Prev: Downsides to madvise/fadvise(willneed) for application startup
Next: futex: FUTEX_LOCK with optional adaptive spinning
From: Andrew Morton on 5 Apr 2010 19:40 On Sun, 4 Apr 2010 23:25:50 +0900 FUJITA Tomonori <fujita.tomonori(a)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 _ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: FUJITA Tomonori on 5 Apr 2010 19:50 On Mon, 5 Apr 2010 16:35:41 -0700 Andrew Morton <akpm(a)linux-foundation.org> wrote: > On Sun, 4 Apr 2010 23:25:50 +0900 > FUJITA Tomonori <fujita.tomonori(a)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, -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: FUJITA Tomonori on 5 Apr 2010 20:10
On Tue, 6 Apr 2010 08:46:38 +0900 FUJITA Tomonori <fujita.tomonori(a)lab.ntt.co.jp> wrote: > On Mon, 5 Apr 2010 16:35:41 -0700 > Andrew Morton <akpm(a)linux-foundation.org> wrote: > > > On Sun, 4 Apr 2010 23:25:50 +0900 > > FUJITA Tomonori <fujita.tomonori(a)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. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |