An awesome app called Qingy (Part 1)

[s4t4n]

now i have compiled kernel with what you told me to. now what should i options in boot loader to?

It depends on the framebuffer driver you compiled into the kernel:

Old VESA driver: see <path_to_kernel_sources>/Documentation/fb/vesafb.txt
(something like 'vga=317' for 1024x768, 16bpp)

VESA tng: don't remember, as I don't currently have access to the machine I'm using it with, anyway search in <path_to_kernel_sources>/Documentation/fb/
and you should find something. Its main advantage it over the old driver is the ability to specify the refresh rate of the framebuffer console (useful for non laptop users).

Trident:
something like 'video=tridentfb:1024x768-8@60', where 8 is the color depth (256 colors here) and 60 the refresh rate. Again, if you do not have a Trident video card this driver is useless to you.

Hope this helps...

[Daniel_walmsley]

Hi
i love qingy it great but i also love doom3 and 3ddesk only problem is i can run them all with out runnig into problems. All tho glxinfo tells me dri is enabled when i go to run doom3 or 3ddesk they dont work (GLXGREARS does work tho).
oh and if i use qingy to loginto a plain console it all works fine.
Im stumped, as i am a recent convert to linux from windows im not yet fully versed on how all this works any one out there able to help me with this??

[akos]

s4t4n, could you take a look at this topic, please?

qingy and DISPLAY=:1.0

Thanks in advance.

Akos

[s4t4n]

s4t4n, could you take a look at this topic, please?

qingy and DISPLAY=:1.0

Thanks in advance.

Akos

Done...

[s4t4n]

Hi
i love qingy it great but i also love doom3 and 3ddesk only problem is i can run them all with out runnig into problems. All tho glxinfo tells me dri is enabled when i go to run doom3 or 3ddesk they dont work (GLXGREARS does work tho).
oh and if i use qingy to loginto a plain console it all works fine.
Im stumped, as i am a recent convert to linux from windows im not yet fully versed on how all this works any one out there able to help me with this??

The issue about GLX and qingy is a long standing one, but I was never able to find a solution, because I could not reproduce the effects. I own a GeForce2 and, alas, it works with full acceleration on my machine, so I really don't know where to look at...

As a workaround, you can create a text-mode session in qingy that actually calls 'startx'. Ugly, but it should work...

[GlooM]

I have a strange GLX issue.
I tried running X without qingy and it works well there.
The problem is GLX takes long time to initialize, for example i type
glxinfo , 10 seconds pass and it prints out the info.
Or i start enemy-territory , 10 seconds pass and the game starts as normal.
Help !

[fastMoon]

Hi, I'm a long time fan of qingy and I was happy to see the autologin feature added.
but.....it doesn't workl...it crasches qingy completely (also on the other tty's)
here's my settings:

x_sessions = "/etc/X11/Sessions/"
text_sessions = "/etc/qingy/sessions/"
xinit = "/usr/X11R6/bin/xinit"
screensavers_dir = "/usr/lib/qingy/screensavers"
screensaver "running_time"#="%H:%S:%M"
themes_dir = "/usr/share/qingy/themes"
theme = "mm"
shutdown_policy = everyone
tty = 1
{
theme = "mm"
screensaver "running_time"="00:03:00"
qingy_DirectFB = "/sbin/qingy-DirectFB"
autologin
{
username = "root"
password = "mypasswd"
session = "Gnome"
#session = lastsession
relogin = yes
}
}


any thoughts ??

[s4t4n]

Hi, I'm a long time fan of qingy and I was happy to see the autologin feature added.
but.....it doesn't workl...it crasches qingy completely (also on the other tty's)
[...]
any thoughts ??

Er... do notice that the autologin feature exists only in CVS, there is no released version of qingy that has it, so you cannot test it out with the version in Portage. You will have to fetch the sources from CVS yourself, and compile them by hand.

