Re: virtio(-scsi) vs. chained sg_lists (was Re: [PATCH] scsi: virtio-scsi:Fix address translation failure of HighMem pages used by sg list)

[Paolo Bonzini]

This is a multi-part message in MIME format.Il 26/07/2012 09:58, Paolo Bonzini ha scritto:
> 
>> > Please CC me on the "convert to sg copy-less" patches, It looks interesting
> Sure.

Well, here is the gist of it (note it won't apply on any public tree,
hence no SoB yet).  It should be split in multiple changesets and you
can make more simplifications on top of it, because
virtio_scsi_target_state is not anymore variable-sized, but that's
secondary.

The patch includes the conversion of virtio_ring.c to use sg_next.
It is a bit ugly because of backwards-compatibility, it can be fixed
by going through all drivers; not that there are many.

I'm still a bit worried of the future of this feature though.  I would
be the first and (because of the non-portability) presumably sole user
for some time.  Someone axing chained sg-lists in the future does not
seem too unlikely.  So in practice I would rather just add to virtio a
function that takes two struct scatterlist ** (or an array of struct
scatterlist * + size pairs).  Converting to sg_chain once it takes off
would be trivial.

Paolo