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

Re: Reminder: Branch 1.6 next Wednesday

From: David Glasser <glasser_at_davidglasser.net>
Date: Thu, 30 Oct 2008 11:24:43 -0700

On Thu, Oct 30, 2008 at 11:21 AM, Hyrum K. Wright
<hyrum_wright_at_mail.utexas.edu> wrote:
> David Glasser wrote:
>> On Thu, Oct 30, 2008 at 11:13 AM, Hyrum K. Wright
>> <hyrum_wright_at_mail.utexas.edu> wrote:
>>> David Glasser wrote:
>>>> On Thu, Oct 30, 2008 at 10:03 AM, Hyrum K. Wright
>>>> <hyrum_wright_at_mail.utexas.edu> wrote:
>>>>> There's been a bit of discussion in response to my mail of last week about
>>>>> branching 1.6 on Wednesday (see
>>>>> http://svn.haxx.se/dev/archive-2008-10/0818.shtml). However, I haven't heard
>>>>> any screaming complaints or needs to push the date back, so I'm still planning
>>>>> on branching 1.6 on Nov. 5, 1700 UTC.
>>>>>
>>>>> As I mentioned earlier, if people feel they need another week or so be get stuff
>>>>> in for 1.6, I don't think that's too much of a problem, but please make it known
>>>>> soon!
>>>> I assume you're aware of my opinions on the current state of FSFS and
>>>> its releasibility. (This doesn't necessary mean it can't get fixed on
>>>> the branch, of course, but that's more effort.)
>>> I know about your concerns about rep-sharing performance, and I've got a patch
>>> to make rep-sharing configurable. I seem to recall you mentioning a few things
>>> about FSFS over the past couple of weeks, but could you enumerate them here,
>>> just so we've got a good checklist of items to tackle? (Or, perhaps put them in
>>> TODO-1.6?)
>>
>> It wasn't just performance (and the sha issue): there were a bunch of
>> serious concurrency bugs that will lead to repository corruption if
>> not fixed. I guess I can copy and paste into TODO-1.6.
>
> That'd be great, thanks.

r33965.

By the way, I hope that the concurrency issues that I found don't
exist in fs_base, but I'm not familiar enough with BDB
transactionality to know if that's the case. Has the fs_base code
been reviewed for the same concurrency issues? I assume that BDB
would be in better shape here because it's all one type of DB.

--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 2008-10-30 19:24:57 CET

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

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.