TARGET_PAGE_SIZE;
Avi Kivity - Jan. 3, 2012, 10:04 a.m.
On 01/02/2012 07:49 PM, Andreas Färber wrote:
> Am 01.01.2012 18:23, schrieb Avi Kivity:
> > cpu_register_physical_memory_log() does not update region_offset
> > if a page was previously registered for the same address.  This
> > could cause mmio accesses going to the wrong place, by using the
> > old region_offset.
> > 
> > Signed-off-by: Avi Kivity <avi@redhat.com>
>
> Line numbers don't match master ;) but applies cleanly and looks good.
>

It's part of my MemoryRegion dispatch patchset.  It may be applied
separately or as part of that patchset, but since it's a bugfix I
thought I'd highlight it.

Patch

diff --git a/exec.c b/exec.c
index 8a3f621..c366835 100644
--- a/exec.c
+++ b/exec.c
@@ -2542,6 +2542,7 @@  void cpu_register_physical_memory_log(target_phys_addr_t start_addr,
                 p->region_offset = 0;
             } else {
                 p->phys_offset = phys_offset;
+                p->region_offset = region_offset;
                 if ((phys_offset & ~TARGET_PAGE_MASK) <= IO_MEM_ROM ||
                     (phys_offset & IO_MEM_ROMD))
                     phys_offset += TARGET_PAGE_SIZE;