`dumpkeys' and `loadkeys' to redefine your keymap. Like any
good application, the kernel keys are configurable.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
g.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
echnology (and
putting X in a separate server that logically sits on top of it), it
would be orders of magnitude smaller (and 100% reliable).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
:
GGI_LITTLE_ENDIAN
GGI_BIG_ENDIAN
One of those will be defined.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
this in:
file_vis = ggiOpen("display-file:/tmp/foo.ppm", NULL);
ggiSetMode(file_vis ...);
ggiCrossBlit(file_vis - normal_vis);
ggiGetc(normal_vis);
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
and FILE_BMP defines -- are they just temporary ?
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
devices files in /dev).
I can do this:
demos/flying-ggis -m 1024x760x16 -t file:/dev/fb0
(I boot with vesafb into 1024x768x16 mode). It works. It's ultra
silly too. No input. Needed SAK to kill :-
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
g-interchange
Using LibGFFD, this new picture library could detect the filetype, and
then look it up in the gpf.conf file to find which DLL to load to
handle the image.
Any interest in this approach ?
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
o be found on the GGI FTP site though.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
:5:0
Try booting with the following kernel cmdline option:
video=vesa:ypan
It enables vertical pan support in the VESAfb driver, which LibGGI can
use for multiple frames.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
presume), but you'll want it bigger.
Joseph, any console command for setting FOV ?
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
?
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
bbing" the
mouse is a very windowing-system-ish concept).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
fbcon to bugger off).
But then you run fbset or the fbdev target, and it the kernel tries to
program the hardware (using wrong data, since the X server has changed
everything). Boom. Video card hosed, and neither the kernel nor the
X server have enough info to fix it...
Cheers,
__
\/ Andrew
structure
(basicly just width, height, format and a pointer to the data). It
should also be able to convert images between different formats.
libmagick ? It's free software, can be used without X, and supports a
sheet-load of image formats.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
ied though by having a "current painted
set" (not unlike libggi's current write frame). Maybe use integer
handles for them (e.g. you could allocate or delete a whole range
at once), not sure if its better than pointers though.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
nd restore the mode if it succeeds. That is not
a nice solution, but IMO it is better than the current situation where
the fbdev target does not report correct info via ggiCheckMode.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
), with very unsatisfactory results. This should also
prevent problems when someone tries to run the fbdev target from a
normal X server (I'm pretty sure they set KD_GRAPHICS mode too).
Comments ?
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
working on
commercial OSes (like a certain mysterious one).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
Christoph Egger writes:
I just found an typo error in the memory-target.
The fix is attached.
Anyone here to commit it, please?
Both fixes (memory tele) look OK to me, and have been committed.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
as changed something or not. The only solution is to
make a smarter X server.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
for doing swapping stuff).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
chanism to "join" input queues. Or poll. Or broaden the
concept of "focus" to allow multiple actual windows.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
. Or a simple game where each object is a painted set.
Maybe even use them for font glyphs (might be too much overhead for
that). Not exactly compelling reasons for multiple painted sets per
ggi_visual_t, but it is a nice concept.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
be a separate
problem).
A workaround to try (until this gets fixed) is to use XGGI on the
display-tile target, as this will make XGGI draw in memory and the
tile target will blit to the screen (make sure no memsets in the tile
target though :-).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
Christoph Egger writes:
On Sat, 29 Apr 2000, Andrew Apted wrote:
The number of characters actually drawn.
Huh? The man-pages of ggiPuts says, it returns 0 if OK, otherwise an error
occured. So I want to know, which kind of error means 29 and 30.
If you are sure, you're right
mapping routines. You'll get better performance by implementing the
other drawing routines too (drawhline, puthline etc), but that can
come later.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
Andreas Beck writes:
.../libgic/programs/demos/confmgrdemo.c (c) Andrew Apted
Could the respective copyright holders please fix this ?
Done.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
iated.
Tested it here, works nicely, and no segfaults.
Had a problem compiling though, the demos check for a HAVE_RANDOM
define, but as far as I can tell there is no check for this in
configure.in (thus doesn't appear in config.h). This caused a problem
because `random' gets #defined to `rand', but
(but doesn't have to, if
vis-visible.y == vis-virt.y) be much faster.
I think Marcus's point is: XMI cannot assume that the application
won't be panning around the virtual area.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
e instead of:
miRectangle rect;
rect.x = x;
rect.y = y;
rect.width = w;
rect.height = h;
miFillRectangles(vis, paintedSet, pGC, 1, rect);
allow this:
miFillRectangle(vis, paintedSet, pGC, x, y, w, h);
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
the following kernel
command line option (in LILO or whatever) :
video=vesa:ypan
You can check that it works by running the libggi `pageflip' demo.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
than the default fbcon)?
Only if there is a working kgicon driver for your chipset, but as far
as I know we don't have any drivers for Trident cards.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
ices, give them group write access, and put yourself in the video
group (with adduser).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
Martin Schulze writes:
Hmm ... still I don't get it - "./inputdump --input linux-mouse,ps2,/dev/mouse"
inputdump --input 'linux-mouse:ps2,/dev/mouse'
^
^
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
format's bitmasks will be
ignored by LibGGI and probably the hardware too. There is an alpha
mask in the ggi_pixelformat but AFAIK it never does anything (in
LibGGI or in hardware).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
is that -novt causes
the second framebuffer to be used without having _any_ consoles on it
(and if you normally add consoles onto fb1, you should remove them
before trying this). Hope that helps.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
).
I've found a way around this, by writing a bunch of scripts, but having the
option in XGGI would be substantially cleaner.
Yep, IMHO it would be a worthwhile addition to XGGI.
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
one on the stack. A third source of weirdness
is changing a structure in a header file but not recompiling the
source files that use it (it sounds obvious, but I've been bitten by
that a couple of times, looking for a "bug" that wasn't there).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
rd doesn't directly support -- that's
what the real driver is for).
Food for thought at least :-).
Cheers,
__
\/ Andrew Apted [EMAIL PROTECTED]
//---
// interface IMda : IUnknown
//
// GUID: 374581C
40 matches
Mail list logo
