lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

amd-iommu: enlarge the aperture dynamically

Browse Source 
By dynamically increasing the aperture the extended allocator is now
ready for use.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This commit is contained in:
Joerg Roedel
2009-05-19 10:23:15 +02:00
parent 00cd122ae5
commit 11b83888ae
1 changed files with 18 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
arch/x86/kernel/amd_iommu.c
View File

@@ -1403,10 +1403,26 @@ static dma_addr_t __map_single(struct device *dev,
if (align) if (align)
align_mask = (1UL << get_order(size)) - 1; align_mask = (1UL << get_order(size)) - 1;
retry:
address = dma_ops_alloc_addresses(dev, dma_dom, pages, align_mask, address = dma_ops_alloc_addresses(dev, dma_dom, pages, align_mask,
dma_mask); dma_mask);
if (unlikely(address == bad_dma_address)) if (unlikely(address == bad_dma_address)) {
goto out; /*
* setting next_address here will let the address
* allocator only scan the new allocated range in the
* first run. This is a small optimization.
*/
dma_dom->next_address = dma_dom->aperture_size;
if (alloc_new_range(iommu, dma_dom, false, GFP_ATOMIC))
goto out;
/*
* aperture was sucessfully enlarged by 128 MB, try
* allocation again
*/
goto retry;
}
start = address; start = address;
for (i = 0; i < pages; ++i) { for (i = 0; i < pages; ++i) {