On Sun, Jun 24, 2001 at 01:09:11AM -0700, Justin Erenkrantz wrote:
> On Sat, Jun 23, 2001 at 07:42:28PM -0700, Justin Erenkrantz wrote:
> > I just ran across a compiler bug when dealing with long long in the GCC
> > shipped with Mandrake 8.0 (Intel). When -O2 is specified (it's in the
> > default)
A mini-patch to fix 'make -j2' builds in apr-util since the dependency
of apr_util.lo on uri_delims.h was missing (it only worked because the
targets were ordered correctly):
Index: Makefile.in
===
RCS file: /home/joe/lib/cvsroot/apr-
The header doesn't mention what the function should return.
Index: apr_tables.h
===
RCS file: /home/joe/lib/cvsroot/apr/include/apr_tables.h,v
retrieving revision 1.18
diff -u -r1.18 apr_tables.h
--- apr_tables.h29 Jun 2001 09
On Tue, Sep 04, 2001 at 09:28:16AM -0500, Ben Collins-Sussman wrote:
>
> A friend of mine is trying to write a FreeBSD port for Subversion. It
> has a lot of external dependencies; specifically, the port wants to
> download a tarball of APR.
Isn't it building from subversion-M3.tar.gz? (which
apr_dir_read (with something like APR_FINFO_TYPE in wanted) will return
APR_INCOMPLETE if it encounters a broken symlink... bug or feature?
I say bug... the caller can quite happily cope with symlinks, broken or
not, as they wish, they just wanted to know what is in the directory.
Regards,
joe
autoconf 2.52 creates a temporary 'autom4te.cache' directory, which
shouldn't get shipped in a release tarball.
Index: buildconf
===
RCS file: /home/cvspublic/apr/buildconf,v
retrieving revision 1.21
diff -u -r1.21 buildconf
--- build
The warning was:
mmap.c: In function `apr_mmap_dup':
mmap.c:180: warning: implicit declaration of function `apr_pmemdup'
and here's a fix:
Index: mmap.c
===
RCS file: /home/cvspublic/apr/mmap/unix/mmap.c,v
retrieving revision 1.37
d
On Mon, Dec 03, 2001 at 03:42:44PM -0800, Ian Holsman wrote:
>
> apparently Mandrake has it's own unique method of naming the db
> files.
> how about we add 2 new config options
> --with-berkeley-db-include
> --with-berkeley-db-lib
> and let the user specify?
Generally those types of option are r
Remove the annoying autoconf 2.5x cache directory in buildconf:
Index: buildconf
===
RCS file: /home/cvspublic/apr-util/buildconf,v
retrieving revision 1.4
diff -u -r1.4 buildconf
--- buildconf 26 Dec 2001 20:45:45 - 1.4
++
The expat detection was adding -I/usr/include to INCLUDES when a
/usr/include/expat.h is found; this is redundant, can mess up include
path ordering, and bleeding edge gcc's even give warnings for it.
(it looks like the code will also do the same for -L/usr/lib in some
circumstances)
Index: build
There are issues with use of asm/atomic.h on Linux: according to the
kernel gurus here this is a header for internal use of the kernel,
and shouldn't be used from userspace.
The problems are, if I understand them correctly:
- with newer kernel/glibc combinations (e.g the RHL7.3 beta) this stuff
r
On Thu, Apr 18, 2002 at 01:58:29PM +0100, Joe Orton wrote:
> There are issues with use of asm/atomic.h on Linux: according to the
> kernel gurus here this is a header for internal use of the kernel,
> and shouldn't be used from userspace.
>
> The problems are, if I under
On Thu, Apr 18, 2002 at 08:38:31AM -0700, Brian Pane wrote:
> Joe Orton wrote:
>
> >There are issues with use of asm/atomic.h on Linux: according to the
> >kernel gurus here this is a header for internal use of the kernel,
> >and shouldn't be used from userspace.
On Thu, Apr 18, 2002 at 09:59:49AM -0700, Justin Erenkrantz wrote:
> On Thu, Apr 18, 2002 at 12:36:08PM -0400, Cliff Woolley wrote:
> > On Thu, 18 Apr 2002, Joe Orton wrote:
> >
> > > Sure - I'd just hope the change could be made for the time being, until
> >
This is useful for when autoconf 2.13 and 2.5x are both in $PATH, but
2.5x is configured using --progam-suffix=-2.53, a la Red Hat Linux 7.3.
And clean up the 'autom4te-2.53.cache' directory produced by that
configuration.
Two patches below: one for apr, one for apr-util (expat portion
submitted
Any problems with these patches or could they be applied?
- Forwarded message from Joe Orton <[EMAIL PROTECTED]> -
From: Joe Orton <[EMAIL PROTECTED]>
To: dev@apr.apache.org
Mail-Followup-To: dev@apr.apache.org
Date: Wed, 8 May 2002 10:30:28 +0100
Subject: [PATCH] support $
On Tue, Apr 30, 2002 at 11:43:29PM -, Justin Erenkrantz wrote:
> Modified:include apr_atomic.h
> Log:
> Remove Linux atomic support as this is most definitely not meant for
> userspace applications. We will have to come up with processor-specific
> implementations, but we can no
Hi folks. For creating binary packages of APR and APR-util, would it be
acceptable to use package versions of "0.9" until an official APR
release is made, since I've seen talk of the first release being "1.0"?
Regards,
joe
This patch changes many AC_DEFINE's to use the 3-argument form, and
removes the corresponding #undef from acconfig.h, since acconfig.h is
deprecated in autoconf 2.5x (and presumably the support for it may
eventually disappear).
Fixing the rest would be easiest using AH_VERBATIM and friends, but I
Second configure patch: these three variables are set but never used:
(these patches may not apply out-of-order, I'll resend if necessary)
--- configure.in.first Fri Sep 13 09:40:40 2002
+++ configure.inFri Sep 13 09:43:59 2002
@@ -562,9 +562,6 @@
fi
fi
-ac_cv_define_READDIR_IS_T
On Fri, Sep 13, 2002 at 10:38:25AM -0700, Justin Erenkrantz wrote:
> On Fri, Sep 13, 2002 at 09:55:13AM +0100, Joe Orton wrote:
> > Second configure patch: these three variables are set but never used:
>
> Shouldn't these values be used by the AC_DEFINE calls? -- justin
I
INT64_C is not being detected if defined in stdint.h; the code path
for that case goes:
ac_cv_define_INT64_C=no
APR_CHECK_DEFINE(INT64_C, stdint.h)
The AC_CACHE_CHECK in APR_CHECK_DEFINE sees that ac_cv_define_INT64_C is
already defined, and fails:
checking for INT64_C... no
checking for IN
Resend:
This patch fixes detection of INT64_C() when defined in stdint.h;
previously the use of APR_CHECK_DEFINE was breaking by picking up the
result of the previous test:
checking for INT64_C... no
checking for INT64_C in stdint.h... (cached) no
--- configure.in3 Oct 2002 15:31:49 -000
Including whitespace before C preprocessor directives breaks with some
older preprocessors; a particularly nasty break on Tru64 4.0f where
APR_CHECK_DEFINE will always suceed.
Index: configure.in
===
RCS file: /home/cvspublic/apr/conf
This just fixes some AC_MSG_RESULT()s to not include double quotes, e.g.
checking if APR supports IPv6... "yes"
and also removes the stray semi-colons.
--- configure.in.msg2002-10-15 21:10:07.0 +0100
+++ configure.in2002-10-15 21:13:35.0 +0100
@@ -1771,25 +1771,25 @@
In Autoconf, it's good practice to use "dnl" only when commenting on an
m4 construct, or something which doesn't end up in the configure script.
But otherwise, '#' should be used, so that the configure script can be
debugged more easily.
This fixes some uses of "dnl" which should be "#", and remov
This patch compresses a few expansions of AC_CHECK_{FUNCS,HEADERS} into
single expansions, since both these macros are implemented as for loops;
this reduces the size of the generated configure script (possibly making
it faster).
--- configure.in16 Oct 2002 12:49:18 - 1.488
+++ co
This patch:
* simplifies APR_CHECK_FILE to use an m4 translit() to avoid the evals
* uses AC_DEFUN to define macros in autoconf, not define()
* drops support for the second and third arguments to the macro, since
these aren't used anywhere. (checked apr-util, httpd-2.0, SVN)
--- build/apr_common.m
On Fri, Oct 18, 2002 at 06:58:47AM -0400, Jeff Trawick wrote:
> [EMAIL PROTECTED] writes:
>
> > wrowe 2002/10/17 11:22:17
> >
> > Modified:buildapr_hints.m4
> > Log:
> > When we build HP-UX applications, they must be able to follow the
> > SHLIB_PATH path variable due to
On Thu, Nov 14, 2002 at 03:25:47PM -0500, Jeff Trawick wrote:
> RedHat 7.3, just did cvs update and picked up Joe's changes
> no local changes
>
> 1) autoconf 2.13
...
> Creating include/arch/unix/apr_private.h.in ...
> /usr/bin/autoheader: Symbol `DEV_RANDOM' is not covered by
> /usr/share/autoco
On some platforms rmdir(2) fails with EEXIST rather than ENOTEMPTY when
trying to delete a non-empty directory; in fact POSIX specifies that
both EEXIST and ENOTEMPTY are allowed for this case.
The test_removeall_fail() test uses APR_STATUS_IS_ENOTEMPTY() for this
case. Is it okay to extend APR_ST
On Sat, Nov 23, 2002 at 09:30:39PM -0800, Ryan Bloom wrote:
> 4) The tests need to be run everywhere so that we can make sure that we
> are actually passing our tests.
One problem with this is that test_ctime and test_localstr functions in
testtime.c are tied to a particular timezone - is that th
On Wed, Dec 04, 2002 at 08:21:36AM -0500, Jeff Trawick wrote:
> Colm MacCárthaigh <[EMAIL PROTECTED]> writes:
>
> > On Wed, Dec 04, 2002 at 12:25:49PM +, Colm MacCárthaigh wrote:
> > > My tests (and patch) were based on apr and apr-util from CVS ,
> > > with the 2.0.43 codebase, because CVS s
On Wed, Dec 04, 2002 at 11:38:06PM +, Colm MacCárthaigh wrote:
> On Wed, Dec 04, 2002 at 11:05:21PM +0000, Joe Orton wrote:
> > Colm, can you try running the apr/test/sendfile binary with your
> > machines?
>
> It was one of the first things I tried when I was debugging
On Thu, Dec 05, 2002 at 10:54:53AM +, Colm MacCárthaigh wrote:
..
> telnet [v6addr] 80
> GET / HTTP/1.1
> Host: madeup.tld
>
>
Have you verified with tcpdump/ethereal etc that this hang is because
the server is not sending the packets? What does netstat -t show for
this connection?
That tes
To summarize some off-list dicussion - the kernel guys have said that
using sendfile on IPv6 sockets may trigger bugs in cards which do
hardware TCP checksumming for card/driver/OS combinations which support
that. (since the cards have to know about IPv6, and that probably
doesn't get tested much).
On Sun, Dec 08, 2002 at 08:50:22PM +0100, Oden Eriksson wrote:
> I'm a rookie at this, but here's the output from "./testall -v" using
> apache2-20021208 (STRIKER_2_0_44_PRE1) on Mandrake Linux 9.0:
>
> All APR Tests:
> Strings:
> Time: ..FF..
On Mon, Dec 09, 2002 at 08:17:28AM -0800, Ryan Bloom wrote:
> On Sun, 8 Dec 2002, Oden Eriksson wrote:
>
> > sïndagen den 8 december 2002 21.14 skrev Joe Orton:
> > > On Sun, Dec 08, 2002 at 08:50:22PM +0100, Oden Eriksson wrote:
> >
> > Socket Creation:
For the implementation of name-based apr_shm_create using shmget (i.e.
the filename != NULL case), I've got two questions:
1) why is APR_EXCL passed to the apr_file_open call? This is annoying
for Apache since if the server crashes, you have to remove the shmem
segment *and* the filename before r
On Fri, Dec 13, 2002 at 10:12:17AM -0800, Ryan Bloom wrote:
> So, I continue to wonder, how is it useful to have this in a _portable_
> run-time, when the concept isn't at all portable?
