qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Benoît Canet" <benoit.canet@irqsave.net>
To: Kevin Wolf <kwolf@redhat.com>
Cc: famz@redhat.com, jcody@redhat.com, qemu-devel@nongnu.org,
	armbru@redhat.com, stefanha@redhat.com
Subject: Re: [Qemu-devel] [PATCH V4 2/7] block: Allow the user to define "node-name" option.
Date: Wed, 11 Dec 2013 15:42:56 +0100	[thread overview]
Message-ID: <20131211144256.GB2947@irqsave.net> (raw)
In-Reply-To: <20131209161526.GI3549@dhcp-200-207.str.redhat.com>

Le Monday 09 Dec 2013 à 17:15:26 (+0100), Kevin Wolf a écrit :
> Am 05.12.2013 um 18:14 hat Benoît Canet geschrieben:
> > Signed-off-by: Benoit Canet <benoit@irqsave.net>
> > ---
> >  block.c | 44 +++++++++++++++++++++++++++++++++++++++-----
> >  1 file changed, 39 insertions(+), 5 deletions(-)
> > 
> > diff --git a/block.c b/block.c
> > index 4f6b36a..61f5ba0 100644
> > --- a/block.c
> > +++ b/block.c
> > @@ -873,6 +873,7 @@ int bdrv_file_open(BlockDriverState **pbs, const char *filename,
> >      const char *drvname;
> >      bool allow_protocol_prefix = false;
> >      Error *local_err = NULL;
> > +    const char *node_name = NULL;
> >      int ret;
> >  
> >      /* NULL means an empty set of options */
> > @@ -880,7 +881,15 @@ int bdrv_file_open(BlockDriverState **pbs, const char *filename,
> >          options = qdict_new();
> >      }
> >  
> > -    bs = bdrv_new("", "");
> > +    node_name = qdict_get_try_str(options, "node-name");
> > +    if (node_name && bdrv_find_node(node_name)) {
> > +        error_setg(errp, "Duplicate node name");
> > +        QDECREF(options);
> > +        return -EINVAL;
> > +    }
> > +    bs = bdrv_new("", node_name ? node_name : "");
> > +    qdict_del(options, "node-name");
> > +
> >      bs->options = options;
> >      options = qdict_clone_shallow(options);
> >  
> > @@ -980,6 +989,7 @@ int bdrv_open_backing_file(BlockDriverState *bs, QDict *options, Error **errp)
> >      int back_flags, ret;
> >      BlockDriver *back_drv = NULL;
> >      Error *local_err = NULL;
> > +    const char *node_name = NULL;
> >  
> >      if (bs->backing_hd != NULL) {
> >          QDECREF(options);
> > @@ -1002,7 +1012,14 @@ int bdrv_open_backing_file(BlockDriverState *bs, QDict *options, Error **errp)
> >                                         sizeof(backing_filename));
> >      }
> >  
> > -    bs->backing_hd = bdrv_new("", "");
> > +    node_name = qdict_get_try_str(options, "node-name");
> > +    if (node_name && bdrv_find_node(node_name)) {
> > +        error_setg(errp, "Duplicate node name");
> > +        QDECREF(options);
> > +        return -EINVAL;
> > +    }
> > +    bs->backing_hd = bdrv_new("", node_name ? node_name : "");
> > +    qdict_del(options, "node-name");
> >  
> >      if (bs->backing_format[0] != '\0') {
> >          back_drv = bdrv_find_format(bs->backing_format);
> > @@ -1046,6 +1063,7 @@ int bdrv_open(BlockDriverState *bs, const char *filename, QDict *options,
> >      BlockDriverState *file = NULL;
> >      QDict *file_options = NULL;
> >      const char *drvname;
> > +    const char *node_name = NULL;
> >      Error *local_err = NULL;
> >  
> >      /* NULL means an empty set of options */
> > @@ -1053,6 +1071,22 @@ int bdrv_open(BlockDriverState *bs, const char *filename, QDict *options,
> >          options = qdict_new();
> >      }
> >  
> > +    node_name = qdict_get_try_str(options, "node-name");
> > +    if (node_name && bdrv_find_node(node_name)) {
> > +        error_setg(errp, "Duplicate node name");
> > +        QDECREF(options);
> > +        return -EINVAL;
> > +    }
> > +
> > +    if (node_name) {
> > +        pstrcpy(bs->node_name, sizeof(bs->node_name), node_name);
> > +        if (node_name[0] != '\0') {
> > +            QTAILQ_INSERT_TAIL(&graph_bdrv_states, bs, node_list);
> > +        }
> > +    }
> > +
> > +    qdict_del(options, "node-name");
> 
> We duplicate some code all over the place. In general you seem to be
> trying to let the caller of bdrv_new() already figure out what the node
> name is by parsing the options QDict; here however, you do it after
> bdrv_new().
> 
> Can't we consolidate this and only ever set the node name in
> bdrv_open_common(), so that the option is parsed only once, there is
> only once place adding BDSes to the list, and there is only one place
> checking for duplicates?