[feffi]

Hi, I don't know, if this question was covered before:

I installed qingy again (after being patient for bugfixes) and it runs smoothly despite one thing: I've got a resolution of 1400x1050 on my laptop using the radeon kernel module. My theme is also layouted to be this size BUT when qingy is fired up, it seems like the resolution qingy determined is wrong. The background picture is FAR too big, fonts too. I can barely see a quarter of the original login-screen. radeon tells me, it's on 1400x1050. DirectFB tells me the same. the picture is good, also the theme settings.
I also tried the "-r" switch with different resolutions to tell qingy the right way of behaviour, but nothing happened.

Maybe someone can look after this? If you need any Information regarding configs, just PM me.

[s4t4n]

I installed qingy again (after being patient for bugfixes) and it runs smoothly despite one thing: I've got a resolution of 1400x1050 on my laptop using the radeon kernel module. My theme is also layouted to be this size BUT when qingy is fired up, it seems like the resolution qingy determined is wrong. The background picture is FAR too big, fonts too. I can barely see a quarter of the original login-screen. radeon tells me, it's on 1400x1050. DirectFB tells me the same. the picture is good, also the theme settings.
I also tried the "-r" switch with different resolutions to tell qingy the right way of behaviour, but nothing happened.

When you layout your theme, keep in mind that qingy, in order to (try to) keep a theme look uniform across different resolutions, uses a virtualized resolution of 800x600. So what you really want is tailoring your theme for that resolution, and make small (pixel-wide) adjustments afterwards, to put things exactly where you want them. This design isn't perfect, though, as images will be placed according to the virtual res, but their size will be the actual size. Hope this helps

[feffi]

thanks for your answer, i'll give it a try this weekend and keep you informed

[s4t4n]

Er... do notice that the autologin feature exists only in CVS, there is no released version of qingy that has it, so you cannot test it out with the version in Portage. You will have to fetch the sources from CVS yourself, and compile them by hand.

Quoting myself ;-P
qingy 0.5.3 is online (and in Portage) now, and sports this feature...

[pqwy]

And a damn good Qingy it is, Sir!

Two complaints, still: first, it adds the Emacs session in spite of the use flag being off. No biggie.

Second, I just don't understand why did you have to remove the system-auth stacking from the pam config and opted for the explicit modules? Some of the pam modules, especially pam_console, cause random cardiac arrests and brain strokes, and the system-auth is a convenient place to manage them. Not that it's difficult to put your modules in another file, but I'd also guess not everyone out there who's in for a sleek login manager is familiar with pam, which does tend to put people off -- as the numerous posts suggest. Further example, your arg to pam_unix doesn't include 'nullok' so no empty passwords.

I really think It's nicer for everyone to have it all in one place....

[s4t4n]

And a damn good Qingy it is, Sir!

Two complaints, still: first, it adds the Emacs session in spite of the use flag being off. No biggie.

Well, actually the emacs USE flag makes qingy ebuild not to install qingy major mode (which gives emacs syntax highlighting for qingy config files). The text session file is there mostly as an example on how to produce your own. Neverthless, for consistency sake, I'll fix this, too...

Second, I just don't understand why did you have to remove the system-auth stacking from the pam config and opted for the explicit modules?

Because system-auth stacking in PAM is simply not availavle on all distros, debian being a significant example. And I want qingy to work our of the box on as many environments as possible...

Some of the pam modules, especially pam_console, cause random cardiac arrests and brain strokes,

WTF ???

and the system-auth is a convenient place to manage them. Not that it's difficult to put your modules in another file, but I'd also guess not everyone out there who's in for a sleek login manager is familiar with pam, which does tend to put people off -- as the numerous posts suggest. Further example, your arg to pam_unix doesn't include 'nullok' so no empty passwords. I really think It's nicer for everyone to have it all in one place....

