qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Nikolay Borisov <nborisov@suse.com>,
	dgilbert@redhat.com, qemu-devel@nongnu.org, jfehlig@suse.com,
	Claudio.Fontana@suse.com, dfaggioli@suse.com
Subject: Re: [PATCH v3 04/14] io: Add generic pwritev/preadv interface
Date: Fri, 10 Feb 2023 16:58:04 +0000	[thread overview]
Message-ID: <Y+Z3nGUC/Xmw6ULQ@redhat.com> (raw)
In-Reply-To: <Y+ZwLm52JzNaGT8N@redhat.com>

On Fri, Feb 10, 2023 at 04:26:22PM +0000, Daniel P. Berrangé wrote:
> On Fri, Oct 28, 2022 at 01:39:04PM +0300, Nikolay Borisov wrote:
> > Introduce basic pwriteve/preadv support in the generic channel layer.
> > SPecific implementation will follow for the file channel as this is
> > required in order to support migration streams with fixed location of
> > each ram page.
> > 
> > Signed-off-by: Nikolay Borisov <nborisov@suse.com>
> > ---
> >  include/io/channel.h | 49 ++++++++++++++++++++++++++++++++++++++++++++
> >  io/channel.c         | 26 +++++++++++++++++++++++
> >  2 files changed, 75 insertions(+)
> > 
> > diff --git a/include/io/channel.h b/include/io/channel.h
> > index c680ee748021..6b10bce8bbdf 100644
> > --- a/include/io/channel.h
> > +++ b/include/io/channel.h
> > @@ -124,6 +124,16 @@ struct QIOChannelClass {
> >                             Error **errp);
> >  
> >      /* Optional callbacks */
> > +    ssize_t (*io_pwritev)(QIOChannel *ioc,
> > +                       const struct iovec *iov,
> > +                       size_t niov,
> > +                       off_t offset,
> > +                       Error **errp);
> > +    ssize_t (*io_preadv)(QIOChannel *ioc,
> > +                      const struct iovec *iov,
> > +                      size_t niov,
> > +                      off_t offset,
> > +                      Error **errp);
> 
> nit-pick the alignment of the 2nd line of parameters onwards,
> needs to be indented by 3 more spaces.
> 
> 
> > +/**
> > + * qio_channel_io_pwritev
> > + * @ioc: the channel object
> > + * @iov: the array of memory regions to write data from
> > + * @niov: the length of the @iov array
> > + * @offset: offset in the channel where writes should begin
> > + * @errp: pointer to a NULL-initialized error object
> > + *
> > + * Not all implementations will support this facility, so may report an error.
> > + * To avoid errors, the caller may check for the feature flag
> > + * QIO_CHANNEL_FEATURE_SEEKABLE prior to calling this method.
> > + *
> > + * Behaves as qio_channel_writev_full, apart from not supporting sending of file
> 
> Given this, we should have  "_full" suffix on these methods
> too for consistent naming policy.
> 
> > + * handles as well as beginning the write at the passed @offset
> > + *
> > + */
> > +ssize_t qio_channel_io_pwritev(QIOChannel *ioc, const struct iovec *iov,
> > +                               size_t niov, off_t offset, Error **errp);

In addition to giving this method the '_full' suffix, we should also
add the qio_channel_io_pwritev entrypoint which takes a single
buffer instead of iovec. The migration code in QEMUFile that is
added in a later patch in this series doesn't actually want to
use iovecs in the first place. So having the non-iov entrypoints
in QIOChannel will simplify the migration code.

