[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [Qemu-block] [RFC] transactions: add transaction-wide p
From: |
Fam Zheng |
Subject: |
Re: [Qemu-devel] [Qemu-block] [RFC] transactions: add transaction-wide property |
Date: |
Tue, 20 Oct 2015 13:16:15 +0800 |
User-agent: |
Mutt/1.5.21 (2010-09-15) |
On Mon, 10/19 09:27, Markus Armbruster wrote:
> John Snow <address@hidden> writes:
>
> > On 10/16/2015 08:23 AM, Stefan Hajnoczi wrote:
> >> On Mon, Oct 12, 2015 at 12:50:20PM -0400, John Snow wrote:
> >>> Ping -- any consensus on how we should implement the "do-or-die"
> >>> argument for transactions that start block jobs? :)
> >>>
> >>> This patch may look a little hokey in how it boxes arguments, but I can
> >>> re-do it on top of Eric Blake's very official way of boxing arguments,
> >>> when the QAPI dust settles.
> >>
> >> I don't understand what you are trying to do after staring at the email
> >> for 5 minutes. Maybe the other reviewers hit the same problem and
> >> haven't responded.
> >>
> >> What is the problem you're trying to solve?
> >>
> >> Stefan
> >>
> >
> > Sorry...
> >
> > What I am trying to do is to add the transactional blocker property to
> > the *transaction* command and not as an argument to each individual action.
> >
> > There was some discussion on this so I wanted to just send an RFC to
> > show what I had in mind.
>
> Was it the discussion on @transactional-cancel? I'm on record
> supporting it per transaction rather than per action:
> Message-ID: <address@hidden>
> http://lists.gnu.org/archive/html/qemu-devel/2015-09/msg05948.html
I prefer we start with a per-transaction flag as in this patch. Any
fine-grained arguments could be added in the future if it turns out to be
useful.
I'll take a look at the implementation later.
Fam
>
> > This series applies on top of Fam's latest series and moves the
> > arguments from each action to a transaction-wide property.