Will you also be removing threads, IPv6, and all the other stuff which
isn't implemented on every single platfor
On Fri, Dec 13, 2002 at 09:41:51AM -0800, Aaron Bannert wrote:
> If your application is sure that the segment is not in use, then remove
> the file.
OK, fair enough, thanks.
> >2) why is the file used to store the segment size, rather than using
> >the
> >shm_segsz member of the shmid_ds structu
On Fri, Dec 13, 2002 at 03:40:51PM -0800, Ryan Bloom wrote:
>
> On Fri, 13 Dec 2002, Blair Zajac wrote:
>
> > Directory:...F.F
> > Random: N
> > DSO: .
>
> <.>
>
> > 190 tests run: 176 passed, 10 failed, 4 n
On Sun, Dec 15, 2002 at 12:09:38PM -0800, Kean Johnston wrote:
> >The second part is a little iffy. Is there no header in SCO which
> > defines SHUT_RDWR? The library (-lsocket) that provides shutdown()
> > doesn't provide SHUT_RDWR? Does the manual for that function really
> > tell you t
On Sun, Dec 29, 2002 at 10:48:24AM -0800, Ryan Bloom wrote:
>
>
> On Sun, 29 Dec 2002, William A. Rowe, Jr. wrote:
>
> > At 12:21 AM 12/29/2002, [EMAIL PROTECTED] wrote:
> >
> > >On 29 Dec 2002 [EMAIL PROTECTED] wrote:
> > >
> > >> wrowe 2002/12/28 21:44:02
> > >>
> > >> First; once an
On Tue, Dec 31, 2002 at 05:10:40PM -, William Rowe wrote:
> Modified:poll/unix poll.c
> Log:
> Fix apr_poll behavior on Darwin/Win32 (now passing testpoll.)
> We were getting spurious returned events because the select-based
> poll implementation wasn't zeroing out previous
On Fri, Jan 10, 2003 at 05:29:05PM -0600, William Rowe wrote:
...
> My recollection of OS/X is that it's nearly as clean as Win32 at this point,
> and suffers from the same bogus test_localstr and test_ctime tests.
> Fortunately, I'm in CST where I notice these sorts of Pacific fubars.
Is it possi
On Wed, Jan 22, 2003 at 11:44:08PM -0500, Eric Gillespie wrote:
> As an attempt to make this on-topic for both lists, i won't go
> into how i discovered the bug. I don't think it's necessary.
>
> Correctly using waitpid(2) involves checking for EINTR and trying
> again. That leads to common usag
On Mon, Feb 03, 2003 at 11:51:56PM -, Branko ïibej wrote:
> brane 2003/02/03 15:51:56
>
> Modified:test testfileinfo.c
> Log:
> Added a new test to check the behaviour of apr_file_info_get in
> combination with buffered writes. At the moment, this test will
> probably f
On Sat, Feb 08, 2003 at 02:36:01AM +0100, Branko Äibej wrote:
> [EMAIL PROTECTED] wrote:
>
> >jorton 2003/02/07 04:33:00
> >
> > Modified:test testfileinfo.c
> > Log:
> > test_buffered_write_size: Pass TRUNCATE not EXCL to open so test isn't
> > guaranteed to fail if run more than
Hi,
On Fri, Feb 07, 2003 at 12:23:09PM -0500, Arliss, Noah wrote:
> I'll try to clarify things as best I can. First off, BIND_VERBOSE would be
> fantastic. I had to add it in order to get my module to load at all with
> shl_load, even though it still would not function. If there was a good
> reaso
On Sat, Feb 08, 2003 at 02:21:28AM +0100, Branko Äibej wrote:
> Joe Orton wrote:
> >It does indeed, is this the kind of fix you were thinking of?
> >
> I expect that change would avoid the problem, yes. (In fact, in
> Subversion, I did an explicit fluxh befode calling apr_file
On Tue, Feb 11, 2003 at 08:51:26AM -0800, Justin Erenkrantz wrote:
> [ Resend from correct address... ]
>
> --On Tuesday, February 11, 2003 3:36 PM + [EMAIL PROTECTED] wrote:
>
> >jorton 2003/02/11 07:36:56
> >
> > Modified:.configure.in
> > Log:
> > Capitalize some "Posix
On Tue, Feb 11, 2003 at 06:41:36PM +0100, Branko Äibej wrote:
> Joe Orton wrote:
>
> >On Sat, Feb 08, 2003 at 02:21:28AM +0100, Branko Äibej wrote:
> >>I expect that change would avoid the problem, yes. (In fact, in
> >>Subversion, I did an explicit fluxh befode call
On Thu, Feb 20, 2003 at 01:06:55PM -0500, Jeff Trawick wrote:
> Network apps can all too often be tricked by clients into segfaulting or
> worse. When this occurs the first time, it can be important to know
> what data was received and with what boundaries (i.e., how much was
> passed up at a t
On Fri, Feb 21, 2003 at 07:58:16AM -0500, Jeff Trawick wrote:
> Joe Orton wrote:
>
> >On Thu, Feb 20, 2003 at 01:06:55PM -0500, Jeff Trawick wrote:
> >
> >>Network apps can all too often be tricked by clients into segfaulting or
> >>worse. When this occurs th
On Fri, Feb 21, 2003 at 02:27:08PM -0500, Jeff Trawick wrote:
...persuasively and at length
Consider my misgivings misguided ;) You might want to swap
stddef.h+offsetof for apr_general.h+APR_OFFSETOF in the patch.
joe
On Thu, Feb 27, 2003 at 02:35:51PM +0100, Sander Striker wrote:
...
> Is there a known issue with mmapping 0 sized files?
>
> FYI, I'm on Linux 2.4.20, glibc 2.3.1.
>From a quick check, unless I'm screwing something up, it appears that
passing a zero length argument to mmap() doesn't work on seve
On Thu, Feb 27, 2003 at 05:10:53PM -0500, Garrett Rooney wrote:
> Joe Orton wrote:
> >>From a quick check, unless I'm screwing something up, it appears that
> >passing a zero length argument to mmap() doesn't work on several
> >platforms (nor does it make a whole
Pipes created by the Unix implementation of apr_file_pipe_create are
inherited by default, but you can't make them uninherited without
registering a custom cleanup handler. Any objections to this patch,
which lets apr_file_inherit_unset work on a pipe for Unix? (it looks
like it might already work
On Wed, Mar 05, 2003 at 08:03:51PM +0100, Christian Kratzer wrote:
> Hi,
>
> On Wed, 5 Mar 2003, Joe Orton wrote:
>
> > Pipes created by the Unix implementation of apr_file_pipe_create are
> > inherited by default, but you can't make them uninherited without
>
On Fri, Mar 07, 2003 at 12:12:43PM -, Jeff Trawick wrote:
> -dnl APR_CHECK_APR_DEFINE( symbol, path_to_apr )
> +dnl APR_CHECK_APR_DEFINE( symbol )
>dnl
>AC_DEFUN(APR_CHECK_APR_DEFINE,[
> -AC_EGREP_CPP(YES_IS_DEFINED, [
> -#include "$2/include/apr.h"
> -#if $1
> -YES_IS_D
Hi,
On Fri, Feb 14, 2003 at 04:07:20PM -, Brad Nicholes wrote:
> --- apr_ldap.h.in 1 Jan 2003 00:02:20 - 1.6
> +++ apr_ldap.h.in 14 Feb 2003 16:07:19 - 1.7
...
> +#define APR_HAS_NETSCAPE_LDAPSDK@apu_has_ldap_netscape@
> +#define APR_HAS_NOVELL_LDAPSDK @
On Mon, Mar 10, 2003 at 02:34:13PM -0700, Brad Nicholes wrote:
> No, since I am only able to compile for NetWare, I need a little help in
> getting the makefiles for the other platforms fixed up. I posted a
> message regarding this after I committed this patch originally. Several
> people replied
On Tue, Mar 11, 2003 at 02:31:32PM -0800, Ben Collins-Sussman wrote:
> RedHat 7.2 machine, I just built the latest Subverison
> 0.19 sourceball: I can't create a new repository now. 'svnadmin
> create' just sits forever, waiting for /dev/random to decide it has
> enough entropy.
