On Fri, Apr 04, 2025 at 01:49:44PM +0530, Prasad Pandit wrote:
> On Thu, 27 Mar 2025 at 20:03, Marco Cavenati <marco.caven...@eurecom.fr>
> wrote:
> > Enable the use of the mapped-ram migration feature with savevm/loadvm
> > snapshots by adding the QIO_CHANNEL_FEATURE_SEEKABLE feature to
> > QIOChannelBlock. Implement io_preadv and io_pwritev methods to provide
> > positioned I/O capabilities that don't modify the channel's position
> > pointer.
> >
> > migration/channel-block.c | 48 +++++++++++++++++++++++++++++++++++++++
> > 1 file changed, 48 insertions(+)
> >
> > +++ b/migration/channel-block.c
> > @@ -30,6 +30,7 @@ qio_channel_block_new(BlockDriverState *bs)
> > QIOChannelBlock *ioc;
> >
> > ioc = QIO_CHANNEL_BLOCK(object_new(TYPE_QIO_CHANNEL_BLOCK));
> > + qio_channel_set_feature(QIO_CHANNEL(ioc),
> > QIO_CHANNEL_FEATURE_SEEKABLE);
> >
>
> * IIUC, when _FEATURE_SEEKABLE is set, the channel I/O sequence
> eventually makes underlying preadv(2)/pwritev(2) calls, which use
> lseek(2) to adjust the stream r/w pointer with the given offset,
> before doing the r/w operation.
>
> * QIOChannelBlock has '.io_seek' routine corresponding to the lseek(2)
> call. It is not clear how setting QIO_CHANNEL_FEATURE_SEEKABLE helps
> for QIOChannelBlock streams.
In QIOChanel API specification, having QIO_CHANNEL_FEATURE_SEEKABLE
set is a pre-requisite for use of qio_channel_{pread,preadv,pwrite,pwritev}.
It is actually NOT really connected to lseek, and as such
QIO_CHANNEL_FEATURE_SEEKABLE is probably a bad choice of name
in retrospect.
With regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
