You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Daniel Shahaf <d....@daniel.shahaf.co.il> on 2008/06/06 04:01:47 UTC
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Branko Čibej wrote on Tue, 27 May 2008 at 12:52 +0200:
> Bernard Blackham wrote:
> > On Tue, May 27, 2008 at 10:57:18AM +0200, Branko Čibej wrote:
> > Anyway, new patches are attached for trunk/1.4.6 and 1.4.5 again. As
> > before, I've only tested the 1.4.5 patch (it's still in the middle
> > of a multi-gigabyte checkout, but it would usually have failed by
> > now :)
> >
>
> Looks good. Someone with a Windows box should test this (on both Cygwin and
> native), but +1 as far as I'm concerned.
>
Is anyone able to test this on both cygwin and native Windows? (Branko,
what do you mean by "testing" --- do you refer to running a large checkout
or to running 'make check'?)
The patch is here:
http://permalink.gmane.org/gmane.comp.version-control.subversion.devel/100518
Daniel
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Posted by David Rothenberger <da...@acm.org>.
On 6/5/2008 9:01 PM, Daniel Shahaf wrote:
> Branko Čibej wrote on Tue, 27 May 2008 at 12:52 +0200:
>> Bernard Blackham wrote:
>>> On Tue, May 27, 2008 at 10:57:18AM +0200, Branko Čibej wrote:
>>> Anyway, new patches are attached for trunk/1.4.6 and 1.4.5 again. As
>>> before, I've only tested the 1.4.5 patch (it's still in the middle
>>> of a multi-gigabyte checkout, but it would usually have failed by
>>> now :)
>>>
>> Looks good. Someone with a Windows box should test this (on both Cygwin and
>> native), but +1 as far as I'm concerned.
>>
>
> Is anyone able to test this on both cygwin and native Windows? (Branko,
> what do you mean by "testing" --- do you refer to running a large checkout
> or to running 'make check'?)
>
> The patch is here:
> http://permalink.gmane.org/gmane.comp.version-control.subversion.devel/100518
I've been testing this with 1.4.6 on Cygwin with good success. I have
successfully run "make check" and have also not had any problems using
the client in my day-to-day work.
Without the patch, I would frequently encounter problems when my disk
defragmenter was enabled and TortoiseSVN overlays were enabled. I've had
both enabled for the last week and have not encountered any problems.
--
David Rothenberger ---- daveroth@acm.org
Carson's Observation on Footwear:
If the shoe fits, buy the other one too.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Posted by Daniel Shahaf <d....@daniel.shahaf.co.il>.
(patch manager hat on)
I've filed this patch as issue #3218. (Bernard, David, you can CC
yourself if you have a tigris username.)
http://subversion.tigris.org/issues/show_bug.cgi?id=3218
Thanks for the patch,
Daniel
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Posted by Daniel Shahaf <d....@daniel.shahaf.co.il>.
Bernard Blackham wrote on Sat, 7 Jun 2008 at 01:35 +0800:
> Daniel Shahaf wrote:
> > Branko Čibej wrote on Fri, 6 Jun 2008 at 11:53 +0200:
> > > Daniel Shahaf wrote:
> > > > Branko Čibej wrote on Tue, 27 May 2008 at 12:52 +0200:
> > > >
> > > > > Bernard Blackham wrote:
> > > > >
> > > > > > On Tue, May 27, 2008 at 10:57:18AM +0200, Branko Čibej wrote:
> > > > > > Anyway, new patches are attached for trunk/1.4.6 and 1.4.5 again. As
> > > > > > before, I've only tested the 1.4.5 patch (it's still in the middle
> > > > > > of a multi-gigabyte checkout, but it would usually have failed by
> > > > > > now :)
> > > > > >
> > > > > Looks good. Someone with a Windows box should test this (on
> > > > > both Cygwin and native), but +1 as far as I'm concerned.
> > > > >
> > > > >
> > > > Is anyone able to test this on both cygwin and native Windows?
> > > > (Branko, what do you mean by "testing" --- do you refer to
> > > > running a large checkout or to running 'make check'?)
> > > >
> > > Both, ideally.
> > >
> >
> > Okay. With the former, though, I can't help, since I don't have a multi-GB
> > repository here to test with :(
>
> If you could send me a binary, I'd be (foolishly) quite happy to test it here.
> I just don't have an environment setup to build native Win32 binaries.
>
Hmm. I just looked again at the patch, and it doesn't change the native
(non-cygwin) logic at all -- it only changes the Cygwin-specific logic --
so don't see a need to test it on native (non-cygwin).