The APR behavio
On Tue, Mar 11, 2003 at 10:48:05PM +0100, Branko Äibej wrote:
> Ben Collins-Sussman wrote:
>
> >Joe Orton <[EMAIL PROTECTED]> writes:
> >
> >
> >
> >>>Sander says there was a long thread about this, about how people
> >>>wanted
On Tue, Mar 11, 2003 at 05:16:23PM -0600, Karl Fogel wrote:
> Branko Äibej <[EMAIL PROTECTED]> writes:
> > >Or gstein has suggested that apr_generate_random_bytes() can grow a
> > >new flag, indicating urandom is preferred.
> >
> > That would look weird to APR users on systems that have never heard
On Mon, Mar 17, 2003 at 11:59:42PM -0600, William Rowe wrote:
...
> If Brad or Brian are available - I will need your eyes on the Unix
> patches - and I don't want to go weeks before we release 0.9.2.
> If we can address these issues on those platforms by Wed that
> would be *really* terrific!
So
On Tue, Mar 18, 2003 at 12:37:16PM -0600, William Rowe wrote:
> At 04:03 AM 3/18/2003, Joe Orton wrote:
> >On Mon, Mar 17, 2003 at 11:59:42PM -0600, William Rowe wrote:
> >...
> >> If Brad or Brian are available - I will need your eyes on the Unix
> >> patches - an
On Tue, Mar 18, 2003 at 09:56:04PM +, Bjoern A. Zeeb wrote:
> On Tue, 18 Mar 2003, Joe Orton wrote:
> > 2. otherwise: even when you set CLOEXEC, the arbitrary binary which
> > can be fork/exec from an untrusted PHP script can then just use
> > ptrace() to arrange for a
On Mon, Mar 17, 2003 at 07:19:22PM +1100, Stas Bekman wrote:
> I'm painfully trying to figure out how to resolve all the dependencies of
> mod_perl 2.0 on aix. Currently I'm stuck at startup time:
>
> Can't load
> '/build/mod_perl/modperl-2.0/blib/arch/Apache2/auto/Apache/RequestUtil/RequestUtil
On Tue, Mar 18, 2003 at 11:18:12PM -0600, William Rowe wrote:
> Thanks to the feedback from Joe and Bjoern, I have committed a ton and 1/2
> of bug fixes to apr's open.c and dupfile.c.
>
> After those changes, this patch is much easier to read. Please take a look
> and comment on anything you see
On Fri, Mar 21, 2003 at 10:49:08AM -0500, Jeff Trawick wrote:
> apr_file_io.h says
>
> /**
> * Unset a file from being inherited by child processes.
> */
>
> should it instead say something like this?
>
> /**
> * Unset a file from being inherited by subprograms
> */
>
> where subprogram is
On Fri, Mar 21, 2003 at 01:56:52PM -0600, William Rowe wrote:
> At 01:09 PM 3/21/2003, Jeff Trawick wrote:
>
> >I'm a bit nervous about the existing/new special handling for descriptors
> >0-2.
>
> Ok, here is a patch that uses the target file's inherit and noclose
> flag bits for apr_file_dup2(
On Thu, Mar 27, 2003 at 01:10:26PM -0500, Jim Jagielski wrote:
> Comments?? :)
Wow, this code is already broken on LP64 platforms (sizeof(sem_t *) !=
sizeof(int)) and has dodgy error handling (#17186) - why add even more
features?
joe
On Tue, Apr 15, 2003 at 01:58:23PM -0400, Jeff Trawick wrote:
...
> #define APR_UINT64_T_FMT_LEN 3
> +#define APR_UINT64_T_HEX_FMT "llx"
> +#define APR_UINT64_T_HEX_FMT_LEN 3
> #define APR_TIME_T_FMT APR_INT64_T_FMT
There's no need to add another _FMT_LEN #define, they are not really
nec
On Tue, Apr 15, 2003 at 04:15:32PM -0400, Jeff Trawick wrote:
> Joe Orton wrote:
> >On Tue, Apr 15, 2003 at 01:58:23PM -0400, Jeff Trawick wrote:
> >...
