Commit 4b0228fa authored by Vishal Verma's avatar Vishal Verma
Browse files

dax: for truncate/hole-punch, do zeroing through the driver if possible 



In the truncate or hole-punch path in dax, we clear out sub-page ranges.
If these sub-page ranges are sector aligned and sized, we can do the
zeroing through the driver instead so that error-clearing is handled
automatically.

For sub-sector ranges, we still have to rely on clear_pmem and have the
possibility of tripping over errors.

Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Jan Kara <jack@suse.cz>
Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
Reviewed-by: default avatarJan Kara <jack@suse.cz>
Signed-off-by: default avatarVishal Verma <vishal.l.verma@intel.com>
parent 679c8bd3

Documentation/filesystems/dax.txt

+32 −0
Original line number Diff line number Diff line
@@ -79,6 +79,38 @@ These filesystems may be used for inspiration: 
- ext4: the fourth extended filesystem, see Documentation/filesystems/ext4.txt





Handling Media Errors

---------------------



The libnvdimm subsystem stores a record of known media error locations for

each pmem block device (in gendisk->badblocks). If we fault at such location,

or one with a latent error not yet discovered, the application can expect

to receive a SIGBUS. Libnvdimm also allows clearing of these errors by simply

writing the affected sectors (through the pmem driver, and if the underlying

NVDIMM supports the clear_poison DSM defined by ACPI).



Since DAX IO normally doesn't go through the driver/bio path, applications or

sysadmins have an option to restore the lost data from a prior backup/inbuilt

redundancy in the following ways:



1. Delete the affected file, and restore from a backup (sysadmin route):

   This will free the file system blocks that were being used by the file,

   and the next time they're allocated, they will be zeroed first, which

   happens through the driver, and will clear bad sectors.



2. Truncate or hole-punch the part of the file that has a bad-block (at least

   an entire aligned sector has to be hole-punched, but not necessarily an

   entire filesystem block).



These are the two basic paths that allow DAX filesystems to continue operating

in the presence of media errors. More robust error recovery mechanisms can be

built on top of this in the future, for example, involving redundancy/mirroring

provided at the block layer through DM, or additionally, at the filesystem

level. These would have to rely on the above two tenets, that error clearing

can happen either by sending an IO through the driver, or zeroing (also through

the driver).





Shortcomings

------------

fs/dax.c

+25 −5
Original line number Diff line number Diff line
@@ -947,6 +947,19 @@ int dax_pfn_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) 
}

EXPORT_SYMBOL_GPL(dax_pfn_mkwrite);



static bool dax_range_is_aligned(struct block_device *bdev,

				 unsigned int offset, unsigned int length)

{

	unsigned short sector_size = bdev_logical_block_size(bdev);



	if (!IS_ALIGNED(offset, sector_size))

		return false;

	if (!IS_ALIGNED(length, sector_size))

		return false;



	return true;

}



int __dax_zero_page_range(struct block_device *bdev, sector_t sector,

		unsigned int offset, unsigned int length)

{
@@ -955,11 +968,18 @@ int __dax_zero_page_range(struct block_device *bdev, sector_t sector, 
		.size		= PAGE_SIZE,

	};



	if (dax_range_is_aligned(bdev, offset, length)) {

		sector_t start_sector = dax.sector + (offset >> 9);



		return blkdev_issue_zeroout(bdev, start_sector,

				length >> 9, GFP_NOFS, true);

	} else {

		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);

	}

	return 0;

}

EXPORT_SYMBOL_GPL(__dax_zero_page_range);

CRA Git | Maintained and supported by SUSTech CRA and CCSE