Package: libnewlib-dev
Version: 1.18.0-6
Severity: wishlist
I have a really hard time compiling anything against libnewlib-dev,
for example, stddef.h is missing and wint_t (sys/_types.h) missing,
/usr/lib/newlib/i486-linux-gnu/include/sys/types.h includes a file
called features.h which isn’t
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA384
Build instructions:
• my stdio-c99, arc4random¹ and mkstemp patches still aren’t merged,
which means we must disable C99 mode in gcc (see below) and add
an either tempnam or, preferred, mkstemp function
① while mksh doesn’t use arc4random(3)
Russ Allbery dixit:
I agree with others in this thread that having a UTF-8 locale without the
collation changes implied by en_US is very useful for various software
packages such as automated test suites that want reproducible results and
were originally written for the C locale.
Same for
Samuel Thibault dixit:
LC_CTYPE has differences between locales, transliterations notably. For
Oh, okay – good to know…
I'd say go on :)
OK.
(of course we'll need to wait for libc to provide the locale
(post-squeeze I guess) before changing the policy).
Sure. Maybe think of something to
Samuel Thibault dixit:
believe that's something that shouldn't break Squeeze at all.
I also believe it cannot possibly do that.
bye,
//mirabilos
--
“It is inappropriate to require that a time represented as
seconds since the Epoch precisely represent the number of
seconds between the
be undefined
+
+ -- Thorsten Glaser t...@mirbsd.de Sat, 04 Sep 2010 12:39:07 +
+
elfutils (0.148-1) unstable; urgency=low
* New upstream release
diff -Nru elfutils-0.148/debian/patches/debian-changes-0.148-1+m68k elfutils-0.148/debian/patches/debian-changes-0.148-1+m68k
--- elfutils-0.148
On Wed, 15 Sep 2010, Olivier Berger wrote:
Creating other includes
invoke-rc.d: unknown initscript, /etc/init.d/postgresql-8.4 not found.
dpkg: error processing gforge-db-postgresql (--configure):
Fixed in r10368 for FF trunk and 5.0 branch; as Roland tagged
5.02 recently, I suppose this may
On Wed, 15 Sep 2010, Olivier Berger wrote:
I'm not sure the whole of 5.0.2 would require freeze exception, and I
guess this only fix could just be backported to squeeze's version in any
case.
As it’s only bugfixes, we could at least try.
//mirabilos
--
tarent Gesellschaft für
On Wed, 22 Sep 2010, Olivier Berger wrote:
Still, the gforge-web-apache2 package should ship an empty
/usr/share/gforge/plugins/ placeholder maybe.
IIRC Debian packages are not allowed to ship empty directories…
bye,
//mirabilos
--
tarent Gesellschaft für Softwareentwicklung und IT-Beratung
On Thu, 23 Sep 2010, Olivier Berger wrote:
or a file could be there to
explain the purpose of the dir : manual installation of plugins ? ;)
I’d say that would work, yes.
bye,
//mirabilos
--
tarent Gesellschaft für Softwareentwicklung und IT-Beratung mbH
Geschäftsführer: Boris Esser, Elmar
tags 601925 + pending
thanks
Clytie Siddall dixit:
The updated Vietnamese translation for the debconf file: mksh
Thanks, committed into CVS, will be integrated with the
next upload.
Note: please use the po-debconf utility:
http://people.debian.org/~jfs/debconf6/html/x847.html
which will
URL from fthain.
+ * debian/rules.patch: Add pr37053 (m68k only, to prevent a potential
+regression on cris) and pr46179, pr39531 on m68k.
+ * debian/rules.defs: remove m68k from locale_no_cpus.
+
+ -- Thorsten Glaser t...@mirbsd.de Sat, 30 Oct 2010 14:45:32 +
+
gcc-4.4 (4.4.5-5
/kd.c: if KDGHWCLK is not defined, even on m68k,
+do not build this deprecated interface; the fallback code
+was wrong and did not compile anyway. Closes: #578168
+
+ -- Thorsten Glaser t...@mirbsd.de Mon, 01 Nov 2010 15:05:20 +
+
util-linux (2.17.2-3.3) unstable; urgency=low
* Non
Hi,
I was told in http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37053#c27
that we want to add another backport¹ to the patches list. Will do so
in my next build, if you happen to upload before that I’d be glad to
have these patches accepted, otherwise I’ll provide an updated debdiff.
