On Tue, 2016-05-10 at 12:49 -0600, Vishal Verma wrote:
>Â
...
> @@ -1240,11 +1254,16 @@ int dax_zero_page_range(struct inode *inode,
> loff_t from, unsigned length,
> Â .size = PAGE_SIZE,
> Â };
> Â
> - if (dax_map_atomic(bdev, &dax) < 0)
> - return PTR_ERR(dax.addr);
> - clear_pmem(dax.addr + offset, length);
> - wmb_pmem();
> - dax_unmap_atomic(bdev, &dax);
> + if (dax_range_is_aligned(bdev, &dax, offset, length))
> + return blkdev_issue_zeroout(bdev, dax.sector,
> + length >> 9, GFP_NOFS, true);
Found another bug here while testing. The zeroout needs to be done for
sector + (offset >> 9). The above just zeroed out the first sector of
the page irrespective of offset, which is wrong.
|