Re: [PATCH 5/5] memory: Make sure root MR won't be added as subregion

[Philippe Mathieu-Daudé]

Hi Peter,

On 3/12/21 7:28 PM, Philippe Mathieu-Daudé wrote:
> From: Peter Xu <peterx@redhat.com>
> 
> Add a bool for MR to mark whether this MR is a root MR of an AS.  We bail out
> asap if this MR is added as a subregion of another MR.
> 
> Signed-off-by: Peter Xu <peterx@redhat.com>
> ---
>  include/exec/memory.h | 1 +
>  softmmu/memory.c      | 2 ++
>  2 files changed, 3 insertions(+)
> 
> diff --git a/include/exec/memory.h b/include/exec/memory.h
> index 54ccf1a5f09..8137ad3a9f6 100644
> --- a/include/exec/memory.h
> +++ b/include/exec/memory.h
> @@ -477,6 +477,7 @@ struct MemoryRegion {
>      bool ram_device;
>      bool enabled;
>      bool warning_printed; /* For reservations */
> +    bool is_root_mr;
>      uint8_t vga_logging_count;
>      MemoryRegion *alias;
>      hwaddr alias_offset;
> diff --git a/softmmu/memory.c b/softmmu/memory.c
> index 9db47b7db6b..ce322ff3d6e 100644
> --- a/softmmu/memory.c
> +++ b/softmmu/memory.c
> @@ -2442,6 +2442,7 @@ static void 
> memory_region_add_subregion_common(MemoryRegion *mr,
>                                                 MemoryRegion *subregion)
>  {
>      assert(!subregion->container);
> +    assert(!subregion->is_root_mr);

Not all containers have to be backed by AddressSpace.

This check is too strict IMO.

>      subregion->container = mr;
>      subregion->addr = offset;
>      memory_region_update_container_subregions(subregion);
> @@ -2819,6 +2820,7 @@ void address_space_init(AddressSpace *as, MemoryRegion 
> *root, const char *name)
>  {
>      memory_region_ref(root);
>      as->root = root;
> +    root->is_root_mr = true;
>      as->current_map = NULL;
>      as->ioeventfd_nb = 0;
>      as->ioeventfds = NULL;
>