This is the mail archive of the
cygwin-developers
mailing list for the Cygwin project.
Re: findutils support in 1.7.0
- From: Corinna Vinschen <corinna-cygwin at cygwin dot com>
- To: cygwin-developers at cygwin dot com
- Date: Sat, 22 Nov 2008 11:11:35 +0100
- Subject: Re: findutils support in 1.7.0
- References: <20080428121307.GZ23852@calimero.vinschen.de> <20080428130432.GA23852@calimero.vinschen.de> <ba40711f0804280739w768be856u264b786b9c640b0a@mail.gmail.com> <20080428145700.GB23852@calimero.vinschen.de> <ba40711f0804280805h2835a9f9q9bdb2666a0031910@mail.gmail.com> <20080428155859.GD23852@calimero.vinschen.de> <20080429091938.GI23852@calimero.vinschen.de> <20080509153834.GA18799@calimero.vinschen.de> <20080516145313.GD32409@calimero.vinschen.de> <20080630162552.GD5038@calimero.vinschen.de>
- Reply-to: cygwin-developers at cygwin dot com
On Jun 30 18:25, Corinna Vinschen wrote:
> On May 16 16:53, Corinna Vinschen wrote:
> > On May 9 17:38, Corinna Vinschen wrote:
> > > On Apr 29 11:19, Corinna Vinschen wrote:
> > > > Looks like Microsoft introduced a bug with the new NTFS implementation
> > > > in Windows 2000 and only fixed that with the new NTFS implementation in
> > > > Vista. If you're interested in the full table of values, see
> > > > http://cygwin.de/foo-values.csv
> > >
> > > Just FYI, I reported this behaviour including a simple testcase(TM)
> > > into the newsgroup microsoft.public.win32.programmer.kernel but got
> > > the reply I should better open a support case:
> > >
> > > http://tinyurl.com/49pgzw
> > >
> > > I opened a support case today. Now I'm just curious if this will
> > > get a fix or not :)
> >
> > One week later I got the exciting message that the problem can be
> > reproduced...
>
> Case closed. No fix. The guys at Microsoft responsible for the file
> system are of the opinion that the bad timing behaviour is the result
> of a natural evolution of the filing system over time, and not actually
> a bug.
While venting about the Microsoft Support in the main list it occured to
me that I never followed up on the quadrratic timing behaviour case
on this list.
After a short discussion on the aforementioend newsgroup, the guys there
convinced me that it might be worth to reopen the case and to dispute
the result. So I reopened the case and complained and argued again.
Only two months later (while I was on vacation) I got a response from
the escalation engineer. It was very disappointing. I got a lecture on
correct testing (don't use VMs, etc). I got told I should switch off
8.3 filename generation (but not how to do that) and I got again told
that this problem is by design and won't be fixed.
I complained again and last week I had a call with a Microsoft guy from
the business relations department. There's no chance for a fix at all
because there exists a workaround. That made me curious of course.
So, here's the official workaround for the quadratic timing problem in
long path name access on Windows 2000/XP and Server 2003.
Hold your breath, here we go...
.
.
.
.
.
.
.
.
.
.
.
.
Update to Vista or Server 2008!
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat