On Wed, Aug 22, 2012 at 02:07:40PM -0700, Tejun Heo wrote:

On Wed, Aug 22, 2012 at 10:04:09AM -0700, Kent Overstreet wrote:

quoted

Previously, there was bio_clone() but it only allocated from the fs bio
set; as a result various users were open coding it and using
__bio_clone().

This changes bio_clone() to become bio_clone_bioset(), and then we add
bio_clone() and bio_clone_kmalloc() as wrappers around it, making use of
the functionality the last patch adedd.

This will also help in a later patch changing how bio cloning works.

I'd prefer simply adding @bioset to bio_clone() so that the caller
always has to make the choice consciously.  We're updating all the
callers anyway.

Possibly, but the btrfs code uses bio_clone() and there fs_bio_set may
be correct (will have to look at what it's doing, if it's cloning a bio
that was allocated out of fs_bio_set that would be bad..)

I would also prefer to simply drop bio_clone() so that
bio_clone_bioset() matches bio_alloc_bioset(), but regardless that'll
have to be a different patch (and I don't think I've had to update any
of the bio_clone() callers in this patch series anyways).