[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Bug in multipath unlock - fails on unrelated working copy locks

From: David Glasser <glasser_at_davidglasser.net>
Date: Fri, 15 Feb 2008 14:47:42 -0800

On Fri, Feb 15, 2008 at 1:57 PM, Steven Bakke <steven.bakke_at_amd.com> wrote:
>
> On Feb 15, 2008, at 3:20 PM, David Glasser wrote:
>
> > On Fri, Feb 15, 2008 at 11:17 AM, Miller, Eric <Eric.Miller_at_amd.com>
> > wrote:
> >>> -----Original Message-----
> >>> From: dglasser_at_gmail.com [mailto:dglasser_at_gmail.com] On Behalf Of
> >> David
> >>> Glasser
> >>
> >>> I dunno. "L" really only happens if you have concurrent access to
> >>> the
> >>> working copy or a crash to recover from. In your real cases, which
> >>> one is it?
> >>
> >> Not important, lets assume both cases happen.
> >
> > Eric, I am still curious about your use case.
> >
> > Specifically, I've been brainstorming designs for a new working copy
> > library. One thing I've been trying to figure out would be whether or
> > not it would be OK to lock the entire working copy for operations (as
> > opposed to just a subdirectory). This is impractical with the current
> > implementation (since every subdirectory has its own lock file) but
> > could be done efficiently with a monolithic-working-copy design. Thus
> > I'm curious about your use case!
>
> If you were to have only one lock, wouldn't that eliminate the
> possibility of multiple users interacting with the working copy?

I'm not convinced that this is a use model that Subversion has ever recommended.

--dave

-- 
David Glasser | glasser@davidglasser.net | http://www.davidglasser.net/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-02-15 23:48:08 CET

This is an archived mail posted to the Subversion Dev mailing list.