30 May
2012
30 May
'12
7:22 a.m.
(5/30/12 3:15 AM), Marek Szyprowski wrote:
Hello,
On Wednesday, May 30, 2012 2:11 AM KOSAKI Motohiro wrote:
static void * __dma_alloc_remap(struct page *page, size_t size, gfp_t gfp, pgprot_t prot, const void *caller) {
- struct arm_vmregion *c;
- size_t align;
- int bit;
- struct vm_struct *area;
- unsigned long addr;
- if (!consistent_pte) {
printk(KERN_ERR "%s: not initialised\n", __func__);
- area = get_vm_area_caller(size, VM_DMA | VM_USERMAP, caller);
In this patch, VM_DMA is only used here. So, is this no effect?
I introduced it mainly to let user know which areas have been allocated by the dma-mapping api.
vma->flags are limited resource, it has only 32 (or 64) bits. Please don't use it for such unimportant thing.
I also plan to add a check suggested by Minchan Kim in __dma_free_remap() if the vmalloc area have been in fact allocated with VM_DMA set.