Hi,
I am planning to work on issue 3690. Before starting with this I have
few questions.
1. Hyrum updated the issue with his comment stating that already there
is work going on in the branch ignore-mergeinfo which addresses
subset of issue 3690, i.e ignoring changes to svn:mergeinfo. Is
I'm looking for something to dig back into wc-ng. People should grab
this rather than wait for me, but I'll see what I can do in the next
few days...
On Tue, Jan 18, 2011 at 10:58, Julian Foad wrote:
> Can anyone spare a bit of time to take this forward? I'll tell you what
> I'm planning to do n
On Thu, Jan 27, 2011 at 03:59:49AM +0200, Daniel Shahaf wrote:
> Is anyone testing the "SVN_UNALIGNED_ACCESS_IS_OK=0" code regularly?
> On a quick inspection, all the bots describe themselves as x64, not ia64.
I could do test runs on OpenBSD/sparc64 if that helps.
Stefan
Is anyone testing the "SVN_UNALIGNED_ACCESS_IS_OK=0" code regularly?
On a quick inspection, all the bots describe themselves as x64, not ia64.
/me would add -DSVN_UNALIGNED_ACCESS_IS_OK=0 to his CFLAGS, but he isn't
building too often these days.
stef...@apache.org wrote on Sat, Jan 01, 2011 at 1
On 01/26/2011 11:39 AM, Blair Zajac wrote:
On 01/26/2011 11:15 AM, Philip Martin wrote:
Blair Zajac writes:
I'm now thinking of putting the retry in svn_io_file_lock2() instead
of handling a deadlock in libsvn_fs_fs itself. It shouldn't hurt any
other use cases and be a general, defensive code
On Wed, Jan 26, 2011 at 10:50 PM, Hyrum K Wright wrote:
> On Tue, Jan 25, 2011 at 8:31 PM, Johan Corveleyn wrote:
> ...
>>> * revv svn_diff.h#svn_diff_fns_t []
>>>
>>> It looks like, for the most part, any destabilizing functionality is
>>> completed, and what remains are simply opti
On Wed, Jan 26, 2011 at 7:18 PM, Daniel Shahaf wrote:
> Johan Corveleyn wrote on Wed, Jan 26, 2011 at 15:18:24 +0100:
>> - The only internal caller of the "old" function 'datasource_open'
>> (for a single datasource) doesn't call it anymore
>> (token.c#svn_diff__get_tokens) (there is no need anymo
On Wed, Jan 26, 2011 at 11:42 PM, Stefan Sperling wrote:
> On Wed, Jan 26, 2011 at 11:25:59PM +0100, Stefan Fuhrmann wrote:
>> Hi @all,
>>
>> I'm planning to merge said branch Monday 7th.
>> Speak now or forever hold your peace.
>>
>> Rationale:
>>
>> I've been testing / using that code for a whil
This also needs a back-port to the 1.6.x branch.
[[[
Fix issue #3789: Correctly ignore missing locations when a renamed file
has more than MAX_OPEN_HISTORIES.
* subversion/libsvn_repos/log.c
(get_path_histories): Ignore more bogus repository locations to restore
On Wed, Jan 26, 2011 at 11:25:59PM +0100, Stefan Fuhrmann wrote:
> Hi @all,
>
> I'm planning to merge said branch Monday 7th.
> Speak now or forever hold your peace.
>
> Rationale:
>
> I've been testing / using that code for a while now
> and am reasonably confident that it works. And it
> is no
Hi @all,
I'm planning to merge said branch Monday 7th.
Speak now or forever hold your peace.
Rationale:
I've been testing / using that code for a while now
and am reasonably confident that it works. And it
is not exactly rocket science, either.
Because a large portion of my other performance
w
On Tue, Jan 25, 2011 at 8:31 PM, Johan Corveleyn wrote:
...
>> * revv svn_diff.h#svn_diff_fns_t []
>>
>> It looks like, for the most part, any destabilizing functionality is
>> completed, and what remains are simply optimizations. This
>> optimization work can probably be performed o
On 01/26/2011 04:33 PM, C. Michael Pilato wrote:
> On 01/26/2011 04:25 PM, svn@coelho.net wrote:
>> ISTM that there is an issue since octobre because of the introduction of a
>> macro SVN_CL__OPTION_CONTINUATION_INDENT in "svn/main.c".
