Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[PATCH -mm 3/5] powerpc: use asm-generic/scatterlist.h

1 view
Skip to first unread message

FUJITA Tomonori

unread,
Apr 4, 2010, 10:50:02 AM4/4/10
to
Signed-off-by: FUJITA Tomonori <fujita....@lab.ntt.co.jp>
Cc: Benjamin Herrenschmidt <be...@kernel.crashing.org>
Cc: Paul Mackerras <pau...@samba.org>
---
arch/powerpc/Kconfig | 3 +++
arch/powerpc/include/asm/scatterlist.h | 29 +----------------------------
2 files changed, 4 insertions(+), 28 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 669a1b1..0133cf9 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -659,6 +659,9 @@ config ZONE_DMA
config NEED_DMA_MAP_STATE
def_bool (PPC64 || NOT_COHERENT_CACHE)

+config NEED_SG_DMA_LENGTH
+ def_bool y
+
config GENERIC_ISA_DMA
bool
depends on PPC64 || POWER4 || 6xx && !CPM2
diff --git a/arch/powerpc/include/asm/scatterlist.h b/arch/powerpc/include/asm/scatterlist.h
index 912bf59..4ae35da 100644
--- a/arch/powerpc/include/asm/scatterlist.h
+++ b/arch/powerpc/include/asm/scatterlist.h
@@ -9,38 +9,11 @@
* 2 of the License, or (at your option) any later version.
*/

-#ifdef __KERNEL__
-#include <linux/types.h>
#include <asm/dma.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
- unsigned long sg_magic;
-#endif
- unsigned long page_link;
- unsigned int offset;
- unsigned int length;
-
- /* For TCE or SWIOTLB support */
- dma_addr_t dma_address;
- u32 dma_length;
-};
-
-/*
- * These macros should be used after a dma_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns, or alternatively stop on the first sg_dma_len(sg) which
- * is 0.
- */
-#define sg_dma_address(sg) ((sg)->dma_address)
-#define sg_dma_len(sg) ((sg)->dma_length)
+#include <asm-generic/scatterlist.h>

#ifdef __powerpc64__
#define ISA_DMA_THRESHOLD (~0UL)
#endif

-#define ARCH_HAS_SG_CHAIN
-
-#endif /* __KERNEL__ */
#endif /* _ASM_POWERPC_SCATTERLIST_H */
--
1.7.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

FUJITA Tomonori

unread,
Apr 4, 2010, 10:50:02 AM4/4/10
to
There are only two ways to define sg_dma_len(); use sg->dma_length or
sg->length. This patch introduces NEED_SG_DMA_LENGTH that enables
architectures to choose sg->dma_length or sg->length.

Signed-off-by: FUJITA Tomonori <fujita....@lab.ntt.co.jp>
Cc: Arnd Bergmann <ar...@arndb.de>
---
arch/ia64/Kconfig | 3 +++
arch/sh/Kconfig | 3 +++
arch/sparc/Kconfig | 3 +++
arch/sparc/include/asm/scatterlist.h | 2 --
arch/x86/Kconfig | 3 +++
include/asm-generic/scatterlist.h | 13 +++++--------
6 files changed, 17 insertions(+), 10 deletions(-)

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 0e5919f..0873a10 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -56,6 +56,9 @@ config MMU
config NEED_DMA_MAP_STATE
def_bool y



+config NEED_SG_DMA_LENGTH
+ def_bool y
+

config SWIOTLB
bool

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index e2e5b3b..5e2ee0c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -183,6 +183,9 @@ config DMA_NONCOHERENT
config NEED_DMA_MAP_STATE
def_bool DMA_NONCOHERENT



+config NEED_SG_DMA_LENGTH
+ def_bool y
+

source "init/Kconfig"

source "kernel/Kconfig.freezer"
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 54ce709..5b751a5 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -127,6 +127,9 @@ config ZONE_DMA
config NEED_DMA_MAP_STATE
def_bool y



+config NEED_SG_DMA_LENGTH
+ def_bool y
+
config GENERIC_ISA_DMA
bool

default y if SPARC32
diff --git a/arch/sparc/include/asm/scatterlist.h b/arch/sparc/include/asm/scatterlist.h
index 0fa0d6d..69d21bb 100644
--- a/arch/sparc/include/asm/scatterlist.h
+++ b/arch/sparc/include/asm/scatterlist.h
@@ -1,8 +1,6 @@
#ifndef _SPARC_SCATTERLIST_H
#define _SPARC_SCATTERLIST_H

-#define sg_dma_len(sg) ((sg)->dma_length)
-
#define ISA_DMA_THRESHOLD (~0UL)

#include <asm-generic/scatterlist.h>
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 8b182ea..6dc69c4 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -102,6 +102,9 @@ config SBUS
config NEED_DMA_MAP_STATE
def_bool (X86_64 || DMAR || DMA_API_DEBUG)

