Michel DÃnzer wrote:
Ok, that was one of my doubts.
+ // allocate the surface
+ for(i=0;i<8;i++)
+ if (!(dev_priv->surfaces&(1<<i)))
+ break;
+
+ if (i>=8)
+ return DRM_ERR(ENOMEM);
+ else
+ dev_priv->surfaces&=(1<<i);
+
+ if ( DRM_COPY_TO_USER( &alloc.surface, &i, + sizeof(int) ) ) {
+ DRM_ERROR( "copy_to_user\n" );
+ return DRM_ERR(EFAULT);
IMHO it should also manage the ranges (prevent overlapping, ...) and parameters of the surfaces.
So if we go that route, it would manage the ranges, prevent overlapping, and also try to "spare" the surfaces by merging adjacent ones with similar properties.
+ DRM_COPY_FROM_USER_IOCTL( memfree, (drm_radeon_mem_free_t __user *)data, + sizeof(memfree) ); + + dev_priv->surfaces&= (~(1<<memfree.surface));
It should definitely ensure that only the owner can free a surface though. It would also need to free a client's surfaces if it dies, etc.
How do you "know" the owner ? I'm not sure the pid would be enough.
Stephane
------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