Nice idea it will make the patches much smaller and cleaner.

Best regards

Benoît

> 
> Kevin

  reply	other threads:[~2013-12-11 14:43 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-05 17:14 [Qemu-devel] [PATCH V4 0/7] Giving names to BlockDriverState graph nodes Benoît Canet
2013-12-05 17:14 ` [Qemu-devel] [PATCH V4 1/7] block: Add bs->node_name to hold the name of a bs node of the bs graph Benoît Canet
2013-12-09 16:04   ` Kevin Wolf
2013-12-09 16:11   ` Kevin Wolf
2013-12-05 17:14 ` [Qemu-devel] [PATCH V4 2/7] block: Allow the user to define "node-name" option Benoît Canet
2013-12-06 15:35   ` Eric Blake
2013-12-09 16:15   ` Kevin Wolf
2013-12-11 14:42     ` Benoît Canet [this message]
2013-12-05 17:14 ` [Qemu-devel] [PATCH V4 3/7] qmp: Add a command to list the named BlockDriverState nodes Benoît Canet
2013-12-06 15:59   ` Eric Blake
2013-12-09 15:46     ` Benoît Canet
2013-12-10 15:22       ` Eric Blake
2013-12-05 17:15 ` [Qemu-devel] [PATCH V4 4/7] qmp: Allow to change password on names block driver states Benoît Canet
2013-12-06 14:27   ` Luiz Capitulino
2013-12-06 15:24     ` Eric Blake
2013-12-06 16:52       ` Luiz Capitulino
2013-12-09 13:35         ` Benoît Canet
2013-12-09 16:23         ` Kevin Wolf
2013-12-09 16:41           ` Luiz Capitulino
2013-12-09 16:48             ` Benoît Canet
2013-12-09 17:03               ` Luiz Capitulino
2013-12-09 17:16                 ` Benoît Canet
2013-12-10  9:57             ` Kevin Wolf
2013-12-10 14:06               ` Luiz Capitulino
2013-12-10 14:25                 ` Kevin Wolf
2013-12-10 15:16                   ` Luiz Capitulino
2013-12-10 15:54                     ` Kevin Wolf
2013-12-10 16:15                     ` Eric Blake
2013-12-11  3:52                     ` Fam Zheng
2013-12-11 13:20                       ` Luiz Capitulino
2013-12-05 17:15 ` [Qemu-devel] [PATCH V4 5/7] qmp: Allow block_resize to manipulate bs graph nodes Benoît Canet
2013-12-09 16:24   ` Kevin Wolf
2013-12-09 16:41     ` Benoît Canet
2013-12-10  9:59       ` Kevin Wolf
2013-12-05 17:15 ` [Qemu-devel] [PATCH V4 6/7] block: Create authorizations mechanism for external snapshots Benoît Canet
2013-12-05 17:15 ` [Qemu-devel] [PATCH V4 7/7] qmp: Allow to take external snapshots on bs graphs node Benoît Canet
2013-12-09 15:32 ` [Qemu-devel] [PATCH V4 0/7] Giving names to BlockDriverState graph nodes Kevin Wolf
2013-12-09 15:52   ` Benoît Canet

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=20131211144256.GB2947@irqsave.net \
    --to=benoit.canet@irqsave.net \
    --cc=armbru@redhat.com \
    --cc=famz@redhat.com \
    --cc=jcody@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /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).