Re: [Xen-users] XEN/arm XENFB support

[Ian Campbell <Ian.Campbell@citrix.com>]

On Tue, 2013-12-17 at 16:42 +0000, Stefano Stabellini wrote:

> diff --git a/xen-all.c b/xen-all.c
> index 4a594bd..d45180d 100644
> --- a/xen-all.c
> +++ b/xen-all.c
> @@ -381,8 +381,6 @@ static int xen_remove_from_physmap(XenIOState *state,
>  
>          rc = xc_domain_add_to_physmap(xen_xc, xen_domid, XENMAPSPACE_gmfn, idx, gpfn);
>          if (rc) {
> -            fprintf(stderr, "add_to_physmap MFN %"PRI_xen_pfn" to PFN %"
> -                    PRI_xen_pfn" failed: %d\n", idx, gpfn, rc);

This message seemed useful to me...

>              return -rc;
>          }
>      }
> 
> diff --git a/xen-mapcache.c b/xen-mapcache.c
> index eda914a..baea376 100644
> --- a/xen-mapcache.c
> +++ b/xen-mapcache.c
> @@ -33,13 +33,8 @@
>  #  define DPRINTF(fmt, ...) do { } while (0)
>  #endif
>  
> -#if defined(__i386__)
>  #  define MCACHE_BUCKET_SHIFT 16
>  #  define MCACHE_MAX_SIZE     (1UL<<31) /* 2GB Cap */
> -#elif defined(__x86_64__)
> -#  define MCACHE_BUCKET_SHIFT 20
> -#  define MCACHE_MAX_SIZE     (1UL<<35) /* 32GB Cap */
> -#endif

This bit seems a unrelated? Or should it really be adding __arm__ and
__aarch64__ bits and this was just a quick hack?

It'd be good if someone would pick up Daniel De Graaf's frontend patches
(see earlier in the thread) to use grants instead of priv mappings and
implement the matching backend stuff in qemu...

Ian.