①
which would be correct/desired
+ * debian/debhelper.in/libc.preinst: require 2.6.32 kernel on m68k
+
+ -- Thorsten Glaser t...@mirbsd.de Mon, 01 Nov 2010 15:09:16 +
+
eglibc (2.11.2-7) unstable; urgency=low
[ Samuel Thibault ]
diff -u eglibc-2.11.2/debian/debhelper.in/libc.preinst
eglibc
Dixi quod…
I was told in http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37053#c27
that we want to add another backport¹ to the patches list. Will do so
Resynched my changes against 4.4.5-6, will provide updated debdiff in
some 3-4 days, once it compiled ;-)
bye,
//mirabilos
--
I believe no one
together to avoid regressions.)
+ * debian/rules.defs: remove m68k from locale_no_cpus.
+
+ -- Thorsten Glaser t...@mirbsd.de Wed, 03 Nov 2010 21:57:06 +
+
gcc-4.4 (4.4.5-6) unstable; urgency=low
* Update to SVN 20101029 from the gcc-4_4-branch (r166075).
diff -u gcc-4.4-4.4.5/debian
Lasse Collin dixit:
default memory usage limit for compression (but not decompression), but
I don't know what that should be (40 %, 80 %, 95 % of RAM? max(80 % of
RAM, RAM - 256 MiB)?). Some default limit may be needed in the future
RAM is irrelevant, datasize ulimit (soft, maybe hard if you
On Tue, 28 Sep 2010, Roland Mas wrote:
,
| v=0 ; invoke-rc.d postgresql restart || v=$?
`
Ouch, sorry. My bad. Thanks Roland for spotting this.
//mirabilos
--
tarent Gesellschaft für Softwareentwicklung und IT-Beratung mbH
Geschäftsführer: Boris Esser, Elmar Geese
HRB AG Bonn 5168 -
Source: linux-2.6
Version: 2.6.32-24
Severity: wishlist
Tags: sid
Please see to that -fno-omit-frame-pointer is used during compilation
of a Debian m68k kernel as long as it is built with gcc-4.3 (I think
defining CONFIG_FRAME_POINTER should be enough, I have yet to build
any kernel at all though
Dixi quod…
Please see to that -fno-omit-frame-pointer is used during compilation
of a Debian m68k kernel as long as it is built with gcc-4.3 (I think
defining CONFIG_FRAME_POINTER should be enough, I have yet to build
any kernel at all though and have not tested this).
ARCH_WANT_FRAME_POINTERS
+CONFIG_ARCH_WANT_FRAME_POINTERS=y
+CONFIG_FRAME_POINTER=y
Corresponding changelog entry:
[ Thorsten Glaser ]
* debian/config/m68k/config: define CONFIG_ARCH_WANT_FRAME_POINTERS
and CONFIG_FRAME_POINTER to yes (Closes: #598893)
Note that this *only* applies to the sid kernel
Source: linux-2.6
Version: 2.6.32-24
Severity: important
Tags: experimental sid
After compiling an m68k kernel for ARAnyM using the atari platform,
requiring the fix from Debian #598893 as well as a patch I've sub-
mitted upstream¹ in order to allow inclusion into the Debian kernel,
and getting
On Wed, 6 Oct 2010, Olivier Berger wrote:
Local customizations (that one would think are naturally welcome in
local.inc, from the file's name) in local.inc are lost upon package
upgrade.
Yeah, it’s sadly not. /etc/gforge/local.d/* with files copied
from /usr/share/gforge/etc/local.d/* and
Bastian Blank dixit:
m68k is not a release architecture. Also all newer kernels uses gcc 4.4.
It doesn’t happen in gcc-4.4 but the kernel currently in sid uses gcc-4.3.
bye,
//mirabilos
--
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or
Ben Hutchings dixit:
On Wed, Oct 06, 2010 at 04:25:43PM +, Thorsten Glaser wrote:
Bastian Blank dixit:
m68k is not a release architecture. Also all newer kernels uses gcc 4.4.
It doesnb
But we aren't going to release with m68k, so why not concentrate on
trunk/experimental?
Makes
BZ.
+ * debian/rules.patch: Add pr41302, pr43804 on m68k.
+
+ -- Thorsten Glaser t...@mirbsd.de Mon, 18 Oct 2010 16:08:12 +
+
gcc-4.4 (4.4.5-3) unstable; urgency=medium
* Update to SVN 20101011 from the gcc-4_4-branch (r164607, 4.4.5 release
diff -u gcc-4.4-4.4.5/debian/rules2 gcc-4.4
+ and tls now; sanity checks were only disabled for linuxthreads)
+- raise minimum kernel version to 2.6.16 and document why we can't
+ set it to 2.6.32 (Debian) yet which would be correct/desired
+ * debian/debhelper.in/libc.preinst: require 2.6.32 kernel on m68k
+
+ -- Thorsten Glaser t
Applying patch pr41302.diff
patching file src/gcc/ChangeLog
Hunk #1 FAILED at 1.
1 out of 1 hunk FAILED -- rejects in file src/gcc/ChangeLog
patching file src/gcc/config/m68k/m68k.c
patching file src/gcc/testsuite/ChangeLog
Hunk #1 FAILED at 1.
1 out of 1 hunk FAILED -- rejects in file
… with eglibc and gcc-4.4 so it might not be an issue of having too
old a toolchain (these are current but not quite uploadable, besides
needing elfutils to compile newer kernels).
Why this error is thrown, I have no idea.
bye,
//mirabilos
--
I believe no one can invent an algorithm. One just
OK, more details:
linux-2.6 build-depends on libelf-dev which is built and installable.
elfutils is not installable because libasm1 isn’t built (but that
doesn’t matter for compiling a linux kernel). This is probably
what you were hinting at with saying it has never worked on m68k.
Hope that
:00:00 2001
From: Thorsten Glaser t...@mirbsd.org
Date: Sun, 24 Oct 2010 16:42:55 +
Subject: [PATCH] add whitespace in output (cosmetic fix) to avoid confusion
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
With pbuilder-satisfydepends-classic, output
I can indeed confirm that just adding the dependency to the util-linux
binary package fixes this very error. (I can’t recompile util-linux on
m68k at the moment, due to a gcc bug, so I had to hack the binary pak-
kage to make cowbuilder work.)
HTH HAND,
//mirabilos
--
FWIW, I'm quite impressed
+dfsg/debian/changelog
@@ -1,3 +1,9 @@
+gmp (2:4.3.2+dfsg-1+m68k) unreleased; urgency=low
+
+ * Add m68k support.
+
+ -- Thorsten Glaser t...@mirbsd.de Sun, 24 Oct 2010 20:36:22 +
+
gmp (2:4.3.2+dfsg-1) unstable; urgency=low
* New upstream version.
Probably the result of a too old libc, but I’m held up by a gcc-4.4
bug at the moment, so I cannot confirm/rebuild. Just FYI.
If you want, close, I’ll reopen if needed then.
//mirabilos
--
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or only
Source: ppl
Version: 0.10.2-8
Severity: important
Hi,
apparently, the build dependencies are not tight enough.
Running cowbuilder --debbuildopts -B --binary-arch --build
ppl_0.10.2-8.dsc fails (after a day and a half, mind you)
with:
[…]
ppl_prolog_Octagonal_Shape_mpz_class.o
Hi again,
the problems with util-linux on m68k go deeper than I had thought,
and are in no way related to libc:
The failure is in hwclock/kd.c, which tries to use the keyboard
interface to access the clock on m68k, and only on m68k (the
entire source code is wrapped in a huge ifdef). However,
Finn Thain dixit:
[ use rtc ]
@util-linux maintainers: could we please get an upload to unstable
where kd.c is not built at all or the “#ifndef __m68k__” is repla-
ced with an “#if 1” then?
CONFIG_GEN_RTC=m
I think Stephen set it to y.
Try this:
r...@aranym:/var/cache/pbuilder/result #
Package: swig1.3
Severity: wishlist
Hi,
I’m currently in the process of sort-of rebootstrapping Debian/m68k.
Could you please do me a favour in your next uploads of swig1.3 and
make m68k a java-less architecture, as I’d like to avoid it this
early in the process (especially considering gcc build
found 363193 pbuilder/0.199
thanks
glib2.0_2.24.2-1.dsc (emphasis mine):
Build-Depends: debhelper ( 5.0.22), pkg-config (= 0.14.0), gettext,
autotools-dev, gnome-pkg-tools (= 0.11), quilt, dpkg-dev (= 1.14.13),
libpcre3-dev (= 7.4-1~), desktop-file-utils, gtk-doc-tools,
libselinux1-dev
Thanks,
building gcc-4.4_4.4.5-7 now. Will close if it works.
bye,
//mirabilos
--
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or only God invents algorithms, we merely copy them.
If you don't believe in God, just consider God as Nature if
You might know better what to do with this. I have no idea, but
looks legit…
[…]
dh_makeshlibs -plibgcc2 -plibgcc2-dbg -- -v4.4.5-7
dpkg-gensymbols: warning: some new symbols appeared in the symbols file: see
diff output below
dpkg-gensymbols: warning: debian/libgcc2/DEBIAN/symbols doesn't match
Besides that, I got:
dpkg-gensymbols: warning: new libraries appeared in the symbols file:
libobjc_gc.so.2
That are the only symbols warnings I got. The changes file is
generated, now I can upload to unstable for the first time.
Thanks for the quick responses/uploads, this helped immensely.
tags 603801 = upstream wontfix
thanks
Jörg-Volker Peetz dixit:
With mksh the second (nth) M-. makes the first insertion of the filename
Sorry, this was implemented as a feature request to make things more
zsh-like: http://article.gmane.org/gmane.os.miros.general/8565
bye,
//mirabilos
--
I
Hi,
similar solution for different problem. I have installed fakeroot as
extra package during cowbuilder --create, but pbuilder insists on
installing it on each build.
You might not notice it, but on m68k (Atari) it’s another minute or
longer (on a very fast machine – slower
tags 603801 - wontfix
thanks
J�rg-Volker Peetz dixit:
It would be only a modification of this requested feature: As long as the
cursor
is not moved it will work as requested, similar to zsh and bash. Only in the
case the cursor is moved inbetween, the next M-. should insert at the current
Bastien ROUCARIES dixit:
to reproduce:
Thanks for the detailed reproduce information. I believe I have never
used ssh-add -c myself but see how it would be useful, especially in
combination with kwalletcli.
I will have to see how ssh calls ssh-askpass when it wants only a
confirmation (instead
Jonathan Nieder dixit:
[1] except that preventing sysadmins from making /bin/sh point where
they want is pretty unfriendly.
ACK.
Thoughts?
See my signature. Anyway, I think we should have some common, shared
code across all shells that are eligible as /bin/sh in Debian, and
maybe manage
Jonathan Nieder dixit:
$ sh -c 'trap echo hi; exit EXIT; exit 9'; echo $?
^
This “exit” command above will return the errorlevel of the echo call.
I don’t believe I will change this, at first glance (it’s past mid-
night though, but now you can work on it).
This
Jonathan Nieder dixit:
Justification: posix [1]
It is handy to be able to use the old exit status from an EXIT handler
by using bare exit:
$ sh -c 'trap echo hi; exit EXIT; exit 9'; echo $?
hi
9
[1]
http://www.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_21_14
(Note
Jonathan Nieder dixit:
Thoughts? Questions? Advice?
Other shells would need to be changed to cope with that as well.
The situation is a bit tricky at the moment, because mksh also
contains code to manage a diversion, which only works if dash
doesn’t, though.
Worse, on upgrade from lenny to
I wonder…
with what scripts exactly this would help?
I’ve never run into such a problem.
bye,
//mirabilos
--
«MyISAM tables -will- get corrupted eventually. This is a fact of life. »
“mysql is about as much database as ms access” – “MSSQL at least descends
from a database” “it's a rebranded
Finn Thain dixit:
I guess you patched your compilers as discussed on linux-m68k?
Interestingly enough, I was not hit by GCC PR/41302 because I built
with CONFIG_ARCH_WANT_FRAME_POINTERS=y and CONFIG_FRAME_POINTER=y
to work around GCC PR/37052 already which defines not only
Package: sks
Version: 1.1.0-4
Severity: normal
We’re running a private keyserver (publicly accessible, but not
synchronised with the keyserver network, and forbidden for search
engines to index via robots.txt, for eMail address privacy) with
the stock Lenny sks package and see occasional database
Hi,
because some of our project management guys at work needed it,
I’ve taken Christophe’s package and polished it a little, for
better debian-policy conformance etc. (for example, startscript
uses now /bin/sh; desktop, menu and mime stuff are added; bugs
and warnings fixed)
I’ve uploaded a
Hi again!
Hm. Maybe I should mention that *.gan files still show the text icon in
Konqueror and are associated with some KDE text editor, despite my tries
w.r.t. the MIME thing.
bye,
//mirabilos
--
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.]
Package: wnnp
Severity: wishlist
I stumbled upon ↓ and would like to use it at work, therefore a
Debian package is needed ☺
http://devzone.zend.com/article/1712
This is basically a way to use Perl from within PHP. I used the
dh-make-pecl thing to generate a skeleton, fixed debian/rules’
「clean」
Dixi quod…
This is basically a way to use Perl from within PHP. I used the
dh-make-pecl thing to generate a skeleton, fixed debian/rules’
「clean」 target, ran it through lintian, etc.
RFC is here:
https://eurynome.mirbsd.org/debs/dists/sid/wtf/pkgs/php-perl/php-perl_1.0.0-1.dsc
Built on a sid
Okay, again; waldi told me there are missing information,
and that debian-de...@lists.d.o has to be Cc’d.
Package name: php-perl (Binary: php5-perl)
Version:1.0.0
Upstream Author:Dmitry Stogov
Licence:PHP 3.0 (upstream), GPL (packaging)
Language:
Hi again,
looks like it works now. I assume you can close the bug?
bye,
//mirabilos
--
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting
Try mksh ;)
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
HTEncodingToCompressType has this redundant, should maybe fixed too:
• the { } are not needed
• assigning to result is redundant, it’s already initialised to cftNone
• we can shortcut the return
Index: HTFile.c
===
RCS file:
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
I would also like to see support for these two formats, because my
(currently) two packages could benefit from it somewhat.
The problem is mostly that, while one can keep the debian/ stuff in
CVS, dpkg-source does *not* ignore the CVS/
Package: libklibc-dev
Version: 1.5.12-2
Severity: important
Excerpt from dietlibc sys/resource.h:
│int getpriority(int which, int who) __THROW;
│int setpriority(int which, int who, int prio) __THROW;
Package: libklibc-dev
Version: 1.5.12-2
Severity: important
SUSv3 function killpg(3) is missing from signal.h in klcc
-- System Information:
Debian Release: 5.0
APT prefers stable
APT policy: (500, 'stable')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.26-1-xen-amd64 (SMP w/1 CPU core)
Package: libklibc-dev
Version: 1.5.12-2
Severity: wishlist
To compile mksh against klibc, I need at least the standardised
tempnam function, but preferably, the mkstemp extension would be
desired.
Example implementation:
http://www.mirbsd.org/cvs.cgi/src/lib/libc/stdio/mktemp.c
(in klibc, you'd
Package: libklibc-dev
Version: 1.5.12-2
Severity: normal
t...@lenny:~ $ fgrep -r killpg /usr/lib/klibc/include | wc -l
fgrep: /usr/lib/klibc/include/asm/asm-x86_64: Too many levels of symbolic links
0
t...@lenny:~ $ ls -laFd /usr/lib/klibc/include/asm/asm-x86_64
lrwxrwxrwx 1 root root 10 Feb 19
Package: libc6-dev
Version: 2.9-6
Severity: wishlist
(mass filing with linux-libc-dev and libc6-dev)
/usr/include/aio.h: char __unused[32];
/usr/include/aio.h: char __unused[32];
/usr/include/bits/stat.h:long int __unused[3];
/usr/include/bits/stat.h:long int __unused[3];
Package: debian-policy
Version: 3.8.1.0
Severity: wishlist
For the mksh regression tests, I need a UTF-8 locale working; most
systems either provide “en_US.UTF-8” or “en_US.utf8” with the former
being recommended.
Build-depending on locales-all has worked for me so far, except it
won’t do in
Package: linux-libc-dev
Version: 2.6.29-2
Severity: wishlist
(mass filing with linux-libc-dev and libc6-dev)
/usr/include/asm/stat.h:long__unused[3];
/usr/include/linux/icmp.h: __be16 __unused;
/usr/include/linux/sysctl.h:unsigned long __unused[4];
These
Package: mksh
Version: 37.2-1
Severity: serious
Justification: no longer builds from source
http://buildd.debian-ports.org/fetch.php?pkg=mksharch=hurd-i386ver=37.2-1stamp=1238962232file=logas=raw
mksh cannot build on hurd-i386 because locales-all fails to install.
This either needs to be fixed
Package: mksh
Version: 37.2-1
Severity: important
https://buildd.debian.org/fetch.cgi?pkg=mksh;ver=37.2-1;arch=ia64;stamp=1238953108
While the package builds, it is unusable, as the regression tests
show: it SIGBUSes – on IA64 only (not Alpha, which is normally quite
picky about alignment too).
Package: mksh
Version: 37.2-1
Severity: wishlist
https://buildd.debian.org/fetch.cgi?pkg=mksh;ver=37.2-1;arch=ia64;stamp=1238953108
Regression test failures are not fatal to the build.
While this is intentional, since some tests are known to fail
on some environment, such as the slow MIPS
block 522777 by 274699 522776
thanks
Note: strictly spoken, only one of the two needs to be fixed
to unblock this one.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Package: locales-all
Version: 2.9-6
Severity: normal
This is probably the same as #274699 as the error is the same,
and I’ll mark it as blocking this bug after reporting.
mksh FTBFS on hurd-i386 because its dependency locales-all is
not installable: #522777
Giacomo A. Catenazzi dixit:
If you need a specific locale (as seems from mksh, not
sure if it is a bug in that program), you need to set it.
You can only set a locale on a glibc-based system if it’s
installed beforehand, which root needs to do.
Why does mksh need UTF-8?
The regression tests
Bill Allombert dixit:
What about LC_COLLATE (which is a major problem with sort(1)) ?
1:1, just like the C locale does.
What about packages that run before /usr is mounted ?
They do not have /usr/*/locale/ anyway. This is a glibc problem.
What about embedded systems with tight space
ANALYSIS OF THE BUG
───
Thorsten Glaser dixit:
Package: mksh
Version: 37.2-1
Severity: important
https://buildd.debian.org/fetch.cgi?pkg=mksh;ver=37.2-1;arch=ia64;stamp=1238953108
While the package builds, it is unusable, as the regression tests
show: it SIGBUSes bpicky about
https://buildd.debian.org/fetch.cgi?pkg=mksh;ver=37.2-1;arch=s390;stamp=1238952678
diet mksh segfaults, but normal mksh doesn’t
might be related though…
//mirabilos
--
“It is inappropriate to require that a time represented as
seconds since the Epoch precisely represent the number of
seconds
Adeodato Simó dixit:
+ Thorsten Glaser (Tue, 07 Apr 2009 18:54:59 +):
Except the ton which sets LC_ALL=C to get sane (parsable,
dependable, historically compatible) output.
These would then unset all other LC_* and LANG and LANGUAGE,
and only set LC_CTYPE to C.UTF-8 to get old behaviour
Bill Allombert dixit:
Fortunately, since Sarge, debian-installer set LANG in
/etc/environment so programs almost never run under C locale anymore.
Except the ton which sets LC_ALL=C to get sane (parsable,
dependable, historically compatible) output.
These would then unset all other LC_* and
Roger Leigh dixit:
However, I would ideally like the C/POSIX locales to be UTF-8
by default as on other systems (with a C.ASCII variant if required).
No, this has the potential to break, for example, tr(1).
I lived through that on MirBSD.
//mirabilos
--
“It is inappropriate to require that a
Adeodato Simó dixit:
I would go as far as suggesting that some package like libc6 itself
FWIW:
-rw-r--r-- 1 tg tg 238336 Apr 7 22:59 en_US.UTF-8/LC_CTYPE
It's not *that* much...
Finally, this stuff that Roger proposes about making “C” be UTF-8, and
create some C.ASCII for people needing
Roger Leigh dixit:
But, does it not reject non 7-bit input in the C
locale for completeness?
No, it doesn't - we (before my time though, I think) fought
hard for eight-bit transparence and eight-bit cleanliness.
Should tools doing raw I/O not be using lower level interfaces
such as fread() and
Roger Leigh dixit:
Are you sure?
Not entirely, but I recall fgetc (or was it fgetwc?)
being affected.
//mirabilos
--
“It is inappropriate to require that a time represented as
seconds since the Epoch precisely represent the number of
seconds between the referenced time and the Epoch.”
Package: mksh
Version: 37.3-1
Severity: wishlist
FAIL ../../mksh/check.t:history-subst-4
Description:
Global substitutions work
(ksh88 and ksh93 do not have -g option)
unexpected exit status 256 (exit-code 1), expected 0
unexpected stdout -
Package: mksh
Version: 37.3-1
https://buildd.debian.org/fetch.cgi?pkg=mksh;ver=37.3-1;arch=s390;stamp=1239145276
./mksh -c true
make[1]: *** [do-build] Illegal instruction
make[1]: Leaving directory `/build/buildd/mksh-37.3/build/small'
This error might be in dietlibc too; mksh linked against
Package: dietlibc-dev
Version: 0.31-1.2
Severity: normal
This is a reduced testcase from the mksh source code.
Public domain, as it’s not complex enough for copyright protection…
The system I took the below snippet on is maintained by waldi,
but the s390 buildd has the same problem.
Interestingly enough, the dietlibc and glibc ideas of sigjmp_buf
deviate slightly: if __WORDSIZE == 64 the glibc one is bigger;
otherwise they are identical.
On debian04v2.zseries.org however __WORDSIZE == 32 so this is
not the cause of the problem.
bye,
//mirabilos
--
“It is inappropriate to
Dixi quod…
waldi requested disassembly.
t...@debian04v2:~$ diet -Os gcc testcase.c ./a.out
/usr/lib/diet/lib-s390/libc.a(vprintf.o): In function printf':
vprintf.c:(.text+0x28): warning: warning: the printf functions add several
kilobytes of bloat.
Segmentation fault
(gdb) r
Starting
Giacomo A. Catenazzi dixit:
The locale C is already a UTF-8 compatible locale.
It is UTF-8 transparent but that's its pro and con.
It does not tell the system that UTF-8 encoding is to be used.
It basically says the encoding is none/unknown.
Why build need to depend to a locale?
[...]
For
tags 518359 confirmed fixed-upstream
thanks
I got a fix in mksh-current and am testing that right now; code similar
to the one in question from resolvconf has made it into an mksh testcase.
mksh now adheres more strict to the standard, so that scripts exploiting
them should work as expected (“set
Niko Tyni dixit:
I believe this happens when dash is not installed so that dash/sh
is unknown. With SHELLOPTS=xtrace:
Thanks, didn’t know about it.
Hope this helps,
Maybe like so?
--- mksh.config 14 Dec 2008 16:51:02 - 1.2
+++ mksh.config 23 Mar 2009 10:34:07 -
@@ -10,3 +10,3 @@
Package: lintian
Version: 2.2.8
Severity: normal
E: mksh: diversion-for-unknown-file $2 postinst:16
E: mksh: package-uses-local-diversion postinst:24
E: mksh: package-uses-local-diversion prerm:16
E: mksh: orphaned-diversion $2 postinst
I have the same problem (although I'm not exactly sure
Package: mksh
Version: 35.2-3
Severity: normal
mksh fails to preconfigure with exit code 10 when selecting
true for debconf:mksh/sh
However, installing mksh as /bin/sh manually with ln(1) works
just fine.
-- System Information:
Debian Release: 5.0
APT prefers stable
APT policy: (500,
Package: resolvconf
Version: 1.42
Severity: grave
Tags: patch
Justification: renders package unusable
Hello,
the /etc/resolvconf/update.d/libc script contains set -e, which
means the shell aborts execution if one of the commands returns
a non-zero exit status.
The aforementioned script contains
Package: libengine-tpm-openssl
Version: 0.4.1+20071221-4
Severity: normal
$ openssl rand -engine tpm -out x 128
can't use that engine
4012:error:25066067:DSO support routines:DLFCN_LOAD:could not load the shared
library:dso_dlfcn.c:162:filename(libtspi.so): libtspi.so: cannot open shared
object
Dixi quod…
I will have to dig out the change which was first applied
then reverted from mksh upstream, to see if this is indeed
what I fear it is.
Okay. I now tested:
• mksh R23 (which contained exec.c,v 1.5)
• mksh R23 with exec.c,v 1.4
The actual patch came in at revision 1.3 from Debian
Hi,
any progress (forwarded to)?
There’s another patch at
‣ http://lists.samba.org/archive/rsync/2005-February/011653.html
also attached in case it goes “bad”. This seems to use a different
checksum (seems to be the one rsync itself uses?) but can be
parametrised a lot… he does say he has
Mark Brown dixit:
If you're interested I suggest working directly upstream to try to
push whatever implementation you prefer; an enthusiastic user can't
Mh. I’m not yet a user though.
do something smart with the compression level options (which was the
main problem with the previous patch
301 - 400 of 4571 matches
Mail list logo
