Folks are already working on ports for other systems, so ai object to these
files in this form (work for FreeBSD and OpenBSD is going on)
And sorry for topposting...
--
Marc Balmer
micro systems, http://www.msys.ch/
Tel. +41 61 383 05 10, Fax +41 61 383 05 12
Am 07.08.2012 um 23:25 schrieb
After building CDE, and starting dtlogin for the first time, I had a
deja-vu of the bad kind... The fonts as is are true ugly, because they
are non-antialiased. But since CDE builds on top of OpenMotif, which
has had antialiased fonts (and UTF8) support for quite some time, I
changed the Dtlogin
Am 08.08.12 09:05, schrieb Frederic Koehler:
> Thanks for the feedback; here are some revised patches
> and a third group of patches which fix enough segfaults
> by removing implicit function definitions
> to allow CDE to startup on x64, albeit to a very buggy desktop.
>
> ==
> Av
Use strlen, not sizeof, here. Fixes a segfault on Debian squeeze 64 bit
and most probably other systems, too.
(If the mailer mangles the diff, I can resend it as attachment)
diff --git a/cde/lib/DtSvc/DtUtil1/DtsMM.c b/cde/lib/DtSvc/DtUtil1/DtsMM.c
index dd82d6f..0004afa 100644
--- a/cde/lib/DtS
ached patch fixes this, plus one other occurence of sprintf to
illustrate the concept.
>From 3b4f65887cc75464814d499dfb6882635c85026f Mon Sep 17 00:00:00 2001
From: Marc Balmer
Date: Thu, 9 Aug 2012 07:08:05 +0200
Subject: [PATCH] When converting sprintf() to snprintf(), don't use the id
is.
>From 68b8ac11f519bf13e947fda3992e55bb69855e77 Mon Sep 17 00:00:00 2001
From: Marc Balmer
Date: Thu, 9 Aug 2012 07:03:46 +0200
Subject: [PATCH] iBuild with debug symbols enabled.
---
cde/config/cf/linux.cf |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/cde/con
A few more sprintf() to snprintf() conversion.
We need to find a proper way to replace strcpy() and strcat(), maybe
keep a local copy of strlcpy() and strlcat() from OpenBSD around
somewhere? Other suggestions?
>From 6aba055101e8b7c2f0155d76e872125dfd69ef8c Mon Sep 17 00:00:00 2001
From: M
Am 09.08.12 09:54, schrieb Pascal Stumpf:
[...]
>> Perhaps you and others working on the BSD ports could contribute to a
>> shared repo (maybe yours, I don't know) and arrange this.
>
> I'm not fond of that, since that means there will be even more
> fragmentation and probably merge problems and
Am 09.08.12 10:36, schrieb Pascal Stumpf:
> The official POSIX name for this signal is SIGCHLD. Linux probably has
> SIGCLD only for SysV compatibility, but BSD does not.
On Linux, SIGCLD is a synonym for SIGCHLD; if signal(7) can be trusted,
only for mips.
definitely this should be changed to S
Am 09.08.12 11:16, schrieb Pascal Stumpf:
> This should use socklen_t where available, really ...
this reminds that CDE should eventually be IPv6 ready..
>
>
>
> diff --git a/cde/lib/tt/bin/ttdbserverd/db_server_svc.C
> b/cde/lib/tt/bin/ttdbserverd/db_server_svc.C
> index dd6355b..67b1ce0 1006
Am 09.08.2012 um 20:56 schrieb Alex Kornilov :
> On 8/9/12 8:35 PM, Jon Trulson wrote:
>> Not that I know of... Is it even possible? ie: can OSX run X11/Motif?
> X11 is available in all OS X versions. But I don't know about Motif
>
>> brew search motif
> No formula found for "motif". Searchin
Am 09.08.12 20:02, schrieb Jon Trulson:
> On Thu, 9 Aug 2012, Marc Balmer wrote:
>
>> A few more sprintf() to snprintf() conversion.
>>
>
> Applied, thanks!
>
> ---
>
>> We need to find a proper way to replace strcpy() and strcat(), maybe
>> keep
See comment below.
