Re: [PATCH] Converter for svn_wc_status3_t to svn_wc_status2_t

On Mon, Apr 26, 2010 at 10:53:04PM +0200, Daniel Näslund wrote: > > [[[ > Implmenent converter function for converting svn_wc_status3_t to > svn_wc_status2_t. > > * subversion/include/private/svn_wc_private.h > (svn_wc__status2_from_3): Return svn_error_t * instead of > svn_wc_status2_t *.

Re: [PATCH] Saving a few cycles, part 2/2

On Monday 26 April 2010 05:11:25 pm Stefan Fuhrmann wrote: > > I saw in a later part of the patch that this is so you can make the > > function 'decode_instruction()' rely on those values. As you know, > > adding the explicit "=0", "=1", "=2" on the enum definition does not > > make any difference

Re: [PATCH v3] Fix O(n) runtime in cache lookup, part 1/2

Philip Martin wrote: Stefan Fuhrmann writes: The improvements: * allow for arbitrary (but preferably very small) number of pinned entries per cache * support pinned entries from multiple threads (i.e. don't use hold the mutex across API calls) * added atomic set&pin operation * check for

Re: [PATCH] Saving a few cycles, part 2/2

On Mon, Apr 26, 2010 at 7:11 PM, Stefan Fuhrmann < stefanfuhrm...@alice-dsl.de> wrote: > Julian Foad wrote: > >> On Sun, 2010-04-25, Stefan Fuhrmann wrote: >> >> >>> here comes the second set of local optimizations. >>> It contains a number of simple improvements over >>> the original code. >>> >>

Re: [PATCH] Saving a few cycles, part 2/2

Julian Foad wrote: On Sun, 2010-04-25, Stefan Fuhrmann wrote: here comes the second set of local optimizations. It contains a number of simple improvements over the original code. Hi Stefan. Thanks for working on optimization. Saving a few cycles can be useful if it's in frequently-e

Re: [PATCH] Saving a few cycles, part 1/2

Philip Martin wrote: Stefan Fuhrmann writes: further reducing my backlog of patches sitting in my working copy, this and the next patch optimize code locally - shaving off cycles here and there. The net effect is somewhere between 3 and 10 percent for repository access (ls, export, etc.).

Re: [PATCH] Saving a few cycles, part 1/2

On Mon, Apr 26, 2010 at 5:23 AM, Philip Martin wrote: > Stefan Fuhrmann writes: > >> further reducing my backlog of patches sitting in my >> working copy, this and the next patch optimize code >> locally - shaving off cycles here and there. The net >> effect is somewhere between 3 and 10 percent

Re: hosting old svn releases on archive.apache.org

+1 Roy

Re: [PATCH] Saving a few cycles, part 1/2

Bert Huijben wrote: -Original Message- From: Stefan Fuhrmann [mailto:stefanfuhrm...@alice-dsl.de] In this patch, I eliminated calls to memcpy for small copies as they are particularly expensive in the MS CRT. Which CRT did you use for these measurements? (2005, 2008, 2010, Debug

Re: [PATCH] issue #3620: svn add ^/ triggers assertion failure

On 04/26/10 05:27 PM, C. Michael Pilato wrote: Hyrum K. Wright wrote: On Fri, Apr 23, 2010 at 4:24 PM, Stefan Sperling [...] I think it's best to consistently expand "^/". In my mind the above should be interpreted as: $ svn add http://svn.example.com/repos/ which would then generate whateve

[PATCH] Converter for svn_wc_status3_t to svn_wc_status2_t

Hi! Since I can't test it, and know too little of libsvn_wc, I ask if any of you guys think it looks ok? [[[ Implmenent converter function for converting svn_wc_status3_t to svn_wc_status2_t. * subversion/include/private/svn_wc_private.h (svn_wc__status2_from_3): Return svn_error_t * instead o

Re: svn looses history when moving directories

On Mon, Apr 26, 2010 at 4:57 PM, Bert Huijben wrote: >> -Original Message- >> From: Mark Phippard [mailto:mphipp...@collab.net] >> Sent: maandag 26 april 2010 16:03 >> To: CollabNet Subversion Client >> Subject: Re: svn looses history when moving directories >> >> On 4/26/10 9:42 AM, "Piet

Re: Hosting tarballs on ASF infrastructure

Manuel Vacelet wrote on Mon, 26 Apr 2010 at 20:03 +0200: > My 2 cts as time-to-time compiler (user in your list): -deps are one > of things I just find perfect when I have to build subversion: no need > to worry and to spend time to find all the dependencies: juste > download 2 tarballs, unpack & g

RE: Feature idea: user-configurable post-commit notifications

Bolstridge, Andrew wrote on Fri, 23 Apr 2010 at 13:52 +0100: > So imagine if I (as a server admin, of course) could configure a new > command ("svn review" or "svn special-cmd review") that uses the > same mechanism as pre-commit, and calls the server hook > "special-review.

