Re: [Qemu-devel] [PATCH RFC] xen/mapcache: store dma information in revmapcache entries for debugging

[Paolo Bonzini <pbonzini@redhat.com>]

On 03/05/2017 02:08, Stefano Stabellini wrote:
> The Xen mapcache is able to create long term mappings, they are called
> "locked" mappings. The third parameter of the xen_map_cache call
> specifies if a mapping is a "locked" mapping.
> 
> 
> From the QEMU point of view there are two kinds of long term mappings:
> 
> [a] device memory mappings, such as option roms and video memory
> [b] dma mappings, created by dma_memory_map & friends
> 
> After certain operations, ballooning a VM in particular, Xen asks QEMU
> kindly to destroy all mappings. However, certainly [a] mappings are
> present and cannot be removed. That's not a problem as they are not
> affected by balloonning. The *real* problem is that if there are any
> mappings of type [b], any outstanding dma operations could fail. This is
> a known shortcoming. In other words, when Xen asks QEMU to destroy all
> mappings, it is an error if any [b] mappings exist.
> 
> However today we have no way of distinguishing [a] from [b]. Because of
> that, we cannot even print a decent warning.
> 
> This patch introduces a new "dma" bool field to MapCacheRev entires, to
> remember if a given mapping is for dma or is a long term device memory
> mapping. When xen_invalidate_map_cache is called, we print a warning if
> any [b] mappings exist. We ignore [a] mappings.
> 
> Mappings created by qemu_map_ram_ptr are assumed to be [a], while
> mappings created by address_space_map->qemu_ram_ptr_length are assumed
> to be [b].
> 
> The goal of the patch is to make debugging and system understanding
> easier.

Sure, why not.

Acked-by: Paolo Bonzini <pbonzini@redhat.com>

Paolo

> Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
> 
> diff --git a/exec.c b/exec.c
> index eac6085..85769e1 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -2084,10 +2084,10 @@ void *qemu_map_ram_ptr(RAMBlock *ram_block, ram_addr_t addr)
>           * In that case just map until the end of the page.
>           */
>          if (block->offset == 0) {
> -            return xen_map_cache(addr, 0, 0);
> +            return xen_map_cache(addr, 0, 0, false);
>          }
>  
> -        block->host = xen_map_cache(block->offset, block->max_length, 1);
> +        block->host = xen_map_cache(block->offset, block->max_length, 1, false);
>      }
>      return ramblock_ptr(block, addr);
>  }
> @@ -2117,10 +2117,10 @@ static void *qemu_ram_ptr_length(RAMBlock *ram_block, ram_addr_t addr,
>           * In that case just map the requested area.
>           */
>          if (block->offset == 0) {
> -            return xen_map_cache(addr, *size, 1);
> +            return xen_map_cache(addr, *size, 1, true);
>          }
>  
> -        block->host = xen_map_cache(block->offset, block->max_length, 1);
> +        block->host = xen_map_cache(block->offset, block->max_length, 1, true);
>      }
>  
>      return ramblock_ptr(block, addr);
> diff --git a/hw/i386/xen/xen-mapcache.c b/hw/i386/xen/xen-mapcache.c
> index 31debdf..fa4282a 100644
> --- a/hw/i386/xen/xen-mapcache.c
> +++ b/hw/i386/xen/xen-mapcache.c
> @@ -62,6 +62,7 @@ typedef struct MapCacheRev {
>      hwaddr paddr_index;
>      hwaddr size;
>      QTAILQ_ENTRY(MapCacheRev) next;
> +    bool dma;
>  } MapCacheRev;
>  
>  typedef struct MapCache {
> @@ -202,7 +203,7 @@ static void xen_remap_bucket(MapCacheEntry *entry,
>  }
>  
>  static uint8_t *xen_map_cache_unlocked(hwaddr phys_addr, hwaddr size,
> -                                       uint8_t lock)
> +                                       uint8_t lock, bool dma)
>  {
>      MapCacheEntry *entry, *pentry = NULL;
>      hwaddr address_index;
> @@ -289,6 +290,7 @@ tryagain:
>      if (lock) {
>          MapCacheRev *reventry = g_malloc0(sizeof(MapCacheRev));
>          entry->lock++;
> +        reventry->dma = dma;
>          reventry->vaddr_req = mapcache->last_entry->vaddr_base + address_offset;
>          reventry->paddr_index = mapcache->last_entry->paddr_index;
>          reventry->size = entry->size;
> @@ -300,12 +302,12 @@ tryagain:
>  }
>  
>  uint8_t *xen_map_cache(hwaddr phys_addr, hwaddr size,
> -                       uint8_t lock)
> +                       uint8_t lock, bool dma)
>  {
>      uint8_t *p;
>  
>      mapcache_lock();
> -    p = xen_map_cache_unlocked(phys_addr, size, lock);
> +    p = xen_map_cache_unlocked(phys_addr, size, lock, dma);
>      mapcache_unlock();
>      return p;
>  }
> @@ -426,8 +428,10 @@ void xen_invalidate_map_cache(void)
>      mapcache_lock();
>  
>      QTAILQ_FOREACH(reventry, &mapcache->locked_entries, next) {
> -        DPRINTF("There should be no locked mappings at this time, "
> -                "but "TARGET_FMT_plx" -> %p is present\n",
> +        if (!reventry->dma)
> +            continue;
> +        fprintf(stderr, "Locked DMA mapping while invalidating mapcache!"
> +                " "TARGET_FMT_plx" -> %p is present\n",
>                  reventry->paddr_index, reventry->vaddr_req);
>      }
>  
> diff --git a/include/sysemu/xen-mapcache.h b/include/sysemu/xen-mapcache.h
> index b8c93b9..01daaad 100644
> --- a/include/sysemu/xen-mapcache.h
> +++ b/include/sysemu/xen-mapcache.h
> @@ -17,7 +17,7 @@ typedef hwaddr (*phys_offset_to_gaddr_t)(hwaddr start_addr,
>  void xen_map_cache_init(phys_offset_to_gaddr_t f,
>                          void *opaque);
>  uint8_t *xen_map_cache(hwaddr phys_addr, hwaddr size,
> -                       uint8_t lock);
> +                       uint8_t lock, bool dma);
>  ram_addr_t xen_ram_addr_from_mapcache(void *ptr);
>  void xen_invalidate_map_cache_entry(uint8_t *buffer);
>  void xen_invalidate_map_cache(void);
> @@ -31,7 +31,8 @@ static inline void xen_map_cache_init(phys_offset_to_gaddr_t f,
>  
>  static inline uint8_t *xen_map_cache(hwaddr phys_addr,
>                                       hwaddr size,
> -                                     uint8_t lock)
> +                                     uint8_t lock,
> +                                     bool dma)
>  {
>      abort();
>  }
>