Re: [PATCH -mm 09/12] blackfin: use use asm-generic/scatterlist.h

From: Mike Frysinger
Date: Wed Apr 07 2010 - 23:51:45 EST


On Mon, Apr 5, 2010 at 19:52, FUJITA Tomonori wrote:
> Signed-off-by: FUJITA Tomonori <fujita.tomonori@xxxxxxxxxxxxx>
> Cc: Mike Frysinger <vapier@xxxxxxxxxx>
> ---
> Âarch/blackfin/include/asm/scatterlist.h | Â 22 +---------------------
> Â1 files changed, 1 insertions(+), 21 deletions(-)
>
> diff --git a/arch/blackfin/include/asm/scatterlist.h b/arch/blackfin/include/asm/scatterlist.h
> index 04f4487..64d41d3 100644
> --- a/arch/blackfin/include/asm/scatterlist.h
> +++ b/arch/blackfin/include/asm/scatterlist.h
> @@ -1,27 +1,7 @@
> Â#ifndef _BLACKFIN_SCATTERLIST_H
> Â#define _BLACKFIN_SCATTERLIST_H
>
> -#include <linux/mm.h>
> -
> -struct scatterlist {
> -#ifdef CONFIG_DEBUG_SG
> - Â Â Â unsigned long sg_magic;
> -#endif
> - Â Â Â unsigned long page_link;
> - Â Â Â unsigned int offset;
> - Â Â Â dma_addr_t dma_address;
> - Â Â Â unsigned int length;
> -};
> -
> -/*
> - * These macros should be used after a pci_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) Â Â Â((sg)->dma_address)
> -#define sg_dma_len(sg) Â Â Â Â Â((sg)->length)
> +#include <asm-generic/scatterlist.h>
>
> Â#define ISA_DMA_THRESHOLD Â Â Â(0xffffffff)

this is the same value as the common asm-generic one, so this define
should be punted, especially since this would probably introduce
redefined warnings.

once that is done, this header is simply a redirect, so it'd be better
if it was just one line:
#include <asm-generic/scatterlist.h>

one of the reasons i hadnt converted sooner was that it seems like the
common header introduces bloat. the Blackfin header has just "length"
while the asm-generic has both "length" and "dma_length". unless i
missed something, this hasnt been a problem for us. perhaps the
common header should have something like:
#if __BITS_PER_LONG == 64
unsigned int dma_length;
#endif
this would match the existing sg_dma_len code. or perhaps declare the
two lengths as an anonymous union.
-mike
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/