aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorDan Williams <dan.j.williams@intel.com>2009-07-01 19:12:53 -0400
committerDan Williams <dan.j.williams@intel.com>2009-07-01 19:12:53 -0400
commitdaf4219dbcbb2efcd638fcd3c29a622e1c18cc38 (patch)
tree8683d86a8a8de2060c83aed7efe5f66b94027a2b
parent43a1a3ed6bf5a1b9ae197b4f5f20033baf19db61 (diff)
dmaengine: move HIGHMEM64G restriction to ASYNC_TX_DMA
On HIGHMEM64G systems dma_addr_t is known to be larger than (void *) which precludes async_xor from performing dma address conversions by reusing the input parameter address list. However, other parts of the dmaengine infrastructure do not suffer this constraint, so the HIGHMEM64G restriction can be down-levelled. Signed-off-by: Dan Williams <dan.j.williams@intel.com>
-rw-r--r--crypto/async_tx/async_xor.c2
-rw-r--r--drivers/dma/Kconfig4
2 files changed, 3 insertions, 3 deletions
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 95fe2c8d6c51..90dd3f8bd283 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -300,7 +300,7 @@ EXPORT_SYMBOL_GPL(async_xor_zero_sum);
300 300
301static int __init async_xor_init(void) 301static int __init async_xor_init(void)
302{ 302{
303 #ifdef CONFIG_DMA_ENGINE 303 #ifdef CONFIG_ASYNC_TX_DMA
304 /* To conserve stack space the input src_list (array of page pointers) 304 /* To conserve stack space the input src_list (array of page pointers)
305 * is reused to hold the array of dma addresses passed to the driver. 305 * is reused to hold the array of dma addresses passed to the driver.
306 * This conversion is only possible when dma_addr_t is less than the 306 * This conversion is only possible when dma_addr_t is less than the
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 3b3c01b6f1ee..babf214a509b 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -4,7 +4,7 @@
4 4
5menuconfig DMADEVICES 5menuconfig DMADEVICES
6 bool "DMA Engine support" 6 bool "DMA Engine support"
7 depends on !HIGHMEM64G && HAS_DMA 7 depends on HAS_DMA
8 help 8 help
9 DMA engines can do asynchronous data transfers without 9 DMA engines can do asynchronous data transfers without
10 involving the host CPU. Currently, this framework can be 10 involving the host CPU. Currently, this framework can be
@@ -100,7 +100,7 @@ config NET_DMA
100 100
101config ASYNC_TX_DMA 101config ASYNC_TX_DMA
102 bool "Async_tx: Offload support for the async_tx api" 102 bool "Async_tx: Offload support for the async_tx api"
103 depends on DMA_ENGINE 103 depends on DMA_ENGINE && !HIGHMEM64G
104 help 104 help
105 This allows the async_tx api to take advantage of offload engines for 105 This allows the async_tx api to take advantage of offload engines for
106 memcpy, memset, xor, and raid6 p+q operations. If your platform has 106 memcpy, memset, xor, and raid6 p+q operations. If your platform has