>
> That's my fault. I didn't realize the effect of the #
On 01/26/2011 04:25 PM, svn@coelho.net wrote:
>
> Hello,
>
> ISTM that there is an issue since octobre because of the introduction of a
> macro SVN_CL__OPTION_CONTINUATION_INDENT in "svn/main.c".
>
> Now gettext extracts lines instead of paragraphs,
>
> #: ../svn/main.c:226
> msgid "accept
Hello,
ISTM that there is an issue since octobre because of the introduction of
a macro SVN_CL__OPTION_CONTINUATION_INDENT in "svn/main.c".
Now gettext extracts lines instead of paragraphs,
#: ../svn/main.c:226
msgid "accept unknown SSL server certificates without\n"
msgstr ""
How am I expe
On 26.01.2011 03:09, Johan Corveleyn wrote:
On Tue, Jan 25, 2011 at 1:37 AM, Stefan Fuhrmann wrote:
[ ... snip ...]
And, as promised, here some ideas how to get more
speed from the generic code. Your latest commit:
+#if SVN_UNALIGNED_ACCESS_IS_OK
+
+ /* Skip quickly over the stuff between
On 25.01.2011 16:58, Hyrum K Wright wrote:
Johan (and other interested parties),
I've been following some of the commits to the
diff-optimizations-branch with interest. While I've not reviewed them
for technical merit, it appears that others have, and that there is
good work going on in the bran
> -Original Message-
> From: Julian Foad [mailto:julian.f...@wandisco.com]
> Sent: woensdag 26 januari 2011 19:11
> To: Branko Čibej
> Cc: dev@subversion.apache.org
> Subject: Re: Ref-counting for pristine texts
>
> Can anyone help me work out the rules for guaranteeing consistency of
>
On 01/26/2011 11:15 AM, Philip Martin wrote:
Blair Zajac writes:
I'm now thinking of putting the retry in svn_io_file_lock2() instead
of handling a deadlock in libsvn_fs_fs itself. It shouldn't hurt any
other use cases and be a general, defensive code.
Should retry be conditional on a threa
Blair Zajac writes:
> I'm now thinking of putting the retry in svn_io_file_lock2() instead
> of handling a deadlock in libsvn_fs_fs itself. It shouldn't hurt any
> other use cases and be a general, defensive code.
Should retry be conditional on a threaded build? Can this problem even
occur in
Blair Zajac writes:
>> --- subversion/libsvn_fs_fs/fs_fs.c (revision 1063629)
>> +++ subversion/libsvn_fs_fs/fs_fs.c (working copy)
>> @@ -583,6 +583,14 @@
>> err = body(baton, subpool);
>> }
>>
>> + if (!strcmp(lock_filename, path_txn_current_lock(fs, pool)))
>> +{
>> +
On 01/26/2011 02:56 AM, Stefan Sperling wrote:
On Tue, Jan 25, 2011 at 11:00:31PM -0800, Blair Zajac wrote:
We're seeing deadlocks in our Subversion multithreaded server when
two distinct processes try to fcntl(F_SETLKW) on two fsfs
repositories' db/txn-current-lock, when the processes begin
tra
On 01/26/2011 12:38 PM, C. Michael Pilato wrote:
> On 01/24/2011 08:52 PM, Justin Erenkrantz wrote:
>> On Mon, Jan 24, 2011 at 2:22 PM, C. Michael Pilato
>> wrote:
>>> [Using dev@ as a public TODO list to avoid pushing stack on a task.]
>>>
>>> In mod_dav_svn/mirror.c:dav_svn__location_body_filte
On 01/26/2011 05:21 AM, Philip Martin wrote:
Philip Martin writes:
Philip Martin writes:
Create two repositories with a pre-commit hook that sleeps for 20
seconds. Start two threaded svnserve processes listening on different
ports.
svnserve -Tdr.
svnserve -Tdr. --listen-port 3691
Commit:
Julian Foad wrote on Wed, Jan 26, 2011 at 17:35:43 +:
> Two other suggestions for what we could do to improve the situation.
>
> * Make "svn" print a consistent error string for "does not exist in
> revision R" errors, regardless of subcommand. At present, the error
> message differs unnece
Johan Corveleyn wrote on Wed, Jan 26, 2011 at 15:18:24 +0100:
> - The only internal caller of the "old" function 'datasource_open'
> (for a single datasource) doesn't call it anymore
> (token.c#svn_diff__get_tokens) (there is no need anymore, since the
> callers in diff.c, diff3.c and diff4.c alrea
Can anyone help me work out the rules for guaranteeing consistency of
the pristine text store?
