Re: gtk amd64 mini.iso
Sven Luther wrote:
On Sun, Nov 13, 2005 at 04:10:53PM +0100, Davide Viti wrote:
ALT+F[1-7] do not seems works :-(
same here; I use qemu to emulate an AMD64 machine, so I probably
can't help very much with debugging (it' too slow)
You need to do Ctr+Alt+F[2-4].
I don't think who brings in this Alt -> Ctr+Alt, i believe it may be a gtk
thingy, since this is the X combo needing to switch consoles, but it would be
nice to have the old behavior back.
actually I was meaning ctrl+alt+F[2-4]: If I use such a combination of keys
the frontend is restarted (probably it crashes before)
Yes, indeed.
Maybe we could solve this by "freezing" the frontend when leaving the console
? I understand this only happens if you are updating the frontend while you
are away from the console, and i suppose the crash happens while writing the
framebuffer while you are away from it.
today i've reproduced the bug running cdebconf from my installed debian
system, this is the output
attilaptop:/home/attilio/cdebconf/cdebconf/src/gtk_test# ../debconf
test.config (*) DirectFB/Config: Parsing config file '/etc/directfbrc'.
---------------------- DirectFB v0.9.22 ---------------------
(c) 2000-2002 convergence integrated media GmbH
(c) 2002-2004 convergence GmbH
-----------------------------------------------------------
(*) DirectFB/Core: Single Application Core. (2005-10-06 20:07)
(*) Direct/Memcpy: Using SSE optimized memcpy()
(!) DirectFB/FBDev/vt: FBIOGET_CON2FBMAP failed!
--> Invalid argument
(!) DirectFB/FBDev/vt: FBIOPUT_CON2FBMAP failed!
--> Invalid argument
(*) Direct/Thread: Running 'VT Switcher' (CRITICAL, 4245)...
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4246)...
(!!!) *** UNIMPLEMENTED [fusion_reactor_set_lock] ***
[../../../lib/fusion/rea ctor.c:802]
(*) DirectFB/Input: AT Translated Set 2 keyboard (1) 0.1 (convergence
integrated media GmbH)
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4247)...
(*) DirectFB/Input: PC Speaker (2) 0.1 (convergence integrated media GmbH)
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4248)...
(*) DirectFB/Input: SynPS/2 Synaptics TouchPad (3) 0.1 (convergence
integrated m edia GmbH)
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4249)...
(*) DirectFB/Input: USB Mouse (4) 0.1 (convergence integrated media GmbH)
(*) Direct/Thread: Running 'Keyboard Input' (INPUT, 4250)...
(*) DirectFB/Input: Keyboard 0.9 (convergence integrated media GmbH)
(*) Direct/Thread: Running 'PS/2 Input' (INPUT, 4253)...
(*) DirectFB/Input: IMPS/2 Mouse 1.0 (Convergence GmbH)
(*) DirectFB/Genefx: MMX detected and enabled
(*) DirectFB/Graphics: MMX Software Rasterizer 0.6 (convergence
integrated media GmbH)
(*) DirectFB/Core/WM: Default 0.2 (Convergence GmbH)
(!) DirectFB/FBDev: Panning display failed!
--> Invalid argument
ANS: 0
ANS: 0 question will be asked
(debconf:4228): Gtk-WARNING **: non use_align not implemented yet
ANS: 0 ok
RET CODE: 0
CHOICE : progressbar
ANS: 0 OK
PROGRESS STOP: 0 OK
ANS: 0 question will be asked
(debconf:4228): Gtk-WARNING **: non use_align not implemented yet
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4261)...
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4262)...
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4263)...
(*) Direct/Thread: Running 'Linux Input' (INPUT, 4264)...
(*) Direct/Thread: Running 'Keyboard Input' (INPUT, 4265)...
(*) Direct/Thread: Running 'PS/2 Input' (INPUT, 4266)...
(!) DirectFB/FBDev: Panning display failed!
--> Invalid argument
ANS: 0 ok
RET CODE: 0
CHOICE : progressbar
ANS: 0 OK
ANS: 20 Incorrect number of arguments
ANS: 0 OK
ANS: 0 OK
(!) [ 4228: 0.000] --> Caught signal 11 (at 0x403f16d0, invalid
address) <--
*** glibc detected *** double free or corruption (!prev): 0x0807ae78 ***
(!) [ 4228: 0.000] --> Caught signal 6 (unknown origin) <--
Killed
i've also mailed the bugreport to mike emmel togheter with instructions
on how to build cdebconf: ATM he's busy with CairoDFB but he told that
as soon CairoDFB is merged into Cairo mainline he'll be able to have a
closer look to GTKDFB (ha has also detected a couple of bugs we didn't).
I've also tried to specify that this bugs seems to be the most important
that affect mini.iso ISOs.
ciao
Attilio
ps
does the bad colors bug under PPC goes away if you use a 256 colours
logo_debian.png
Reply to: