headers
Thorsten Glaser | 1 Jul 2006 17:01
Picon
Gravatar

Re: CVS: herc.66h.42h.de: src

Thorsten Glaser dixit:

>Modified files:
>	usr.bin/elf2aout: elf2aout.c
>
>Log message:
>don't make this crash converting symbol tables;

this also replaces the one strcpy(3) with strlcpy(3)
which was a requirement for fixing these segfaults

>rather, use realloc(3) correctly to kludge around.
>also sprinkle another endian conversion and note
>that this code is not 64-bit safe.

hto[bl]e32 just isn't hto[bl]el on LP64 arches
this requires even greater changes of the code
plus an evaluation whether the target is elf32
or elf64

in addition I don't think this code handles an
elf64 file *at all* - still would be useful to
convert sparc(32) kernels on an amd64 box

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
(Continue reading)

Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 2 Jul 2006 02:56
Picon
Gravatar

Re: CVS: herc.66h.42h.de: X11

Thorsten Glaser dixit:

>Log message:
>revert to xterm-200-based (plus fix an openbsd glitch)
>xterm-215-based does not work correctly:
>* uxterm does not honour its class (e.g. my font is too small)
>* utmp is never updated

I even tried to build it autoconf-based, but even that
suffered from the same symptoms. Something is horridly
broken in there.

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 3 Jul 2006 01:12
Picon
Gravatar

Re: CVS: herc.66h.42h.de: X11

Thorsten Glaser dixit:

>Commit ID:	10044A8012E31E153E7

>prevent crash with dark garden font (still fails for mherrb tho)

I can now run the command
$ xterm -fn '-misc-dark garden-medium-r-normal--0-0-0-0-p-0-iso10646-1'
just fine without it crashing. Yay!

Reference: http://article.gmane.org/gmane.os.openbsd.tech/11268

bye,
//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 3 Jul 2006 01:58
Picon
Gravatar

Re: CVS: herc.66h.42h.de: X11

Thorsten Glaser dixit:

>Log message:
>fix DESTDIR install

cought by systrace, thanks Niels

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 3 Jul 2006 02:19
Picon
Gravatar

Desktop Environment chooser for live CD

Hi people,

I intend to amend the Live CD's .xsession by code similar to the following:

xmessage -button twm:0,evilwm:101,failsafe:1 \
    -center -default twm -timeout 12 \
    'Choose window manager^JWait 12 seconds for default'
case $? {
(0)	WINDOWMANAGER=twm ;;
(1)	WINDOWMANAGER=xterm ;;
(101)	WINDOWMANAGER=evilwm ;;
(102)	WINDOWMANAGER=icewm-session ;;
# ...
}

Reasons for choosing the codes:
- 0 is returned for the -timeout option (so make that the default)
- 1 is returned in case of an error (so make that failsafe)
- from 101 onward, we can just count

Opinions? That'd be good for #9bis even.

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
(Continue reading)

Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 3 Jul 2006 04:27
Picon
Gravatar

RFD: rpath / dt_need for libc?

Hi,

libkvm contains
  NEEDED      libz.so.5.3
  RPATH       /usr/lib
but no
  NEEDED      libc.so.38.4

Should we add '-lc' to the default link command line in <bsd.lib.mk>
and why / why not? Which other systems would be impacted?

(NB: libc version bumps affect other shlibs anyway, whether it's used
or not, because executables link against libc usually, and most libs
do in fact pull in libc.)

Probably it'd be better to nm all libs, scan if we need libc, and add
that to LDADD in the respective Makefile.

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 4 Jul 2006 03:17
Picon
Gravatar

Re: CVS: herc.66h.42h.de: (multiple)

Thorsten Glaser dixit:

>Modified files:
>	shells/mksh/pkg: DESCR

I know I ought to bump the package patchlevel here, but truly,
a change to DESCR isn't a reason to recompile the whole bitch
on a 75 MHz machine, is it?

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 5 Jul 2006 07:02
Picon
Gravatar

Re: CVS: herc.66h.42h.de: src

Thorsten Glaser dixit:

>* sync sparc autoiso with i386

This is all preliminary. Both kernels build, now I'm fascinated
about whether the mr.fs can be built and fits.

>* don't attempt to build a miniroot (yet)

I wonder if TNF makefs(8) can be used these days...

//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Thorsten Glaser | 5 Jul 2006 23:49
Picon
Gravatar

Re: CVS: herc.66h.42h.de: src

Thorsten Glaser dixit:

>	distrib/i386/floppy: list.local
>	distrib/miniroot: mtree.conf.small

>and optionally use makefs(8) for generating the floppy9.fs image
>so that it fits (more to come later)

I think when using makefs(8) for the mr.fs we might have better
compression of the kernels. I'll investigate that later, after
that tool has been throughoutly tested.

We still cannot get rid of the vnd(4)-using process because there
is no option in makefs(8) to display the blocks a file uses, nor
in ffs nor in iso9660. This must be fixed. I'll write TNF.

bye,
//mirabile
--

-- 
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 won't deny
existence.		-- Coywolf Qi Hunt
Permalink | Reply | 0 comments |
headers
Benny Siegert | 12 Jul 2006 11:47
Picon
Gravatar

Fwd: [ wxwindows-Patches-1520208 ] add MirOS as a supported platform

Hi,

Just to let you know that there _are_ projects who can take a patch in
less than one year's time ... (yes, cairo developers, I am looking at
you) ;)

tg: I will merge your 2.6.1 port with the pkgsrc one (2.6.3).
Actually, I did not see that you had already done 2.6.1, I thought we
still had 2.2 or something. Default will be gtk+2 (now strongly
recommended by upstream), 2.2 and 2.4 compat, and a full unicode
build. Is this OK for you?

--Benny.

---------- Forwarded message ----------
From: SourceForge.net <noreply@...>
Date: Jul 12, 2006 11:01 AM
Subject: [ wxwindows-Patches-1520208 ] add MirOS as a supported platform
To: noreply@...

Patches item #1520208, was opened at 2006-07-10 20:45
Message generated for change (Comment added) made by neis
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=309863&aid=1520208&group_id=9863

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Build -- unix
Group: new feature
(Continue reading)

Permalink | Reply | 10 comments |

Gmane