Jan Kara <jack@xxxxxxx> writes:
> On Sun 16-04-23 15:38:40, Ritesh Harjani (IBM) wrote:
>> This patch converts ext2 direct-io path to iomap interface.
>> - This also takes care of DIO_SKIP_HOLES part in which we return -ENOTBLK
>> from ext2_iomap_begin(), in case if the write is done on a hole.
>> - This fallbacks to buffered-io in case of DIO_SKIP_HOLES or in case of
>> a partial write or if any error is detected in ext2_iomap_end().
>> We try to return -ENOTBLK in such cases.
>> - For any unaligned or extending DIO writes, we pass
>> IOMAP_DIO_FORCE_WAIT flag to ensure synchronous writes.
>> - For extending writes we set IOMAP_F_DIRTY in ext2_iomap_begin because
>> otherwise with dsync writes on devices that support FUA, generic_write_sync
>> won't be called and we might miss inode metadata updates.
>> - Since ext2 already now uses _nolock vartiant of sync write. Hence
>> there is no inode lock problem with iomap in this patch.
>> - ext2_iomap_ops are now being shared by DIO, DAX & fiemap path
>>
>> Tested-by: Disha Goel <disgoel@xxxxxxxxxxxxx>
>> Reviewed-by: Christoph Hellwig <hch@xxxxxx>
>> Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@xxxxxxxxx>
>
> One comment below:
>
>> @@ -844,6 +868,13 @@ static int
>> ext2_iomap_end(struct inode *inode, loff_t offset, loff_t length,
>> ssize_t written, unsigned flags, struct iomap *iomap)
>> {
>> + /*
>> + * Switch to buffered-io in case of any error.
>> + * Blocks allocated can be used by the buffered-io path.
>> + */
>> + if ((flags & IOMAP_DIRECT) && (flags & IOMAP_WRITE) && written == 0)
>> + return -ENOTBLK;
>> +
>> if (iomap->type == IOMAP_MAPPED &&
>> written < length &&
>> (flags & IOMAP_WRITE))
>
> Is this really needed? What for?
>
Sorry Jan, I got caught into something else so couldn't respond on this
earlier. Thanks a lot for review.
I don't think this will be called for IOMAP_DIRECT for write case.
I mostly see this code was already present for IOMAP_DAX path.
It is to truncate the blocks in case if the iomap dax write failed to
write but the blocks might have been allocated in ->iomap_begin
function.
Is there a specific query that you would like me to check and verify?
I can check more by probing this path to see what happens when this gets
called. But my understanding was it is used for truncating blocks as I
mentioned above.
Thanks
-ritesh
