summaryrefslogtreecommitdiff
path: root/shared/radeon_mem.c
diff options
context:
space:
mode:
Diffstat (limited to 'shared/radeon_mem.c')
-rw-r--r--shared/radeon_mem.c55
1 files changed, 27 insertions, 28 deletions
diff --git a/shared/radeon_mem.c b/shared/radeon_mem.c
index 7ca10753..c3cbd3a9 100644
--- a/shared/radeon_mem.c
+++ b/shared/radeon_mem.c
@@ -40,7 +40,7 @@
*/
static struct mem_block *split_block(struct mem_block *p, int start, int size,
- int pid )
+ DRMFILE filp )
{
/* Maybe cut off the start of an existing block */
if (start > p->start) {
@@ -49,7 +49,7 @@ static struct mem_block *split_block(struct mem_block *p, int start, int size,
goto out;
newblock->start = start;
newblock->size = p->size - (start - p->start);
- newblock->pid = 0;
+ newblock->filp = 0;
newblock->next = p->next;
newblock->prev = p;
p->next->prev = newblock;
@@ -65,7 +65,7 @@ static struct mem_block *split_block(struct mem_block *p, int start, int size,
goto out;
newblock->start = start + size;
newblock->size = p->size - size;
- newblock->pid = 0;
+ newblock->filp = 0;
newblock->next = p->next;
newblock->prev = p;
p->next->prev = newblock;
@@ -75,20 +75,20 @@ static struct mem_block *split_block(struct mem_block *p, int start, int size,
out:
/* Our block is in the middle */
- p->pid = pid;
+ p->filp = filp;
return p;
}
static struct mem_block *alloc_block( struct mem_block *heap, int size,
- int align2, int pid )
+ int align2, DRMFILE filp )
{
struct mem_block *p;
int mask = (1 << align2)-1;
for (p = heap->next ; p != heap ; p = p->next) {
int start = (p->start + mask) & ~mask;
- if (p->pid == 0 && start + size <= p->start + p->size)
- return split_block( p, start, size, pid );
+ if (p->filp == 0 && start + size <= p->start + p->size)
+ return split_block( p, start, size, filp );
}
return NULL;
@@ -108,25 +108,25 @@ static struct mem_block *find_block( struct mem_block *heap, int start )
static void free_block( struct mem_block *p )
{
- p->pid = 0;
+ p->filp = 0;
- /* Assumes a single contiguous range. Needs a special pid in
+ /* Assumes a single contiguous range. Needs a special filp in
* 'heap' to stop it being subsumed.
*/
- if (p->next->pid == 0) {
+ if (p->next->filp == 0) {
struct mem_block *q = p->next;
p->size += q->size;
p->next = q->next;
p->next->prev = p;
- DRM_FREE(q);
+ DRM_FREE(q, sizeof(*q));
}
- if (p->prev->pid == 0) {
+ if (p->prev->filp == 0) {
struct mem_block *q = p->prev;
q->size += p->size;
q->next = p->next;
q->next->prev = q;
- DRM_FREE(p);
+ DRM_FREE(p, sizeof(*q));
}
}
@@ -141,47 +141,46 @@ static int init_heap(struct mem_block **heap, int start, int size)
*heap = DRM_MALLOC(sizeof(**heap));
if (!*heap) {
- DRM_FREE( blocks );
+ DRM_FREE( blocks, sizeof(*blocks) );
return -ENOMEM;
}
blocks->start = start;
blocks->size = size;
- blocks->pid = 0;
+ blocks->filp = 0;
blocks->next = blocks->prev = *heap;
memset( *heap, 0, sizeof(**heap) );
- (*heap)->pid = -1;
+ (*heap)->filp = (DRMFILE) -1;
(*heap)->next = (*heap)->prev = blocks;
return 0;
}
-/* Free all blocks associated with the releasing pid.
+/* Free all blocks associated with the releasing file.
*/
-void radeon_mem_release( struct mem_block *heap )
+void radeon_mem_release( DRMFILE filp, struct mem_block *heap )
{
- int pid = DRM_CURRENTPID;
struct mem_block *p;
if (!heap || !heap->next)
return;
for (p = heap->next ; p != heap ; p = p->next) {
- if (p->pid == pid)
- p->pid = 0;
+ if (p->filp == filp)
+ p->filp = 0;
}
- /* Assumes a single contiguous range. Needs a special pid in
+ /* Assumes a single contiguous range. Needs a special filp in
* 'heap' to stop it being subsumed.
*/
for (p = heap->next ; p != heap ; p = p->next) {
- while (p->pid == 0 && p->next->pid == 0) {
+ while (p->filp == 0 && p->next->filp == 0) {
struct mem_block *q = p->next;
p->size += q->size;
p->next = q->next;
p->next->prev = p;
- DRM_FREE(q);
+ DRM_FREE(q, sizeof(*q));
}
}
}
@@ -198,10 +197,10 @@ void radeon_mem_takedown( struct mem_block **heap )
for (p = (*heap)->next ; p != *heap ; ) {
struct mem_block *q = p;
p = p->next;
- DRM_FREE(q);
+ DRM_FREE(q, sizeof(*q));
}
- DRM_FREE( *heap );
+ DRM_FREE( *heap, sizeof(**heap) );
*heap = 0;
}
@@ -248,7 +247,7 @@ int radeon_mem_alloc( DRM_IOCTL_ARGS )
alloc.alignment = 12;
block = alloc_block( *heap, alloc.size, alloc.alignment,
- DRM_CURRENTPID );
+ filp );
if (!block)
return DRM_ERR(ENOMEM);
@@ -287,7 +286,7 @@ int radeon_mem_free( DRM_IOCTL_ARGS )
if (!block)
return DRM_ERR(EFAULT);
- if (block->pid != DRM_CURRENTPID)
+ if (block->filp != filp)
return DRM_ERR(EPERM);
free_block( block );