Am 09.08.12 21:19, schrieb Jon Trulson:
> On Thu, 9 Aug 2012, ch...@chriswareham.net wrote:
>
>> Hi folks,
>>
>
> Hi, a couple of things: in the future, could you use git format patch
> format for any further patch submissions? Also, please don't compress
> them, I think mos
Am 09.08.12 21:25, schrieb Jon Trulson:
> On Thu, 9 Aug 2012, Marc Balmer wrote:
>
>>
>>
>> Am 09.08.2012 um 20:56 schrieb Alex Kornilov :
>>
>>> On 8/9/12 8:35 PM, Jon Trulson wrote:
>>>> Not that I know of... Is it even possible? ie: can
Am 09.08.12 22:29, schrieb Brent Busby:
> On Thu, 9 Aug 2012, Robert Tomsick wrote:
>
>> My $0.02: part of the value of CDE is that we can produce a more
>> polished version of CDE for older systems (ones long since abandoned
>> by KDE/GNOME/etc. but still in use), so maybe we shouldn't set the
Am 10.08.12 04:29, schrieb Aaron W. Hsu:
> Jon Trulson writes:
>
>> Hmmm... I'm leary of introducing further dependancies... Is it really
>> worth it? Do you have any idea how many strcpy et. al. are in CDE? :)
>
> I agree that there is no need to go on a hunt to replace all strcpy with
> strl
Am 10.08.12 07:57, schrieb Marc Balmer:
> Am 10.08.12 04:29, schrieb Aaron W. Hsu:
>> Jon Trulson writes:
>>
>>> Hmmm... I'm leary of introducing further dependancies... Is it really
>>> worth it? Do you have any idea how many strcpy et. al. are in CDE? :)
Am 10.08.2012 um 15:55 schrieb "ch...@chriswareham.net"
:
> Hi folks,
>
> I've taken into account peoples comments with regard to patch format and not
> including changes that are just reformatting. The attached patch is solely
> for imake, adding function prototypes where they are missing,
A few comments:
Since sizeof(char) is 1, it is not needed to write
malloc(... + 5 * sizeof(char))
but just
malloc(... + 5)
And actually, only 4 extra characters are needed, so don't waste that
byte ;)
And never use a multiplication in a malloc, never, ever. It has been
the source of remote-
Wouldn't it be better to avoid the multiplication in malloc here, too?
i.e. instead of
> sets = (std::ostrstream **)
> + malloc(sizeof(std::ostrstream *) * sets_max);
use
> sets = (std::ostrstream **)
> + calloc(sizeof(std::ostrstream *), sets_max);
(to
Am 12.08.12 08:32, schrieb Frederic Koehler:
> This patch makes tooltalk useful, letting many features work: logout,
> virtual desktop switching,
> and calling the dtfile command while it is running (which before just
> created a hung process).
> Before all tooltalk messages failed during sending,
I goind through the output of a make World, trying to fix warnings and
potential bugs (at least the low haning fruit).
One problem on 64 bit platforms seems to be that a lot of pointers
(which are 8 bytes on 64 bit linux) are assigned to int's (which are 4
bytes on 64 bit linux), thus loosing data
A first round to fix some compile time warnings.
>From c507d00e861b5b8a5a7df97d9b115ac7742bfeea Mon Sep 17 00:00:00 2001
From: Marc Balmer
Date: Sun, 12 Aug 2012 11:50:45 +0200
Subject: [PATCH] Fix compile time warning, mostly missing include files, but
also
integer issues on 64 bit linux
Am 12.08.12 11:54, schrieb Marc Balmer:
> A first round to fix some compile time warnings.
Please don't apply this patch. There is bug lurking in it...
>
>
>
> --
> Live Security Virtual Conf
Am 12.08.12 17:08, schrieb Douglas Mencken:
> http://ftp.osuosl.org/pub/manulix/CDE/cde-configRun_tweaks.patch
what is the gain of replacing echo with printf? is printf also builtin
in the shell or is /usr/bin/printf used?
>
>
>
> ---
Am 12.08.12 18:44, schrieb Aaron W. Hsu:
> Marc Balmer writes:
>
>> A possible solution could be to use long instead of int, as sizeof(long)
>> == sizeof(void *) on 32bit and 64bit linux. But if such values are used
>> in externalized binary form somewhere (too
echo -n works on all modern platforms (linux, bsds)
--
Marc Balmer
micro systems, http://www.msys.ch/
Tel. +41 61 383 05 10, Fax +41 61 383 05 12
Am 12.08.2012 um 22:34 schrieb Douglas Mencken :
>> I am pretty sure that 'echo -n' and 'echo -e' work everywhere. Wh
Since today, the script in linux/configRun calls a binary lpstat, which
is not installed on my Debian squeeze 32 bit system.
I did not find out which package does provide lpstat, can someone
enlighten me, please?
--
Live
yes, why make it conditional if the new code works on all platforms?
--
Marc Balmer
micro systems, http://www.msys.ch/
Tel. +41 61 383 05 10, Fax +41 61 383 05 12
Am 10.09.2012 um 20:06 schrieb Jon Trulson :
> On Mon, 10 Sep 2012, Marcin Cieslak wrote:
>
>> On Sun, 9 Sep 2012,
for patch 1: don't roll your own strfunctions, use strlcpy or so
for patch 2: hunk 1 and 3 are not needed
--
Marc Balmer
micro systems, http://www.msys.ch/
Tel. +41 61 383 05 10, Fax +41 61 383 05 12
Am 28.09.2012 um 20:27 schrieb Frederic Koehler :
> Pretty self-explanatory. Progress
Am 29.09.12 03:29, schrieb Jon Trulson:
> On Fri, 28 Sep 2012, Marc Balmer wrote:
>
>> for patch 1: don't roll your own strfunctions, use strlcpy or so
>
> I'm inclined to apply this - strlcpy isn't available on all systems
> (on linux, requires bsd lib I thi
31 matches
Mail list logo
