Is there any thing I left out in xfs related to cache coherency.
On Thu, Apr 28, 2011 at 12:59 PM, Ajeet Yadav <ajeet.yadav.77@xxxxxxxxx> wrote:
> MIPS32Â 34Kâ Core
> It does not provide Âinvalidate_kernel_vmap_range() /
> flush_kernel_vmap_range() to deal with cache coherency problem
> Therefore we provided dma_cache_inv() / dma_cache_wback_inv() , after
> that we did not had any coherency problem
> But as you saying it still has cache coherency problem, wrt xfstests 001 ?
>
> diff -Nurp -X linux-2.6.35.11/Documentation/dontdiff
> linux-2.6.35.11/fs/xfs/linux-2.6/xfs_buf.c
> linux-2.6.35.11-dirty/fs/xfs/linux-2.6/xfs_buf.c
> --- linux-2.6.35.11/fs/xfs/linux-2.6/xfs_buf.c Â2011-02-07
> 04:04:07.000000000 +0900
> +++ linux-2.6.35.11-dirty/fs/xfs/linux-2.6/xfs_buf.c  Â2011-03-22
> 18:29:09.000000000 +0900
> @@ -1192,7 +1192,7 @@ xfs_buf_bio_end_io(
> Â Â Â Âxfs_buf_ioerror(bp, -error);
>
> Â Â Â Âif (!error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ))
> - Â Â Â Â Â Â Â invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp));
> + Â Â Â Â Â Â Â dma_cache_inv((unsigned long)bp->b_addr, xfs_buf_vmap_len(bp));
>
> Â Â Â Âdo {
>        Âstruct page   *page = bvec->bv_page;
> @@ -1304,7 +1304,7 @@ next_chunk:
> Âsubmit_io:
> Â Â Â Âif (likely(bio->bi_size)) {
> Â Â Â Â Â Â Â Âif (xfs_buf_is_vmapped(bp)) {
> - Â Â Â Â Â Â Â Â Â Â Â flush_kernel_vmap_range(bp->b_addr,
> + Â Â Â Â Â Â Â Â Â Â Â dma_cache_wback_inv((unsigned long)bp->b_addr,
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Âxfs_buf_vmap_len(bp));
> Â Â Â Â Â Â Â Â}
> Â Â Â Â Â Â Â Âsubmit_bio(rw, bio);
>
> On Wed, Apr 27, 2011 at 10:41 PM, Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote:
>> Does 001 also fail for say ext2 and ext3?
>>
>>
>
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
