[PATCH v4 2/3] common: DMA-mapping: add DMA_ATTR_NOHUGEPAGE attribute

[Douglas Anderson]

This patch adds the DMA_ATTR_NOHUGEPAGE attribute to the DMA-mapping
subsystem.

This attribute can be used as a hint to the DMA-mapping subsystem that
it's likely not worth it to try to allocation large pages behind the
scenes.  Large pages are likely to make an IOMMU TLB work more
efficiently but may not be worth it.  See the Documentation contained in
this patch for more details about this attribute and when to use it.

Note that the name of the hint (DMA_ATTR_NOHUGEPAGE) is based on the
name MADV_NOHUGEPAGE, which has the same meaning.  If we have expected
users, we could also add MADV_HUGEPAGE which has the opposite meaning of
this hint.

Signed-off-by: Douglas Anderson <dianders@xxxxxxxxxxxx>
Acked-by: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx>
---
Changes in v4:
- renamed DMA_ATTR_SEQUENTIAL to DMA_ATTR_NOHUGEPAGE
- added Marek's ack

Changes in v3:
- add DMA_ATTR_SEQUENTIAL attribute new for v3

Changes in v2: None

 Documentation/DMA-attributes.txt | 23 +++++++++++++++++++++++
 include/linux/dma-attrs.h        |  1 +
 2 files changed, 24 insertions(+)

diff --git a/Documentation/DMA-attributes.txt b/Documentation/DMA-attributes.txt
index 18dc52c4f2a0..0a2f56e9c5bd 100644
--- a/Documentation/DMA-attributes.txt
+++ b/Documentation/DMA-attributes.txt
@@ -100,3 +100,26 @@ allocated by dma_alloc_attrs() function from individual pages if it can
 be mapped as contiguous chunk into device dma address space. By
 specifying this attribute the allocated buffer is forced to be contiguous
 also in physical memory.
+
+DMA_ATTR_NOHUGEPAGE
+-------------------
+
+This is a hint to the DMA-mapping subsystem that it's probably not worth
+the time to try to allocate memory to in a way that gives better TLB
+efficiency (AKA it's not worth trying to build the mapping out of larger
+pages).  You might want to specify this if:
+- You know that the accesses to this memory won't thrash the TLB.
+  You might know that the accesses are likely to be sequential or
+  that they aren't sequential but it's unlikely you'll ping-ping
+  between many addresses that are likely to be in different physical
+  pages.
+- You know that the penalty of TLB misses while accessing the
+  memory will be small enough to be inconsequential.  If you are
+  doing a heavy operation like decryption or decompression this
+  might be the case.
+- You know that the DMA mapping is fairly transitory.  If you expect
+  the mapping to have a short lifetime then it may be worth it to
+  optimize allocation (avoid coming up with large pages) instead of
+  getting the slight performance win of larger pages.
+Setting this hint doesn't guarantee that you won't get huge pages, but it
+means that we won't try quite as hard to get them.
diff --git a/include/linux/dma-attrs.h b/include/linux/dma-attrs.h
index 99c0be00b47c..678662a235d1 100644
--- a/include/linux/dma-attrs.h
+++ b/include/linux/dma-attrs.h
@@ -18,6 +18,7 @@ enum dma_attr {
 	DMA_ATTR_NO_KERNEL_MAPPING,
 	DMA_ATTR_SKIP_CPU_SYNC,
 	DMA_ATTR_FORCE_CONTIGUOUS,
+	DMA_ATTR_NOHUGEPAGE,
 	DMA_ATTR_MAX,
 };
 
-- 
2.6.0.rc2.230.g3dd15c0