> >
> >>#define APR_UINT64_T_FMT_LEN 3
> >>+#define APR_UINT64_T_HEX_FMT "llx"
&g
Has 0.9.3 actually been released? I see the tag, but can't find tarballs
for it. Is it just waiting for someone to make tarballs?
Recently APR HEAD has been trying to keep backwards compatibility with
0.9.2, with the proviso that this will be dropped "for 1.0". It would
seem sensible to allow HE
On Tue, Apr 15, 2003 at 05:22:22PM -0500, William Rowe wrote:
> At 04:58 PM 4/15/2003, Joe Orton wrote:
> >An APR "stable" branch will thus be needed which can be followed by the
> >httpd-2.0 stable branch; this could be called APR_0_9_BRANCH and forked
> >of
On Wed, Apr 16, 2003 at 10:50:17AM +0100, David Reid wrote:
> > "David Reid" <[EMAIL PROTECTED]> writes:
> >
> > > Ben's point is more (I think) that if we say we're providing randomness
> then
> > > we should provide that.
> >
> > But a caller isn't allowed to *explicitly* ask for poor-quality,
>
On Tue, Apr 15, 2003 at 09:31:04PM -0500, William Rowe wrote:
> A few developers keep suggesting "we need to break things"
>
> Nobody has ever provided a compelling example since the polls API change.
> And stubbing old functions certainly shouldn't be "getting in the way" of
> further progress tw
I was a bit worried by the copyright notice in this file:
!* This code is based on the UltraSPARC atomics library by Mike Bennett
!* The Initial Developer of the Original Code is Mike Bennett,
!* [EMAIL PROTECTED], Copyright (C) 1999. All Rights Reserved.
!* This code is based on the sparc arc
On Wed, Apr 30, 2003 at 11:20:57AM -0700, Roy Fielding wrote:
> > This code was licensed under version 1.0 of the Mozilla Public
> >License
> > (which is a copyleft and a lot more restrictive than the ASL).
>
> And not approved for distribution by Apache projects. You will need to
> get the ve
On Wed, May 28, 2003 at 04:45:46AM -, Justin Erenkrantz wrote:
> jerenkrantz2003/05/27 21:45:46
>
> Modified:.CHANGES configure.in
>buildapu-hints.m4
>crypto apr_md5.c
>test .cvsignore Makefile.in
> Added: test
On Wed, May 28, 2003 at 02:46:28PM -0700, MATHIHALLI,MADHUSUDAN
(HP-Cupertino,ex1) wrote:
> 'wondering if anybody has objections for the below patch. By having
> the BIND_VERBOSE option, we can know what really failed during the shl_load.
Hi Madhu, BIND_VERBOSE was removed on purpose - the
On Wed, May 28, 2003 at 04:44:53PM -0700, MATHIHALLI,MADHUSUDAN
(HP-Cupertino,ex1) wrote:
>
> >-Original Message-
> >From: Justin Erenkrantz [mailto:[EMAIL PROTECTED]
> [SNIP]
>
> >Is it possible for HP-UX's DSO's error to be captured in a
> >string rather than forcibly sent to stderr?
On Thu, May 29, 2003 at 07:38:29AM -0400, Jeff Trawick wrote:
> [EMAIL PROTECTED] wrote:
> >trawick 2003/05/28 11:24:13
> >
> > Modified:test testdso.c
> > Log:
> > get testdso to work with the vendor's compiler on HP-UX, which
> > defines a different symbol than gcc
>
> #if APR_PL
On Thu, May 29, 2003 at 12:02:00PM +0100, Max Bowsher wrote:
> Below are 2 patches, one for apr, one for the embedded expat in apr-util.
>
> They add a fallback method to locating libtool.m4.
>
> This is important for systems which use auto-version selecting redirector
> scripts for the autotools
On Fri, May 30, 2003 at 09:23:11PM +0200, Sascha Schumann wrote:
> Hi there,
>
> the sed on IRIX does not cooperate with the complex sed
> expressions used to cut out and transform the proper layout
> section. Thus configure has been failing since 2.0.40 or
> so on IRIX.