Re: Hosting tarballs on ASF infrastructure

On Mon, Apr 26, 2010 at 6:20 PM, Hyrum K. Wright wrote: > I would *love* to stop shipping the deps, and would be happy to make this > happen as part of the 1.7 release cycle.  It would be nice to hear from our > users / packages / 3rd-party-clients to find out if they actually use or > care about

Re: svn commit: r938068 - in /subversion/trunk/subversion: include/svn_wc.h libsvn_client/commit_util.c libsvn_wc/adm_crawler.c libsvn_wc/deprecated.c libsvn_wc/wc.h

Greg Stein wrote: > On Mon, Apr 26, 2010 at 10:45, wrote: > >... > > +++ subversion/trunk/subversion/libsvn_wc/adm_crawler.c Mon Apr 26 14:45:55 > > 2010 > >... > > @@ -1057,7 +1058,8 @@ svn_wc__internal_transmit_text_deltas(co > > void *wh_baton; > > const svn_checksum_t *expected_md5_check

Re: Hosting tarballs on ASF infrastructure

On Mon, Apr 26, 2010 at 11:20:52AM -0500, Hyrum K. Wright wrote: > On Mon, Apr 26, 2010 at 11:06 AM, Mark Phippard wrote: > > > On Mon, Apr 26, 2010 at 11:44 AM, Hyrum K. Wright > > wrote: > > I was under the impression we couldn't post it unless we remove Neon > > and maybe other parts. > > > >

Re: Hosting tarballs on ASF infrastructure

On Mon, Apr 26, 2010 at 11:06 AM, Mark Phippard wrote: > On Mon, Apr 26, 2010 at 11:44 AM, Hyrum K. Wright > wrote: > > This looks like a good location, do you know what needs to happen so I > can > > get the appropriate privs to upload to the subversion/ directory there? > > > > Also, I'm plann

Re: Hosting tarballs on ASF infrastructure

On Mon, Apr 26, 2010 at 11:05 AM, Greg Stein wrote: > On Mon, Apr 26, 2010 at 11:44, Hyrum K. Wright > wrote: > > > > > > On Sat, Apr 17, 2010 at 8:24 AM, Justin Erenkrantz < > jus...@erenkrantz.com> > > wrote: > >> > >> On Fri, Apr 16, 2010 at 10:02 AM, Hyrum K. Wright > >> wrote: > >> > 1) Is

hosting old svn releases on archive.apache.org

Hey all, We are interested in hosting old svn releases (1.0 thru 1.6.current, maybe pre-1.0 tarballs if we can find them) on the archive.apache.org host. There are two salient points here: 1) the license is effectively Apache 1.1 (with a text replace of the name), thus it matches the "no more res

Re: [PATCH] Saving a few cycles, part 2/2

[Philip Martin] > We have svn_ctype_isalpha which is a macro that does a table lookup. > We don't currently have svn_ctype_isxdigit but it could be added. > > > I'd rather see the whole thing become a single table lookup. Untested > > patch appended. > > Is this a bottleneck? I think using the

Re: Hosting tarballs on ASF infrastructure

On Mon, Apr 26, 2010 at 11:44 AM, Hyrum K. Wright wrote: > This looks like a good location, do you know what needs to happen so I can > get the appropriate privs to upload to the subversion/ directory there? > > Also, I'm planning on just putting the release tarballs, not the deps > tarballs.  Any

Re: Hosting tarballs on ASF infrastructure

On Mon, Apr 26, 2010 at 11:44, Hyrum K. Wright wrote: > > > On Sat, Apr 17, 2010 at 8:24 AM, Justin Erenkrantz > wrote: >> >> On Fri, Apr 16, 2010 at 10:02 AM, Hyrum K. Wright >> wrote: >> > 1) Is there a guide somewhere about how to use the ASF mirrors and >> > release >> > infrastructure? >> >

Re: Hosting tarballs on ASF infrastructure

On Sat, Apr 17, 2010 at 8:24 AM, Justin Erenkrantz wrote: > On Fri, Apr 16, 2010 at 10:02 AM, Hyrum K. Wright > wrote: > > 1) Is there a guide somewhere about how to use the ASF mirrors and > release > > infrastructure? > > http://www.apache.org/dev/mirrors.html > > > 2) Is it possible for us to

Re: [PATCH] Saving a few cycles, part 2/2

Peter Samuelson writes: > [Stefan Fuhrmann] >> --- subversion/libsvn_subr/checksum.c(revision 937673) >> +++ subversion/libsvn_subr/checksum.c(working copy) >> @@ -206,12 +206,39 @@ >> >>for (i = 0; i < len; i++) >> { >> - if ((! isxdigit(hex[i * 2])) || (! isxdigi