+config NEED_SG_DMA_LENGTH
+ def_bool X86_64
+
config GENERIC_ISA_DMA
def_bool y

diff --git a/include/asm-generic/scatterlist.h b/include/asm-generic/scatterlist.h
index 51a7a43..5e08794 100644
--- a/include/asm-generic/scatterlist.h
+++ b/include/asm-generic/scatterlist.h
@@ -11,7 +11,9 @@ struct scatterlist {
unsigned int offset;
unsigned int length;
dma_addr_t dma_address;
+#ifdef CONFIG_NEED_SG_DMA_LENGTH
unsigned int dma_length;
+#endif
};

/*
@@ -22,17 +24,12 @@ struct scatterlist {
* is 0.
*/
#define sg_dma_address(sg) ((sg)->dma_address)
-#ifndef sg_dma_len
-/*
- * Normally, you have an iommu on 64 bit machines, but not on 32 bit
- * machines. Architectures that are differnt should override this.
- */
-#if __BITS_PER_LONG == 64
+
+#ifdef CONFIG_NEED_SG_DMA_LENGTH
#define sg_dma_len(sg) ((sg)->dma_length)
#else
#define sg_dma_len(sg) ((sg)->length)
-#endif /* 64 bit */
-#endif /* sg_dma_len */
+#endif

#define ARCH_HAS_SG_CHAIN

Andrew Morton

unread,
Apr 5, 2010, 7:40:01 PM4/5/10
to
On Sun, 4 Apr 2010 23:25:50 +0900
FUJITA Tomonori <fujita....@lab.ntt.co.jp> wrote:

> There are only two ways to define sg_dma_len(); use sg->dma_length or
> sg->length. This patch introduces NEED_SG_DMA_LENGTH that enables
> architectures to choose sg->dma_length or sg->length.

i386 allnoconfig:

arch/x86/kernel/pci-nommu.c: In function 'nommu_map_sg':
arch/x86/kernel/pci-nommu.c:71: error: 'struct scatterlist' has no member named 'dma_length'

So I enabled NEED_SG_DMA_LENGTH on x86_32 as well. It compiles...

--- a/arch/x86/Kconfig~asm-generic-add-need_sg_dma_length-to-define-sg_dma_len-fix
+++ a/arch/x86/Kconfig
@@ -103,7 +103,7 @@ config NEED_DMA_MAP_STATE
def_bool (X86_64 || DMAR || DMA_API_DEBUG)

config NEED_SG_DMA_LENGTH
- def_bool X86_64
+ def_bool y

config GENERIC_ISA_DMA
def_bool y
diff -puN include/asm-generic/scatterlist.h~asm-generic-add-need_sg_dma_length-to-define-sg_dma_len-fix include/asm-generic/scatterlist.h
_

FUJITA Tomonori

unread,
Apr 5, 2010, 7:50:02 PM4/5/10
to
On Mon, 5 Apr 2010 16:35:41 -0700
Andrew Morton <ak...@linux-foundation.org> wrote:

> On Sun, 4 Apr 2010 23:25:50 +0900
> FUJITA Tomonori <fujita....@lab.ntt.co.jp> wrote:
>
> > There are only two ways to define sg_dma_len(); use sg->dma_length or
> > sg->length. This patch introduces NEED_SG_DMA_LENGTH that enables
> > architectures to choose sg->dma_length or sg->length.
>
> i386 allnoconfig:
>
> arch/x86/kernel/pci-nommu.c: In function 'nommu_map_sg':
> arch/x86/kernel/pci-nommu.c:71: error: 'struct scatterlist' has no member named 'dma_length'
>
> So I enabled NEED_SG_DMA_LENGTH on x86_32 as well. It compiles...

Sorry about that. The fix is fine by me. We can remove dma_length in
pci-nommu.c though (there is no point of setting dma_length here).

Thanks,

FUJITA Tomonori

unread,
Apr 5, 2010, 8:10:02 PM4/5/10
to
On Tue, 6 Apr 2010 08:46:38 +0900
FUJITA Tomonori <fujita....@lab.ntt.co.jp> wrote:

> On Mon, 5 Apr 2010 16:35:41 -0700
> Andrew Morton <ak...@linux-foundation.org> wrote:
>
> > On Sun, 4 Apr 2010 23:25:50 +0900
> > FUJITA Tomonori <fujita....@lab.ntt.co.jp> wrote:
> >
> > > There are only two ways to define sg_dma_len(); use sg->dma_length or
> > > sg->length. This patch introduces NEED_SG_DMA_LENGTH that enables
> > > architectures to choose sg->dma_length or sg->length.
> >
> > i386 allnoconfig:
> >
> > arch/x86/kernel/pci-nommu.c: In function 'nommu_map_sg':
> > arch/x86/kernel/pci-nommu.c:71: error: 'struct scatterlist' has no member named 'dma_length'
> >
> > So I enabled NEED_SG_DMA_LENGTH on x86_32 as well. It compiles...
>
> Sorry about that. The fix is fine by me. We can remove dma_length in
> pci-nommu.c though (there is no point of setting dma_length here).

Oops, I forgot that pci-nommu.c is also used on x86_64.

There is no point of setting dma_length here on x86_32 but we need to
do on x86_64 since x86_64 uses dma_length. Your fix is correct.

Sorry again.

0 new messages