Re: [PATCH v1 1/2] vhost: Defer filtering memory sections until building the vhost memory structure

["Michael S. Tsirkin" <mst@redhat.com>]

On Thu, Feb 16, 2023 at 12:47:51PM +0100, David Hildenbrand wrote:
> Having multiple devices, some filtering memslots and some not filtering
> memslots, messes up the "used_memslot" accounting. If we'd have a device
> the filters out less memory sections after a device that filters out more,
> we'd be in trouble, because our memslot checks stop working reliably.
> For example, hotplugging a device that filters out less memslots might end
> up passing the checks based on max vs. used memslots, but can run out of
> memslots when getting notified about all memory sections.
> 
> Further, it will be helpful in memory device context in the near future
> to know that a RAM memory region section will consume a memslot, and be
> accounted for in the used vs. free memslots, such that we can implement
> reservation of memslots for memory devices properly. Whether a device
> filters this out and would theoretically still have a free memslot is
> then hidden internally, making overall vhost memslot accounting easier.
> 
> Let's filter the memslots when creating the vhost memory array,
> accounting all RAM && !ROM memory regions as "used_memslots" even if
> vhost_user isn't interested in anonymous RAM regions, because it needs
> an fd.
> 
> When a device actually filters out regions (which should happen rarely
> in practice), we might detect a layout change although only filtered
> regions changed. We won't bother about optimizing that for now.

That caused trouble in the past when using VGA because it is playing
with mappings in weird ways.
I think we have to optimize it, sorry.


> Note: we cannot simply filter out the region and count them as
> "filtered" to add them to used, because filtered regions could get
> merged and result in a smaller effective number of memslots. Further,
> we won't touch the hmp/qmp virtio introspection output.
> 
> Fixes: 988a27754bbb ("vhost: allow backends to filter memory sections")
> Cc: Tiwei Bie <tiwei.bie@intel.com>
> Signed-off-by: David Hildenbrand <david@redhat.com>

I didn't review this yet but maybe you can answer:
will this create more slots for the backend?
Because some backends are limited in # of slots and breaking them is
not a good idea.

Thanks!

> ---
>  hw/virtio/vhost.c | 79 +++++++++++++++++++++++++++++++++--------------
>  1 file changed, 55 insertions(+), 24 deletions(-)
> 
> diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
> index eb8c4c378c..b7fb960fa9 100644
> --- a/hw/virtio/vhost.c
> +++ b/hw/virtio/vhost.c
> @@ -219,8 +219,13 @@ static void vhost_log_sync_range(struct vhost_dev *dev,
>      int i;
>      /* FIXME: this is N^2 in number of sections */
>      for (i = 0; i < dev->n_mem_sections; ++i) {
> -        MemoryRegionSection *section = &dev->mem_sections[i];
> -        vhost_sync_dirty_bitmap(dev, section, first, last);
> +        MemoryRegionSection *mrs = &dev->mem_sections[i];
> +
> +        if (dev->vhost_ops->vhost_backend_mem_section_filter &&
> +            !dev->vhost_ops->vhost_backend_mem_section_filter(dev, mrs)) {
> +            continue;
> +        }
> +        vhost_sync_dirty_bitmap(dev, mrs, first, last);
>      }
>  }
>  
> @@ -503,12 +508,6 @@ static bool vhost_section(struct vhost_dev *dev, MemoryRegionSection *section)
>              return false;
>          }
>  
> -        if (dev->vhost_ops->vhost_backend_mem_section_filter &&
> -            !dev->vhost_ops->vhost_backend_mem_section_filter(dev, section)) {
> -            trace_vhost_reject_section(mr->name, 2);
> -            return false;
> -        }
> -
>          trace_vhost_section(mr->name);
>          return true;
>      } else {
> @@ -525,6 +524,43 @@ static void vhost_begin(MemoryListener *listener)
>      dev->n_tmp_sections = 0;
>  }
>  
> +static void vhost_realloc_vhost_memory(struct vhost_dev *dev,
> +                                       unsigned int nregions)
> +{
> +    const size_t size = offsetof(struct vhost_memory, regions) +
> +                        nregions * sizeof dev->mem->regions[0];
> +
> +    dev->mem = g_realloc(dev->mem, size);
> +    dev->mem->nregions = nregions;
> +}
> +
> +static void vhost_rebuild_vhost_memory(struct vhost_dev *dev)
> +{
> +    unsigned int nregions = 0;
> +    int i;
> +
> +    vhost_realloc_vhost_memory(dev, dev->n_mem_sections);
> +    for (i = 0; i < dev->n_mem_sections; i++) {
> +        struct MemoryRegionSection *mrs = dev->mem_sections + i;
> +        struct vhost_memory_region *cur_vmr;
> +
> +        if (dev->vhost_ops->vhost_backend_mem_section_filter &&
> +            !dev->vhost_ops->vhost_backend_mem_section_filter(dev, mrs)) {
> +            continue;
> +        }
> +        cur_vmr = dev->mem->regions + nregions;
> +        nregions++;
> +
> +        cur_vmr->guest_phys_addr = mrs->offset_within_address_space;
> +        cur_vmr->memory_size     = int128_get64(mrs->size);
> +        cur_vmr->userspace_addr  =
> +            (uintptr_t)memory_region_get_ram_ptr(mrs->mr) +
> +            mrs->offset_within_region;
> +        cur_vmr->flags_padding   = 0;
> +    }
> +    vhost_realloc_vhost_memory(dev, nregions);
> +}
> +
>  static void vhost_commit(MemoryListener *listener)
>  {
>      struct vhost_dev *dev = container_of(listener, struct vhost_dev,
> @@ -532,7 +568,6 @@ static void vhost_commit(MemoryListener *listener)
>      MemoryRegionSection *old_sections;
>      int n_old_sections;
>      uint64_t log_size;
> -    size_t regions_size;
>      int r;
>      int i;
>      bool changed = false;
> @@ -564,23 +599,19 @@ static void vhost_commit(MemoryListener *listener)
>          goto out;
>      }
>  
> -    /* Rebuild the regions list from the new sections list */
> -    regions_size = offsetof(struct vhost_memory, regions) +
> -                       dev->n_mem_sections * sizeof dev->mem->regions[0];
> -    dev->mem = g_realloc(dev->mem, regions_size);
> -    dev->mem->nregions = dev->n_mem_sections;
> +    /*
> +     * Globally track the used memslots *without* device specific
> +     * filtering. This way, we always know how many memslots are required
> +     * when devices with differing filtering requirements get mixed, and
> +     * all RAM memory regions of memory devices will consume memslots.
> +     */
>      used_memslots = dev->mem->nregions;
> -    for (i = 0; i < dev->n_mem_sections; i++) {
> -        struct vhost_memory_region *cur_vmr = dev->mem->regions + i;
> -        struct MemoryRegionSection *mrs = dev->mem_sections + i;
>  
> -        cur_vmr->guest_phys_addr = mrs->offset_within_address_space;
> -        cur_vmr->memory_size     = int128_get64(mrs->size);
> -        cur_vmr->userspace_addr  =
> -            (uintptr_t)memory_region_get_ram_ptr(mrs->mr) +
> -            mrs->offset_within_region;
> -        cur_vmr->flags_padding   = 0;
> -    }
> +    /*
> +     * Rebuild the regions list from the new sections list, filtering out all
> +     * sections that this device is not interested in.
> +     */
> +    vhost_rebuild_vhost_memory(dev);
>  
>      if (!dev->started) {
>          goto out;
> -- 
> 2.39.1