Bernard, David, if you can try current trunk under Cygwin and report if
the patch has the intended effect (i.e. works with the patch, fails
without it) it would be great.
Meanwhile, won't we also need a committer with a Cygwin build environment
to commit this change? How do we usually handle Cygwin-specific patches?
> I also tried building trunk under Cygwin but it failed due to a missing
> apr-related symbol which I suspect means I need to update my APR, at which
> point I got nervous about stepping on Cygwin packages and stopped. I will try
> again and persist with that next week. Though again, if there's a statically
> linked Cygwin svn binary, I'd be happy to test it.
>
> Cheers,
> Bernard.
>
>
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Posted by Bernard Blackham <be...@largestprime.net>.
Daniel Shahaf wrote:
> Branko Čibej wrote on Fri, 6 Jun 2008 at 11:53 +0200:
>> Daniel Shahaf wrote:
>>> Branko Čibej wrote on Tue, 27 May 2008 at 12:52 +0200:
>>>
>>>> Bernard Blackham wrote:
>>>>
>>>>> On Tue, May 27, 2008 at 10:57:18AM +0200, Branko Čibej wrote:
>>>>> Anyway, new patches are attached for trunk/1.4.6 and 1.4.5 again. As
>>>>> before, I've only tested the 1.4.5 patch (it's still in the middle
>>>>> of a multi-gigabyte checkout, but it would usually have failed by
>>>>> now :)
>>>>>
>>>> Looks good. Someone with a Windows box should test this (on both Cygwin
>>>> and
>>>> native), but +1 as far as I'm concerned.
>>>>
>>>>
>>> Is anyone able to test this on both cygwin and native Windows? (Branko,
>>> what do you mean by "testing" --- do you refer to running a large checkout
>>> or to running 'make check'?)
>>>
>> Both, ideally.
>>
>
> Okay. With the former, though, I can't help, since I don't have a
> multi-GB repository here to test with :(
If you could send me a binary, I'd be (foolishly) quite happy to test it
here. I just don't have an environment setup to build native Win32 binaries.
I also tried building trunk under Cygwin but it failed due to a missing
apr-related symbol which I suspect means I need to update my APR, at
which point I got nervous about stepping on Cygwin packages and stopped.
I will try again and persist with that next week. Though again, if
there's a statically linked Cygwin svn binary, I'd be happy to test it.
Cheers,
Bernard.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Posted by Daniel Shahaf <d....@daniel.shahaf.co.il>.
Branko Čibej wrote on Fri, 6 Jun 2008 at 11:53 +0200:
> Daniel Shahaf wrote:
> > Branko Čibej wrote on Tue, 27 May 2008 at 12:52 +0200:
> >
> > > Bernard Blackham wrote:
> > >
> > > > On Tue, May 27, 2008 at 10:57:18AM +0200, Branko Čibej wrote:
> > > > Anyway, new patches are attached for trunk/1.4.6 and 1.4.5 again. As
> > > > before, I've only tested the 1.4.5 patch (it's still in the middle
> > > > of a multi-gigabyte checkout, but it would usually have failed by
> > > > now :)
> > > >
> > > Looks good. Someone with a Windows box should test this (on both Cygwin
> > > and
> > > native), but +1 as far as I'm concerned.
> > >
> > >
> >
> > Is anyone able to test this on both cygwin and native Windows? (Branko,
> > what do you mean by "testing" --- do you refer to running a large checkout
> > or to running 'make check'?)
> >
>
> Both, ideally.
>
Okay. With the former, though, I can't help, since I don't have a
multi-GB repository here to test with :(
Re: [PATCH RFC v2] WIN32_RETRY_LOOP on Cygwin
Posted by Branko Čibej <br...@xbc.nu>.
Daniel Shahaf wrote:
> Branko Čibej wrote on Tue, 27 May 2008 at 12:52 +0200:
>
>> Bernard Blackham wrote:
>>
>>> On Tue, May 27, 2008 at 10:57:18AM +0200, Branko Čibej wrote:
>>> Anyway, new patches are attached for trunk/1.4.6 and 1.4.5 again. As
>>> before, I've only tested the 1.4.5 patch (it's still in the middle
>>> of a multi-gigabyte checkout, but it would usually have failed by
>>> now :)
>>>
>>>
>> Looks good. Someone with a Windows box should test this (on both Cygwin and
>> native), but +1 as far as I'm concerned.
>>
>>
>
> Is anyone able to test this on both cygwin and native Windows? (Branko,
> what do you mean by "testing" --- do you refer to running a large checkout
> or to running 'make check'?)
>
Both, ideally.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org