Re: svn commit: r937275 - in /subversion/trunk/subversion: include/svn_subst.h libsvn_subr/subst.c

On Mon, Apr 26, 2010 at 06:54, Julian Foad wrote: > On Fri, 2010-04-23, Stefan Sperling wrote: >> On Fri, Apr 23, 2010 at 01:01:24PM -, s...@apache.org wrote: >... >> > +++ subversion/trunk/subversion/include/svn_subst.h Fri Apr 23 13:01:23 >> > 2010 >> > @@ -301,7 +301,9 @@ svn_subst_transla

Re: svn commit: r938068 - in /subversion/trunk/subversion: include/svn_wc.h libsvn_client/commit_util.c libsvn_wc/adm_crawler.c libsvn_wc/deprecated.c libsvn_wc/wc.h

On Mon, Apr 26, 2010 at 10:45, wrote: >... > +++ subversion/trunk/subversion/libsvn_wc/adm_crawler.c Mon Apr 26 14:45:55 > 2010 >... > @@ -1057,7 +1058,8 @@ svn_wc__internal_transmit_text_deltas(co >   void *wh_baton; >   const svn_checksum_t *expected_md5_checksum; >   svn_checksum_t *verify_ch

Re: [PATCH] issue #3620: svn add ^/ triggers assertion failure

Hyrum K. Wright wrote: > > On Fri, Apr 23, 2010 at 4:24 PM, Stefan Sperling > wrote: > > ... > > We should also consider changing the way "^/" is interpreted. > Maybe only commands that expect URLs as arguments should expand "^/" > to a URL? That way the ab

RE: svn looses history when moving directories

> -Original Message- > From: Mark Phippard [mailto:mphipp...@collab.net] > Sent: maandag 26 april 2010 16:03 > To: CollabNet Subversion Client > Subject: Re: svn looses history when moving directories > > On 4/26/10 9:42 AM, "Piet-Hein Peeters" > wrote: > > > That's still not the same us

Re: [PATCH] Saving a few cycles, part 2/2

[Stefan Fuhrmann] > --- subversion/libsvn_subr/checksum.c (revision 937673) > +++ subversion/libsvn_subr/checksum.c (working copy) > @@ -206,12 +206,39 @@ > >for (i = 0; i < len; i++) > { > - if ((! isxdigit(hex[i * 2])) || (! isxdigit(hex[i * 2 + 1]))) > + char c1 = hex[i * 2

Re: svn commit: r937452 - in /subversion/trunk/subversion: include/private/svn_wc_private.h libsvn_client/status.c libsvn_wc/node.c libsvn_wc/questions.c

Oops. Will fix. Julian Foad wrote: > "subversion/libsvn_client/status.c:409: 'replaced' may be used > uninitialized in this function" > > - Julian > > > On Fri, 2010-04-23, cmpil...@apache.org wrote: > [...] >> + if (added) >> +{ >> + svn_boolean_t replaced; >

Re: [PATCH v3] Fix O(n) runtime in cache lookup, part 1/2

Stefan Fuhrmann writes: > The improvements: > > * allow for arbitrary (but preferably very small) number > of pinned entries per cache > * support pinned entries from multiple threads > (i.e. don't use hold the mutex across API calls) > * added atomic set&pin operation > * check for pin leaks b

Re: svn commit: r937275 - in /subversion/trunk/subversion: include/svn_subst.h libsvn_subr/subst.c

On Fri, 2010-04-23, Stefan Sperling wrote: > On Fri, Apr 23, 2010 at 01:01:24PM -, s...@apache.org wrote: > > Author: stsp > > Date: Fri Apr 23 13:01:23 2010 > > New Revision: 937275 > > > > URL: http://svn.apache.org/viewvc?rev=937275&view=rev > > Log: > > Pool usage fixes in the translation

Re: [PATCH] Saving a few cycles, part 1/2

On Sun, 2010-04-25, Stefan Fuhrmann wrote: > In this patch, I eliminated calls to memcpy for small > copies as they are particularly expensive in the MS CRT. If we do decide it is worth avoiding calls to 'memcpy' for small sizes, please don't put the expanded code in-line at the points of use, bec

Re: [PATCH] Saving a few cycles, part 2/2

On Sun, 2010-04-25, Stefan Fuhrmann wrote: > here comes the second set of local optimizations. > It contains a number of simple improvements over > the original code. Hi Stefan. Thanks for working on optimization. Saving a few cycles can be useful if it's in frequently-executed code so that the

Re: [PATCH] Saving a few cycles, part 1/2

Stefan Fuhrmann writes: > further reducing my backlog of patches sitting in my > working copy, this and the next patch optimize code > locally - shaving off cycles here and there. The net > effect is somewhere between 3 and 10 percent > for repository access (ls, export, etc.). > > In this patch,