On Wed, 2011-01-19, Branko Čibej wrote:
> On 18.01.2011 16:58, Julian Foad wrote:
> > On Thu, 2011-01-13, Branko Čibej wrote:
> >> This would indicate that the reference counting happens too soon ... i
On 01/24/2011 08:52 PM, Justin Erenkrantz wrote:
> On Mon, Jan 24, 2011 at 2:22 PM, C. Michael Pilato
> wrote:
>> [Using dev@ as a public TODO list to avoid pushing stack on a task.]
>>
>> In mod_dav_svn/mirror.c:dav_svn__location_body_filter() and
>> dav_svn__location_in_filter() are code blocks
Two other suggestions for what we could do to improve the situation.
* Make "svn" print a consistent error string for "does not exist in
revision R" errors, regardless of subcommand. At present, the error
message differs unnecessarily among subcommands.
* Make "svn" include a unique identifi
"C. Michael Pilato" writes:
> Would it be possible to, say, provide a mapping between the apr_err code
> space and errorcodes? I'm not sure.
Not a unique one.
> I was thinking that the errorcode
> space was limited to 32k distinct values, whereas the apr_err code space is
> much, much bigger.
On 01/25/2011 05:00 PM, anatoly techtonik wrote:
> The issues with return codes from command line utility previously
> raised in Bitten project also. I can't understand why you still do not
> think that this is a valid case for a ticket, i.e. that command line
> client doesn't allow to see if a pat
On Wed, Jan 26, 2011 at 2:46 PM, Hyrum K Wright wrote:
> On Tue, Jan 25, 2011 at 11:41 PM, Daniel Shahaf
> wrote:
>> Johan Corveleyn wrote on Wed, Jan 26, 2011 at 03:31:11 +0100:
>>> Revving svn_diff_fns_t: what do you mean with parallelizing it? I must
>>> admit that I don't really know (yet) h
On Tue, Jan 25, 2011 at 11:41 PM, Daniel Shahaf wrote:
> Johan Corveleyn wrote on Wed, Jan 26, 2011 at 03:31:11 +0100:
>> Revving svn_diff_fns_t: what do you mean with parallelizing it? I must
>> admit that I don't really know (yet) how to go about that. Very early
>> during the branch work, danie
Philip Martin writes:
>>> svn mkdir -mm svn://localhost/repoA/X1
>>> svn mkdir -mm svn://localhost:3691/repoB/X1
>>> svn mkdir -mm svn://localhost/repoB/X2
>>> svn mkdir -mm svn://localhost:3691/repoA/X2
> I start the four commits above, one after the other with a small delay
> between them. Th
Philip Martin writes:
> Philip Martin writes:
>
>> Create two repositories with a pre-commit hook that sleeps for 20
>> seconds. Start two threaded svnserve processes listening on different
>> ports.
>>
>> svnserve -Tdr.
>> svnserve -Tdr. --listen-port 3691
>>
>> Commit:
>>
>> svn mkdir -mm svn
Philip Martin writes:
> Create two repositories with a pre-commit hook that sleeps for 20
> seconds. Start two threaded svnserve processes listening on different
> ports.
>
> svnserve -Tdr.
> svnserve -Tdr. --listen-port 3691
>
> Commit:
>
> svn mkdir -mm svn://localhost/repoA/X1
> svn mkdir -mm
On Tue, Jan 25, 2011 at 11:00:31PM -0800, Blair Zajac wrote:
> We're seeing deadlocks in our Subversion multithreaded server when
> two distinct processes try to fcntl(F_SETLKW) on two fsfs
> repositories' db/txn-current-lock, when the processes begin
> transactions in reverse order.
>
> Process 1
Blair Zajac writes:
> We're seeing deadlocks in our Subversion multithreaded server when two
> distinct processes try to fcntl(F_SETLKW) on two fsfs repositories'
> db/txn-current-lock, when the processes begin transactions in reverse
> order.
>
> Process 1 Process 2
>
hwri...@apache.org writes:
> Author: hwright
> Date: Tue Jan 25 17:46:38 2011
> New Revision: 1063375
>
> @@ -2341,8 +2341,9 @@ svn_client_status(svn_revnum_t *result_r
> * If @a include_merged_revisions is set, log information for revisions
> * which have been merged to @a targets will also b
39 matches
Mail list logo