> > +
> > +
> > +/**
> > + * qio_channel_io_preadv
> > + * @ioc: the channel object
> > + * @iov: the array of memory regions to read data into
> > + * @niov: the length of the @iov array
> > + * @offset: offset in the channel where writes should begin
> > + * @errp: pointer to a NULL-initialized error object
> > + *
> > + * Not all implementations will support this facility, so may report an error.
> > + * To avoid errors, the caller may check for the feature flag
> > + * QIO_CHANNEL_FEATURE_SEEKABLE prior to calling this method.
> > + *
> > + * Behaves as qio_channel_readv_full, apart from not supporting receiving of file
> > + * handles as well as beginning the read at the passed @offset
> > + *
> > + */
> > +ssize_t qio_channel_io_preadv(QIOChannel *ioc, const struct iovec *iov,
> > +                             size_t niov, off_t offset, Error **errp);
> >  /**
> >   * qio_channel_shutdown:
> >   * @ioc: the channel object
> > diff --git a/io/channel.c b/io/channel.c
> > index 0640941ac573..f5ac9499a7ad 100644
> > --- a/io/channel.c
> > +++ b/io/channel.c
> > @@ -437,6 +437,32 @@ GSource *qio_channel_add_watch_source(QIOChannel *ioc,
> >  }
> >  
> >  
> > +ssize_t qio_channel_io_pwritev(QIOChannel *ioc, const struct iovec *iov,
> > +                               size_t niov, off_t offset, Error **errp)
> > +{
> > +    QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
> > +
> > +    if (!klass->io_pwritev) {
> > +        error_setg(errp, "Channel does not support pwritev");
> > +        return -1;
> > +    }
> 
> This also possibly benefits from a validation check
> 
>     if (!qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_SEEKABLE)) {
>         error_setg_errno(errp, EINVAL,
>                          "Requested channel is not seekable")
>         return -1;
>     }
> 
> because the QIOChannelFile impl will support io_pwritev callback,
> but not all the FDs it can use will support seeking.
> 
> Same check for preadv
> 
> > +
> > +    return klass->io_pwritev(ioc, iov, niov, offset, errp);
> > +}
> > +
> > +ssize_t qio_channel_io_preadv(QIOChannel *ioc, const struct iovec *iov,
> > +                               size_t niov, off_t offset, Error **errp)
> > +{
> > +    QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
> > +
> > +    if (!klass->io_preadv) {
> > +        error_setg(errp, "Channel does not support preadv");
> > +        return -1;
> > +    }
> > +
> > +    return klass->io_preadv(ioc, iov, niov, offset, errp);
> > +}
> > +
> >  int qio_channel_shutdown(QIOChannel *ioc,
> >                           QIOChannelShutdown how,
> >                           Error **errp)
> > -- 
> > 2.34.1
> > 
> 
> 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 :|
> 
> 

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 :|



  reply	other threads:[~2023-02-10 16:59 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-28 10:39 [PATCH v3 00/14] File-based migration support and fixed-ram features Nikolay Borisov
2022-10-28 10:39 ` [PATCH v3 01/14] migration: support file: uri for source migration Nikolay Borisov
2022-10-28 10:39 ` [PATCH v3 02/14] migration: Add support for 'file:' uri for incoming migration Nikolay Borisov
2023-02-10 15:58   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 03/14] migration: Initial support of fixed-ram feature for analyze-migration.py Nikolay Borisov
2023-02-10 16:13   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 04/14] io: Add generic pwritev/preadv interface Nikolay Borisov
2023-02-10 16:26   ` Daniel P. Berrangé
2023-02-10 16:58     ` Daniel P. Berrangé [this message]
2022-10-28 10:39 ` [PATCH v3 05/14] io: implement io_pwritev for QIOChannelFile Nikolay Borisov
2023-02-10 16:34   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 06/14] io: add and implement QIO_CHANNEL_FEATURE_SEEKABLE for channel file Nikolay Borisov
2023-02-10 16:38   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 07/14] migration/qemu-file: add utility methods for working with seekable channels Nikolay Borisov
2022-10-28 10:39 ` [PATCH v3 08/14] io: Add preadv support to QIOChannelFile Nikolay Borisov
2023-02-10 16:59   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 09/14] migration: add qemu_get_buffer_at Nikolay Borisov
2023-02-10 16:59   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 10/14] migration/ram: Introduce 'fixed-ram' migration stream capability Nikolay Borisov
2023-02-10 17:11   ` Daniel P. Berrangé
2023-03-20 11:05     ` Claudio Fontana
2023-03-20 11:08       ` Claudio Fontana
2022-10-28 10:39 ` [PATCH v3 11/14] migration: Refactor precopy ram loading code Nikolay Borisov
2022-10-28 10:39 ` [PATCH v3 12/14] migration: Add support for 'fixed-ram' migration restore Nikolay Borisov
2022-10-28 10:39 ` [PATCH v3 13/14] tests: Add migrate_incoming_qmp helper Nikolay Borisov
2023-02-10 17:13   ` Daniel P. Berrangé
2022-10-28 10:39 ` [PATCH v3 14/14] tests/qtest: migration-test: Add tests for file-based migration Nikolay Borisov
2023-02-10 17:17   ` Daniel P. Berrangé
2023-02-09 13:32 ` [PATCH v3 00/14] File-based migration support and fixed-ram features Claudio Fontana
2023-02-10 15:35   ` Daniel P. Berrangé
2023-03-20 11:14     ` Claudio Fontana
2023-03-20 11:25       ` Daniel P. Berrangé

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Y+Z3nGUC/Xmw6ULQ@redhat.com \
    --to=berrange@redhat.com \
    --cc=Claudio.Fontana@suse.com \
    --cc=dfaggioli@suse.com \
    --cc=dgilbert@redhat.com \
    --cc=jfehlig@suse.com \
    --cc=nborisov@suse.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).