Well, if you think you can provide me with a better pam config, feel free, and I'll ship it with next version of qingy I could even install different pam files for different distros, but you will have to convince me

[hans0r]

i've got a problem when i use qingy to launch X.
i installed the new ati-driver yesterday and since then whenever i launch X using qingy i get this issue when launching 3d-apps:



wai@kellerkind ~ $ glxinfo
name of display: :1.0

10 second pause

server glx extensions:
GLX_ARB_multisample, GLX_EXT_visual_info, GLX_EXT_visual_rating,
GLX_EXT_import_context, GLX_OML_swap_method, GLX_SGI_make_current_read,
GLX_SGIS_multisample, GLX_SGIX_fbconfig
client glx vendor string: ATI
client glx version string: 1.3
client glx extensions:
GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context,
GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_ATI_pixel_format_float,
GLX_ATI_render_texture
[....]


this also happens when i launch other 3d applications/games.
everytime it just stops for approximately 10 seconds before the program actually starts.

when i launch X the usual way via command-line this problem does not occur.

[s4t4n]

i've got a problem when i use qingy to launch X.
i installed the new ati-driver yesterday and since then whenever i launch X using qingy i get this issue when launching 3d-apps:

[SNIP]

this also happens when i launch other 3d applications/games.
everytime it just stops for approximately 10 seconds before the program actually starts.

when i launch X the usual way via command-line this problem does not occur.

This is a long-standing issue that I was never able to replicate. However, I just aquired a new laptop that sports an Ati Radeon 9700; give me a few days to get it up and running with gentoo, and I'll see if *this time* I'll see it...

[pqwy]

Argh, I crave my isp's manager's blood. So, back on track...

Well, actually the emacs USE flag makes qingy ebuild not to install qingy major mode (which gives emacs syntax highlighting for qingy config files).

My mistake, sorry.

Well, if you think you can provide me with a better pam config, feel free, and I'll ship it with next version of qingy I could even install different pam files for different distros, but you will have to convince me

Well, I could give it a shot . See, if you look around the forums, you'll notice pam tends to confuse souls. In particular, the pam_console, which you happily enabled, is completely inconsistent with a udev/groups permission scheme Gentoo seems to be trying to push. Or if it isn't, the console.perms in it's current state definitely is, and there are quite some threads where people complain about mysterious permission loss on device nodes. Besides, that particular module is disabled by default AFAIK. And my nullok also disappeared.

So, a better config would at least be to copy the Gentoo default settings, otherwise you're changing the login system, however subtly. Of course, you wouldn't want to change the login system of the other distros you maintain the package for, so having multiple pam files would be nice to people. And while at it, you could keep the consistency of Gentoo's centralized pam config and stack it all up to system-auth at no cost. Sounds reasonable?

[erb69]

s4t4n,

What do I edit in the inittab file?

[seringen]

s4t4n,

What do I edit in the inittab file?

http://qingy.sourceforge.net/manual.php
hth

[s4t4n]

Well, I could give it a shot .

I'm waiting

Sounds reasonable?

OK, you have convinced me

[rush_ad]

i get errors compiliing directfb on amd64.

Code: Select all

In file included from /usr/include/linux/fb.h:4,
                 from ati128.c:38:
/usr/include/asm/types.h:28: error: conflicting types for 'int64_t'
/usr/include/sys/types.h:194: error: previous declaration of 'int64_t' was here
/usr/include/asm/types.h:29: error: conflicting types for 'u_int64_t'
/usr/include/sys/types.h:200: error: previous declaration of 'u_int64_t' was here
make[3]: *** [ati128.lo] Error 1
make[3]: *** Waiting for unfinished jobs....
 gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../include -I../../lib -I../../lib -I../../src -D_REENTRANT -I/usr/include/libmpeg3 -Wall -O3 -ffast-math -pipe -march=athlon64 -O2 -pipe -D_GNU_SOURCE -Werror-implicit-function-declaration -MT ati128_state.lo -MD -MP -MF .deps/ati128_state.Tpo -c ati128_state.c  -fPIC -DPIC -o .libs/ati128_state.o
make[3]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/gfxdrivers/ati128'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/gfxdrivers'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22'
make: *** [all] Error 2

!!! ERROR: dev-libs/DirectFB-0.9.22 failed.
!!! Function src_compile, Line 86, Exitcode 2
!!! (no error message)
!!! If you need support, post the topmost build error, NOT this status message.

help me out

[hans0r]

if you don't use an ati128 in your amd64-system (i bet you don't...) then don't build the directfb-driver for it.

like this:

Code: Select all

USE_VIDEO_CARDS="nvidia" emerge DirectFB

valid values for 'USE_VIDEO_CARDS' are:

ati128 cle266 cyber5k i810 mach64 matrox neomagic nsc nvidia radeon savage sis315 tdfx unichrome

[rush_ad]

i am using a pci voodoo3 2000, which one shoud i use?

from my knowledge, i should use tdfx, am i right?

