Currently, there's a logic with checks if *count is non-zero,
q->num_buffers is zero and q->memory is different than memory.
That's flawed when the device is initialized, or after the
queues are freed, as it does, unnecessary calls to
__vb2_queue_cancel() and __vb2_queue_free().
That can be avoided by making sure that q->memory is set to
VB2_MEMORY_UNKNOWN at vb2_core_queue_init(), and adding such
check at the loop.
[mchehab@s-opensource.com: fix checkpatch issues and improve the
patch, by setting q->memory to zero at vb2_core_queue_init]
Signed-off-by: Satendra Singh Thakur <satendra.t@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Acked-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
q->num_buffers -= buffers;
if (!q->num_buffers) {
- q->memory = 0;
+ q->memory = VB2_MEMORY_UNKNOWN;
INIT_LIST_HEAD(&q->queued_list);
}
return 0;
return -EBUSY;
}
- if (*count == 0 || q->num_buffers != 0 || q->memory != memory) {
+ if (*count == 0 || q->num_buffers != 0 ||
+ (q->memory != VB2_MEMORY_UNKNOWN && q->memory != memory)) {
/*
* We already have buffers allocated, so first check if they
* are not in use and can be freed.
mutex_init(&q->mmap_lock);
init_waitqueue_head(&q->done_wq);
+ q->memory = VB2_MEMORY_UNKNOWN;
+
if (q->buf_struct_size == 0)
q->buf_struct_size = sizeof(struct vb2_buffer);