>
>
On Fri, May 30, 2003 at 01:54:21PM -0400, MATHIHALLI,MADHUSUDAN
(HP-Cupertino,ex1) wrote:
...
> Now, what should be the behaviour under the following scenarios :
>
> - everything is fine : program runs fine. No problems.
> - Unresolved symbols and stderr is open : program will report the
> unre
Hi Max,
On Tue, Jun 10, 2003 at 07:09:48PM +0100, Max Bowsher wrote:
> Below are the patches, one to apr and one to apr-util, revised to use
> PrintPath.
I think it's useful to allow finding libtool.m4 in odd places without
requiring an aclocal installation - since this can be done with a very
si
On Tue, Jun 17, 2003 at 10:15:32PM +0100, Max Bowsher wrote:
> Joe Orton wrote:
> > Hi Max,
> >
> > On Tue, Jun 10, 2003 at 07:09:48PM +0100, Max Bowsher wrote:
> >> Below are the patches, one to apr and one to apr-util, revised to use
> >> PrintPath.
>
On Wed, Jul 02, 2003 at 08:59:19AM -0400, Jeff Trawick wrote:
> [EMAIL PROTECTED] wrote:
> >jorton 2003/07/02 05:12:30
> >
> > Modified:test Makefile.in test_apr.h testall.c testprocmutex.c
> > Log:
> > - dropped support for specifying a lock filename, hopefully that is
> > not cri
Rationale: Using AF_UNSPEC as the default family passed to getaddrinfo()
is a bad default in many configurations; it means queries will be
performed, regardless of whether any IPv6 addresses are configured on
any interfaces (on most systems). The extra DNS queries mean slower
lookups.
The
On Sat, Jul 12, 2003 at 03:38:15PM +0100, Max Bowsher wrote:
> Unlike most other autoconf defines, this is *defined to 0* when the check
> symbol is not found. (See info autoconf, search on "Macro: AC_CHECK_DECLS")
>
> The usage in threadproc/unix/signals.c:
> #if defined(SYS_SIGLIST_DECLARED) ||
On Mon, Jul 14, 2003 at 10:02:06AM +0100, Joe Orton wrote:
> On Sat, Jul 12, 2003 at 03:38:15PM +0100, Max Bowsher wrote:
> > Unlike most other autoconf defines, this is *defined to 0* when the check
> > symbol is not found. (See info autoconf, search on "Macro: AC_CHECK_DECLS&
On Wed, Jul 23, 2003 at 03:16:07AM +0200, [EMAIL PROTECTED] wrote:
> Hi,
>
> I am writing an Apache 2.0 module and have problems with the mutex_trylock
> functions.
>
> apr_global_mutex_lock() and apr_global_mutex_unlock() work but
> apr_global_mutex_trylock() segfaults the apache2 childs on my
On Thu, Jul 24, 2003 at 11:51:18AM +0100, David Reid wrote:
> > On Wed, Jul 23, 2003 at 03:16:07AM +0200, [EMAIL PROTECTED] wrote:
> > > Hi,
> > >
> > > I am writing an Apache 2.0 module and have problems with the
> mutex_trylock
> > > functions.
> > >
> > > apr_global_mutex_lock() and apr_global_m
On Wed, Jul 30, 2003 at 11:21:21AM -0500, William Rowe wrote:
> Attached is a patch that appears to be as close as thread-safe
> (but altogether not reentrant) as I can accomplish.
>
> The existing unix/thread_mutex.c code is neither thread-safe
> nor reentrant when using the 'nested' (default) th
On Wed, Jul 30, 2003 at 01:34:41PM -0500, William Rowe wrote:
...
> Because the new patch protects the uninitalization of the mutex while
> the lock is still held, the only failure scenario that remains is;
>
> 1. thread is interrupted (e.g. signal handler) in between the unsetting
>of the own
On Sun, Aug 03, 2003 at 01:17:15AM +0200, André Malo wrote:
> Well, I've started with a patch (attached). It works fine for me, but I'm not
> sure with the naming at all. Any comments?
I understand the need for a strcasecmp replacement, but are the
strcmp/!strcmp replacements just for consistency?
1 - 100 of 1017 matches
Mail list logo