[rush_ad]

Code: Select all

make[3]: Entering directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/inputdrivers'
make[3]: Nothing to be done for `all-am'.
make[3]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/inputdrivers'
make[2]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/inputdrivers'
Making all in gfxdrivers
make[2]: Entering directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/gfxdrivers'
Making all in ati128
make[3]: Entering directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/gfxdrivers/ati128'
if /bin/sh ../../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../include -I../../lib -I../../lib -I../../src  -D_REENTRANT  -I/usr/include/libmpeg3  -Wall -O3 -ffast-math -pipe -march=athlon64 -O2 -pipe -D_GNU_SOURCE  -Werror-implicit-function-declaration -MT ati128.lo -MD -MP -MF ".deps/ati128.Tpo" -c -o ati128.lo ati128.c; \
then mv -f ".deps/ati128.Tpo" ".deps/ati128.Plo"; else rm -f ".deps/ati128.Tpo"; exit 1; fi
if /bin/sh ../../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../include -I../../lib -I../../lib -I../../src  -D_REENTRANT  -I/usr/include/libmpeg3  -Wall -O3 -ffast-math -pipe -march=athlon64 -O2 -pipe -D_GNU_SOURCE  -Werror-implicit-function-declaration -MT ati128_overlay.lo -MD -MP -MF ".deps/ati128_overlay.Tpo" -c -o ati128_overlay.lo ati128_overlay.c; \
then mv -f ".deps/ati128_overlay.Tpo" ".deps/ati128_overlay.Plo"; else rm -f ".deps/ati128_overlay.Tpo"; exit 1; fi
mkdir .libs
 gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../include -I../../lib -I../../lib -I../../src -D_REENTRANT -I/usr/include/libmpeg3 -Wall -O3 -ffast-math -pipe -march=athlon64 -O2 -pipe -D_GNU_SOURCE -Werror-implicit-function-declaration -MT ati128.lo -MD -MP -MF .deps/ati128.Tpo -c ati128.c  -fPIC -DPIC -o .libs/ati128.o
 gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../include -I../../lib -I../../lib -I../../src -D_REENTRANT -I/usr/include/libmpeg3 -Wall -O3 -ffast-math -pipe -march=athlon64 -O2 -pipe -D_GNU_SOURCE -Werror-implicit-function-declaration -MT ati128_overlay.lo -MD -MP -MF .deps/ati128_overlay.Tpo -c ati128_overlay.c  -fPIC -DPIC -o .libs/ati128_overlay.o
In file included from /usr/include/linux/fb.h:4,
                 from ati128.c:38:
/usr/include/asm/types.h:28: error: conflicting types for 'int64_t'
/usr/include/sys/types.h:194: error: previous declaration of 'int64_t' was here
/usr/include/asm/types.h:29: error: conflicting types for 'u_int64_t'
/usr/include/sys/types.h:200: error: previous declaration of 'u_int64_t' was here
make[3]: *** [ati128.lo] Error 1
make[3]: *** Waiting for unfinished jobs....
make[3]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/gfxdrivers/ati128'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22/gfxdrivers'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/DirectFB-0.9.22/work/DirectFB-0.9.22'
make: *** [all] Error 2

!!! ERROR: dev-libs/DirectFB-0.9.22 failed.
!!! Function src_compile, Line 86, Exitcode 2
!!! (no error message)
!!! If you need support, post the topmost build error, NOT this status message.

thats what i got after doing

Code: Select all

USE_VIDEO_CARDS="tdfx" emerge DirectFB

[hans0r]

oh, it's just VIDEO_CARDS="..." not USE_VIDEO_CARDS.
my mistake...


yes, tdfx is the right setting for your card.