Re: regex_internal.h modification lead to GCC crash

2005-09-19 Thread Paul Eggert
This is getting a bit complicated and I suspect we will run into further problems like this How about if we do this instead? The basic idea is to shut off __attribute__ entirely for GCC before 3.1. There's little point to figuring out exactly when each attribute was added. We can simply pic

Re: regex_internal.h modification lead to GCC crash

2005-09-19 Thread Ben Pfaff
Derek Price <[EMAIL PROTECTED]> writes: > I can reproduce this on some half-dozen platforms. It has to do with > calling __attribute() without arguments with gcc 2.95.4 or so. They all > seg-fault but one, which manages to print an error. > > I've attached a patch, which defines attribute_always

Re: regex_internal.h modification lead to GCC crash

2005-09-19 Thread Derek Price
Yoann Vandoorselaere wrote: >Hi list, > >Recently introduced modification to regex_internal.h lead to a crash >when compiling GnuLib regex module under FreeBSD-4.11 GCC-2.95.4. > >The same module is reported to compile successfully under FreeBSD-5.4, >GCC-3.4.2. >The GCC crash was traced down to

regex_internal.h modification lead to GCC crash

2005-09-19 Thread Yoann Vandoorselaere
Hi list, Recently introduced modification to regex_internal.h lead to a crash when compiling GnuLib regex module under FreeBSD-4.11 GCC-2.95.4. The same module is reported to compile successfully under FreeBSD-5.4, GCC-3.4.2. The GCC crash was traced down to the following addition: +#if !__GNUC

Re: [bug-gnulib] gnulib-tool --import slowness

2005-09-19 Thread Derek Price
Bruno Haible wrote: >Derek Price wrote: > > >>The --tests-base option is no longer defaulting to "tests" as >>`gnulib-tool --help' specifies, however: >> >>gnulib-tool: *** missing --tests-base option >>gnulib-tool: *** Stop. >> >> > >Fixed. Thanks for reminding me. > > > Looks good. Tha

Re: [bug-gnulib] gnulib-tool --import slowness

2005-09-19 Thread Bruno Haible
Derek Price wrote: > The --tests-base option is no longer defaulting to "tests" as > `gnulib-tool --help' specifies, however: > > gnulib-tool: *** missing --tests-base option > gnulib-tool: *** Stop. Fixed. Thanks for reminding me. Bruno ___ bug-gnul

Re: pathmax module license

2005-09-19 Thread Jim Meyering
Jim Meyering <[EMAIL PROTECTED]> wrote: > Yoann Vandoorselaere <[EMAIL PROTECTED]> wrote: >> On Wed, 2005-08-31 at 08:35 +0200, Jim Meyering wrote: >>> Yoann Vandoorselaere <[EMAIL PROTECTED]> wrote: >>> > Hi Jim, >>> > >>> > Would it be possible to change the pathmax license from GPL to LGPL ? I >

Re: config.h inclusion

2005-09-19 Thread Paul Eggert
Ralf Wildenhues <[EMAIL PROTECTED]> writes: > OK. Here's the second patch updated, then. Thanks. Since there was general agreement I installed all the patches, except for the following files: mkdtemp.c setenv.c unsetenv.c argp-eexst.c argp-fmtstream.c argp-fmtstream.h argp-fs-xinl.c argp-xinl

Re: [bug-gnulib] gnulib-tool --import slowness

2005-09-19 Thread Derek Price
Bruno Haible wrote: >Derek Price wrote: > > >>$ time maint-aux/gnulib-update >> >>real14m42.968s >>user3m45.188s >>sys 10m41.704s >>$ >> >>Just a few weeks back, I think it was taking a minute or three to run >> >> > >Should be better now. > > Much better now, thanks: [EMAIL

socklen.m4 typo

2005-09-19 Thread Simon Josefsson
I installed this. 2005-09-19 Simon Josefsson <[EMAIL PROTECTED]> * socklen.m4: Fix typo. --- socklen.m4 08 Sep 2005 22:06:42 +0200 1.1 +++ socklen.m4 19 Sep 2005 18:09:34 +0200 @@ -18,3 +18,4 @@ # include #endif) ]) +]) ___

gnulib-tool changes

2005-09-19 Thread Bruno Haible
Hi, I made the following changes to gnulib-tool: - Fixed the speed problem that some of you noticed. O(N) instead of O(N²). - Made --with-tests work also with --import, not just with --create-testdir and --create-megatestdir. - Option --dry-run now behaves more reasonably. - The docume

Re: gnulib-tool --import slowness

2005-09-19 Thread Simon Josefsson
Bruno Haible <[EMAIL PROTECTED]> writes: > Derek Price wrote: >> Bruno Haible wrote: >> >files. I'll change this to use "join" and "sort" instead of a double loop. >> >> Did you do this? I'm still getting pretty slow operation with CVS: >> >> $ time maint-aux/gnulib-update >> >> real14m42.968

FYI: GNULIB glob-libc.h path in srclist.txt

2005-09-19 Thread Derek Price
2005-09-19 Derek Price <[EMAIL PROTECTED]> * srclist.txt: glibc's glob.h is now in lib/glob-libc.h. Derek -- Derek R. Price CVS Solutions Architect Ximbiot v: +1 717.579.6168 f: +1 717.234.3125 2005-09-16 Paul Eggert <[EMAIL PROTECT

Re: [bug-gnulib] gnulib-tool: don't ever forget to set $dry again

2005-09-19 Thread Bruno Haible
Ralf Wildenhues wrote: > Hmm. This way forces you to write > if $dry; then > echo do something > else > do something > fi Yes. I think the user of gnulib-tool prefers to read an English summary of the actions that would be performed, rather than a list of shell commands. Bruno _

Re: [PATCH] Re: wctype.h and wchar.h inclusion

2005-09-19 Thread Yoann Vandoorselaere
On Mon, 2005-09-05 at 15:23 -0400, Derek Price wrote: > Yoann Vandoorselaere wrote: > > >On Thu, 2005-09-01 at 12:28 +0200, Yoann Vandoorselaere wrote: > > > > > >>Hi, > >> > >>An OpenBSD Prelude user reported that GnuLib will fail to compile on > >>OpenBSD 3.7 due to the new dependencies of mod

Re: [bug-gnulib] gnulib-tool --import slowness

2005-09-19 Thread Bruno Haible
Derek Price wrote: > Bruno Haible wrote: > >files. I'll change this to use "join" and "sort" instead of a double loop. > > Did you do this? I'm still getting pretty slow operation with CVS: > > $ time maint-aux/gnulib-update > > real14m42.968s > user3m45.188s > sys 10m41.704s > $ > > J