In order to map device memory using superpages, mmap(2) must find a

superpage-aligned virtual address for the mapping.  Revision 1.65
implemented an overly simplistic and generally ineffectual method for
finding a superpage-aligned virtual address.  Specifically, it rounds
the virtual address corresponding to the end of the data segment up to
the next superpage-aligned virtual address.  If this virtual address
is unallocated, then the device will be mapped using superpages.
Unfortunately, in modern times, where applications like the X server
dynamically load much of their code, this virtual address is already
allocated.  In such cases, mmap(2) simply uses the first available
virtual address, which is not necessarily superpage aligned.

This revision changes mmap(2) to use a more robust method,
specifically, the VMFS_ALIGNED_SPACE option that is now implemented by
vm_map_find().

1 files changed, 3 insertions, 5 deletions

diff --git a/sys/vm/vm_mmap.c b/sys/vm/vm_mmap.c
index d5b6e3c..ca86cf9 100644
--- a/sys/vm/vm_mmap.c
+++ b/sys/vm/vm_mmap.c
@@ -1456,15 +1456,13 @@ vm_mmap(vm_map_t map, vm_offset_t *addr, vm_size_t size, vm_prot_t prot,
 		maxprot |= VM_PROT_EXECUTE;
 #endif
 
-	if (fitit)
-		*addr = pmap_addr_hint(object, *addr, size);
-
 	if (flags & MAP_STACK)
 		rv = vm_map_stack(map, *addr, size, prot, maxprot,
 		    docow | MAP_STACK_GROWS_DOWN);
 	else if (fitit)
-		rv = vm_map_find(map, object, foff, addr, size, TRUE,
-				 prot, maxprot, docow);
+		rv = vm_map_find(map, object, foff, addr, size,
+		    object != NULL && object->type == OBJT_DEVICE ?
+		    VMFS_ALIGNED_SPACE : VMFS_ANY_SPACE, prot, maxprot, docow);
 	else
 		rv = vm_map_fixed(map, object, foff, *addr, size,
 				 prot, maxprot, docow);