Thread (16 messages) 16 messages, 4 authors, 2012-08-23

[RFC 2/4] ARM: dma-mapping: IOMMU allocates pages from pool with GFP_ATOMIC

From: KyongHo Cho <hidden>
Date: 2012-08-22 12:47:03
Also in: linux-mm

Hi.

vzalloc() call in __iommu_alloc_buffer() also causes BUG() in atomic
context.


On Wed, Aug 22, 2012 at 7:20 PM, Hiroshi Doyu [off-list ref] wrote:
quoted hunk ↗ jump to hunk
Makes use of the same atomic pool from DMA, and skips kernel page
mapping which can involves sleep'able operation at allocating a kernel
page table.

Signed-off-by: Hiroshi Doyu <redacted>
---
 arch/arm/mm/dma-mapping.c |   22 ++++++++++++++++++----
 1 files changed, 18 insertions(+), 4 deletions(-)
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index aec0c06..9260107 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -1028,7 +1028,6 @@ static struct page **__iommu_alloc_buffer(struct
device *dev, size_t size,
quoted hunk ↗ jump to hunk
        struct page **pages;
        int count = size >> PAGE_SHIFT;
        int array_size = count * sizeof(struct page *);
-       int err;

        if (array_size <= PAGE_SIZE)
                pages = kzalloc(array_size, gfp);
@@ -1037,9 +1036,20 @@ static struct page **__iommu_alloc_buffer(struct
device *dev, size_t size,
quoted hunk ↗ jump to hunk
        if (!pages)
                return NULL;

-       err = __alloc_fill_pages(&pages, count, gfp);
-       if (err)
-               goto error
+       if (gfp & GFP_ATOMIC) {
+               struct page *page;
+               int i;
+               void *addr = __alloc_from_pool(size, &page);
+               if (!addr)
+                       goto err_out;
+
+               for (i = 0; i < count; i++)
+                       pages[i] = page + i;
+       } else {
+               int err = __alloc_fill_pages(&pages, count, gfp);
+               if (err)
+                       goto error;
+       }

        return pages;
 error:
@@ -1055,6 +1065,10 @@ static int __iommu_free_buffer(struct device *dev,
struct page **pages, size_t s
        int count = size >> PAGE_SHIFT;
        int array_size = count * sizeof(struct page *);
        int i;
+
+       if (__free_from_pool(page_address(pages[0]), size))
+               return 0;
+
        for (i = 0; i < count; i++)
                if (pages[i])
                        __free_pages(pages[i], 0);
--
1.7.5.4

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo at kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email at kvack.org </a>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120822/61768214/attachment-0001.html>
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help